case study on sepsis

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

A 26-year-old female arrives with a companion to an urgent care at 0845 by personal vehicle for treatment of suspected foot infection. The patient’s companion (a female roommate) reports to the triage nurse that the patient cut her foot while wading in the ocean over the weekend. They did not initially notice the cut but discovered it while removing tar from the bottom of the right foot. Approximately 24 hours later, her foot became too painful for ambulation, and a “thick, yellowish” discharge began to drain from the cut. Vitals upon arrival at urgent care showed a temperature of 101.5F, heart rate of 130, respiratory rate of 24, and blood pressure of 86/40. Her pain was 9/10 in her right foot and described as throbbing. During a HTT assessment by the PA, the patient is reported to be arousable to voice, oriented to person and place only, and complaining of nausea. The patient reports she took Tylenol that morning to relieve pain and fever. Her skin is pale, diaphoretic, and hot.

The urgent care calls 911, and medics are dispatched to the center for transfer to the local hospital to treat the patient for suspected sepsis. Upon arrival, medics find the patient is still tachycardic, and that her blood pressure has dropped to 80/40. Her respiratory rate has increased to 30. During transport, medics insert a 20 gauge peripheral IVs in the patient’s left antecubital. They infuse a fluid bolus of 500 mL of normal saline to manage her patient’s hypotension, and administer oxygen by simple mask at 4L/min. During the primary assessment, the patient’s right foot reveals a two-inch laceration with no active bleeding that is erythematous, edematous (non-pitting), and radiating heat. Edema is covering the entire bottom of the right foot and extends to the patient’s ankle.

The patient arrives to the emergency room within 15 minutes and is admitted for treatment at 1000. On the unit, Code Sepsis is called, and the agency’s sepsis protocol based on the Surviving Sepsis campaign is implemented. The patient’s vitals are now a temperature of 102F, heart rate of 140, respiratory rate of 34, and blood pressure of 96/42. Lactate levels are immediately measured. A second 20 gauge peripheral IV is inserted into the right antecubital, blood cultures are drawn, and a swab sample is taken of the cut and submitted to the laboratory for a culture and sensitivity test. Broad spectrum antibiotic ceftriaxone (Rocephin) is administered, and patient is given Ibuprofen to manage her fever. The patient is diagnosed with septic shock, and because she is still hypotensive, 30mL/kg of normal saline is infused. The patient’s lactate levels come back as 2.4 mmol/L. Norepinephrine (Levophed) is also hung, and the patient is further monitored. With careful titration and vital monitoring, the use of vasopressors restores the patient’s blood pressure to 101/52. Although fluid resuscitation helps to bring the patient’s heart rate down to 104, Nicardipine (Cardene) was ordered in anticipation of further needs to manage tachycardia. The patient is transferred to the ICU at 1300 for further monitoring and management of her hemodynamic status.

In the ICU, the patient’s vitals stabilize. Her tachypneic state reduces, and respiratory rate is now 18. She no longer requires oxygen supplementation. Her pain is being managed with IV morphine and she rates the pain in her as 3/10. Her IV pump is running 125 mL an hour of normal saline along with piggybacked ceftriaxone (Rocephin), and labs return a lactate level of 1.5 mmol/L. The patient’s roommate arrives. She is tearful and explains to the ICU nurse that she wanted to tell the patient’s parents what happened, but the patient refused. The ICU nurse calls for the case manager and a social service consult to inquire further. The patient’s roommate explains to the interdisciplinary team that the patient does not have insurance because she is 26 and has been removed from her parents’ medical plan. The parents are also currently engaged in a divorce, do not speak to each other, and use their daughter to communicate. The patient is aware of their financial situation and her lack of medical coverage and does not want to worry her parents in spite of her critical medical state.

• What are the priority nursing interventions for this patient in the ICU setting?
• What signs and symptoms in this patient would indicate the need for mechanical ventilation?
• What is the nurse’s role in addressing the patient’s financial concerns?

References:

Gordon, A.C., Mason, A.J., Thirunavukkarasu, N., et al. (2016). Effect of early vasopressin vs norepinephrine on kidney failure in patient with septic shock: The VANISH randomized clinical trial. JAMA, 316 (5), 509–518. doi:10.1001/jama.2016.10485

Hinkle, J. L., & Cheever, K. H. (2014). Brunner & Suddarth’s textbook of medical-surgical nursing. Philadelphia: Lippincott Williams & Wilkins. PulmCCM. (2019, January 14). From the Surviving Sepsis Guidelines: Criteria for diagnosis of  sepsis. Retrieved from https://pulmccm.org/review-articles/surviving-sepsis-guidelines-criteria-diagnosis-sepsis/

Schmidt, G.A., & Mandel, J. (2019, March). Evaluation and management of suspected sepsis  and septic shock in adults. Retrieved from https://www.uptodate.com/contents/evaluation-and-management-of-suspected-sepsis-and-septic-shock-in-adults?search=sepsis treatmentadult&source=search_result&selectedTitle=1~150&usage_type=default&display_rank=1#H465649907

Society of Critical Care Medicine. (2019). Hour-1 bundle: Initial resuscitation for sepsis and  septic shock. Retrieved from http://www.survivingsepsis.org/SiteCollectionDocuments/Surviving-Sepsis-Campaign-Hour-1-Bundle.pdf

Zhang, M., Zheng, Z., & Ma, Y. (2014). Albumin versus other fluids for fluid resuscitation in patients with sepsis: A meta-analysis. PloS one , 9 (12), e114666.

Nursing Case Studies by and for Student Nurses Copyright © by jaimehannans is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License , except where otherwise noted.

Share This Book

Sepsis Patient Case Study

Executive Summary

The percentage of sepsis patient cases meeting bundle requirements was below benchmark and there was opportunity to improve both mortality and length of stay (LOS).

Key Stakeholders

Medical Staff, nursing, performance improvement, virtual sepsis unit (VSU), healthcare informatics, laboratory personnel, pharmacy and patients.

People, Process and Technology

• Interdisciplinary Committee, established December 2017
• Assign patient champions
• Ongoing Physician Education (Team Health EDP group onboard)
• VSU/eICU (PCU/ICU and Med-Surg Units) April 2017; January 2018
• Sepsis Handoff Tool – May 2019
• Sepsis Bundles
• Device integration
• Bedside specimen collection and scanning
• Clinical decision support (CDS)–Alerts
• Care team communication

The sepsis patient mortality rate decreased from as high as 1.91 in Q1 2017 to as low as 0.45 in 2019. Cases meeting the bundle compliance increased from as low as 52% in Jan 2018 to as high as 88% in August 2019. LOS also decreased from a high of 6.83 days on average in January of 2017 to as low as 3.88 days on average in August of 2019.

Lesson Learned

• Teamwork and collaboration were instrumental in the success of bundle build and increasing bundle compliance. Accountability and the ability to measure outcomes and compliance are critical
• Assign a patient champion who is an expert in sepsis management
• Utilization of a VSU such as an eICU provides another layer of surveillance.
• CDS, alerts, dashboards and direct communication with the care teams are part of the direct communication in place to improve care.

Define the Clinical Problem and Pre-Implementation Performance

Local problem.

Our goal was to reduce clinical variation in the care of sepsis patients at Homestead Hospital and throughout the system at Baptist Health South Florida (BHSF). We engaged the care team in improving processes related to the treatment of patients presenting to respective Emergency Departments (ED), via direct admission, or who become septic during their stay.

Sepsis affects over 26 million people worldwide every year, and the organization treats over 3,000 patients annually with sepsis, severe sepsis or septic shock. Sepsis is the body’s response to an infection that has become overwhelming and can lead to tissue damage, organ failure, amputations and death. Mortality increases 8% for every hour that treatment is delayed and as many as 80% of sepsis deaths could have been prevented with rapid diagnosis and treatment. Sepsis patients have the largest cost of hospitalizations in the United States consuming more than $24 billion dollars each year. Sepsis patients also have almost double the average cost per stay at around $18,400 per admission.

Previous work to improve bundle compliance was achieved through the BHSF Accelerated Change Team (ACT), which developed system wide order sets. The reflex lactate was also implemented through the BHSF ACT enabling lab to automatically order a timed lactate to achieve follow-up lactate compliance.

Homestead Hospital has a robust sepsis patient committee that meets monthly, reviewing outcome measures and providing education, mock codes, as well as reviewing cases month to month. The organization partnered with Navigant on the T2020 initiative which includes redesigning care for select DRGs. Navigant and the organization decided upon a structure of a Sepsis Steering Committee as well as two design groups: An ED team and an inpatient and ICU team. These teams collaborated on creating clinical specification to ensure sepsis patients got the same care every patient, every time. BHSF facilities such as Homestead Hospital had varied levels of success in completing the Centers for Medicare and Medicaid Services (CMS) 3-hour and 6-hour bundles for patients identified as septic. The inconsistency of implementing the two bundles in a timely manner led to significant LOS and improved mortality opportunities. Earlier identification and implementation of the interventions described in the bundles led to better outcomes for sepsis patients and a decrease in the LOS.

The mortality rate decreased from as high as 1.91 in Q1 2017 to as low as 0.45 in 2019. Sepsis patient cases meeting bundle compliance increased from as low as 52% in Jan 2018 to as high as 88% in August 2019. LOS also decreased from a high of 6.83 days on average in January of 2017 to as low as 3.88 days on average in August of 2019.

All patients >18 years of age are screened for sepsis, the numerator is the total count of patients treated in compliance with the bundle and the denominator includes all patients with the MS-DRG of sepsis (positive screen). Mortality rates are based on severity adjusted benchmarks and LOS is based on the average LOS against the benchmark of CMS and Premier.

Targeted performance

To meet and/or exceed the benchmark.

Benchmark data

BHSF benchmarks sepsis patient data against CMS, the Acute Physiology And Chronic Health Evaluation (APACHE) IV severity of disease classification system (ICU/PCU), Premier, and internal goals.

Technology initiatives

Electronic health record (EHR) data, CDS such as the St. Johns Sepsis Alert, clinical dashboards, Ascom phones for communication, the eICU for virtual care management, bedside specimen collection scanning, device integration for clinical data and bundle management via PowerPlans™.

People and Process

The evidence-based clinical care (EBCC) committee oversees the organizational structure for process waves which identify areas for improvement. The VSU is a component that enhances people, process and technology. The sepsis patient champions help to optimize infection management and emphasize the importance of early recognition and timely treatment, they also facilitate sepsis patient care and optimize patient outcomes. Ongoing physician and team education are available via lunch and learns with classroom time, elbow to elbow support, web-based learning, online formats on the EBCC website via the intranet and on the Baptist Health South website which is available in the public domain. The continuing education is a vital component to the hospital-wide code rescue response team.

Design and Implementation Model Practices and Governance

Baptist Health South Florida’s EBCC initiative is a strategic system-wide standardization effort to reduce variation and unnecessary costs while focusing on evidence-based, quality care. The process is driven by key stakeholders and is supported by real-time, statistically supported benchmarked data. The charter was signed in 2016 and provides the foundation for a methodical approach to improve patient outcomes (Figure 1).

The methodology begins with a call to action to for an evidenced based care assessment of current and future state, design plan, team approval, development of an implementation plan, measurement and sustainment plans (Figure 2).

Each specific project is supported by a sub-group who are experts in on the focus topic. The Service Line Collaborative includes:

• Cardiac and Vascular
• Critical Care
• Emergency Department
• Gastrointestinal
• Infectious Disease
• Neonatology
• Neuroscience
• Orthopedics
• Surgery/PEI/ERAS/NSQIP

Navigant and BHSF decided upon a structure for a Sepsis Steering Committee to reduce variation in the management of sepsis to improve the sepsis patient mortality rate.

Education was completed via lunch and learns with classroom time, online formats on the EBCC website via the intranet and on the organization’s website which is available in the public domain.

PowerPlan™ education is a consistent part of physician education and CME education is available with every MS-DRG or pathway as it rolls out. The pilot for the sepsis patient go-live began in February 2017 and the hospital wide go-live was April 2017. The iterations over time have continued based on new benchmarks and evidence as it becomes available.

Technologies include: the physiologic data within the EHR, device integration, CDS, bedside specimen collection, sepsis algorithm, support from the eICU/VSU, dashboards and PowerPlans™ for the care bundle.

Clinical Transformation enabled through Information and Technology

To reduce clinical variation in the care of sepsis patients throughout the health system at BHSF, we engaged the care teams to improve processes related to the treatment of patients presenting to respective EDs, via direct admission, or who become septic during their stay. The workflows are geared to meet the care requirements as outlined by the industry in evidenced based research such as CMS and the Society of Critical Care Medicine’s Surviving Sepsis Campaign (Figure 3). While the overarching goal is the same throughout the venues of care, the workflows are created to meet clinical specification to ensure sepsis patients get the same care—every patient, every time. The utilization of the bundle is the foundation for minimizing the variation in care, and the people, process and technology as overseen by the EBCC committee provides the balance to drive action.

The workflows for the ED begin at triage and lead to the engagement of the VSU (Figure 4). The VSU is “air traffic control” for compliance and workflows are also designed for the ICU/PCU and med/surg areas (Figures 5 and 6). The VSU is operated out of the eICU and the virtual team streamlines the workflows to improve compliance to CMS guidelines, improving outcomes and reducing reimbursement penalties.

The algorithm that drives the alert is embedded into the EHR (Figure 7). The EHR supplies the clinical data required for the alert by integrating technologies such as vital sign devices, bedside specimen collection and scanning and lab values. Dashboards are available in the VSU, ED and nursing units to enhance access to the alerts, as well as alerting within the EHR.

Documentation of care takes place in the EHR and the CDS for the algorithm generates the alert (Figure 8).

Managing the alert volume to prevent alert fatigue is a key responsibility of the VSU. The VSU reduces the number of non-actionable alerts going to physicians and nurses. Fewer alerts help to improve the specificity of the alert and provides clinical validation. (Figure 9).

In 2019, new guidelines were released for the recommendation of lactate measures and these recommendations were built into the bundle and the workflows (Figure 10).

Examples of documentation for the bundle includes suspected sepsis patient and a quick bundle (Figure 9). Education for all care is available in the electronic version of ‘what I need to know’ (eWINK), a BHSF online education tool in the public domain which also offers CMEs/CEUs. Education for staff also includes lunch and learns with classroom time, elbow to elbow support and online formats on the EBCC website via the intranet. PowerPlan™ education is a consistent part of physician education and WINK collateral, CME education is available with every MS-DRG or pathway as it rolls out (Figure 11).

Using existing infrastructure of the eICU, virtual sepsis management was incorporated into exiting workflows. PowerPlans™ are used for bundle documentation, integration of clinical data within the EHR is supported by device integration and specimen collection and the EBCC drives pushing the current evidence to the point of care and keeps the educational material up to date.

Improving Adherence to the Standard of Care

All patients >18 years of age are screened for sepsis upon triage in the ED and all inpatients >18 years of age are monitored via CDS surveillance with the sepsis alert running within the EHR. The numerator is the total count of patients treated in compliance with the bundle and the denominator includes all patients with the MS-DRG of sepsis (positive screen). The organization transitioned to the current EHR in September of 2016 and implemented the bundle PowerPlans™ and sepsis initiative in 2017. Prior data indicated the facilities had varied levels of success in completing the bundles for patients identified as septic and the inconsistency led to opportunities to improve LOS and mortality rate.

Over time, at Homestead Hospital the compliance rate for the CMS 3-hour sepsis bundle increased from ~35% in 2015 to >90% in February 2020, with the data steward being CMS (Figure 12).

Homestead Hospital followed the standard process of change management and care redesign as outlined in the EBCC methodology. The EBCC is the governing body driving the utilization of evidence-based care focused on eliminating variation in care delivery.

Improving Patient Outcomes

The sepsis severity adjusted mortality rate decreased from as high as 1.91 in Q1 2017 to as low as 0.45 in 2019 (Figure 13). Average LOS also decreased from a high of 6.83 days on average in January of 2017 to as low as 3.88 days on average in August of 2019 (Figure 14). The risk adjusted mortality and O:E ratio are generated from Premier data.

Accountability and Driving Resilient Care Redesign

BHSF and Homestead Hospital rely on a data driven and evidence based clinical care approach to guide the design and implementation of sepsis patient care bundles. The goals of the organization’s EBCC are to decrease variation across the clinical areas and provide predictable, data-driven high quality, affordable care. Having the tools to collect as close to real-time as possible compliance data and report on that data in near-real-time reflects the ability of the organization to target and successfully improve care delivery, and ultimately improve the clinical outcomes.

In near-real-time, the clinical team document the compliance of bundle utilization and the data is then accessible in their DivePort analytics dashboard (Figures 15 and 16). The dashboard is designed to provide statistical analysis, benchmark information and severity adjusted data with the capability to drill in multiple layers.

PowerPlans™ and bundle utilization is also available in DivePort, lending to the capability of measuring and reporting not only on the outcomes, but also to the compliance of the guidelines (Figure 17).

Using analytics to find variation

• APR-DRG Population Group is identified based on cost opportunity when compared to HCUP 40th percentile (Total Cost Per Case and ALOS).
• EBCC and Analytics Integrity Committee members review MS-DRG specific groupings that correspond with APR-DRG grouping.
• Premier benchmark levels of 50th and 75th percentile variable cost opportunities are used to further validate the data.
• MS-DRGs are recommended for EBCC team redesign based on variable cost opportunity, average LOS and volume.
• EBCC DivePort 7.0 Portal reporting is updated as DRGs waves are defined for tracking outcomes.

The rollout of the VSU is an example of using data to further refine the care redesign to complement the people, process and technology to enhance care delivery (Figure 18).

The views and opinions expressed in this content or by commenters are those of the author and do not necessarily reflect the official policy or position of HIMSS or its affiliates.

HIMSS Davies Awards

The HIMSS Davies Award showcases the thoughtful application of health information and technology to substantially improve clinical care delivery, patient outcomes and population health.

Begin Your Path to a Davies Award

JL / Science Photo Library / Shutterstock.com

Several personal stories of sepsis survivors and deaths from sepsis have been published in the media, highlighting the importance of sepsis as a major cause of preventable deaths in the UK. This article will provide an overview of sepsis with particular focus on the role of pharmacy professionals in recognition and referral.

Pathophysiology

Sepsis is characterised by a systemic inflammatory response to an invasive infection that has become unregulated [1] . White blood cells and pro-inflammatory cytokines are released causing widespread vasodilation and an increase in capillary permeability, resulting in the loss of fluid from circulation. This results in hypovolaemia and a fall in systemic vascular resistance, which in turn leads to a fall in blood pressure and a decrease in organ perfusion, culminating in tissue hypoxia and organ failure [1] .

There are around 250,000 cases of sepsis annually in the UK, 20% of which are fatal [1], [2] . The UK Sepsis Trust estimates that early diagnosis of sepsis and the application of evidence-based treatment could save 14,000 lives per year [2] . Therefore, the timely identification of sepsis can lead to rapid treatment and potential mortality reduction.

Initial recognition of sepsis relies on identifying symptoms, which presents challenges for healthcare professionals as the common symptoms are not specific to sepsis and could be caused by non-infective pathology (e.g. trauma, pancreatitis, burns ) [3] .

The inflammatory-response-induced hypovolaemia affects the brain and can cause confusion, slurred speech and loss of consciousness. Similarly, as the kidneys are affected, there is a reduction in glomerular filtration resulting in a drop in urine output and development of acute kidney injury [1] , [3] .

Gas exchange across the alveoli is compromised as fluid and proteins leak into the lungs, causing a drop in systemic oxygen saturation and a rise in carbon dioxide levels. The body attempts to compensate by increasing its respiratory rate, but the problem is ultimately compounded as the drop in organ perfusion affects the lungs, meaning that even if oxygen-rich air is present, there is little blood flow with which gas exchange can take place [1] , [3] .

Initially, the heart rate increases in an attempt to compensate for the drop in blood pressure. However, the reduction in circulating volume and the fall in systemic vascular resistance ultimately undermine this action. This is because a reduction in venous return prevents the ventricles from properly filling before they contract, reducing the cardiac output [1] .

The typical signs and symptoms of sepsis may vary across different age groups, be general or may not all be present. Examples of moderate- and high-risk symptoms include:

• Reduced urine output (e.g. dry nappies in babies and toddlers);
• Feeling cold with shivering or chills;
• Rapid breathing (increased resting breaths per minute);
• Rapid heart rate (increased resting heart rate per minute);
• Mottled (see Photoguide A) or ashen appearance;
• Cyanosis (blue tint) of skin, lips or tongue (see Photoguide B);
• Non-blanching rash (see Photoguide C) [1] , [4] , [5] .

Additional signs in babies and young children (aged under five years)

• Not responding normally to social cues (e.g. does not smile);
• Visibly unwell (e.g. floppy or overly passive);
• Wakes only with prolonged stimulation or, if roused, does not stay awake;
• Weak high-pitched or continuous cry;
• Parent or carer is concerned that the child is behaving differently from normal;
• Has a seizure or convulsion;
• Pallor of the skin, lips or tongue;
• Cold extremities, but head and torso may be hot to the touch;
• Change in temperature (e.g. red flag temperatures are over 38 o C in those aged under 3 months, 39 o C in those aged 3–6 months and less than 36 o C for any age) [1] , [4] , [5] .

Additional signs that can affect patients aged over five years

• Evidence of new-onset confusion (history may be from a parent, carer, relative or friend);
• Slurred speech;
• Signs of potential infection (e.g. redness, swelling or discharge at surgical site, breakdown of the wound);
• Tympanic temperature less than 36 o C;
• New onset arrhythmia [1] , [4] , [5] .

If sepsis is suspected, the patient should be immediately referred for emergency medical assessment. 

Photoguide: symptoms of sepsis

Source: Science Photo Library / Shutterstock.com

At-risk groups

Sepsis can affect anyone, but there are some patient groups that should be considered to be more susceptible to the development of sepsis, including:

• Very young children (aged under 1 year);
• Frail or older people (aged over 75 years);
• Immunocompromised people (e.g. those being treated for cancer with or without chemotherapy, post-splenectomy, taking long-term steroids or other immunosuppressant drugs);
• People who have had surgery or other invasive procedures in the past six weeks;
• People with any breach of skin integrity (e.g. cuts, burns, blisters or skin infections);
• People who misuse drugs intravenously;
• People with indwelling lines or catheters;
• Pregnant women;
• Women who have given birth or who have had a termination of pregnancy or miscarriage in the past six weeks [4] .

Understanding the patient risk factors may help improve timely diagnosis of suspected sepsis.

Sepsis can occur in response to a wide range of infections, but is most commonly associated with bacterial infection of the lungs, urinary tract, abdomen, central nervous system, or skin and soft tissues [1] . It is primarily diagnosed by a clinical assessment. Any number or combination of signs and symptoms may be present on diagnosis.

The National Institute of Health and Care Excellence (NICE) and the UK Sepsis Trust have published risk stratification tools to facilitate appropriate recognition of sepsis and the level of risk to the patient. Most NHS organisations use these tools or a locally approved variation of them [1] , [4] . These tools cater to a wide range of patients of different ages and the recommended action differs depending on whether patients’ symptoms are recognised in the primary or secondary care setting.

Patients who meet the high-risk criteria (see Box for the criteria for children aged under five years) should be sent urgently for emergency care (at a setting with resuscitation facilities). These patients should receive intravenous antibiotics with an appropriate level of cover within one hour of recognition of sepsis, along with other treatments and investigations [1] , [4] .

Box: high-risk criteria for children aged under five years outside of the hospital setting

• No response to social cues;
• Appears ill to a healthcare professional;
• Does not wake, or if roused, does not stay awake;
• Weak high-pitched or continuous cry.
• Aged under one year: 160 beats per minute or more;
• Aged one to two years: 150 beats per minute or more;
• Aged three to four years: 140 beats per minute or more;
• Heart rate less than 60 beats per minute at any age.

Respiratory rate

• Aged under one year: 60 breaths per minute or more;
• Aged one to two years: 50 breaths per minute or more;
• Aged three to four years: 40 breaths per minute or more;
• Oxygen saturation of less than 90% in air or increased oxygen requirement over baseline.

Temperature

• Aged under three months: 38°C or more;
• Any age: less than 36°C.

To see examples of mottled skin or ashen appearance, non-blanching rash of the skin, and cyanosis of the skin, lips or tongue, see the Photoguide.

Source: National Institute for Health and Care Excellence [8]

Initial blood tests should be requested to aid diagnosis and further inform on the likelihood of infection and prognosis. These should include:

• C-reactive protein — to detect inflammatory response;
• Full blood count — to detect immune response;
• Lactate — to detect tissue hypoxia.

Other investigations, such as a chest X-ray or lumbar puncture, may also be indicated depending on the likely focus of the suspected infection.

Point-of-care testing and future diagnostics

In the future, point-of-care testing facilities for key biomarkers may have a greater role in both primary care and hospital emergency departments, aiding healthcare professionals in diagnosing infection and sepsis. Recently published research into the use of sensor technologies designed to rapidly report on raised levels of biomarkers closely associated with sepsis (notably interleukin-6) may have the potential to aid sepsis diagnosis in the future [6] , [7] .

Case studies

Case study 1: a 12-month-old baby with suspected sepsis.

A mother brings her 12-month-old daughter Alice* into the pharmacy and asks to speak to the pharmacist. The mother clearly appears concerned and expresses that Alice seems very poorly and is not her usual self.

Consultation

The mother explains that her daughter attended nursery today and the nursery staff phoned in the afternoon to report that Alice was not feeling well. As she had a temperature of 39 o C, they administered a dose of paracetamol. The nursery staff also said that Alice had not been eating or drinking well and her nappies were dry all day.

Alice takes no regular medicine and has no ongoing health problems. She seems withdrawn and is not smiling or engaging with anyone, which the mother insists is unusual behaviour.

When assessing the patient, Alice’s hands feel cold despite her body and head feeling hot. During the interaction with Alice, she is noticeably very passive and inactive, and seems floppy in her mother’s arms. Alice does not appear to have a rash, though her skin appears pale.

Alice’s breathing appears very rapid.

Information given in the consultation suggests that Alice may have sepsis. Use the  National Institute for Health and Care Excellence ‘Sepsis risk stratification tool’ for children aged under five years out of hospital and compare it to the information obtained during the consultation [8] . By doing so, it is apparent that the following moderate-to-high risk criteria were demonstrated by the patient:

• Behaviour: parent or carer concern that the child is behaving differently than usual; not responding normally to social cues; no smile; and decreased activity;
• Cold hands or feet;
• Reduced urine output;
• Between 40 and 49 breaths per minute (normal resting respiratory rate for a 12-month-old is typically 20–40 breaths per minute);
• Pallor of skin, lips or tongue.

Although her temperature was high (39 o C), this would only be a moderate-risk criterion if Alice was aged between 3–6 months of age outside the hospital setting.

Advice and recommendations

Alice’s symptoms clearly indicate moderate-to-high risk of sepsis. According to the risk stratification tool, Alice should be referred either for a definitive diagnosis for treatment outside the hospital (i.e. to the patient’s GP), or to hospital for further review.

Considering the risk of sepsis and the importance of timely management, referral to hospital is the best choice for this patient. Calmly and clearly explain to the mother that Alice may have a serious infection and that the best course is for her to go to the hospital immediately.

Case study 2: an 11-year-old child with a viral infection

An 11-year-old boy called Liam* is brought into the pharmacy by his parents. They explain that he has not been feeling well for the past few days. The parents want to know if Liam’s symptoms can be treated with an over-the-counter (OTC) product or whether they should take him to the GP.

Liam has been unwell for the past three or four days and his condition does not seem to be improving. He has not yet taken any medicine for this illness. His temperature was 37.8 o C when most recently checked using an ear thermometer at home.

Liam says that he has a sore throat, a cough and a blocked nose. Liam describes the severity of his throat pain as three out of ten. Liam has been eating and drinking normally, has no long-term medical conditions and takes no regular medicine.

Liam and his parents agree that his behaviour and function are normal and he appears alert and coherent through the consultation. Liam’s breathing rate does not appear to be raised and his skin and lips appear normal, with no signs of a rash.

Liam may have an infection, which is likely to be viral, but does not require medical attention at present. However, to rule out sepsis, use the  National Institute for Health and Care Excellence ‘Sepsis risk stratification tool’ for children aged 5–11 years out of hospital and compare this to the information obtained during the consultation [8] . By doing so, it is apparent that no moderate-to-high risk criteria were demonstrated by the patient.

Liam can be treated with OTC analgesics to alleviate his throat pain and his raised temperature. Liam’s parents should ensure he is taking plenty of fluids and continue to monitor his temperature. If they become concerned about his condition, his behaviour or general functional ability, they should return to or call the pharmacy. Particular symptoms you advise them to look out for include development of a rash, if his skin becomes pale or mottled, his urine output drops, his breathing rate increases, or any general concern that mental state or activity is not normal.

Case study 3: an adult with red flag sepsis

Rahul*, a 28-year-old man, comes into the pharmacy and asks to see the pharmacist as he has begun to feel very unwell.

Rahul works as a driver, and he fell and scraped his leg while getting out of his van the previous day. Rahul explains that at the time he did not think much about the cut, so he did not clean or dress the wound and has not yet taken any medicine for it. Although the wound began to swell and weep overnight, Rahul went into work this morning, but left when he suddenly started to feel seriously unwell — around an hour ago.

Rahul says the wound now looks much worse than it previously did. On examination you find the wound is very red and swollen, the tissue around the wound is blistering and weeping, and there are red track marks extending further up Rahul’s leg. It is clear that this wound is infected.

Rahul uses inhalers for mild asthma, but otherwise takes no regular medicine.

Rahul says he is feeling hot and sluggish, and is clearly struggling to maintain his train of thought. His condition appears to be worsening by the minute. He needs to be referred for further help, but more information is required to decide whether he is referred to his GP or to a hospital emergency department. 

Rahul’s temperature is 35.8 o C. His breathing appears to be rapid. His blood pressure is 92/58mmHg and his heart rate is 140 beats per minute. Rahul does not appear to have a skin rash, although his lips seem to have a blue tint.

Rahul is likely to have a serious infection, which requires urgent attention. Use the National Institute for Health and Care Excellence ‘Sepsis risk stratification tool’ for people aged 18 years and over outside of the hospital setting and compare this to the information obtained during the consultation [8] . The patient has moderate- to high-risk criteria, such as tympanic temperature less than 36 o C and signs of potential infection. He also has several high-risk criteria that require urgent referral to emergency care:

• Altered behaviour or mental state;
• Respiratory rate over 25 breaths per minute;
• Heart rate more than 130 beats per minute;
• Cyanosis of skin, lips or tongue.

Tell Rahul that it is likely that he has a serious infection, and ask him to sit and wait in the pharmacy while an ambulance is called. It is not safe for him to drive to hospital. Share his information with the 999 operator and make another note of the details gathered during the consultation to hand over to the ambulance team upon arrival.

*All cases are fictional

Useful additional resources

• The UK Sepsis Trust. The Sepsis Manual. 2017–2018: Available at:  https://sepsistrust.org/wp-content/uploads/2018/06/Sepsis_Manual_2017_web_download.pdf
• National Institute for Health and Care Excellence. Sepsis: risk stratification tools. Available at:  https://www.nice.org.uk/guidance/ng51/resources/algorithms-and-risk-stratification-tables-compiled-version-2551488301
• For patients who are interested in finding out more about sepsis you can direct them to the NHS website. Available at:  https://www.nhs.uk/conditions/sepsis/

[1] The UK Sepsis Trust. The Sepsis Manual. 2017–2018. 2017. Available at: https://sepsistrust.org/wp-content/uploads/2018/06/Sepsis_Manual_2017_web_download.pdf (accessed October 2019)

[2] The UK Sepsis Trust. Professional Resources. Available at: https://sepsistrust.org/professional-resources/ (accessed October 2019)

[3] Gotts JE & Matthay MA. Sepsis: pathophysiology and clinical management. BMJ 2016;353:i1585.  doi: 10.1136/bmj.i1585

[4] National Institute for Health and Care Excellence. Sepsis: recognition, diagnosis and early management. NICE guideline [NG51]. 2016. Available at: https://www.nice.org.uk/Guidance/NG51 (accessed October 2019)

[5] National Health Service: Overview — Sepsis. 2019. Available at: https://www.nhs.uk/conditions/sepsis/ (accessed October 2019)

[6] Russell C, Ward AC, Vezza V et al . Development of a needle shaped microelectrode for electrochemical detection of the sepsis biomarker interleukin-6 (IL-6) in real time. Biosens Bioelectron 2019;126:806–814. doi: 10.1016/j.bios.2018.11.053

[7] Dolin H, Papadimos T, Stepkowski S et al. A novel combination of biomarkers to herald the onset of sepsis prior to the manifestation of symptoms. Shock 2018;49(4):364–370. doi: 10.1097/SHK.0000000000001010

[8] National Institute for Health and Care Excellence. Sepsis: Risk stratification tools. 2017. Available at: https://www.nice.org.uk/guidance/ng51/resources/algorithms-and-risk-stratification-tables-compiled-version-2551488301 (accessed October 2019)

You might also be interested in…

How the London Ambulance Service is working to transform medicines management

All emergency departments must have dedicated pharmacist, royal college and UKCPA say

Hyponatraemia: clinical presentation, diagnosis and treatment

Septic Shock (Sepsis) Case Study (45 min)

Watch More! Unlock the full videos with a FREE trial

Included In This Lesson

Study tools.

Access More! View the full outline and transcript with a FREE trial

What initial nursing assessments need to be performed for Mr. McMillan?

• Full set vital signs (T, P, RR, BP, SpO 2 )
• OLDCARTS or PQRST assessment of symptoms (urinary burning)
• LOC/orientation assessment
• Heart and lung sounds

Upon further assessment, Mr. McMillan is weak, his face is flushed, his skin is warm and dry. He is oriented to person and place, but states the year is 1952. His vital signs were as follows:

BP 99/60 mmHg Ht 170.2 cm

HR 92 bpm and regular Wt 60 kg 

RR 28 bpm SpO 2 93% on Room Air

Temp 38.9°C

What diagnostic tests should be ordered for Mr. McMillan?

•  Blood Tests – CBC, BMP, ABG, Lactic Acid, Blood Cultures x 2
• Urine Tests – Urinalysis, Urine Culture
• X-rays – Chest, Kidneys/Ureters/Bladder

What nursing actions would you take at this time for Mr. McMillan? Why?

•  Elevate the HOB to improve breathing and oxygenation
• Apply cardiac monitor
• Notify provider of elevated temp and low SpO 2  
• Apply cool washcloth to forehead and/or behind neck for comfort
• Possibly get ice packs to axillae and groin and remove any blankets to help bring the patient’s temperature closer to normal.

The ED provider orders the following:

• Bloodwork – CBC, BMP, ABG, Lactic Acid, Blood Cultures x 2
• Diagnostics – CXR (chest x-ray), KUB (x-ray of kidneys, ureters, and bladder)
• Nasal Cannula to keep SpO 2 > 92%
• Meds – 1L Normal Saline bolus IV x 1, now.  1,500 mg Vancomycin IVPB x 1 dose, now

Which order should you implement first? Why?

• Blood and urine cultures must be drawn before any antibiotics are administered.
• Blood work – urine tests – fluids – antibiotics
• IF the patient’s SpO 2 is below 92%, apply oxygen via nasal cannula – at this time, there is no indication of that, yet.

All blood and urine tests are completed and you initiate the fluid bolus for Mr. McMillan. You are still waiting for the Vancomycin to arrive from the pharmacy. You notice he is more drowsy. He is now only oriented to self and feels warmer. You take another set of vital signs to find the following:

BP 86/50 mmHg MAP 62 mmHg

HR 108 bpm Temp 39.3°C

RR 36 bpm SpO 2 88% on Room Air

Mr. McMillan’s lab results have also resulted, the following abnormal values were reported:

WBC 22,000 / mcL Lactic Acid 3.6 mmol/L

pH 7.22 pCO 2 30 mmHg

HCO 3 16 mEq/L pO 2 64 mmHg

Urine Cloudy with sediment

What action(s) should you take at this time? Why?

•  #1 – apply oxygen via nasal cannula – ensure HOB elevated for easy breathing
• Notify provider of decreasing blood pressure and elevated WBC, lactic acid ANSWER

What orders do you anticipate for Mr. McMillan? (procedures, meds, transfer, etc?)

• Mr. McMillan may need another liter of IV fluids. The guidelines are for patients to receive 30 mL/kg of body weight in the first 6 hours. That means he would need to receive at least 1,800 mL of IV fluid bonuses.
• Mr. McMillan may need vasopressors to improve his blood pressure – in which case he will also need a central line for administration of those medications as well as an arterial line to monitor his MAP.
• Mr. McMillan will need to be transferred to the ICU for close monitoring and management of his drips

Mr. McMillan responds well to the first liter of fluids, and antibiotics are initiated within an hour of arrival. The ED physicians place an arterial line and central line to initiate vasopressors. They order a Norepinephrine infusion to be titrated to keep MAP > 65 mmHg. The Critical Care team asks you to prepare the patient for transfer to the ICU.

Art. Line BP 82/48 mmHg MAP 58 mmHg

HR 122 bpm CVP 4 mmHg

RR 32 bpm SVR 640 dynes/sec/m -5 SpO 2 90% on Room Air

What, physiologically, is going on with Mr. McMillan?

• Mr. McMillan has an infection, likely urinary, and it has created a systemic inflammatory response. That inflammatory response is causing massive peripheral vasodilation so his vital organs are not receiving adequate blood flow
• He is showing signs of decreased perfusion to his brain (↓ LOC) and decreased cardiac output (↓ BP).  
• His skin is warm and flushed and his temperature is elevated because of the vasodilation in the non-vital organs.

What does it mean to titrate an infusion to keep MAP >65?

• Titration means achieving the desired result with the least amount of drug possible. Therefore we would adjust the infusion up or down to maintain the MAP above, but not too far above, 65 mmHg

After 2 days in the ICU, a norepinephrine infusion and a total of two liters of normal saline, Mr. McMillan’s blood pressure is stable, his MAP is 67 mmHg. He is becoming more alert and is now oriented to person, place, and time.  His blood and urine cultures were positive for bacterial growth. He has received multiple doses of Vancomycin as well as antibiotics targeted to his specific bacterial infection. He is being weaned off of the vasopressors, and the providers hope he can transfer out of the ICU tomorrow.

What explanation or education topics would you want to provide to the patient and his caregiver before discharge?

• Sepsis and septic shock are a result of a severe infection that has gotten into the bloodstream and affected the patient’s ability to pump blood to the body. This is what makes their blood pressure drop so low. We treat this condition by getting the infection under control and supporting the patient’s blood pressure.
• Signs and symptoms of infection – in elderly people, one of the first signs of infection is altered mental status. If the patient seems ‘off’ or ‘not themselves’, it is worth notifying a healthcare provider to prevent a worse situation. 
• The patient will need to ensure he is drinking plenty of fluids and practicing good hygiene to prevent urinary tract infections. He may also consider cranberry juice.
• If receiving a PO course of antibiotics – be sure to take the full course and notify HCP of any adverse reactions.

Happy Nursing!

View the FULL Outline

When you start a FREE trial you gain access to the full outline as well as:

• SIMCLEX (NCLEX Simulator)
• 6,500+ Practice NCLEX Questions
• 2,000+ HD Videos
• 300+ Nursing Cheatsheets

“Would suggest to all nursing students . . . Guaranteed to ease the stress!”

Nursing Case Studies

This nursing case study course is designed to help nursing students build critical thinking.  Each case study was written by experienced nurses with first hand knowledge of the “real-world” disease process.  To help you increase your nursing clinical judgement (critical thinking), each unfolding nursing case study includes answers laid out by Blooms Taxonomy  to help you see that you are progressing to clinical analysis.We encourage you to read the case study and really through the “critical thinking checks” as this is where the real learning occurs.  If you get tripped up by a specific question, no worries, just dig into an associated lesson on the topic and reinforce your understanding.  In the end, that is what nursing case studies are all about – growing in your clinical judgement.

Nursing Case Studies Introduction

Cardiac nursing case studies.

• 6 Questions
• 7 Questions
• 5 Questions
• 4 Questions

GI/GU Nursing Case Studies

• 2 Questions
• 8 Questions

Obstetrics Nursing Case Studies

Respiratory nursing case studies.

• 10 Questions

Pediatrics Nursing Case Studies

• 3 Questions
• 12 Questions

Neuro Nursing Case Studies

Mental health nursing case studies.

• 9 Questions

Metabolic/Endocrine Nursing Case Studies

Other nursing case studies.

Case Study: A Systematic Approach to Early Recognition and Treatment of Sepsis

Submitted by Madeleine Augier RN BSN

Tags: assessment Case Study emergency department guidelines mortality prevention risk factors sepsis standard of care treatment

Share Article:

Sepsis is a serious medical condition that affects 30 million people annually, with a mortality rate of approximately 16 percent worldwide (Reinhart, 2017). The severity of this disease process is not well known to the public or health care workers. Often, health care providers find sepsis difficult to diagnose with certainty. Deaths related to sepsis can be prevented with accurate assessments and timely treatment. Sepsis must be considered an immediate life-threatening condition and needs to be treated as a true emergency.

Relevance and Significance

Sepsis is defined as “the life-threatening organ dysfunction resulting from a dysregulated host response to infection” (Kleinpell, Schorr, & Balk, 2016, p. 459). Jones (2017) study of managing sepsis affirms that the presence of sepsis requires a suspected source of infection plus two or more of the following: hyperthermia (>38.1 degrees Celsius) or hypothermia (<36 degrees Celsius), tachycardia (>91 beats per minute), leukocytosis or leukopenia, altered mental status, tachypnea (>21 breaths per minute), or no urine output for 12 hours. If the infection persists, acute organ dysfunction or failure occurs from widespread inflammation, eventually leading to septic shock (Palleschi, Sirianni, O’Connor, Dunn, & Hasenau, 2013).  Palleschi et al.  (2013) states that during septic shock, “the cardiovascular system fails, resulting in hypotension, depriving vitals organs of an adequate supply of oxygenated blood” (p. 23). Ultimately the body can go into multiple organ dysfunction syndrome (MODS), leading to death if there is inaccurate assessment and inadequate treatment.

The purpose of this case study is to make the nurse practitioner aware of the severity sepsis, and how to accurately diagnose and treat using evidence-based data. Sepsis can affect everyone, despite his or her age or comorbidity.  Center for Medicare and Medicaid Services (CMS) has diagnosed this problem as a priority and uses sepsis management in determining payment to providers (Tedesco, Whiteman, Heuston, Swanson-Biearman, & Stephens, 2017). This medical diagnosis is unpredictable and presents a challenge to nurse practitioners worldwide. Early recognition and treatment of sepsis by the nurse practitioner is critical to decrease morbidity and mortality.

After completing this case study, the reader should be able to:

• Identify the risk factors of sepsis
• Identify the signs and symptoms of sepsis
• Identify the treatment course of sepsis

Case Presentation

A 65-year-old Asian female presented to the emergency department accompanied by her husband with a chief complaint of altered mental status. Upon assessment, the patient was lethargic, and alert and oriented to person only. The patient’s heart rate was 136, blood pressure 104/50, oral temperature 99 degrees Fahrenheit, oxygen saturation 97% on 4 liters nasal cannula, and respirations 26 per minute. The patient’s blood glucose was obtained with a result 454.

Further orders, such as labs and imaging were made by the provider to rule out potential diagnoses. A rectal temperature was obtained revealing a fever of 103.7 degrees Fahrenheit. The patient remained restless on the stretcher. After one hour in the emergency department, her heart rate spiked to 203 beats per minute, respirations became more rapid and shallow, and she became more lethargic. The patient’s altered mental status, increasing heart rate and respirations caused the providers to act rapidly.

Medical History

The patient’s husband reports that she is a type one diabetic, he denies any other medical conditions. In addition, the patient’s husband states that she has not been exposed to any sick individuals in the past few weeks. The husband reports a family history of diabetes, other wise no significant familial history. No history of smoking, drinking, or illicit drug use was to be noted.

Physical Assessment Findings

The patient appeared lethargic and confused with a Glasgow Coma Scale of 12. She appeared tachypnic, with shallow respirations, and a rate of 28 breaths per minute. Upon auscultation, breath sounds were coarse. Her abdomen was soft and non-tender, no nausea or vomiting noted. The patient appeared diaphoretic, and her legs were mottled.

Laboratory and Diagnostic Testing and Results

During the initial assessment, a complete blood count (CBC), basic metabolic panel (BMP), and lactic acid level were ordered for blood work. A STAT electrocardiogram (EKG), urinalysis, and a chest X-ray were ordered to differentiate possible diagnoses. The CBC revealed leukocytosis with a white blood cell count of 23,000 and an increased lactic acid level of 4.3. The anion gap and potassium level remained within a normal limit, ruling out the possibility of diabetic ketoacidosis (DKA). The patient’s EKG showed supraventricular tachycardia (SVT). The chest X-ray revealed infiltrates to the right lung. The urinalysis was free from leukocytes or nitrites. Blood cultures were ordered to confirm their hypothesized diagnosis, septicemia.

Pharmacology

The provider initiated intravenous (IV) fluid treatment with Lactated Ringers at a bolus of 30 mL/kg. Because the patient’s heart rate was elevated, 6 mg of adenosine was ordered to combat the SVT. Additionally, broad-spectrum IV antibiotics were initiated. One gram of vancomycin and 3.375 grams of piperacillin-tazobactam were the preferred antibiotics of choice.

Final Diagnosis

Upon arrival, the providers were ruling out DKA and sepsis, given the patient’s history.

The patient’s elevated white blood cell counts, temperature, lactic acid level, heart/respiratory rate, and altered mental status were all clinical indicators of sepsis. The chest X-ray revealed a right lung infiltrate, persuading the providers to diagnose the patient with sepsis secondary to pneumonia.

Patient Management

After sepsis was ruled as the patient’s diagnosis, rapid antibiotic administration and IV fluid treatment became priority after the patient’s heart rate was controlled. A cooling blanket and a temperature sensing urinary catheter was placed to continuously monitor and control the patient’s fever. Later, the patient was transferred to a critical care unit for further treatment. Shortly after being transferred, the patient went into respiratory failure and was placed on a ventilator. After two days in the ICU, the patient remained in septic shock, and died from multisystem organ failure.

When the patient initially presented to the emergency department, accurate and rapid diagnosis of sepsis was critical in order to stabilize the patient and prevent mortality. A challenge was presented to the provider regarding a rapid diagnosis due to the patient’s history and her presenting signs and symptoms. Increased awareness and interprofessional education regarding sepsis and its’ treatment is vital to decrease mortality. Health care providers need to be competent in recognizing and accurately treating sepsis in a rapid manner.

Research shows that outcomes in sepsis are improved with timely recognition and early resuscitation (Javed et al., 2017). It is important for the provider to identify certain risk factors and symptoms to easily diagnose sepsis. A research study by Henriksen et al. (2015) proved that age, and comorbidities including psychotic disorders, immunosuppression, diabetes, and alcohol abuse served as top risk factors for sepsis.

Once the diagnosis of sepsis is determined, rapid treatment must be initiated. The golden standard of treatment consists of a bundle of care that includes blood cultures, broad-spectrum antibiotic agents, and lactate measurement completed within 3 hours as described by Henriksen et al. (2015). A study by Seymour et al. (2017) showed that the more rapid administration of the bundle of care is correlated with a decreased mortality rate. In addition, The Survival of Sepsis Campaign formed a guideline to sepsis treatment; Rhodes et al. (2016) suggests giving a 30 mL/kg of IV crystalloid fluid for hypoperfusion. If hypotension persists (mean arterial pressure <65), vasopressors, preferably norepinephrine, should be initiated (Rhodes et al., 2016). Prompt recognition of sepsis and implementation of the bundle of care can help reduce avoidable deaths.

To increase awareness, interprofessional education regarding sepsis and its’ common signs and symptoms needs to be established. Evidence-based protocols should be utilized in hospital care settings that provide nurse practitioners with a guideline to follow to ensure rapid and accurate treatment is given. Increased awareness and education helps providers and other healthcare workers to properly identify and accurately treat sepsis.

The public and health care providers must become more aware and educated on the severity of sepsis. It is crucial to be able to recognize signs and symptoms of sepsis to prevent further complications such as septic shock and multi-organ failure. Increased awareness, interprofessional education, accurate assessment, and rapid treatment can help reduce incidence and mortality. Sepsis management must focus upon early goal-directed therapy (antibiotic administration, fluid resuscitation, blood cultures, lactate level) and individualized management pertaining to the patient’s history and assessment (Head & Coopersmith, 2016). Misdiagnosis and delay in emergency treatment can result in missed opportunities to save lives.

• Head, L. W., & Coopersmith, C. M. (2016). Evolution of sepsis management:from early goal-directed therapy personalized care. Advances in Surgery, 50 (1), 221-234. doi:10.1016/j.yasu.2016.04.002
• Henriksen, D. P., Pottegard, A., Laursen, C. B., Jensen, T. G., Hallas, J., Pedersen, C., & Lassen, A. T. (2015). Risk factors for hospitalization due to community-acquired sepsis-a population-based case-control study. PLOS ONE, 10 (4), 1-12. doi:10.1371/journal.pone.0124838
• Javed, A., Guirgis, F. W., Sterling, S. A., Puskarich, M. A., Bowman, J., Robinson, T., & Jones, A. E. (2017). Clinical predictors of early death from sepsis. Journal of Critical Care, 42 , 30-34. doi:10.1016/j.jcrc.2017.06.024
• Jones, J. (2017). Managing sepsis effectively with national early warning scores and screening tools. British Journal of Community Nursing, 22 (6), 278-281. doi:10.12968/bjcn.2017.22.6.278
• Kleinpell, R. M., Schorr, C. A., & Balk, R. A. (2016). The new sepsis definitions: Implications for critical care. American Journal of Critical Care, 25 (5), 457-464. doi:10.4037/ajcc2016574
• Palleschi, M. T., Sirianni, S., O'Connor, N., Dunn, D., & Hasenau, S. M. (2013). An interprofessioal process to improve early identification and treatment for sepsis. Journal for Healthcare quality, 36 (4), 23-31. doi:10.1111/jhq.12006
• Reinhart, K., Daniels, R., Kissoon, N., Machado, F. R., Schachter, R. D., & Finfer, S. (2017). Recognizing sepsis as a global health priority-A WHO resolution. The New England Journal of Medicine, 377 (5), 414-417. doi:10.1056/NEJMp1707170
• Rhodes, A., Evans, L. E., Alhazzani, W., Levy, M. M., Anotnelli, M., Ferrer, R.,...Beale, R. (2017). Surviving sepsis campaign: International guidelines for management of sepsis and septic shock: 2016. Intensive Care Medicine, 43 (3), 304-377. doi:10.1007/s00134-017-4683-6
• Seymour, C. W., Gesten, F., Prescott, H. C., Friedrich, M. E., Iwashyna, T. J., Phillips, G. S.,...Levy, M. M. (2017). Time to treatment and mortality during mandated emergency care for sepsis. The New England Journal of Medicine, 376 (23), 2235-2244. doi:10.1056/NEJMoal1703058
• Tedesco, E. R., Whiteman, K., Heuston, M., Swanson-Biearman, B., & Stephens, K. (2017). Interprofessional collaboration to improve sepsis care and survival within a tertiary care emergency department. Journal of Emergency Nursing, 43 (6), 532-538. doi:10.1016/j.jen.2017.04.014

Career Opportunities

More Like This

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Sepsis assessment and management in critically Ill adults: A systematic review

Contributed equally to this work with: Mohammad Rababa, Dania Bani Hamad, Audai A. Hayajneh

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation Adult Health Nursing Department, Faculty of Nursing, Jordan University of Science and Technology, Irbid, Jordan

Roles Conceptualization, Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing

• Mohammad Rababa, 
• Dania Bani Hamad, 
• Audai A. Hayajneh

• Published: July 1, 2022
• https://doi.org/10.1371/journal.pone.0270711
• Reader Comments

Early assessment and management of patients with sepsis can significantly reduce its high mortality rates and improve patient outcomes and quality of life.

The purposes of this review are to: (1) explore nurses’ knowledge, attitude, practice, and perceived barriers and facilitators related to early recognition and management of sepsis, (2) explore different interventions directed at nurses to improve sepsis management.

A systematic review method according to the PRISMA guidelines was used. An electronic search was conducted in March 2021 on several databases using combinations of keywords. Two researchers independently selected and screened the articles according to the eligibility criteria.

Nurses reported an adequate of knowledge in certain areas of sepsis assessment and management in critically ill adult patients. Also, nurses’ attitudes toward sepsis assessment and management were positive in general, but they reported some misconceptions regarding antibiotic use for patients with sepsis, and that sepsis was inevitable for critically ill adult patients. Furthermore, nurses reported they either were not well-prepared or confident enough to effectively recognize and promptly manage sepsis. Also, there are different kinds of nurses’ perceived barriers and facilitators related to sepsis assessment and management: nurse, patient, physician, and system-related. There are different interventions directed at nurses to help in improving nurses’ knowledge, attitudes, and practice of sepsis assessment and management. These interventions include education sessions, simulation, decision support or screening tools for sepsis, and evidence-based treatment protocols/guidelines.

Our findings could help hospital managers in developing continuous education and staff development training programs on assessing and managing sepsis in critical care patients.

Nurses have poor to good knowledge, practices, and attitudes toward sepsis as well as report many barriers related to sepsis management in adult critically ill patients. Despite all education interventions, no study has collectively targeted critical care nurses’ knowledge, attitudes, and practice of sepsis management.

Citation: Rababa M, Bani Hamad D, Hayajneh AA (2022) Sepsis assessment and management in critically Ill adults: A systematic review. PLoS ONE 17(7): e0270711. https://doi.org/10.1371/journal.pone.0270711

Editor: Paavani Atluri, Bay Area Hospital, North Bend Medical Center, UNITED STATES

Received: December 1, 2021; Accepted: June 14, 2022; Published: July 1, 2022

Copyright: © 2022 Rababa et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the article and its files.

Funding: This study was funded by The deanship of research at Jordan University of Science and Technology (grant number 20200668).

Competing interests: The authors have declared that no competing interests exist.

Introduction

Sepsis is a global health problem that increases morbidity and mortality rates worldwide and which is one of the most common complications documented in intensive care units (ICUs) [ 1 ]. About 48.9 million cases of sepsis and 11 million sepsis-related deaths were documented in 2017 worldwide [ 2 ]. Sepsis is an emergency condition leading to several life-threatening complications, such as septic shock and multiple organ dysfunction and failure [ 3 ]. Sepsis has negative physiological, psychological, and economic consequences. Untreated sepsis can lead to septic shock; multiple organ failure, such as acute renal failure [ 4 ]; respiratory distress syndrome [ 5 ]; cardiac arrhythmia (e.g. Atrial Fibrillation) [ 6 ]; and disseminated intravascular coagulation (DIC) [ 7 ]. Also, sepsis is associated with anxiety, depression, and post-traumatic stress disorder [ 8 ]. As for the financial burden of sepsis on the healthcare system, the cost of healthcare services and supplies for ICU critical care patients with sepsis is high [ 1 ]. In 2017, the estimated annual cost of sepsis in the United States (US) was over $24 billion [ 2 ].

Previous studies have shown that among nurses, misunderstanding and misinterpretation of the early clinical manifestations of sepsis, poor knowledge, attitudes, and practices related to sepsis, and inadequate training might lead to delayed assessment and management of sepsis [ 9 – 11 ]. Moreover, the limited numbers of specific and sensitive assessment tools and standard protocols for the early identification and assessment of sepsis in critical care patients leads to delayed management, therefore increasing sepsis-related mortality rates [ 10 ].

Critical care nurses, as frontline providers of patient care, play a vital role in the decision-making process for the early identification and prompt management of sepsis [ 11 ]. Therefore, improving nurses’ knowledge, attitudes, and practices related to the early identification and management of sepsis is associated with improved patient outcomes [ 12 , 13 ]. To date, there remains a wide gap between the findings of previous research and sepsis-related clinical practice in critical care units (CCUs). Furthermore, there is no evidence in the nursing literature regarding nurses’ knowledge, attitudes, and practices related to the early identification and management of sepsis in adult critical care patients and the association of these factors with patient health outcomes. Therefore, summarizing and synthesizing the existing research on sepsis assessment and management among adult critical care patients is needed to guide future directions of sepsis-related clinical practice and research. Accordingly, this review aims to identify nurses’ knowledge, and attitudes, practices related to the early identification and management of sepsis in adult critical care patients.

Materials and methods

The present review used a systematic review design guided by structured questions constructed after reviewing the nursing literature relevant to sepsis assessment and management in adult critical care patients. The authors (MR, DB, AH) carefully reviewed and evaluated the selected articles and synthesized and analyzed their findings to reach a consensus. This review was guided by the following questions: (a) what are nurses’ knowledge, attitudes, and practices related to sepsis assessment and management in adult critical care patients?, (b) what are the perceived facilitators of and barriers to the early identification and effective management of sepsis in adult critical care units?, and (c) what are the interventions directed at improving nurses’ sepsis assessment and management?

Eligibility criteria

The review questions were developed according to the PICOS (Participants, Interventions, Comparisons, Outcome, and Study Design) framework, as displayed in Table 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0270711.t001

Inclusion criteria.

The articles were retrieved and assessed independently by two researchers (MR, DB) according to the following inclusion criteria: (1) being written in English, (2) having an abstract and reference list, (3) having been published during the past 10 years, (4) focusing on critical care nurses as a target population, (5) examining knowledge, attitudes, and practices related to the assessment and management of sepsis, and (6) having been conducted in adult critical care units.

Exclusion criteria.

Studies were excluded if they were (1) written in languages other than English, and (2) conducted in pediatric critical care units or non-ICU. Dissertations, reports, reviews, editorials, and brief communications were also excluded.

Search strategy.

An electronic search of the databases CINAHL, MEDLINE/PubMed, EBSCO, Embase, Cochrane, Scopus, Web of Science, and Google Scholar was conducted using combinations of the following keywords: critical care, intensive care, critically ill, critical illness, knowledge, awareness, perception, understanding, attitudes, opinion, beliefs, thoughts, views, practice, skills, strategies, approaches, barriers, obstacles, challenges, difficulties, issues, problems, limitations, facilitators, motivators, enablers, sepsis, septic, septic shock, and septicemia. The search terms used in this review were described in S1 File . The search was initially conducted in March 2021, and a search re-run was conducted in April 2022. The search was conducted in the selected databases from inception to 4/2022. The initial search, using the keywords independently, resulted in 1579 articles, and after using the keyword combinations, this number was reduced to 241 articles. Then, after applying the inclusion and exclusion criteria, the number of articles was reduced to 92. A manual search of the reference lists of the 92 articles was carried out to identify any relevant publications not identified through the search. The researcher (MR) used the function “cited by” on Google Scholar to explore these publications in more depth. The researchers (MR, DB) then screened the identified citations of these publications, applying the eligibility criteria. In case of discrepancies, the researchers (MR, DB) discussed their conflicting points of view until a consensus was reached. Then, after careful reading of the article abstracts, 61 irrelevant articles were excluded, and a total of 31 articles were included in this review. Fig 1 below shows the Preferred Reporting Items for Meta-Analysis (PRISMA) checklist and flow chart used as a method of screening and selecting the eligible studies.

https://doi.org/10.1371/journal.pone.0270711.g001

Data extraction

The following data were extracted from each of the selected studies: (1) the general features of the article, including the authors and publication year; (2) the characteristics of the study setting (e.g., single vs. multisite); (3) the sociodemographic and clinical characteristics of the target population, including mean age, and medical diagnosis (e.g., sepsis, septic shock, and SIRS); (4) the name of the sepsis protocol used, if any; (5) the characteristics of the study methodology (e.g., sample size and measurements); (7) the main significant findings of the study; and (8) the study strengths and limitations. All extracted data were summarized in an evidence-based table ( Table 2 ). Data extraction was performed by two researchers (MR, DB). An expert third researcher (AH) was consulted to reach a consensus between the two researchers throughout the process of data extraction.

https://doi.org/10.1371/journal.pone.0270711.t002

Ethical considerations

There was no need to obtain ethical approval to conduct this systematic review since no human subjects were involved.

Quality assessment and data synthesis

A quality assessment of the selected studies was performed independently by two researchers based on the guidelines of Melnyk and Fineout-Overholt [ 14 ]. Disagreements between the two researchers (MR, DB) were identified and resolved through a detailed discussion held during a face-to-face meeting. For complicated cases, the researchers (MR, DB) requested a second opinion from a third researcher (AH). According to the guidelines of Melnyk and Fineout-Overholt [ 14 ], twelve of the studies were at level 3 in terms of quality, four studies at level 5, and nine studies at level 6.

A qualitative synthesis was performed to synthesize the findings of the reviewed studies. The following steps were applied throughout the process of data synthesis:

• The data in the selected studies were assessed, evaluated, contrasted, compared, and summarized in a table ( Table 2 ). This data included the design, purpose, sample, main findings, strengths/limitations, and level of evidence for each of the studies.
• The similarities and differences between the main findings of the selected studies were highlighted.
• The strengths and limitations of the reviewed studies were discussed.

Description of the selected studies

Most of the reviewed studies were conducted in Western countries [ 9 , 11 , 12 ], with only one study conducted in Eastern countries [ 1 ], and two in Middle-Eastern countries [ 15 , 16 ]. The detailed geographical distribution of the studies and other characteristics are described in Table 2 .

Nurses’ knowledge, attitudes, and practices

Nine of the selected studies assessed nurses’ knowledge and attitudes related to sepsis assessment and management in critically ill adult patients [ 1 , 9 , 12 , 15 , 17 – 21 ] ( Table 3 ) . Nucera et al. [ 18 ] found that ICU nurses had poor attitudes towards blood culture collection techniques and timing and poor levels of knowledge related to the early identification, diagnosis, and management of sepsis. For example, the majority of nurses reported that there is no need to sterilize the tops of culture bottles, and there is no specific time for specimen collection [ 18 ]. However, the participating nurses reported good levels of knowledge related to blood culture procedures and the risk factors for sepsis. Similarly, R. J. Roberts et al. [ 19 ] found the participating nurses to have good knowledge of septic shock and good attitudes toward the initiation of antibiotics for critically ill adult patients with sepsis. Only two studies assessed nurses’ practices related to sepsis assessment and management [ 15 , 19 ]. For example, in the study of R. J. Roberts et al. [ 19 ], 40% of the nurse participants reported that they were aware of the importance of initiating antibiotics and IV fluid within one hour of septic shock recognition [ 20 ]. Also, Yousefi et al. [ 15 ] found the participating nurses to have good practices related to sepsis assessment and management.

https://doi.org/10.1371/journal.pone.0270711.t003

Barriers to and facilitators of sepsis assessment and management

The reviewed studies identified three types of barriers to the early identification and management of sepsis, namely patient-, nurse-, and system-related barriers ( Table 4 ). Meanwhile, only nurse- and system-related facilitators were reported in the reviewed studies. The most-reported barriers and facilitators were system-related. The reported barriers included (a) the lack of written sepsis treatment protocols or guidelines adopted as hospital policy [ 22 , 23 ]; (b) the complexity and atypical presentation of the early symptoms of sepsis [ 19 ]; (c) nurses’ poor level of education and clinical experience [ 1 , 12 ]; (d) the lack of sepsis educational programs or training workshops for nurses [ 22 , 23 ]; (e) the high comorbid burden among patients with sepsis, which complicates the critical thinking process of sepsis management [ 19 ]; (f) nurses’ deficits in knowledge related to sepsis treatment protocols and guidelines [ 22 – 24 ]; (g) the lack of mentorship programs in which junior nurses’ actions/activities are strictly supervised by experienced nurses [ 17 , 23 ]; (h) heavy workloads or high patient-nurse ratios [ 22 ]; (i) the shortage of well-trained and experienced physicians, particularly in EDs [ 19 , 22 , 23 ]; (j) the lack of awareness related to antibiotic use for patients with sepsis [ 19 , 22 ]; (k) the lack of IV access and unavailability of ICU beds [ 25 ]; (l) the non-use of drug combinations for the treatment of sepsis [ 22 , 26 , 27 ], and (m) poor teamwork and communication skills among healthcare professionals [ 22 , 26 ]. Only three facilitators of sepsis assessment and management were identified in the reviewed studies. These facilitators were (1) nurses’ improved confidence in caring for patients with sepsis, (2) increased consistency in sepsis treatment, and (3) positive enforcement of successful stories of sepsis management [ 22 , 27 ].

https://doi.org/10.1371/journal.pone.0270711.t004

Measurement tools of sepsis-related knowledge, attitudes, and practices

One of the reviewed studies used a Knowledge, Attitudes, and Practice (KAP) questionnaire developed according to the Surviving Sepsis Campaign (SSC) guidelines [ 15 ] to measure nurses’ knowledge, attitudes, and practices related to sepsis assessment and management. Meanwhile, eight studies [ 1 , 9 , 12 , 17 – 21 ] used self-developed questionnaires based on the literature and SSC guidelines and validated by expert panels. Details of these measurement tools and their psychometric properties are summarized in Table 5 .

https://doi.org/10.1371/journal.pone.0270711.t005

Interventions directed at improving nurses’ sepsis assessment and management

Educational programs..

Only four of the selected studies examined the impact of educational programs on nurses’ knowledge, attitudes, and practices related to sepsis management and found significant improvements in nurses’ posttest scores ( Table 6 ) [ 11 , 15 , 28 , 29 ]. For example, Drahnak’s study [ 28 ] implemented an educational program developed by the authors and integrated with patients’ health electronic records (HER) and found significant improvements in nurses’ post-test nursing knowledge scores. Another educational program developed by the authors was implemented to improve ICU nurses’ knowledge, attitudes, and practices related to sepsis and found a significant improvement in posttest scores among the intervention group [ 15 ]. Another study was designed to examine the effectiveness of the Taming Sepsis Educational Program® (TSEP™) in improving nurses’ knowledge of sepsis [ 11 ]. A 15-minute structured educational session was developed to decrease the mean time needed to order a sepsis order set for critically ill patients through improving ER nurses’ knowledge about SSC guidelines and found that the mean time was reduced by 33 minutes among the intervention group [ 29 ].

https://doi.org/10.1371/journal.pone.0270711.t006

Simulation.

Only two studies examined the effect of using simulation in improving the early recognition and prompt treatment of sepsis by critical care nurses ( Table 6 ) [ 30 , 31 ]. Vanderzwan et al. [ 30 ] assessed the effect of a medium-fidelity simulation incorporated into a multimodel nursing pedagogy on nurses’ knowledge of sepsis and showed significant improvements in six of the nine questionnaire items. While Giuliano et al. examined the difference in mean times required for sepsis recognition and treatment initiation between nurses exposed to two different monitor displays in response to simulated case scenarios of sepsis and showed a significant reduction in the mean times required for sepsis recognition and treatment initiation by those nurses who were exposed to enhanced bedside monitor (EBM) display [ 31 ].

Decision support tools.

Four of the selected studies examined the effectiveness of decision support tools, adapted based on the SSC guidelines and the “sepsis alert protocol”, on the early identification and management of sepsis and confirmed the effectiveness of these tools ( Table 7 ) [ 32 – 35 ]. The decision support tools used in three of the studies guided the nurses throughout their decision-making processes to reach effective assessment, high quality and timely management of sepsis, and, in turn, optimal patient outcomes [ 32 , 33 , 35 ]. However, no significant differences in the time of blood culture collection and antibiotic administration were reported between the intervention and control groups in the study of Delawder et al. [ 34 ].

https://doi.org/10.1371/journal.pone.0270711.t007

Sepsis protocols.

Eight of the selected studies examined the effectiveness of sepsis protocols [ 24 , 36 – 38 ] and sepsis screening tools [ 16 , 39 – 41 ] for the early assessment and management of sepsis ( Table 7 ). All of these articles revealed that the implementation of sepsis screening tools or protocols based on the SSC guidelines leads to the early identification and timely management of sepsis, as well as the improvement in nurses’ compliance to the SSC guidelines for the detection and management of sepsis. For example, in one study, patients who received Early Goal-Directed Therapy (EGDT) had a lower mortality rate as compared to patients who received usual care [ 16 ]. The sepsis screening tools and guidelines were also tested to examine their impact on some patient outcomes, and variabilities were identified. For example, the use of the Modified Early Warning Score (MEW-S) tool revealed no significant improvement in patient mortality rate [ 41 ]. In contrast, mortality rates were decreased by using the Nurse Driven Sepsis Protocol (NDS) [ 40 ], Quality Improvement (QI) initiative [ 38 ], and a computerized protocol [ 37 ]. In addition, nurses in the computerized protocol group had better compliance with the SSC guidelines than did nurses in the paper-based group [ 37 ]. One of the selected studies compared between a paper-based sepsis protocol and a computer-based protocol and found that antibiotic administration, blood cultures, and lactate level checks were conducted more often and sooner by nurses in the computerized protocol group [ 37 ]. Two of the selected studies used the EGDT as a screening tool for sepsis and found no significant differences in times of diagnosis, blood culture collection, or lactate measurements between the control and intervention groups [ 16 , 24 ]. However, significant differences were found in the time of antibiotic administration in the study of Oliver et al. [ 24 ]. Although El-khuri et al. [ 16 ] revealed no significant differences in the time of antibiotic administration, the mortality rate among patients in the intervention group declined significantly.

Most of the reviewed studies focused on assessing critical care nurses’ knowledge, attitudes, and practices related to sepsis assessment and management, revealing poor levels of knowledge, moderate attitude levels, and good practices. Also, this review revealed that the three most common barriers to effective sepsis assessment and management were nursing staff shortages, delayed initiation of antibiotics, and poor teamwork skills. Meanwhile, the three most common facilitators of sepsis assessment and management were the presence of standard sepsis management protocols, professional training and staff development, and positive enforcement of successful stories of sepsis treatment. Moreover, this review reported on a wide variety of interventions directed at improving sepsis management among nurses, including educational sessions, simulations, screening or decision support tools, and intervention protocols. The impacts of these interventions on patient outcomes were also explored.

The findings of our review are consistent with the findings of previous studies which have explored critical care nurses’ knowledge related to sepsis assessment and management [ 42 ]. Also, recent studies conducted in different clinical settings support the findings of our review regarding nurses’ knowledge of sepsis. For example, a recent study conducted in a medical-surgical unit revealed that nurses had good knowledge of early sepsis identification in non-ICU adult patients [ 43 ]. The variations in nurses’ levels of knowledge related to sepsis assessment were attributed to variations in educational level and work environment (i.e., ICU vs. non-ICU).

The evidence indicates that the successful treatment of critically ill patients with suspected or actual sepsis requires early identification or assessment [ 44 , 45 ]. Early assessment is a critical step for the initiation of antibiotics for patients with sepsis, leading to improved patient outcomes and a decline in mortality rates [ 44 ]. The current review also revealed the significant role of educational programs in improving nurses’ knowledge, attitudes, and practices related to the early recognition and management of sepsis. These findings are in line with the findings of another study, which tested the impact of e-learning educational modules on pediatric nurses’ retention of knowledge about sepsis [ 45 ]. The study revealed that the educational modules improved the nurses’ knowledge acquisition and retention and clinical performance related to sepsis management [ 45 ]. The findings of our review related to sepsis screening and decision support tools are in congruence with the findings of a previous clinical trial which assessed the impact of a prompt telephone call from a microbiologist upon a positive blood culture test on sepsis management [ 46 ]. The study revealed that this screening tool contributed to the prompt diagnosis of sepsis and antibiotic administration, improved patient outcomes, and reduced healthcare costs [ 46 ]. The findings of our review related to the effectiveness of educational programs in improving the assessment and management of sepsis were consistent with the findings of a recent quasi-experimental study. The study found that incorporating sepsis-related case scenarios in ongoing educational and professional training programs improved nurses’ self-efficacy and led to a prompt and accurate assessment of sepsis [ 47 ]. One of the interventions explored in this review was a simulation that facilitated decision-making related to sepsis management. The simulation was found to be effective in mimicking the real stories of patients with sepsis and proved to be a safe learning environment for inexperienced nurses before encountering real patients, increasing nurses’ competency, self-confidence, and critical thinking skills [ 48 ]. Also, a recent study showed that the combination of different interventions aimed at targeting sepsis assessment and management, including educational programs and simulation, may lead to optimal nurse and patient outcomes [ 49 ].

Limitations

The present review has several limitations. There is limited variability in the findings of the reviewed studies in terms of the main variable, sepsis. Moreover, the review excluded studies written in languages other than English and conducted among populations other than critical care nurses. However, there may be studies written in other languages which may have significant findings not considered in this review. Further, only eight databases were used to search for articles related to the topic of interest, which may have limited the number of retrieved studies. Finally, due to the heterogeneity between the selected studies, a meta-analysis was not performed.

Relevance to clinical practice

Our findings could help hospital managers in developing continuous education and staff development training programs on assessing and managing sepsis for critical care patients. Establishing continuous education, workshops, professional developmental lectures focusing on sepsis assessment and management for critical care nurses, as well as training courses on how to use evidence-based sepsis protocol and decision support and screening tools for sepsis, especially for critical care patients are highly recommended. Also, our findings could be used to development of an evidence-based standard sepsis management protocol tailored to the unmet healthcare need of patients with sepsis.

To date, nurses remain to have poor to good knowledge of and attitudes towards sepsis and report many barriers related to the early recognition and management of sepsis in adult critically ill patients. The most-reported barriers were system-related, pertaining to the implementation of evidence-based sepsis treatment protocols or guidelines. Our review indicated that despite all educational interventions, no study has collectively targeted nurses’ knowledge, attitudes, and practices related to the assessment and treatment of sepsis using a multicomponent interactive teaching method. Such a method would aim to guide nurses’ decision-making and critical thinking step by step until a prompt and effective treatment of sepsis is delivered. Also, despite all available protocols and guidelines, no study has used a multicomponent intervention to improve health outcomes in adult critically ill patients. Future research should focus on sepsis-related nurse and patient outcomes using a multilevel approach, which may include the provision of ongoing education and professional training for nurses and the implementation of a multidisciplinary sepsis treatment protocol.

Supporting information

S1 checklist. prisma 2020 checklist..

https://doi.org/10.1371/journal.pone.0270711.s001

S1 File. Search strategies.

https://doi.org/10.1371/journal.pone.0270711.s002

Acknowledgments

The authors want to thank the Liberian of Jordan University of Science and Technology for his help in conducting this review.

• View Article
• PubMed/NCBI
• Google Scholar
• 14. Melnyk BM, Fineout-Overholt E, editors. Evidence-based practice in nursing & healthcare: A guide to best practice. Lippincott Williams & Wilkins; 2011.

Your Browser is Out of Date

To provide you with the best experience, our websites uses features that are not supported by your current browser. For the best experience on Childrenshospitals.org, please open this site in a different browser such as Google Chrome

Pediatric Sepsis: Nathan's Story

In November 2012, 17-year-old Nathan came home from school and told his parents he wasn’t feeling well. He was still recovering from a virus he had the week before that caused him to miss four days of school. With a history of respiratory infections, Nathan’s mother, Sherry, was vigilant about monitoring his symptoms. She called the pediatrician who said that Nathan should come in in the morning if he wasn’t feeling better.

By 4 a.m., Nathan was struggling to breathe and experiencing severe pain in his back and chest. His parents rushed him to the nearest emergency room, and they transported him to Nationwide Children’s Hospital in an ambulance. Sherry was terrified—she didn’t know what was wrong with her son. But with at least 10 doctors and nurses taking care of Nathan at once, she knew they were doing everything they could.

That first day, the family had a meeting with the lead physician. They had just met, but he gave them hugs at the end of the meeting, giving a small measure of comfort to Sherry and her husband. Nathan’s white blood cell count was through the roof. While doctors waited for the cultures to grow, Nathan was intubated and put on a ventilator. Sherry recalled, "I’ve never seen so many machines and tubes coming out of a person."

Each morning, when the doctors and nurses conducted their rounds, they would invite Sherry to join the discussion about Nathan’s care. Being part of the daily rounds, Sherry said, made a huge difference. "Being right there, as a part of the conversation, made it easier to understand. Not once did we feel like we were in the way. They never asked us to leave the room, and we were included in decision-making. It gave me a sense of control."

Then, Nathan was diagnosed with sepsis. The doctors told Sherry he had an infection running through his blood system that could shut down his organs. "I had heard the word, but had no clue what sepsis was,” Sherry said. “I asked the doctors if he could die, and they told me it was a possibility."

But after three and a half days in the ICU, Nathan began to recover and was transferred to another unit. He was diagnosed with double pneumonia in both lungs, but went home after a week in the hospital.

Sherry wasn’t aware that lung infections are one of the top four infections most often associated with sepsis, or that children with compromised immune systems are more susceptible. Since she wasn’t familiar with sepsis, she didn’t know the signs or to ask the doctors if that’s what he might have had.

"He just went downhill," Sherry said. "We were surprised at how fast it happened. If I had taken him to the emergency room even a few hours earlier, things might not have been as severe."

Four years later, Nathan is healthy. His family takes preventive measures if he experiences respiratory symptoms, and they know to trust their instincts.

Pediatric experts estimate that about 55 percent of patients develop sepsis before they get to the hospital. Even for pediatric clinicians, it can be difficult to recognize in children, as symptoms vary and awareness of the condition is low.

Sherry is grateful for the care Nathan received at Nationwide Children’s. "We’re lucky Nathan was taken to Children’s, otherwise I don’t think he would have survived."

Sepsis is a leading cause of death in hospitalized children, killing almost 5,000 children annually in the U.S.

Sharing these sepsis patient stories is part of the Improving Pediatric Sepsis Outcomes collaborative, a multi-year quality initiative to significantly reduce sepsis-related mortality and morbidity across children’s hospitals.

For more information, connect with us.

About Sepsis

Sepsis is a leading cause of death in hospitalized children, and children’s hospitals are dedicated to improving outcomes through collaboration, early identification and timely treatment.

Related Content

How to Launch a Multi-Year, Multi-Site Quality Improvement Collaborative

Bringing People and Data Together

Defining, Developing and Understanding Pediatric Sepsis Outcomes

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 09 September 2019

Critical Transitions in Intensive Care Units: A Sepsis Case Study

• Pejman F. Ghalati 1   na1 ,
• Satya S. Samal 1   na1   nAff3 ,
• Jayesh S. Bhat 1 ,
• Robert Deisz 2 ,
• Gernot Marx 2 &
• Andreas Schuppert 1  

Scientific Reports volume  9 , Article number:  12888 ( 2019 ) Cite this article

5093 Accesses

12 Citations

3 Altmetric

Metrics details

• Computational science
• Health care

The progression of complex human diseases is associated with critical transitions across dynamical regimes. These transitions often spawn early-warning signals and provide insights into the underlying disease-driving mechanisms. In this paper, we propose a computational method based on surprise loss (SL) to discover data-driven indicators of such transitions in a multivariate time series dataset of septic shock and non-sepsis patient cohorts (MIMIC-III database). The core idea of SL is to train a mathematical model on time series in an unsupervised fashion and to quantify the deterioration of the model’s forecast (out-of-sample) performance relative to its past (in-sample) performance. Considering the highest value of the moving average of SL as a critical transition, our retrospective analysis revealed that critical transitions occurred at a median of over 35 hours before the onset of septic shock, which suggests the applicability of our method as an early-warning indicator. Furthermore, we show that clinical variables at critical-transition regions are significantly different between septic shock and non-sepsis cohorts. Therefore, our paper contributes a critical-transition-based data-sampling strategy that can be utilized for further analysis, such as patient classification. Moreover, our method outperformed other indicators of critical transition in complex systems, such as temporal autocorrelation and variance.

Similar content being viewed by others

Discovery of signatures of fatal neonatal illness in vital signs using highly comparative time-series analysis

Justin C. Niestroy, J. Randall Moorman, … Douglas E. Lake

Data-driven discovery of a novel sepsis pre-shock state predicts impending septic shock in the ICU

Ran Liu, Joseph L. Greenstein, … Raimond L. Winslow

Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation

Jacob Deasy, Pietro Liò & Ari Ercole

Introduction

Certain biological systems exhibit nonlinear dynamics that undergo sudden regime transitions at tipping points 1 , 2 . In a medical context, these transitions often indicate changes in clinical phenotypes, e.g., disease-onset 3 . Such phenomena have been studied mathematically with techniques from the application of singularity theory to dynamical systems 4 , 5 , 6 . In addition, data-driven methods use statistical indicators known as early-warning signals to model the dynamics of systems approaching transitions 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 . Modeling such transitions is beneficial for several applications in systems medicine, such as monitoring health 15 , 16 , predicting disease-onset and gaining an improved understanding of the underlying disease progression 17 .

Our focus is on sepsis, a common complication in the intensive care unit (ICU), and we introduce a notion of regime transition in septic dynamics. As stated in the Third International Consensus Definitions of Sepsis and Septic Shock (Sepsis-3), “sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection”, and “septic shock is a subset of sepsis in which underlying circulatory and cellular/metabolic abnormalities are profound enough to substantially increase mortality 18 ”. Sepsis causes a high rate of in-hospital mortality and costs the healthcare sector billions due to rising incidence rates and prolonged hospital stays 19 , 20 . Accurate diagnosis, however, remains a challenging task for physicians due to the heterogeneity of infectious agents and the frequent presence of multiple comorbidities. Early, aggressive administration of antibiotics is crucial, and delays in this treatment significantly increase mortality 21 , 22 .

To detect signs of sepsis early, numerous illness severity scores or early-warning signals exist: the Acute Physiology and Chronic Health Evaluation (APACHE II), the Simplified Acute Physiology Score (SAPS II), the Sepsis-related Organ Failure Assessment Score (SOFA), the Modified Early Warning Score (MEWS), and the Simple Clinical Score 23 . These scores are good predictors of general disease severity and mortality but cannot estimate the risk of developing sepsis with reasonable sensitivity and specificity 23 .

Numerous machine learning (ML) methods were therefore developed to predict sepsis onset 24 , 25 , 26 . Rothman et al . 27 used structured information from electronic health records (EHRs) to identify sepsis on admission or to predict its onset during hospitalization. For septic shock prediction, Ghosh et al . 28 proposed an integrative model combining sequential contrast patterns with coupled hidden Markov models. Henry et al . 23 developed a targeted real-time early-warning score (TREWScore) by training a Cox regression model to identify patients at high risk of developing septic shock. Additionally, Horng et al . 29 argued that combining free-text patient data with other predictor features significantly improved the performance of ML models. Although these ML approaches have the potential to increase diagnostic accuracy, they involve time-consuming and domain-specific variable/feature selection 30 , 31 . Our proposed method can be considered in the preprocessing stages to select appropriate data for further downstream analysis.

Our computational method aims to identify and characterize signals indicative of critical transitions based on the concept of surprise loss (SL) 32 . SL was originally developed in econometrics to assess forecast breakdown, i.e., instability in the model’s forecasting ability. Such instability was attributed to instability in the underlying data-generating process, whose effects have been studied from a mathematical perspective 33 , 34 . We assume that similar instability occurs in patient data because of changes in the underlying biological mechanism due to medical intervention or disease progression.

We utilize SL to identify regions in the time series where the data-generating process changes and quantify them with a numerical score. The score captures the extent of deviation between the past performance of a model and its future performance. We consider the highest value of such a score to be a putative tipping point in the disease dynamics, and we consider it as a surrogate for critical transition. In addition, we present a critical-transition-based data-sampling strategy is also presented where data are sampled at regions around critical transition; this strategy outperforms random sampling in differentiation between septic shock and non-sepsis patients. We also compare our approach to methods based on autocorrelation and variance 7 , 15 , 16 , 35 , which have been used to identify early-warning signals of critical transitions.

Materials and Methods

Data source.

We sourced patients’ multivariate time series data from the publicly available EHR database, Medical Information Mart for Intensive Care MIMIC-III v1.4 36 , which contained longitudinal data of 46,520 deidentified patients from 58,976 distinct ICU admissions. For ease of interpretation, we treated each admission as a distinct patient.

In the ICU, clinical staff make swift decisions or take prompt actions during patient management. These employees prioritize timely and correct treatment over consistent documentation of their processes, thereby limiting the reliability of clinical annotation for retrospective analysis. Furthermore, the execution of guidelines for identifying imminent disease varies across hospitals. Hence, we restricted our data analysis to predominantly machine-recorded quantitative variables.

Decision rules for retrospective annotation of the sepsis syndrome have evolved over the decades as knowledge of its pathophysiology and epidemiological impact have increased 37 . Whereas earlier definitions (1991 38 , 2001 39 ) focused on uncontrolled systemic inflammation as the major indicator, the latest 2016 18 definition, commonly known as Sepsis-3, emphasizes organ dysfunction as the leading effect of the sepsis syndrome and proposes to update the International Classification of Diseases (ICD) coding system 40 , 41 (ICD-9: 995.92, 785.52; ICD-10: R65.20, R65.21). SOFA scoring system grades the extent of organ dysfunction and is calculated every 24 hours during a patient’s ICU stay 42 , 43 .

Because the ICD-9 codes in our data were not compatible with Sepsis-3, we annotated the patient data in accordance with Table  2 from the 2016 consensus definition 18 . Fig.  1 illustrates a general schematic of our annotation framework.

Over the length of a patient’s ICU stay, all timestamps of body fluid (blood, urine, cerebrospinal fluid) sampling and antibiotic administration were retrieved. For each of the timestamps, an infection was suspected if antibiotics were administered within 72 hours of any prior body fluid sampling (irrespective of culture findings) or if any body fluids were sampled within 24 hours of prior antibiotic administration. Sepsis-3 criteria were independently evaluated over time windows around the infection-suspected timestamps (IST). Each time window began 48 hours prior to IST until 24 hours post IST. If the criteria were satisfied during a given time window, then the beginning of the window was annotated as the onset time. In the schematic, the 2 nd antibiotics administration falls within 72 hours of previous body fluid sampling; thus, an infection is suspected.

The annotation framework was applied to all 58,976 patients, identifying 22,547 (38.2%) sepsis patients and 3208 (5.4%) septic shock patients. Among the 3208 septic shock patients, we analyzed only adults (18+ years old at admission) with at least a 36-hours stay and at most 144 hours spent in the unit before onset, which generated a cohort of 630 patients. Our non-sepsis cohort comprised 6,236 patients who lacked Sepsis-3 annotation or sepsis-specific ICD-9 codes and who stayed between 36 and 144 hours in the ICU. Demographic information on the two cohorts can be found in Supplementary Table  S1 .

We cannot exhaustively evaluate and validate the accuracy of our annotation framework owing to the absence of a manually curated “ground truth” dataset of Sepsis-3 patients. Software implementations with different data cleaning processes and patient exclusion criteria (PEC) from the same annotation framework could result in divergent cohorts. For example, for the same database, another implementation 44 annotated almost half (49.1%) of their analysis cohort (n = 11,791; reasonable PEC) as Sepsis-3, whereas our implementation annotated approximately 38% of the entire population (n = 58,976; no PEC). There may be a high degree of overlap in the annotated cohorts; thus, a comparison of the two implementations is currently under way.

Based on availability and relevance to sepsis, we preselected groups of variables: the laboratory variables included bicarbonate, creatinine, blood urea nitrogen (BUN), hematocrit, hemoglobin, platelet count, white blood cell count (WBC), potassium, and sodium; the vital signs and physiological variables comprised body temperature, heart rate, respiratory rate, oxygen saturation (SpO2), arterial blood pressure (systolic, mean, and diastolic), and urine output; the two septic markers comprised the shock index (ratio of heart rate over systolic blood pressure), and the ratio of BUN to creatinine 23 . Table  1 shows the mean sampling rates of the variables in the respective patient cohorts, and their distribution can be seen in Supplementary Fig.  S1 .

Missing value imputation and time binning

Data representation is a crucial step in analyzing time series. Continuous EHRs suffer from missing values due to insufficient data collection and lack of documentation. Additionally, high heterogeneity in variable type and irregular sampling intervals make such data difficult to handle. To address the problems of missing data and data sparsity, we transformed our time series into 30-minute time bins by imputing values in the bins and averaging measurement values over the bins. We experimented with different imputation methods, such as linear, polynomial and Stineman interpolation 45 . The Stineman method was chosen due to its superior performance in reducing overshoots and handling sharp changes in the imputed values.

Data normalization

Our variables (Table  1 ) had different scales and measurement units. Data normalization was therefore needed for our method. For this purpose, we transformed the observables by Z-score normalization to address the use of different units of measurement.

State space model

To define SL, we require a dynamical mathematical model for our multivariate clinical time series. Here, we consider a state space model (SSM) approach 46 , 47 , which models the data in a hierarchical manner with hidden states that give rise to observables. In our context, the hidden states can be assumed to represent the biological processes, and the observables represent the clinically measured variables. The observables in our SSM are expressed as linear combinations of hidden random states. Such a model incorporates the variations in the biological processes and a measurement noise term. The variations due to biology are modeled by adding a stochastic term to the hidden states, whereas the measurement noise term is added to the observables. Both terms are assumed to follow a multivariate normal (MVN) distribution.

The computation of SL is agnostic to the underlying dynamical model. The SL literature 32 uses a linear dynamical model, whereas we use an SSM for our application. The primary reason to use this type of model is to separate the biological processes from the observables, i.e., to model two sources of variability. Below, we represent such an SSM model.

where the indices of the time series are from t  = 1, …, T ; e is the number of hidden trends; x is an e  ×  T matrix of hidden states; y is an n  ×  T matrix of n observables; and w is an e  ×  T matrix of process error. In general, \(e\ll n\) . The process error at time t follows an MVN distribution with mean 0 and e  ×  e covariance matrix Q ; v is an n  ×  T matrix of observation error. The observation error at time t follows an MVN distribution with mean 0 and n  ×  n covariance matrix R ; Z is an n  ×  e parameter matrix; a is a vector of offsets; π is a matrix of e  × 1 means; ∧ is an e  ×  e covariance matrix. The set of parameters can be represented in compact form as θ  = ( Q , R , Z , x 1 , …, T , π , ∧ ), and their estimate is \(\hat{\theta }\) . \({\hat{y}}_{t}\) and \({\tilde{y}}_{t+\lambda }\) are the estimate and λ -step-ahead forecast, respectively, of the given observables y t .

Our implementation incorporated MARSS 48 , 49 , which is an R package for fitting constrained and unconstrained linear multivariate autoregressive SSMs by maximum likelihood parameter estimation. We utilized MARSS to fit an SSM to our multivariate time series data, using its recommended initial conditions that ensure parameter identifiability. We assumed the presence of multiple hidden states and fixed e  = 3. Furthermore, we evaluated the robustness of our results with respect to the changes in the model parameters (see ‘Robustness of the SSM model’).

Perturbations in the dynamics

Early-warning indicator.

Our proposed computational method based on surprise loss (SL) 32 computes the difference between the forecast error, i.e., out-of-sample error, and the in-sample performance. The out-of-sample error measures the quality of model forecasts, i.e., the prediction of the model for the data that were not used for fitting, whereas the in-sample error quantifies the deviation between the model estimates and the data that were used for model fitting. A high out-of-sample error compared to the in-sample error is suggestive of instability in the patient data. In such a scheme, our model may be a poor fit for the data, but we are interested in evaluating whether the past performance of the model is consistent with future forecasts. The performance is measured for a fixed loss function using a moving time window. Furthermore, the SL computation is unsupervised, i.e., the clinical conditions of patients, such as septic shock or non-sepsis, are not required. Originally, the idea of SL was used to perform a statistical test to determine forecast breakdown in time series, i.e., to determine whether the average of SL is close to zero 32 . However, in our application, the aim is not to test whether a given time series underwent a statistically significant forecast breakdown; rather, it is to identify high SL values in the given time series and later use this information in postprocessing steps (see ‘Data-sampling strategy with SLMean’).

In spirit, this approach is close to the identification of structural breaks or change-points analysis 50 , 51 . However, the SL-based approach has the additional advantage of being robust to model misspecification. Specifically, in practice, the SSM model (i.e., the functional form and variables) is likely to be misspecified and may not be a good approximation of the underlying disease processes. By formalizing SL as the difference between in-sample and out-of-sample performance and not relying on model parameters or error variances, the SL-based approach provides a natural way to handle such scenarios (see ‘Relationship with the literature’ in Giacomini et al . 32 ).

With a moving time window of width m , the SSM model (see equation ( 1 )) was fitted for time indices \(t-m+\mathrm{1,}\ldots ,t\) . \({y}_{t}^{{i}_{c}}\) denotes the observables of a given patient i with clinical condition c at time index t , and \({T}^{{i}_{c}}\) is the length of the corresponding time series. The in-sample error is a quadratic loss function that averages the squared differences between the estimated and the given observables, and it is denoted as \({L}_{j}({\hat{\theta }}_{t}^{{i}_{c}})=\frac{1}{n}{\sum }_{k=1}^{n}\,{(y{(k)}_{j}^{{i}_{c}}-\hat{y}{(k)}_{j}^{{i}_{c}})}^{2}\) where \(y{(k)}_{j}^{{i}_{c}}\) is the k th element of column vector \({y}_{j}^{{i}_{c}}\) . Similarly, the out-of-sample error is a quadratic loss function that averages the squared differences between the λ -step-ahead forecast and the given observables, and it is denoted as \({L}_{t+\lambda }({\hat{\theta }}_{t}^{{i}_{c}})=\frac{1}{n}{\sum }_{k=1}^{n}\,{(y{(k)}_{t+\lambda }^{{i}_{c}}-\tilde{y}{(k)}_{t+\lambda }^{{i}_{c}})}^{2}\) . The SL is the difference between the out-of-sample and the in-sample error:

To remove short-term fluctuations, a moving-average filter (with size δ ) smooths the SL:

For a given patient i , prior to the clinically annotated onset of disease c , a relatively high \(SLMea{n}_{t}^{{i}_{c}}\) suggests putative transitions across dynamical regimes and serves as an early-warning indicator. We consider the maximum of \(SLMea{n}^{{i}_{c}}\) at time index \({t}_{max}^{{i}_{c}}\) to denote a critical transition. Fig.  2a illustrates the calculation of \(S{L}^{{i}_{c}}\) , \(SLMea{n}^{{i}_{c}}\) and \({t}_{max}^{{i}_{c}}\) . A simulated example using synthetic data is shown in Fig.  3 .

( a ) A schematic for the calculation of \(S{L}^{{i}_{c}}\) , \(SLMea{n}^{{i}_{c}}\) , and \({t}_{max}^{{i}_{c}}\) for a given patient i and clinical condition c . The SSM was fitted with a moving time window of length m (as shown in blue) and the \(S{L}^{{i}_{c}}\) was calculated. A second sliding window of length δ was used to compute the \(SLMea{n}^{{i}_{c}}\) (as illustrated in green). The \({T}^{{i}_{c}}\) denotes disease onset in septic shock patients and it represents the time of discharge or death in non-sepsis patients. The \({t}_{max}^{{i}_{c}}\) denotes the time index of the highest \(SLMea{n}^{{i}_{c}}\) and it was used in our data-sampling approach. ( b ) A schematic diagram illustrating our data-sampling strategy using our method. Observables at the time of highest SLMean magnitudes, i.e., critical transition points, were selected from septic shock and non-sepsis patients. The Wilcoxon rank-sum test was used to determine the statistical significance of the changes in the observables.

Artificial example showing the calculation of SLMean from a synthetic dataset that was generated by concatenating 50 points, drawn independently from three univariate normal distributions with different means (5, 10, 15) and a standard deviation of 0.5. Computed with a moving time window of length 30 and the number of hidden states set to 1, the magnitude of SL intensified at the 50 th and 100 th time-points, where the parameters of the data-generating process changed, i.e., a proxy for transitions across different dynamical regimes.

Uncertainty in SLMean

Uncertainty in out-of-sample forecasting and in-sample performance adds noise to the precise location of \({t}_{max}^{{i}_{c}}\) . Let \({t}_{max(up)}^{{i}_{c}}\) and \({t}_{max(low)}^{{i}_{c}}\) , respectively, be the time indices corresponding to the modes of the upper and lower bounds of the 95% prediction interval of SLMean . Our approach is robust if the deviations of \({t}_{max}^{{i}_{c}}\) from \({t}_{max(up)}^{{i}_{c}}\) and \({t}_{max(low)}^{{i}_{c}}\) are close to zero.

Data-sampling strategy with SLMean

Here, we demonstrate a method for sampling data from the critical transition points (derived from SLMean ) to differentiate the septic shock cohort from the non-sepsis cohort (see Fig.  2b ). We also propose a bootstrap test (based on a random sampling of data) to evaluate whether it outperforms the SL-based approach. Such a data selection step can be seen as a preprocessing step for the machine learning-based techniques being developed to study sepsis (as described in ‘Introduction’). The data sampling step is agnostic to the clinical condition of the patient, i.e., data for each patient are based on SL (see ‘Perturbations in the dynamics’), and in a subsequent step, we used the clinical condition to perform statistical tests.

Specifically, we selected the data at \({t}_{max}^{{i}_{c}}\) , i.e., the critical transition points (in the case of multiple \({t}_{max}^{{i}_{c}}\) values, the one closer to the disease-onset was selected), sampled the corresponding data and represented them as an n  ×  v variable matrix \({S}^{c}=[{y}_{{t}_{{\max }}}^{{1}_{c}},\ldots ,{y}_{{t}_{{\max }}}^{{v}_{c}}]\) where c   ∈  {0, 1} i.e., non-sepsis and septic shock conditions, and v is the total number of patients. Thereafter, for each variable, a p-value based on Wilcoxon rank-sum test 52 was calculated, quantifying the significance of differences between the two patient cohorts (as shown in the equation ( 4 )).

where pval (.) returns the p-value based on the Wilcoxon rank-sum test. \({S}_{j}^{0}\) and \({S}_{j}^{1}\) denote the j th row vectors of matrices S 0 and S 1 matrices, respectively. Furthermore, we performed the Benjamini and Hochberg correction method to adjust the p-values 53 accounting for multiple comparisons.

Bootstrapping

Furthermore, a bootstrap test was used to compare the p-values calculated at critical transition points from the p-values that were obtained from random points in our time series. For a randomly selected time index t with its corresponding observation \({y}_{t}^{{i}_{c}}\) , where \(t\in \mathrm{(1,}\,{T}^{{i}_{c}})\) , the t random p-values were calculated by replacing t max with t . The test was repeated 1000 times. Bootstrap frequency (BF) denotes the fraction of replications wherein t max p-values were less than t random p-values. A high BF value indicates that the SL based approach has an advantage over the random approach. In addition to computing the BF on data randomly sampled from all times, we computed BF on randomly sampled data of septic patients from two arbitrary time intervals, 36 hours and 18 hours before the onset of septic shock. This step allows us to test whether merely randomly sampling data close to the onset time can outperform the SL approach.

Autocorrelation and variance as early-warning signals

In the dynamics of a system, increased temporal autocorrelation and increased variance are hypothesized to be two indicators that the system is approaching a state transition 7 . To evaluate the SL concept, we calculated these two presumed early warning signals and compared the results with those obtained from the SL approach. As these measures are both univariate, to apply them to our multivariate time series data, we formulated them as follows:

where AC and AC1 are autocorrelation and variance functions applied on variable y ( k ) for time indices \(t-m+\mathrm{1,}\ldots ,t\) . t is the time index, and m is the width of a moving time window. The first coefficient of auto-correlation \(AC{1}_{t}^{{i}_{c}}\) and variance \(VA{R}_{t}^{{i}_{c}}\) were computed by averaging over N variables. i is the index of a given patient with clinical condition c , and \({T}^{{i}_{c}}\) is the length of the corresponding time series.

Similar to the SL concept, t max is defined as the time index where the highest value of the early-warning signal occurs (here, the largest value of \(AC{1}^{{i}_{c}}\) or \(VA{R}^{{i}_{c}}\) ). P-values and bootstrap frequencies were computed as described in ‘SLMean-based data-sampling strategy’ and ‘Data-sampling strategy with SLMean’.

To support reproducible research, our computational method is available at https://github.com/JRC-COMBINE/SL-MTS .

SLMean as an early-warning indicator

Over a moving time window ( m  = 36, i.e., 18 hours; e  = 3; λ -step-ahead = 1, i.e., 30 minutes; δ  = 6, i.e., 3 hours), the \(SLMea{n}^{{i}_{c}}\) values (‘Perturbations in the dynamics’), as shown in Fig.  4 , were computed. A positive \(SLMea{n}_{t}^{{i}_{c}}\) indicates higher out-of-sample error than in-sample error, signaling putative transitions in the underlying dynamics. The componentwise mean vector and associated standard deviation of all septic shock patients, i.e., \(SLMea{n}^{{1}_{c}},\ldots ,SLMea{n}^{{N}_{c}}\) (where N is the total number of septic shock patients and c is the septic shock clinical condition), intensified as the moving time window approached the disease onset. For the same cohort of septic shock patients, a slight increase in the componentwise mean vector and associated standard deviation of \(VA{R}^{{i}_{c}},\ldots ,VA{R}^{{N}_{c}}\) could be seen, while those of \(AC{1}^{{1}_{c}},\ldots ,AC{1}^{{N}_{c}}\) did not show any changes over time. The findings are summarized in Fig.  5 .

The changes over time in a group of clinical variables used in this study and the corresponding computed \(SLMea{n}^{{i}_{c}}\) of a sample septic patient before the onset of septic shock (violet line). The \(SLMea{n}^{{i}_{c}}\) is calculated over a moving time window ( m  = 36, i.e., 18 hours; e  = 3; λ -step-ahead = 1, i.e., 30 minutes; δ  = 6, i.e., 3 hours). The red line shows the time location ( \({t}_{max}^{{i}_{c}}\) ) of the largest \(SLMea{n}^{{i}_{c}}\) (i.e., the critical transition point).

Componentwise mean (red dots) and ±standard deviation (blue lines) of ( a ) SLMean ( b ) AC1 and ( c ) VAR for all septic shock patients prior to disease-onset (see ‘SLMean as an early-warning indicator’); T is the length of the time series (i.e. max( \({T}^{{1}_{c}},\ldots ,{T}^{{N}_{c}}\) ), where c  = 1 represents septic shock condition and N is the total number of septic shock patients), and t  −  T is the time before the onset of septic shock. The number of samples per time point could be different due to the heterogeneous length of hospitalization (see ‘Data source’). As the maximum length of hospitalization was 144 hours, with a moving time-window length of 18 hours and an average window of 3 hours, the minimum value of t  −  T was −123 hours.

It should be taken into account that the largest \(SLMea{n}_{t}^{{i}_{c}}\) need not necessarily occur exactly at the time of disease onset. For septic shock patients, the location of the time index t max from the onset time ( T ) is shown in Fig.  6b . In the majority of our patients’ data, the highest SLMean occurred near septic shock onset (60% of the patients, the signal occurred less than 48 hours prior to onset, as shown in Fig.  6b ). However, in some patients, the signal was observed beyond onset time. Possible explanations include a lack of records or a low sampling rate of variables a few days before the onset of septic shock, resulting in a nonsignificant SLMean . The highest SLMean , on average, occurred 46 hours (median of 35.6 hours) prior to the appearance of septic shock symptoms. In comparison, TREWScore 23 identified septic patients at a median of 28.2 hours before onset.

( a ) Componentwise mean of SLMean for all septic shock patients prior to disease-onset (see ‘SLMean as an early-warning indicator’); T is the length of the time series, ( b ) Distribution of the times of critical transitions from the onset times of septicshock, i.e. \({t}_{max}^{{1}_{c}}-{T}^{{1}_{c}},\ldots ,{t}_{max}^{{N}_{c}}-{T}^{{N}_{c}}\) , where c  = 1 represents septic shock condition and N is the total number of septic shock patients. \(SLMea{n}^{{i}_{c}}\) reaches a maximum at \({t}_{max}^{{i}_{c}}\) .

While the median time of the peak SLMean occurred at 35.6 hours before the onset of septic shock, visual inspection of the mean and standard deviation of SLMean indicates an upward trend starting from approximately 24 hours (Figs  5a and 6a ). The explanation for the apparent deviation from the baseline is that the highest \(SLMea{n}_{t}^{{i}_{c}}\) values that occurred closer to onset were greater in magnitude.

Furthermore, we determined the uncertainty in SL calculation using prediction intervals (as described in ‘Uncertainty in SLMean’). Our results show negligible deviation in t max i.e., the median deviation is 0, and the interquartile range (IQR) is 5.4 hours.

SLMean -based data-sampling strategy

We compared the p-values for data sampled at t max (i.e., critical transition point) to those obtained via random sampling (see equation ( 4 ) and ‘Data-sampling strategy with SLMean’). The same procedure was implemented for AC1 and VAR , and the bootstrap test was performed for all time indices. The bootstrap frequencies were denoted as BF ( SL ), BF ( AC1 ) and BF ( VAR ), respectively (see Table  2 ). The different BF computations test the association of the bootstrap frequency values of some variables with high SLMean , AC1 and VAR . In 14 out of 19 variables, BF ( SL ) demonstrates superior results. In the next step, in addition to all the time indices, the bootstrap test was performed for time-windows of 18 and 36 hours before the onset of septic shock; the bootstrap frequencies are represented as BF ( Full ), BF (18  hours ), and BF (36  hours ). Fig.  7a plots BF ( Full ) against p-values computed at t random and at high SLMean (i.e., t max ). Most of the variables show a good BF with high log-transformed p-values when sampled at large SLMean , particularly in the case of variables such as blood pressures, temperature and SpO2, where random sampling leads to poor p-values. As the random sampling strategy changed to either to 36 or 18 hours in Fig.  7b , BF reduced for six variables (WBC, diastolic blood pressure, Hemoglobin, SpO2, creatinine, and BUN), but it was preserved for nine variables (respiratory rate, heart rate, potassium, mean blood pressure, hematocrit, shock index, temperature, BUN-creatinine, and systolic blood pressure), i.e., the differences among BF ( Full ), BF (36  hours ), and BF (18  hours ) were small. Four variables, bicarbonate, urine output, platelets and sodium, had low BF ( Full ), BF (36  hours ), and BF (18  hours ).

( a ) A statistical significance test (see ‘Data-sampling strategy with SLMean’) was performed to test whether the values of the clinical variables at largest SLMean were able to differentiate septic shock patients from non-sepsis patients. The − log 10 ( P - value ) of each variable at t max was compared with the median − log 10 ( P - value ) of randomly selected points from the whole sequence. A bootstrap test (denoted as BF Full ) was performed to quantify the number of times the p-values at t max are lower than those at t random . ( b ) Another bootstrap test, performed by randomly sampling points from 18 hours and 36 hours windows prior to the onset of septic shock, tested whether low p-values at t max are an effect of time or a characteristic of regions with high SLMean values.

Robustness of the SSM model

We assessed the robustness of our method to perturbations in the model parameters. We changed the length of the moving time window, m   ∈  (24, 30), and the number of trends in the SSM model ( e   ∈  (4, 5)) and compared the changes in the t max with respect to the reference setting, i.e., m  = 36 and e  = 3. The chosen values of e are based on the assumptions described in ‘State space model’ ( e  = 3 and \(e\ll n\) ). The length of the moving time window was selected with regard to the average variables sampling rate (see Table  1 , as well as the length of hospitalization in the ICU (see ‘Data source’). The differences in t max due to the perturbations are summarized in online Supplementary Fig.  S2 . The zero median of such differences confirmed the robustness of our approach. Due to multiple similar high values SLMean in some patients, alteration of model parameters led to different t max values in these patients, which caused the outliers in online Supplementary Fig.  S2 .

Discussions

Healthcare can benefit from the analysis of continuously monitored health data, which are rapidly growing in quantity due to the increasing availability of long time series collected either by wearables or by monitoring systems such as those established in the ICU. However, significant challenges remain unresolved. A major drawback is the restriction of data availability to variables that are easy to collect by noninvasive sensors. These variables provide only correlated surrogates of the primary disease-driving processes. Hence, sensor signals are rarely specific on their own; advanced computational processing is typically necessary to identify relevant signals to improve therapy.

Focusing data analysis on the prediction and identification of critical transitions, i.e., instabilities in patient data, may complement established scoring methods in the classification of stable states. Although critical transitions differ qualitatively from scores in classifying stable states, the former method provides an independent assessment of health status. Because critical transitions are simply identified through the evolution of individual longitudinal time series, in contrast to established scores based on absolute variable values, markers for the detection of critical transitions are relatively robust to normalization and data standardization issues.

To identify such critical transitions in ICU patients, we applied the concept of surprise loss (SL), which was originally developed for determining instability in a model’s forecasting ability in econometrics. We changed the model in the original SL approach to a multivariate SSM model to model two sources of variability, namely, the hidden underlying biological processes and the observables. Despite a multitude of interventions in the ICU, our moving average SL, SLMean , showed, on average, an increasing signal approximately 24 hours before the expert-annotated onset of septic shock (see Fig.  6a ), thereby indicating its applicability as an early-warning indicator. We utilized such an indicator to devise a critical-transition-based data-sampling strategy for discriminating septic shock from non-sepsis patients. Additionally, through a bootstrap test (quantified through BF ( Full )), the benefit of our method is shown with respect to a random data selection strategy (as summarized in Table  2 and Fig.  7a ). Except for bicarbonate, urine output, platelets and sodium, the SL-based approach results in better p-values and BF ( Full ) than the random strategy. In addition, we selectively sampled random data from 36 hours and 18 hours before the septic onset to compute BF (36  h ) and BF (18  h ), respectively (see Fig.  7b and Supplementary Table  S2 ). Such selective sampling evaluates whether merely sampling data close to the onset time of septic shock outperforms our method in distinguishing sepsis from non-sepsis. These new BF values seem to be well-preserved for most variables that have correspondingly high BF ( Full ). Therefore, an SL-informed sampling strategy may improve the quality of patient classification and eventually enable the reduction of sample sizes.

Moreover, from a systems theory point of view, mechanisms that control the system in homeostasis begin to collapse around a critical transition or tipping point. Consequently, variables that are under tight control within stable states may be more sensitive to systemic variability around an unstable point. Our data analysis supports this hypothesis (see Fig.  7a ): some variables under tight control, e.g., blood pressure and body temperature, showed significant improvement in p-values compared to random sampling. We compared our method with two other univariate early-warning measures for critical transitions in complex systems: temporal autocorrelation and variance 7 , 15 , 16 , 35 . As shown in Fig.  5 , our method outperformed these estimators as an early-warning indicator for septic shock patients. Similarly, the p-values and BF of our method were also more favorable than those of the other methods (Table  2 ).

Conceptually, SL computation is based on the premise that the underlying system has a stable stationary state and that all observed deviations can be explained as responses to stochastic perturbations. The permissible amount of deviation is controlled by the system’s robustness at the time of computation. As a result, SL-based analysis can forewarn of a “loss of stability” even before the underlying system has changed its state. In that sense, SL provides indicators similar to those from the analysis of critical slowing down 35 . One drawback is that local loss of robustness may neither result in a transition to another state nor indicate a new state. SL-based warning systems, in isolation, may thus lead to false alarms and could be improved by combining them with ML classifiers. Additionally, moving-window length restricts the capability of the SL-based warning system, and analysis can only be performed only when sufficient data have been collected. Hence, further evaluations must be performed towards utilization of SL-based analysis in practice. As a high SL is not specific and can be generated by any sudden event affecting the data, either errors in the monitoring system or health-related covariates, a robust characterization of the standard SL patterns characterizing control states is crucial. As sudden, high SL peaks can arise from sudden monitoring aberrations, we expect that a threshold-based alarm system might result in an unacceptable false positive rate. Therefore, emphasis should be placed on the characterization of SL patterns that are representative of the control state, eventually for each individual patient, followed by an AI-based pattern classifier. Effectively, this method will result in significant calibration times to setup the alarm system for each patient, such that effective training procedures for the learning of the control state patterns might be essential for transfer to clinical applications.

Mojtahedi, M. et al . Cell fate decision as high-dimensional critical state transition. PLOS Biol. 14 , 1–28, https://doi.org/10.1371/journal.pbio.2000640 (2016).

Article   CAS   Google Scholar  

Samal, S. S. et al . Metastable regimes and tipping points of biochemical networks with potential applications in precision medicine. bioRxiv , https://doi.org/10.1101/466714 (2018).

Trefois, C., Antony, P. M., Goncalves, J., Skupin, A. & Balling, R. Critical transitions in chronic disease: transferring concepts from ecology to systems medicine. Curr. Opin. Biotechnol. 34 , 48–55, https://doi.org/10.1016/j.copbio.2014.11.020 (2015).

Article   CAS   PubMed   Google Scholar  

Thomas, R. & d’Ari, R. Biological feedback (CRC press, 1990).

Waddington, C. H. The strategy of the genes. London : Allen 86 (1957).

Arnold, V. I. Singularities, Bifurcations, and Catastrophes. In Catastrophe Theory , 1–2, https://doi.org/10.1007/978-3-642-58124-3 (Springer Berlin Heidelberg, Berlin, Heidelberg, 1992).

Book   Google Scholar  

Scheffer, M. et al . Early-warning signals for critical transitions. Nature 461 , 53–59, https://doi.org/10.1038/nature08227 (2009).

Article   ADS   CAS   PubMed   Google Scholar  

Guttal, V. & Jayaprakash, C. Changing skewness: an early warning signal of regime shifts in ecosystems. Ecol. Lett. 11 , 450–460, https://doi.org/10.1111/j.1461-0248.2008.01160.x (2008).

Article   PubMed   Google Scholar  

Dakos, V. et al . Slowing down as an early warning signal for abrupt climate change. Proc. Natl. Acad. Sci. 105 , 14308–14312, https://doi.org/10.1073/pnas.0802430105 (2008).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Dakos, V., van Nes, E. H., Donangelo, R., Fort, H. & Scheffer, M. Spatial correlation as leading indicator of catastrophic shifts. Theor. Ecol. 3 , 163–174, https://doi.org/10.1007/s12080-009-0060-6 (2010).

Article   Google Scholar  

Carpenter, S. R. & Brock, W. A. Rising variance: a leading indicator of ecological transition: Variance and ecological transition. Ecol. Lett. 9 , 311–318, https://doi.org/10.1111/j.1461-0248.2005.00877.x (2006).

Chen, L., Liu, R., Liu, Z.-P., Li, M. & Aihara, K. Detecting early-warning signals for sudden deterioration of complex diseases by dynamical network biomarkers. Sci. reports 2 , 342 (2012).

Lenton, T. M., Livina, V. N., Dakos, V., Van Nes, E. H. & Scheffer, M. Early warning of climate tipping points from critical slowing down: Comparing methods to improve robustness. Philos. Transactions Royal Soc. A: Math. Phys. Eng. Sci. 370 , 1185–1204, https://doi.org/10.1098/rsta.2011.0304 (2012).

Article   ADS   CAS   Google Scholar  

Chen, P., Liu, R., Li, Y. & Chen, L. Detecting critical state before phase transition of complex biological systems by hidden Markov model. Bioinformatics 32 , 2143–2150, https://doi.org/10.1093/bioinformatics/btw154 (2016).

Gijzel, S. M. W. et al . Dynamical resilience indicators in time series of self-rated health correspond to frailty levels in older adults. 72 , 991–996, https://doi.org/10.1093/gerona/glx065 (2017).

Gijzel, S. M. W. et al . Dynamical indicators of resilience in postural balance time series are related to successful aging in high-functioning older adults, https://doi.org/10.1093/gerona/gly170 (2018).

Liu, R., Chen, P., Aihara, K. & Chen, L. Identifying early-warning signals of critical transitions with strong noise by dynamical network markers. Sci . Reports 5 , https://doi.org/10.1038/srep17501 (2015).

Singer, M. et al . The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 315 , 801, https://doi.org/10.1001/jama.2016.0287 (2016).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Fleischmann, C. et al . Epidemiology of sepsis in germany: Incidence, mortality and associated costs of care 2007-2013. Intensive Care Medicine Exp. 3 , A50, https://doi.org/10.1186/2197-425X-3-S1-A50 (2015).

Bhattacharjee, P., Edelson, D. P. & Churpek, M. M. Identifying Patients With Sepsis on the Hospital Wards. Chest 151 , 898–907, https://doi.org/10.1016/j.chest.2016.06.020 (2017).

Gaieski, D. F. et al . Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergency department*. Critical Care Medicine 38 , 1045–1053, https://doi.org/10.1097/CCM.0b013e3181cc4824 (2010).

Kumar, A. et al . Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock*. Critical Care Medicine 34 , 1589–1596, https://doi.org/10.1097/01.CCM.0000217961.75225.E9 (2006).

Henry, K. E., Hager, D. N., Pronovost, P. J. & Saria, S. A targeted real-time early warning score (TREWScore) for septic shock. Sci . Transl . Medicine 7 , 299ra122–299ra122, https://doi.org/10.1126/scitranslmed.aab3719 (2015).

Mitchell, S. et al . Optimization of sepsis risk assessment for ward patients. 107–112, https://doi.org/10.1109/SIEDS.2016.7489280 (IEEE, 2016).

Carrara, M., Baselli, G. & Ferrario, M. Mortality prediction in septic shock patients: Towards new personalized models in critical care. 2792–2795, https://doi.org/10.1109/EMBC.2015.7318971 (IEEE, 2015).

Tang, C. H. H. et al . Non-invasive classification of severe sepsis and systemic inflammatory response syndrome using a nonlinear support vector machine: a preliminary study. Physiol. Meas. 31 , 775–793, https://doi.org/10.1088/0967-3334/31/6/004 (2010).

Rothman, M. et al . Sepsis as 2 problems: Identifying sepsis at admission and predicting onset in the hospital using an electronic medical record–based acuity score. J. Critical Care 38 , 237–244, https://doi.org/10.1016/j.jcrc.2016.11.037 (2017).

Ghosh, S., Li, J., Cao, L. & Ramamohanarao, K. Septic shock prediction for ICU patients via coupled HMM walking on sequential contrast patterns. J. Biomed. Informatics 66 , 19–31, https://doi.org/10.1016/j.jbi.2016.12.010 (2017).

Horng, S. et al . Creating an automated trigger for sepsis clinical decision support at emergency department triage using machine learning. Plos One 12 , e0174708, https://doi.org/10.1371/journal.pone.0174708 (2017).

Cook, R. D. Detection of influential observation in linear regression. Technometrics 19 , 15–18, https://doi.org/10.1080/00401706.1977.10489493 (1977).

Article   MathSciNet   MATH   Google Scholar  

Guyon, I. & Elisseeff, A. An introduction to variable and feature selection. J. machine learning research 3 , 1157–1182 (2003).

MATH   Google Scholar  

Giacomini, R. & Rossi, B. Detecting and predicting forecast breakdowns. Rev. Econ. Stud. 76 , 669–705, https://doi.org/10.1111/j.1467-937X.2009.00545.x (2009).

Clements, M. & Hendry, D. Forecasting economic time series (Cambridge University Press, 1998).

Hendry, D. F. & Doornik, J. A. The implications for econometric modelling of forecast failure. Scott. J. Polit. Econ. 44 , 437–461, https://doi.org/10.1111/1467-9485.t01-1-00065 (1997).

Scheffer, M. et al . Quantifying resilience of humans and other animals. 115 , 11883–11890, https://doi.org/10.1073/pnas.1810630115 (2018).

Johnson, A. E. W. et al . MIMIC-III, a freely accessible critical care database. Sci . Data 3 , sdata201635, https://doi.org/10.1038/sdata.2016.35 (2016).

Abraham, E. New definitions for sepsis and septic shock: continuing evolution but with much still to be done. Jama 315 , 757–759 (2016).

Bone, R. C. et al . Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 101 , 1644–1655 (1992).

Levy, M. M. et al . 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Critical Care Medicine 31 , 1250–1256, https://doi.org/10.1097/01.CCM.0000050454.01978.3B (2003).

Organization, W. H. ICD-10: The ICD-10 Classification of Mental and Behavioural Disorders: diagnostic criteria for research. In ICD-10 : the ICD-10 classification of mental and behavioural disorders : diagnostic criteria for research (1993).

Statistics (US), N. C. f. H. The International Classification of Diseases : 9th Revision , Clinical Modification : ICD-9-CM (1991).

Vincent, J. L. et al . Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Critical care medicine 26 , 1793–1800 (1998).

Vincent, J. L. et al . The SOFA ( Sepsis-related Organ Failure Assessment ) score to describe organ dysfunction/failure (Springer, 1996).

Johnson, A. E. W. et al . A Comparative Analysis of Sepsis Identification Methods in an Electronic Database. Critical care medicine (2018).

Stineman, R. W. A Consistently Well Behaved Method of Interpolation. Creat. Comput. 6 , 54–57 (1980).

Google Scholar  

Harvey, A. C. Forecasting , structural time series models and the Kalman filter (Cambridge university press, 1990).

Auger-Méthé, M. et al . State-space models’ dirty little secrets: even simple linear Gaussian models can have estimation problems. Sci. Reports 6 , 26677, https://doi.org/10.1038/srep26677 (2016).

Holmes, E., Ward, E. & Wills, K. MARSS : Multivariate Autoregressive State-Space Modeling (2018).

Holmes, E. E., Ward, E. J. & Wills, K. MARSS: Multivariate autoregressive state-space models for analyzing time-series data. The R J. 4 , 30 (2012).

Andrews, D. W. K. Tests for parameter instability and structural change with unknown change point. Econometrica 61 , 821–856 (1993).

Article   MathSciNet   Google Scholar  

Chow, G. C. Tests of equality between sets of coefficients in two linear regressions. Econometrica 28 , 591–605 (1960).

Hollander, M., A. Wolfe, D. & Chicken, E. Nonparametric Statistical Methods : Hollander/Nonparametric Statistical Methods (John Wiley & Sons, Inc., Hoboken, NJ, USA, 2015).

Benjamini, Y. & Hochberg, Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J. Royal Stat. Soc. Ser. B (Methodological) 57 , 289–300 (1995).

MathSciNet   MATH   Google Scholar  

Download references

Acknowledgements

The computing resources were granted by RWTH Aachen University under project rwth0260. S.S.S. was supported by funding from CompSE profile area, RWTH Aachen University. We wish to thank the anonymous reviewers whose constructive comments helped to improve the manuscript.

Author information

Satya S. Samal

Present address: BASF SE, Carl-Bosch-Strasse 38, 67056, Ludwigshafen am Rhein, Germany

Pejman F. Ghalati and Satya S. Samal contributed equally.

Authors and Affiliations

Joint Research Center for Computational Biomedicine, RWTH Aachen University, 52074, Aachen, Germany

Pejman F. Ghalati, Satya S. Samal, Jayesh S. Bhat & Andreas Schuppert

Klinik für Operative Intensivmedizin und Intermediate Care, Universitätsklinikum Aachen, 52074, Aachen, Germany

Robert Deisz & Gernot Marx

You can also search for this author in PubMed   Google Scholar

Contributions

P.F.G. and S.S.S. developed the idea, conducted the research, and implemented the algorithms. J.S.B. helped in the preparation of the data and in proofreading of the article. R.D. and G.M. provided the clinical insights and interpreted the findings. A.S. supervised and supported the research project. All authors have reviewed the manuscript.

Corresponding author

Correspondence to Andreas Schuppert .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Ghalati, P.F., Samal, S.S., Bhat, J.S. et al. Critical Transitions in Intensive Care Units: A Sepsis Case Study. Sci Rep 9 , 12888 (2019). https://doi.org/10.1038/s41598-019-49006-2

Download citation

Received : 21 February 2019

Accepted : 14 August 2019

Published : 09 September 2019

DOI : https://doi.org/10.1038/s41598-019-49006-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Ground truth labels challenge the validity of sepsis consensus definitions in critical illness.

• Holger A. Lindner
• Shigehiko Schamoni
• Verena Schneider-Lindner

Journal of Translational Medicine (2022)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

• school Campus Bookshelves
• menu_book Bookshelves
• perm_media Learning Objects
• login Login
• how_to_reg Request Instructor Account
• hub Instructor Commons
• Download Page (PDF)
• Download Full Book (PDF)
• Periodic Table
• Physics Constants
• Scientific Calculator
• Reference & Cite
• Tools expand_more
• Readability

selected template will load here

This action is not available.

7: Case Study #6- Sepsis

• Last updated
• Save as PDF
• Page ID 9901
• 7.1: Learning Objectives
• 7.2: Patient- George Thomas
• 7.3: Sleepy Hollow Care Facility
• 7.4: Emergency Room
• 7.5: Day 1- Medical Ward
• 7.6: Day 2- Medical Ward

3 scenarios to train for diagnosis, treatment of sepsis

Use simulation education to teach ems providers to recognize and treat sepsis.

BLS providers practice capnography monitoring with a high-fidelity patient simulator.

Photo/Aaron Dix

By Aaron Dix

EMS is dispatched to a local residence for a 20-year-old female with foot pain. Medics find the patient sitting in a recliner with her left foot supported by a pillow.

She spent the weekend at the beach and cut her foot on a piece of glass while walking in the ocean. Over the past several days her foot has become increasingly painful, and this morning she developed a fever that hasn’t responded to acetaminophen. She is conscious and alert but weak. Skin is pale, dry and hot. Heart rate 110, respiratory rate 24, blood pressure 100/60. An oral thermometer is not available.

Focused assessment of the patient’s foot reveals a one-inch laceration with no active bleeding that is red, swollen and painful. Swelling is present throughout the bottom of her foot and ankle. Since the patient does not appear to be in any acute distress, the medics suggest that she go to the local urgent care center in the morning and a refusal is obtained.

The next morning, medics are dispatched to the urgent care center for a transfer to the local hospital for hypotension. Upon arrival, the same medics find the above-mentioned patient with a blood pressure of 86/40 and a heart rate of 140. Tympanic temperature of 101.5 was obtained by the urgent care staff. The patient is urgently transported to the local emergency department. During transport, medics infuse 500 mL of normal saline per protocol for the management of medical hypotension.

At the emergency department, blood cultures are drawn, ibuprofen is administered for her fever, and she is admitted to the hospitalist service. Antibiotics for cellulitis from an infected wound are started once she is admitted. Her blood pressure continues to worsen throughout the night, and she is intubated and moved to the ICU. Despite blood pressure support and aggressive antibiotic treatment, the patient dies 24 hours after arriving at the hospital. The young patient in the above description died from septic shock secondary to her foot wound.

Sepsis is manageable with early diagnosis and treatment

A leading cause of death in the United States, sepsis carries a significantly higher mortality rate than both stroke and STEMI [1]. But when identified and treated with antibiotics at the onset of symptoms, sepsis is manageable and survivable.

Early antibiotic treatment is associated with drastically improved outcomes, yet many EMS providers lack the training to readily recognize sepsis and septic shock. Mortality increases an estimated nine percent per hour when antibiotics are delayed after hypotension [2].

Unlike STEMI and stroke, sepsis recognition does not require a specific exam or new equipment. As a recent study demonstrated, effective EMS sepsis recognition only requires the provider to evaluate respiratory rate, heart rate, temperature and the possibility of an infection [3]. Measuring lactate, while useful in determining the severity of sepsis, is not necessary in the recognition phase.

Simulation education for sepsis recognition

Simulation education can be useful in educating EMS providers, both basic and advanced, in both sepsis recognition and treatment. Here are four learning objectives to use or modify for an EMS sepsis training:

1. Discuss the SIRS criteria and how it relates to sepsis recognition. 2. Demonstrate an appropriate sepsis assessment. 3. Identify patients who have a high probability of being septic. 4. Differentiate between the flu and pneumonia.

Simulation tip: Some simulators have limitations and certain vital signs such as temperature and glucose will not be obtainable utilizing standard EMS equipment. The facilitator will need to provide the correct information when promoted by the participant’s actions, either verbally or through the simulator’s patient monitor. However, all high-fidelity simulators have the ability to provide respiratory rate, heart rate, blood pressure and lung sounds. Providers should be prompted to assess the simulator as a real patient and gather vital signs and history in real time.

Here are three simulation scenarios to use or modify for your EMS training program.

Scenario 1: Standard Sepsis

Facilitators should begin with a non-complicated scenario that easily demonstrates the signs and symptoms of sepsis. The debriefing should concentrate on ensuring that a proper assessment will yield all the necessary information a provider needs to determine sepsis: two or more SIRS criteria and a known or suspected infection. Two common causes of sepsis EMS providers are likely to encounter are pneumonia and urinary tract infections.

Overview: Crew arrives to find a 68-year-old male complaining of shortness of breath. His shortness of breath began after waking six hours earlier and is progressively getting worse. Since lunch he has been unable to ambulate without becoming significantly short of breath. He hasn’t been feeling great over the past several days and has had a productive cough that has occasionally awoken him from sleep.

History: HTN, previous MI (2002), hypothyroid

Allergies: none

Medications: lisinopril, levothyroxine, warfarin

Patient weight/height: 180 pounds/5 feet, 11 inches

Vitals: HR is 130, BP is 106/60, RR is 26/min, SpO2 is 90 percent, glucose 250 mg/dl, temp 101.3 F, ETCO2 30 mm Hg with a normal waveform, and lung sounds are bilateral rhonchi.

Treatment should include oxygenation administration, fluid replacement, sepsis alert and antibiotic therapy if available.

Scenario 2: Differentiating sepsis/pneumonia from the flu

EMS providers must be capable of maintaining a high sensitivity for sepsis patients while limiting false positives. The flu can easily mimic pneumonia, making sepsis recognition more difficult. In this case, the rapid onset, non-productive cough and clear lung sounds make the argument for a flu diagnosis over pneumonia.

Overview: Crew arrives to find a 56-year-old female with sudden onset of high fever, general malaise and a non-productive cough. She woke this morning feeling normal. Her symptoms began suddenly right after lunch and worsened rapidly. She has a frequent non-productive cough, and her fever has not responded to acetaminophen.

History: hyperlipidemia and type 2 diabetes

Allergies: penicillin and naproxen

Medications: simvastatin and metformin

Patient weight/height: 120 pounds/5 feet, 6 inches

Vitals: HR is 100, BP is 118/70, RR is 20/min, SpO2 is 98 percent, glucose 140 mg/dl, temp 103.5 F, ETCO2 40 mm Hg with a normal waveform, and lung sounds are clear.

Scenario 3: Septic shock

Septic shock has a mortality rate near 50 percent and requires aggressive treatment. This case has two main objectives: aggressively treating septic shock and understanding that sepsis can occur in the presence of hypothermia. While approximately 80 percent of septic patients will show hyperthermia, temperature dysregulation, not fever, is the hallmark sign.

ETCO2 can also be discussed as an identifier of severe sepsis or septic shock . Decreases in ETCO2 correlate with elevated levels of lactate and increases in mortality.

Overview: EMS responds to a local nursing home for altered mental status. Patient is a 72-year-old male who was admitted to a skilled nursing facility for rehabilitation following a total hip replacement. Staff states he became altered this afternoon and was unable to ambulate this evening. He was admitted to the facility last night, and very little information is known.

History: hypertension, atrial-fibrillation, and type 2 diabetes

Medications: metformin, lisinopril, amiodarone and warfarin

Patient weight/height: 220 pounds and 5ft 9in

Vitals: HR is 150, BP is 84/50, RR is 22/min, O2 is 96 percent, glucose 280 mg/dl, temp 94.8 F, ETCO2 20 mm Hg with a normal waveform, and lung sounds are clear.

Additional info: Swollen and red surgical incision site on the right hip covered by the gown visible only if inspected.

Treatment: High volumes of normal saline, pressor support (norepinephrine preferred), sepsis alert and antibiotic therapy if available.

In conclusion, simulation training can improve the ability of both advanced and basic providers to diagnose and treat sepsis. Facilitators should concentrate on ensuring a comprehensive patient assessment to identify and treat patients who have a high probability of sepsis. Assess respiratory rate, heart rate, temperature and the possibility of an infection to make an accurate determination of sepsis. Faster recognition and treatment by EMS providers will lead to improved patient outcomes.

1. Cronshaw, 2011. Impact of surviving sepsis campaign on the recognition and management of severe sepsis in the emergency department: Are we failing? EMJ, Volume 12, pp. 296-327.

2. Kumar et al, 2006. Duration of hypotension before initiation of effective antimicrobial therapy is the determinant of survival in human septic shock. Critical Care Medicine, Volume 34, pp. 589-596.

3. Walchok et al, 2016. Paramedic-Initiated CMS Sepsis Core Measure Bundle Prior to Hospital Arrival: A Stepwise Approach, Prehospital Emergency Care, DOI: 10.1080/10903127.2016.1254694

Sentiment Analysis of Patient- and Family-Related Sepsis Events: Exploratory Study

Affiliation.

• 1 University of Cincinnati, Cincinnati, OH, United States.
• PMID: 38557694
• DOI: 10.2196/51720

Background: Despite the life-threatening nature of sepsis, little is known about the emotional experiences of patients and their families during sepsis events. We conducted a sentiment analysis pertaining to sepsis incidents involving patients and families, leveraging textual data retrieved from a publicly available blog post disseminated by the Centers for Disease Control and Prevention (CDC).

Objective: This investigation involved a sentiment analysis of patient- and family-related sepsis events, leveraging text responses sourced from a publicly accessible blog post disseminated by the CDC. Driven by the imperative to elucidate the emotional dynamics encountered by patients and their families throughout sepsis incidents, the overarching aims centered on elucidating the emotional ramifications of sepsis on both patients and their families and discerning potential avenues for enhancing the quality of sepsis care.

Methods: The research used a cross-sectional data mining methodology to investigate the sentiments and emotional aspects linked to sepsis, using a data set sourced from the CDC, which encompasses 170 responses from both patients and caregivers, spanning the period between September 2014 and September 2020. This investigation used the National Research Council Canada Emotion Lexicon for sentiment analysis, coupled with a combination of manual and automated techniques to extract salient features from textual responses. The study used negative binomial least absolute shrinkage and selection operator regressions to ascertain significant textual features that correlated with specific emotional states. Moreover, the visualization of Plutchik's Wheel of Emotions facilitated the discernment of prevailing emotions within the data set.

Results: The results showed that patients and their families experienced a range of emotions during sepsis events, including fear, anxiety, sadness, and gratitude. Our analyses revealed an estimated incidence rate ratio (IRR) of 1.35 for fear-related words and a 1.51 IRR for sadness-related words when mentioning "hospital" in sepsis-related experiences. Similarly, mentions of "intensive care unit" were associated with an average occurrence of 12.3 fear-related words and 10.8 sadness-related words. Surviving patients' experiences had an estimated 1.15 IRR for joy-related words, contrasting with discussions around organ failure, which were associated with multiple negative emotions including disgust, anger, fear, and sadness. Furthermore, mentions of "death" were linked to more fear and anger words but fewer joy-related words. Conversely, longer timelines in sepsis events were associated with more joy-related words and fewer fear-related words, potentially indicating improved emotional adaptation over time.

Conclusions: The study's outcomes underscore the imperative for health care providers to integrate emotional support alongside medical interventions for patients and families affected by sepsis, emphasizing the emotional toll incurred and highlighting the necessity of acknowledgment and resolution, advocating for the use of sentiment analysis as a means to tailor personalized emotional aid, and thereby potentially augmenting both patient and family welfare and overall outcomes.

Keywords: families; patients; sentiment analysis; sepsis.

©Mabel Ntiamoah, Teenu Xavier, Joshua Lambert. Originally published in JMIR Nursing (https://nursing.jmir.org), 01.04.2024.

• Open access
• Published: 27 November 2023

Extending the ‘host response’ paradigm from sepsis to cardiogenic shock: evidence, limitations and opportunities

• Marie Buckel 1   na1 ,
• Patrick Maclean 2   na1 ,
• Julian C. Knight 2 , 3 ,
• Patrick R. Lawler 4 , 5 &
• Alastair G. Proudfoot 1 , 6  

Critical Care volume  27 , Article number:  460 ( 2023 ) Cite this article

2953 Accesses

11 Altmetric

Metrics details

Recent clinical and research efforts in cardiogenic shock (CS) have largely focussed on the restoration of the low cardiac output state that is the conditio sine qua non of the clinical syndrome. This approach has failed to translate into improved outcomes, and mortality has remained static at 30–50%. There is an unmet need to better delineate the pathobiology of CS to understand the observed heterogeneity of presentation and treatment effect and to identify novel therapeutic targets. Despite data in other critical illness syndromes, specifically sepsis, the role of dysregulated inflammation and immunity is hitherto poorly described in CS. High-dimensional molecular profiling, particularly through leukocyte transcriptomics, may afford opportunity to better characterise subgroups of patients with shared mechanisms of immune dysregulation. In this state-of-the-art review, we outline the rationale for considering molecular subtypes of CS. We describe how high-dimensional molecular technologies can be used to identify these subtypes, and whether they share biological features with sepsis and other critical illness states. Finally, we propose how the identification of molecular subtypes of patients may enrich future clinical trial design and identification of novel therapies for CS.

Graphical Abstract

Introduction

Cardiogenic shock (CS) is a complex syndrome of hypoperfusion resulting from cardiac dysfunction. Historically, a simple model of the pathophysiology of CS has focused on a reduction in cardiac output leading to reduced end-organ perfusion with systemic tissue oxygen starvation which culminates in death in 30–50% of patients [ 1 ]. As such, beyond specific interventions such as culprit-vessel percutaneous coronary intervention (PCI) in acute myocardial infarction-CS (AMI-CS), current best care is supportive, targeting normalisation of hemodynamic (cardiac output and blood pressure) and biochemical perturbations through the use of inopressors and/or mechanical circulatory support (MCS) [ 2 , 3 ]. However, augmentation of cardiac output has yet to demonstrate a survival benefit in clinical trials [ 4 , 5 , 6 , 7 ]. Recently published randomised control trial (RCT) [ 7 ] and meta-analysis [ 8 ] data illustrate no mortality benefit with the use of MCS in CS. This highlights the importance of finding alternative or complementary therapeutic strategies which have the potential to modify the syndrome itself once established.

One of the many challenges that CS presents, and in parallel with other critical illness syndromes such as sepsis [ 9 , 10 ], is significant heterogeneity. The magnitude and haemodynamic presentation of shock differs considerably between patients, contributing to a range of clinical phenotypes, often despite similar aetiological insults [ 11 ]. Further heterogeneity is observed in the individual response to supportive measures which likely impacts disease severity, prognosis and treatment response. Whilst this variation has been described clinically [ 11 , 12 , 13 ], its pathobiological basis has hitherto been poorly delineated. In parallel with the response to infection in sepsis, cardiomyocyte necrosis and tissue hypoxia in CS may activate an inflammatory response. Although intended to be reparative and restorative, as in sepsis and ARDS, it is plausible that the ‘host response’ to myocardial injury and end organ failure may be misaligned, maladaptive or even injurious. Potential adverse sequelae of maladaptive inflammatory and immune activation include impaired microcirculation, inappropriate vasodilatation further compromising tissue perfusion, impaired cardiomyocyte recovery and compounded organ hypoperfusion through inappropriate vasodilatation and progression of the shock state. Importantly, the extent of inflammatory activation in response to AMI varies from patient-to-patient [ 14 ] with greater inflammation linked to short- and longer-term adverse outcomes [ 15 ]. This emerging ‘host-response’-oriented model is now established in inflammatory critical illness but has yet to be fully appreciated in CS. Accordingly, in this review we consider the relevant basis for such a model in CS, focusing primarily on AMI-CS where the pathophysiology and outcomes have been better studied to date. We theorise how this may enhance current research endeavours to improve patient outcomes.

Cardiogenic shock subphenotypes

In recognition of a need to relate clinical care and triage with CS severity, recent societal and registry efforts have attempted to stratify patients into groups defined by clinical, haemodynamic, metabolic and biochemical parameters [ 11 , 12 , 13 ]. The Society for Cardiovascular Angiography and Interventions (SCAI) Shock Classification, proposed by an expert panel, groups patients into stages A (“At risk”) to E (“Extremis”) [ 12 ]. The SCAI classification is currently the most well-validated in terms of mortality prediction [ 16 , 17 , 18 , 19 ] and has been adopted into clinical guidelines and clinical practice [ 20 , 21 ]. Extending this, an unsupervised machine learning analysis of registry data identified three distinct clinical CS phenotypes, termed ‘non-congested’, ‘cardiorenal’ and ‘cardio-metabolic’ [ 11 ]. These phenotypes had distinct haemodynamic and biochemical profiles with reproducible associations with shock severity and mortality [ 22 ].

Whilst such groupings may inform CS severity, there is evidence that the patients defined within these cohorts, their associated mortality risk and treatment responses remain highly heterogeneous [ 23 , 24 , 25 ]. These classifications also provide no additional information on pathobiology to guide treatments. Delineating patient groups based on combinations of clinical manifestations and biomarker patterns that link with disease pathobiology, so-called sub-phenotyping, has been proposed to bridge underlying disease mechanisms with clinical phenotype [ 26 ]. Notably, in the recent ‘Extra-Corporeal Life Support in Infarct Related Shock’ (ECLS-SHOCK) study of 420 patients with AMI-CS, mortality was approximately 50% and of those that died, more than half died of refractory cardiogenic shock [ 7 ]. This finding suggests that there are drivers of persistent shock and organ failure that extend beyond cardiac output. Vasodilatory CS is defined by a haemodynamic profile characterised by low systemic vascular resistance and reduced cardiac output (CO) with or without infection. This subphenotype appears to have greater shock severity, more organ failures and a worse prognosis [ 27 , 28 ]. Similarly, patients with microcirculatory dysfunction, or the uncoupling of macro- and peripheral microcirculatory blood flow in CS, may represent an additional sub-phenotype reflecting immune-mediated endothelial dysfunction and microvascular thrombosis identifiable by impairments in capillary refill time [ 29 , 30 ].

The ‘host response’ to cardiogenic shock

The role of inflammation in cardiovascular disease [ 31 ] and the immune response to myocardial infarction is well established [ 32 , 33 ]. Data in CS are largely derived from small-scale observational studies or sub-studies of randomised controlled trials examining circulating cytokines and proteins, with limited mechanistic data. Higher levels of biomarkers of systemic inflammation, such as C-reactive protein (CRP), procalcitonin (PCT) and IL-6, are associated with more severe hypoperfusion in CS, whilst levels of PCT and IL-6 correlate with multi-organ failure (MOF) [ 34 ] and mortality [ 35 ]. In a small single-centre study of AMI-CS, admission IL-6 was more strongly associated with 30-day mortality than more traditionally cardiac-specific markers, such as N-terminal pro-brain natriuretic peptide (NT-pro-BNP) [ 36 ]. Table 1 presents a comprehensive, though not exhaustive, list of inflammatory biomarkers implicated in the pathophysiology of CS.

The occurrence of sepsis super-imposed on CS, and CS secondary to sepsis (septic cardiomyopathy), further drives the hypothesis that dysregulated immunity may contribute to CS pathobiology [ 53 , 54 , 55 ]. Patients with CS have multiple risk factors for increased risk of infection including preceding cardiac arrest, gut hypo-perfusion and risk of bacterial translocation, use of multiple invasive central venous catheters and prolonged mechanical ventilation and intensive care unit (ICU) stay. Estimates of the incidence of concomitant or secondary sepsis range widely, from 6% of patients with AMI-CS in a large US payer database to around 50% in single-centre observational cohort studies [ 45 , 55 ]. Reasons for this wide range likely relate to overlap between the traditional clinical and biochemical markers used to diagnose CS and sepsis [ 45 , 53 , 54 ] as well as variability in the requirement for detection of a causative microorganism. The presence of concomitant sepsis is associated with increased shock severity [ 28 ], an increased risk of organ failures [ 55 , 56 ] and higher mortality [ 45 , 55 ].

Whether infection is a driver of dysregulated immunity or secondary to it, however, remains unclear. One potential source of sterile inflammation is endotoxin translocation from either digestive tract hypoperfusion or ischemia–reperfusion in CS. Endotoxemia reduces cardiac performance [ 57 ] and can propagate a low cardiac output state [ 58 ]. Despite sound rationale, three small studies [ 59 , 60 , 61 ] have failed to identify direct proof of endotoxemia in patients with CS with the caveat that no study has hitherto reported quantitative measurement (i.e. mass) of circulating endotoxins and many endotoxin assays have inherent detection limitations. An alternative explanation for the occurrence of secondary sepsis in AMI-CS is the development of maladaptive immune response, or even an acquired immunodeficiency. In a large cohort of critically ill patients with diffuse insults, after an initial phase of adaptive injury-induced immune response, a persistence of altered T-cell and monocyte response at one week was associated with secondary infection [ 62 ].

Septic cardiomyopathy typifies the overlap between septic and cardiogenic shock. Similar to CS, the precise mechanisms of cardiac dysfunction remain poorly elucidated. There is overlap in the observed immune response [ 38 , 44 , 47 , 63 , 64 ] to both AMI-CS and septic cardiomyopathy, despite the absence of a primary cardiac insult in the former. It is postulated that the systolic dysfunction of septic cardiomyopathy is an adaptive response which manifests as more classical cardiogenic shock when there is maladaptation and associated cellular dysfunction [ 65 , 66 ]; a comprehensive review is beyond the scope of this article but is covered elsewhere [ 67 ]. Further research into potentially shared pathobiology between septic cardiomyopathy and cardiogenic shock syndromes through comparisons of existing multi-modal sepsis datasets with emerging CS datasets will hopefully be mechanistically and therapeutically revealing.

An emerging potential modulator of CS pathobiology is the circulating enzyme dipeptidyl peptidase 3 (DPP3). This zinc-dependent metallopeptidase is found intracellularly throughout all the body’s organ systems [ 68 ] and is released into the circulation during cell death [ 69 , 70 ]. High levels of circulating DPP3 (cDPP3) have been found to be associated with increased severity in a range of shocked states [ 71 ]. Deletion of DPP3 impacts production of both proinflammatory and anti-inflammatory cytokines [ 72 ]. Injection of DPP3 in a murine model produced myocardial depression, whilst administration of a specific antibody targeted against cDPP3 normalised haemodynamics [ 73 ]. In vivo cDPP3 levels were measured in patients recruited to the multi-centre Optima CC trial [ 46 ] comparing epinephrine versus norepinephrine for haemodynamic support in CS [ 74 ]. High levels of cDPP3 (> 51.9 ng/ml) were associated with greater risk of death at 90 days, greater organ dysfunction and lower cardiac index, findings which have been confirmed in subsequent observational studies [ 75 , 76 ]. A large multi-centre, prospective study investigating the role of cDPP3 in acute coronary syndromes (ACS) found that high levels were independently predictive of the development of in-hospital CS (HR 1.49, 95% CI 1.14–1.96, P  = 0.004) [ 52 ]. Whilst promising, further data to clarify the precise biological role of DPP3 in CS are needed.

The contribution of baseline inflammation in the system-wide pathogenesis of CS has been further highlighted by a potential role for clonal haematopoiesis (CH) [ 77 , 78 ]. CH is the acquisition of somatic mutations of potentially oncogenic genes in haematopoietic stem cells (HSC) that results in distinct immune cell clones with dysregulated function. The carriage of these mutations has been associated with increased risk of atherosclerotic cardiovascular disease and cardiac failure [ 79 ]. Some in vitro studies have suggested that macrophages and monocytes with clonal features have a hyper-inflammatory phenotype [ 79 , 80 ]. In a biomarker sub-study of the multi-centre “Culprit Lesion Only PCI Versus Multivessel PCI in Cardiogenic Shock—CULPRIT-SHOCK” trial [ 81 ] a correlation between the burden of CH and risk of death or requirement for renal replacement therapy was observed [ 77 ]. CH was also associated with increased levels of the inflammatory cytokines IL-6 and IL1-beta. Similarly, in a single-centre matched retrospective cohort study comparing the presence of CH in patients with CS versus those with stable heart failure (HF), CS patients had a 50% higher prevalence of CH-related mutations (odds ratio 1.5; p = 0.02) [ 78 ]. Higher CH was associated with reduced survival and dysregulation of circulating inflammatory cytokines, particularly in those patients with mutations of tet methylcytosine dioxygenase 2 ( TET2 ). Collectively these data suggest that CH may augment the acute inflammatory state in CS, contributing to the development of superimposed vasodilatory shock, and in turn to worse outcomes.

Given the assertion that immune activation and dysregulated inflammation are implicated in the pathogenesis of CS, it would follow that immunomodulation may improve clinical outcomes. To date, despite trials in patients with heart failure [ 82 , 83 , 84 ], there have been no clinical trials specifically testing immunomodulatory therapy in patients with CS. This likely reflects both challenges of studies in CS patients per se as well as the rudimentary state of understanding of the inflammatory response as a therapeutic target. One ongoing study will assess the effects of the IL-6 monoclonal antibody tocilizumab on the development of CS after MI (ClinicalTrials.gov Identifier: NCT05350592), testing the importance of inflammation and the neurohormonal response to the development of CS. Another ongoing study will test the Oxiris membrane™, which removes circulating pro-inflammatory cytokines and lipopolysaccharides (ClinicalTrials.gov Identifiers: NCT05642273 and NCT04886180), in the most severe cohort of CS patients supported with venoarterial extracorporeal membrane oxygenation. Figure  1 illustrates how the past and current understanding of CS pathophysiology has shaped both treatment goals and clinical trial design.

Past and present risk stratification in CS. + Severity staging pyramid adopted from Society of Cardiovascular Angiography & Intervention (SCAI) [ 12 ]. Historically the management of cardiogenic shock (CS) has focussed on the normalisation of haemodynamic and biochemical parameters. As such, early research has compared the optimal modality to achieve this i.e. pharmacotherapy versus mechanical circulatory support. More recently, investigators have aimed phenotype patients into risk categories using both clinical parameters and measurements of inflammatory mediators to define severity of shock or risk of death

Future State of the ‘Host Response’ to Cardiogenic Shock: sub-phenotypes, endotyping and treatable traits

The observations describing the inflammatory response to CS above highlight the need for future studies which capture rich data on immune phenotypes. The immune system is highly complex, characterised by multidimensional relationships across multiple scales. Genes, cells and whole organs function together to preserve homeostasis often with multiple layers of redundancy. This complexity has driven the use of high-dimensional readouts, based on genetics, transcriptomics (RNA-sequencing), proteomics and metabolomics, to collect measurements informative for different features of the host immune state. In CS, there is the opportunity to discover molecular features or mechanisms associated with observed clinical traits, either individually in the form of single associated genetic loci, genes or proteins, or in ensemble gene sets or co-expression modules. Such molecular features, as well as clinical patient characteristics, can enable the discovery and definition of sub-phenotypes (subgroups). A disease endotype refers to instances where sub-phenotype (subgroup) characteristics/biomarkers define or associate with a specific pathophysiological mechanism. In terms of clinical utility, there is interest in establishing treatable traits, whereby sub-phenotype characteristics/biomarkers identify a group of patients with a specific pathophysiological derangement which manifests a predictable response to a specific therapy.

The most common approach to sub-phenotype and endotype discovery is peripheral blood leukocyte transcriptomics using RNA-sequencing. Whole blood provides a snapshot of the gene expression and abundance of cell types at different stages of haematopoiesis. Whilst peripheral blood sampling is logistically simple, the cellular composition of peripheral blood may change significantly over the natural history of critical illness and in response to critical care interventions and therapies [ 85 ]. Whole blood isolation of peripheral blood mononuclear cells (PBMCs) is informative but limits analysis to lymphocyte and monocyte populations [ 14 ]. This approach has the advantage of isolating cell types with greater relevance to the adaptive immune system but is relatively laborious and omits the key granulocyte populations which are the major effectors of the acute host response. The granularity of RNA-sequencing data acquisition and cellular heterogeneity can be further enhanced by quantification of the transcriptome of individual cells (single-cell RNA-sequencing) as opposed to measurement of average gene expression measured across a large population of cells (bulk RNA-sequencing).

The largest whole-blood transcriptomic studies, predominantly performed in sepsis populations [ 86 , 87 , 88 , 89 ], have used unsupervised clustering approaches to partition patients into discrete categories. The hypothesis is that these gene-expression groupings reflect fundamental differences in the host response, which are not better explained by known clinical covariates like age, sex, blood cell proportions, causative organism or immunosuppression. Clinical outcomes can then be compared across the subgroups, with association found for mortality after adjusting for known confounders in sepsis by different research teams [ 87 ] or worsening of a clinical score [ 90 , 91 ].

Identified genes or other molecular features can be further investigated using established biological pathway datasets to understand their biological function. For example, recent work identified a poor outcome sepsis endotype, broadly replicated across different infectious disease contexts [ 86 ], that had features of maladaptive ‘emergency myelopoiesis’, with increased abundance of activated neutrophils, so-called myeloid-derived suppressor cells (MDSCs), haematopoietic stem and progenitor cells (HSPCs) and specific immature neutrophil populations. A patient’s membership in a subphenotype or endotype can be modelled as a continuous trait, rather than discrete categories [ 92 ] which increases the power to detect dynamic changes in immune status over the course of disease or in response to therapy. Conceptually, endotyping can also support the identification of existing or emerging animal model systems that best represent clinical CS subphenotypes to support preclinical testing [ 93 , 94 , 95 , 96 ].

The majority of such host response profiling has been performed in infectious diseases and sepsis where the relevance of peripheral blood is intuitive with a paucity of large-scale work performed using tissue samples of relevance to CS. The ShockOmics consortium contrasted leukocyte gene expression at three different time-points (days 1, 2 and 7) in 21 patients with septic shock and 11 patients with CS [ 97 ]. Patients were matched for demographics and illness severity and inclusion required survival at seven days which may have enriched for a cohort with a favourable trajectory. Overall gene expression features were found to be similar across the 2 shock sub-phenotypes. The major source of between-sample variation in this dataset corresponded to the time from disease onset to sampling, suggesting that patients with either septic shock or CS follow broadly similar host response trajectories from the first to the seventh day of intensive care unit admission.

As the most prevalent aetiology of CS, study of the ‘host response’ to AMI may provide insights into the pathobiology and heterogeneity of patients who progress towards CS. Hence, study of leukocyte gene expression in over 100 patients with AMI identified 2 sub-phenotypes which coded for proteins related to platelet function. Patients with sub-phenotype 2 exhibited higher CRP on admission than those with sub-phenotype 1. Gene set enrichment analysis of 139 patients in eleven datasets of PBMCs from AMI patients highlighted extensive changes characterised by pro-inflammatory activation and enhanced leukocyte-platelet interactions with one-third of patients classified into a hyper-inflammatory group [ 98 ]. Stratification by consensus clustering suggested AMI patients differ in the severity of inflammatory activation; however, there were no data to associate this with progression to or severity of CS. Speculatively, these data suggest that differential pathway activation may contribute to different CS sub-phenotypes. The Prospective Observational Study Investigating Genomic Determinants of Outcome From Cardiogenic Shock (GOlDilOCS, ClinicalTrials.gov Identifier: NCT05728359) and VANQUISH Shock [ 99 ] will analyse these associations at a gene expression and proteome level in prospective study of 300 and 600 CS patients, respectively.

Whilst conceptually appealing, the approach of whole blood sampling may not, however, be a relevant proxy measure for host response patterns in remote organs or cells [ 100 ]. In the context of CS, targeted collection of coronary endothelial or even myocardial/endocardial samples may be a useful addition to the more accessible peripheral blood samples which remain the mainstay of studies of the host response. Because these tissue samples are difficult to acquire, emerging modalities which provide information on damage occurring in remote organs are likely to be of interest.

One example of this is cell-free DNA methylation sequencing that is a developing technique which can provide information on damage occurring in remote tissues [ 101 ]. This extracellular DNA is released by dying cells and then passes into the peripheral blood compartment, where it circulates with a half-life of 1–2 h. DNA methylation patterns are tissue-specific, in most instances due to conserved epigenetic enhancer usage across cell types. Sequencing and deconvolution of the circulating cell-free DNA provides an indirect index of the degree of cell death occurring in organs which are difficult to sample. This approach may complement the use of existing clinical tests for organ-specific damage in CS such as cardiomyocyte troponin and hepatocellular aminotransferases. It also offers an opportunity to investigate the inconsistently observed association between increased levels of circulating cell-free DNA and adverse outcomes in critical illness [ 102 , 103 , 104 ], which could be related to differences in the tissue-of-origin of the circulating DNA. Proof-of-concept studies have demonstrated cardiomyocyte-specific release in heart failure [ 105 ] and after MI [ 106 ]. Investigation of the dynamic release of cfDNA specifically derived from the vascular endothelial cells (VECs) of different organs demonstrated that organ-specific VEC cfDNA can be detected in plasma during clinical illness—for example, lung-derived VEC dfDNA during sepsis, exacerbations of chronic respiratory disease and after cardiac catheterisation [ 107 ].

Cell-free DNA can be readily isolated from frozen plasma samples and can be amplified inexpensively for a small number of cell-type defining methylation sites or sequenced in toto to assess for a wide range of cell-type-specific patterns. As this method matures, it is conceivable that this may become a useful tool for dynamic assessment of organ function, for example before and after administration of a drug or initiation of MCS support in CS. This could be of particular interest in CS, where understanding the effect of tissue hypoperfusion on specific organs is likely to be a useful tool for patient phenotyping.

Parallels with other critical illness syndromes: exotyping

Given the absence of therapies that improve outcome in CS and the observation that restoration of the low cardiac output that is the conditio sine qua non for CS has not improved clinical outcomes [ 108 , 109 ], it serves to reason that the current concepts of CS may not adequately capture the complexity of what is essentially a critical illness syndrome.

Critical illness states such as sepsis and CS have historically been defined by the primary organ dysfunction combined with constellations of non-specific clinical, biochemical and physiological abnormalities. It is apparent that many of these abnormalities are shared across critical illness syndromes regardless of the initial insult [ 10 , 22 ]. Whilst there is clearly heterogeneity both within critical illness syndromes and between them, this apparent homogeneity raises the prospect that despite disparate insults, similar underlying biological mechanisms or molecular signatures exist across critical illness syndromes as drivers of a common physiological derangement. This is the concept of exotypes, defined as endotypes conserved across different syndromes and sub-phenotypes [ 26 ].

Hence, the genomic response to trauma, severe burns and endotoxin exposure differed largely only in the duration rather than the response itself [ 110 , 111 ]. The observed transcriptional up-regulation of inflammatory mediators mirrors that in ARDS and pancreatitis [ 111 ]. Comprehensive, longitudinal immune profiling in patients with infectious (sepsis) and sterile (traumatic injury and post-surgery) injury demonstrated common signatures of pro-/anti-inflammatory, innate and adaptive immune responses [ 62 ]. Whilst coronary ischaemia–reperfusion will be exclusive to AMI-CS, the pathobiological drivers of the systemic endothelial dysfunction and organ dysfunction observed in sepsis may be contributory in CS. For example, perturbations of the angiopoietin-2 -Tie 2 axis, a regulator of capillary permeability, have been observed in both sepsis [ 112 , 113 ] and AMI-CS [ 114 ] with elevations of angiopoietin-2 associated with both poor outcome and coronary reperfusion success. It is therefore conceivable that other mechanistic drivers of sepsis, namely immune activation/dysfunction, mitochondrial dysfunction, complement activation, renin–angiotensin–aldosterone system activation and microcirculatory dysfunction [ 115 , 116 , 117 ], represent sub-phenotypes of AMI-CS. The identification and validation of biomarkers of AMI-CS sub-phenotypes, such as bio-adrenomedullin for severe or refractory vasoplegia [ 118 ], raises the enticing prospect of future targeted therapies across a range of shock states including AMI-CS. Future multi-omic preclinical and clinical study should be undertaken to identify conserved molecular responses across sub-phenotypes of differing shock states, specifically those patients with refractory shock who appear to be at highest risk of death (Graphical Abstract).

Future directions

Precision medicine in cardiogenic shock.

Secondary analyses from prior clinical trials in CS have failed to identify subgroups with clear differential treatment effects. This may reflect limitations in conventional one-at-a-time subgroup analysis which may be mitigated by machine learning methods such as causal forests or risk-based modelling, as recently demonstrated [ 119 , 120 , 121 ]. This may also reflect the choice to partition based on clinical groups as opposed to either molecular sub-phenotypes or endotypes described herein. Whilst advances continue to be made in clinical phenotyping of CS, greater precision, specifically linking clinical traits with pathobiology, is required to identify populations who will predictably respond to existing or emerging therapies. The identification of CS endotypes and treatable traits [ 122 ] offers the potential to enrich future clinical trial design and interventions through a more personalised or precision medicine approach (Fig.  2 ).

Endotype enrichment of future clinical trials in AMI-CS. AMI-CS = Acute myocardial infarction cardiogenic shock. Future research into the host response to cardiogenic shock should aim to identify endotypes which can then be used to enrich clinical trials to increase the likelihood of positive trial results

An endotype or treatable trait could be incorporated into a clinical trial as either a stratifying variable at recruitment, or in the case of dynamic and especially quantitative endotype systems, as a response measure. The first scenario is oftentimes divided into ‘prognostic’ and ‘predictive’ enrichment [ 123 ]. In prognostic enrichment, a set of predictors thought to have prognostic value, generally for mortality, are used as a screening mechanism at trial recruitment. Limiting recruitment to the cohort at highest risk of adverse outcome leverages the assumption that individuals with the most extreme risk may experience differential treatment effects. This assumption should be subject to some scrutiny given how frequently the opposite interpretation, that a trial has returned negative results because its population is too unwell to have the possibility to benefit (e.g. later-stage, unmodifiable risk), is also offered in the literature [ 124 ]. Risk-based heterogeneity of treatment response may be nonlinear, with greatest benefit in those at sufficient risk of the outcome to benefit, but not so sick as to be past the point where treatment is beneficial. Nonetheless, recent work in sepsis has demonstrated the potential for endotype-based stratification and quantitative scoring of patients with acute infection at point of care [ 92 ] by providing a framework that can be used with existing rapid turnaround methods (real-time quantitative reverse transcription PCR, qRT-PCR), as well as full-transcriptome technologies. This opens up the potential for future bedside, prognostic enrolment into clinical trials and even personalized therapeutic decision-making.

An alternative endotype-based strategy is so-called predictive enrichment, where a measurable trait, thought to have a biological relationship with experimental treatment, is used to select patients with a higher expected likelihood of benefit. Endotype stratification has been used in post hoc analyses of sepsis clinical trials. Whilst the VANISH randomised controlled trial [ 125 ] showed no mortality effect associated with corticosteroid treatment in sepsis, an interaction was found between sepsis endotype at baseline and mortality, with patients assigned to the lower-risk ‘immunocompetent’ endotype shown to have poorer survival [ 85 ]. Similar results were produced in a separate re-analysis using a separate endotype classification [ 126 ]. These associations are yet to be tested prospectively but suggest a route for developing endotype assignments as a stratifying factor for clinical trial design. Although the initial trial showed no mortality benefit with the use of polymyxin B haemofiltration versus sham [ 127 ], the post hoc division of patients by endotoxin activity showed differential effects on ventilatory-free days, mean arterial pressure (MAP) and mortality [ 128 ]. A large ongoing sepsis/ARDS trial is using 2 endotype classifications as pre-specified randomisation strata [ 129 ], will test these tools in the prospective setting.

The major challenge is finding an appropriate stratifying tool; to a large extent, traits which robustly predict treatment response are unknown before they are tested in trials. Post hoc stratification of trials with broad recruitment criteria can provide clues, but findings should be replicated in additional cohorts and ideally in preclinical model systems [ 130 ].

Caution should be exercised when estimating the likelihood that a null overall treatment response is masking unobserved heterogeneity of treatment effect across the trial population, and that high-dimensional assays can consequently be used to distinguish individual ‘responders’ from ‘non-responders’. Most applications in critical care allow only a single measure of treatment response for each patient [ 131 ]. When observing statistical heterogeneity in a response variable after a single episode of treatment, this may potentially be explained by statistical noise rather than a true biological difference and can easily be compounded by arbitrary dichotomization of a continuous variable to distinguish ‘responders’ from ‘non-responders’ [ 132 ]. The often-made statement that null overall treatment responses in critical illness therapeutic trials could reflect unmeasured heterogeneity in either the clinical phenotype or in the treatment response should be tested, not simply assumed to be true [ 133 ]. Emerging applications of adaptive clinical trial designs are being deployed in acute and critical illnesses that may be more amenable than conventional designs to the prospective identification of heterogeneity of treatment effect [ 134 ].

New therapies and new approaches to clinical trial design are an urgent and unmet need if we are to improve the current lethality of CS. AMI-CS is increasingly recognised as encompassing features of systemic inflammation in addition to the systemic hypoperfusion due to pump failure. The holy grail of CS management is a granular understanding of disease heterogeneity as it relates to specific disease mechanisms and physiologic responses that would afford the opportunity to identify bespoke treatments with predictable responses that improve patient outcomes. Using the more nuanced approaches outlined herein may offer insights into the underlying molecular mechanisms of AMI-CS, with potential parallels to other critical illness syndromes and allow a transition from the current risk-based approach towards a mechanistic approach that embraces the heterogeneity within the CS population.

Abbreviations

Acute coronary syndrome

Acute myocardial infarction

Acute respiratory distress syndrome

Cardiac critical care unit

Cardiac output

• Cardiogenic shock

Cell-free DNA

Circulating dipeptidyl peptidase

Clonal haematopoiesis

C-Reactive protein

Deoxyribonucleic acid

Dipeptidyl peptidase 3

Extra-corporeal life support

Extra-corporeal membrane oxygenation

Granulocyte-colony-stimulating factor

Growth differentiation factor-15

Haematopoietic stem cells

Haematopoietic stem and progenitor cells

Heart failure

Heterogeneity of treatment effect

Interferon gamma

Intensive care unit

Interleukin-1

Interleukin-1 beta

Interleukin-1 receptor antagonist

Interleukin-5

Interleukin-6

Interleukin-7

Interleukin-8

Interleukin-10

Macrophage inflammatory protein-1 beta

Mechanical circulatory support

Monocyte chemoattractant protein-1

Monocyte chemoattractant protein-1 beta

Multi-organ failure

Myeloid-derived suppressor cells

Myocardial infarction

Neutrophil-to-leucocyte ratio

Non-ST elevation myocardial infarction

N-terminal pro-brain natriuretic peptide

Pentraxin 3

Percutaneous coronary intervention

Peripheral blood mononuclear cells

Peroxiredoxin-1

Platelet-to-lymphocyte ratio

Post-cardiac arrest syndrome

Procalcitonin

Regulated on activation, normal T cell expressed and secreted

Ribonucleic acid

Selenoprotein P

Septic shock

Society for Cardiovascular Angiography and Interventions

ST-elevation myocardial infarction

Tet methylcytosine dioxygenase 2

Tumour necrosis factor alpha

United States

Vascular endothelial cells

White blood cells

Lawler PR, Berg DD, Park JG, Katz JN, Baird-Zars VM, Barsness GW, et al. The range of cardiogenic shock survival by clinical stage: data from the critical care cardiology trials network registry. Crit Care Med. 2021;49(8):1293–302.

Article   PubMed   Google Scholar  

Byrne RA, Rossello X, Coughlan JJ, Barbato E, Berry C, Chieffo A, et al. 2023 ESC guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44:3720–826.

van Diepen S, Katz JN, Albert NM, Henry TD, Jacobs AK, Kapur NK, et al. Contemporary management of cardiogenic shock: a scientific statement from the american heart association. Circulation. 2017. https://doi.org/10.1161/CIR.0000000000000525 .

Jung C, Fuernau G, De Waha S, Eitel I, Desch S, Schuler G, et al. Intraaortic balloon counterpulsation and microcirculation in cardiogenic shock complicating myocardial infarction: an IABP-SHOCK II substudy. Clin Res Cardiol. 2015;104(8):679–87.

Ostadal P, Rokyta R, Kruger A, Vondrakova D, Janotka M, Smíd O, et al. Extra corporeal membrane oxygenation in the therapy of cardiogenic shock (ECMO-CS): rationale and design of the multicenter randomized trial: extra corporeal membrane oxygenation in the therapy of cardiogenic Shock (ECMO-CS). Eur J Heart Fail. 2017;19:124–7.

Article   CAS   PubMed   Google Scholar  

Ouweneel DM, Eriksen E, Sjauw KD, Van Dongen IM, Hirsch A, Packer EJS, et al. Percutaneous mechanical circulatory support versus intra-aortic balloon pump in cardiogenic shock after acute myocardial infarction. J Am Coll Cardiol. 2017;69(3):278–87.

Thiele H, Zeymer U, Akin I, Behnes M, Rassaf T, Mahabadi AA, et al. Extracorporeal life support in infarct-related cardiogenic shock. N Engl J Med. 2023;389:1286–97.

Zeymer U, Freund A, Hochadel M, Ostadal P, Belohlavek J, Rokyta R, et al. Venoarterial extracorporeal membrane oxygenation in patients with infarct-related cardiogenic shock: an individual patient data meta-analysis of randomised trials. The Lancet. 2023;402:1338–46.

Article   Google Scholar  

Lawler PR, Fan E. Heterogeneity and phenotypic stratification in acute respiratory distress syndrome. Lancet Respir Med. 2018;6(9):651–3.

Maslove DM, Tang B, Shankar-Hari M, Lawler PR, Angus DC, Baillie JK, et al. Redefining critical illness. Nat Med. 2022;28(6):1141–8.

Zweck E, Thayer KL, Helgestad OKL, Kanwar M, Ayouty M, Garan AR, et al. Phenotyping cardiogenic shock. JAHA. 2021;10(14): e020085.

Article   PubMed   PubMed Central   Google Scholar  

Baran DA, Grines CL, Bailey S, Burkhoff D, Hall SA, Henry TD, et al. SCAI clinical expert consensus statement on the classification of cardiogenic shock: this document was endorsed by the American College of Cardiology (ACC), the American Heart Association (AHA), the Society of Critical Care Medicine (SCCM), and the Society of Thoracic Surgeons (STS) in April 2019. Catheter Cardiovasc Interv. 2019;94:29–37.

Lim HS. Phenotyping and hemodynamic assessment in cardiogenic shock: from physiology to clinical application. Cardiol Ther. 2022;11(4):509–22.

Toma A, Dos Santos C, Burzyńska B, Góra M, Kiliszek M, Stickle N, et al. Diversity in the expressed genomic host response to myocardial infarction. Circ Res. 2022;131(1):106–8.

Lawler PR, Bhatt DL, Godoy LC, Lüscher TF, Bonow RO, Verma S, et al. Targeting cardiovascular inflammation: next steps in clinical translation. Eur Heart J. 2021;42(1):113–31.

Baran DA, Long A, Badiye AP, Stelling K. Prospective validation of the SCAI shock classification: single center analysis. Catheter Cardiovasc Interv. 2020;96(7):1339–47.

Jentzer JC, Soussi S, Lawler PR, Kennedy JN, Kashani KB. Validation of cardiogenic shock phenotypes in a mixed cardiac intensive care unit population. Cathet Cardio Intervent. 2022;99(4):1006–14.

Naidu SS, Baran DA, Jentzer JC, Hollenberg SM, van Diepen S, Basir MB, et al. SCAI SHOCK stage classification expert consensus update: a review and incorporation of validation studies. J Am Coll Cardiol. 2022;79(9):933–46.

Zweck E, Kanwar M, Li S, Sinha SS, Garan AR, Hernandez-Montfort J, et al. Clinical course of patients in cardiogenic shock stratified by phenotype. JACC Heart Failure. 2023

McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021;42(36):3599–726.

Heidenreich PA, Bozkurt B, Aguilar D, Allen LA, Byun JJ, Colvin MM, et al. AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2022. https://doi.org/10.1161/CIR.0000000000001063

Soussi S, Dos Santos C, Jentzer JC, Mebazaa A, Gayat E, Pöss J, et al. Distinct host-response signatures in circulatory shock: a narrative review. ICMx. 2023;11(1):50.

Jentzer JC. Understanding cardiogenic shock severity and mortality risk assessment. Circ Heart Failure. 2020;13(9):e007568.

Burstein B, Van Diepen S, Wiley BM, Anavekar NS, Jentzer JC. Biventricular function and shock severity predict mortality in cardiac ICU patients. Chest. 2022;161(3):697–709.

Thayer KL, Zweck E, Ayouty M, Garan AR, Hernandez-Montfort J, Mahr C, et al. Invasive hemodynamic assessment and classification of in-hospital mortality risk among patients with cardiogenic shock. Circ Heart Failure. 2020;13(9):1007099.

Jentzer JC, Rayfield C, Soussi S, Berg DD, Kennedy JN, Sinha SS, et al. Advances in the staging and phenotyping of cardiogenic shock. JACC Adv. 2022;1(4):100120.

Kohsaka S. Systemic inflammatory response syndrome after acute myocardial infarction complicated by cardiogenic shock. Arch Intern Med. 2005;165(14):1643.

Jentzer JC, Lawler PR, Van Diepen S, Henry TD, Menon V, Baran DA, et al. Systemic inflammatory response syndrome is associated with increased mortality across the spectrum of shock severity in cardiac intensive care patients. Circ Cardiovasc Qual Outcomes. 2020;13(12):1006956.

Merdji H, Curtiaud A, Aheto A, Studer A, Harjola VP, Monnier A, et al. Performance of early capillary refill time measurement on outcomes in cardiogenic shock: an observational, prospective multicentric study. Am J Respir Crit Care Med. 2022;206(10):1230–8.

Lawler PR, van Diepen S. Toward a broader characterization of macro- and microcirculatory uncoupling in cardiogenic shock. Am J Respir Crit Care Med. 2022;206(10):1192–3.

Boyalla V, Gallego-Colon E, Spartalis M. Immunity and inflammation in cardiovascular disorders. BMC Cardiovasc Disord. 2023;23(1):148.

Zlatanova I, Pinto C, Silvestre JS. Immune modulation of cardiac repair and regeneration: the art of mending broken hearts. Front Cardiovasc Med. 2016. https://doi.org/10.3389/fcvm.2016.00040/full .

Gentek R, Hoeffel G. The innate immune response in myocardial infarction, repair, and regeneration. In: Sattler S, Kennedy-Lydon T, editors. The immunology of cardiovascular homeostasis and pathology. Cham: Springer International Publishing; 2017 [cited 2023 Jul 7]. p. 251–72. (Advances in Experimental Medicine and Biology; vol. 1003). https://doi.org/10.1007/978-3-319-57613-8_12

Geppert A, Steiner A, Zorn G, Delle-Karth G, Koreny M, Haumer M, et al. Multiple organ failure in patients with cardiogenic shock is associated with high plasma levels of interleukin-6. Crit Care Med. 2002;30(9):1987–94.

Kataja A, Tarvasmäki T, Lassus J, Sionis A, Mebazaa A, Pulkki K, et al. Kinetics of procalcitonin, C-reactive protein and interleukin-6 in cardiogenic shock—Insights from the CardShock study. Int J Cardiol. 2020; https://www.internationaljournalofcardiology.com/article/S0167-5273(20)33652-4/fulltext

Andrié RP, Becher UM, Frommold R, Tiyerili V, Schrickel JW, Nickenig G, et al. Interleukin-6 is the strongest predictor of 30-day mortality in patients with cardiogenic shock due to myocardial infarction. Crit Care. 2012;16(4):R152.

Geppert A, Steiner A, Delle-Karth G, Heinz G, Huber K. Usefulness of procalcitonin for diagnosing complicating sepsis in patients with cardiogenic shock. Intensive Care Med. 2003;29(8):1384–9.

Debrunner M, Schuiki E, Minder E, Straumann E, Naegeli B, Mury R, et al. Proinflammatory cytokines in acute myocardial infarction with and without cardiogenic shock. Clin Res Cardiol. 2008;97(5):298–305.

Picariello C, Lazzeri C, Chiostri M, Gensini G, Valente S. Procalcitonin in patients with acute coronary syndromes and cardiogenic shock submitted to percutaneous coronary intervention. Intern Emerg Med. 2009;4(5):403–8.

Prondzinsky R, Unverzagt S, Lemm H, Wegener NA, Schlitt A, Heinroth KM, et al. Interleukin-6, -7, -8 and -10 predict outcome in acute myocardial infarction complicated by cardiogenic shock. Clin Res Cardiol. 2012;101(5):375–84.

Prondzinsky R, Unverzagt S, Lemm H, Wegener N, Heinroth K, Buerke U, et al. Acute myocardial infarction and cardiogenic shock: Prognostic impact of cytokines: INF-γ, TNF-α, MIP-1β, G-CSF, and MCP-1β. Med Klin Intensivmed Notfmed. 2012;107(6):476–84.

Fuernau G, Poenisch C, Eitel I, De Waha S, Desch S, Schuler G, et al. Growth-differentiation factor 15 and osteoprotegerin in acute myocardial infarction complicated by cardiogenic shock: a biomarker substudy of the IABP-SHOCK II-trial. Eur J Heart Fail. 2014;16(8):880–7.

Lipkova J, Parenica J, Duris K, Helanova K, Tomandl J, Kubkova L, et al. Association of circulating levels of RANTES and −403G/A promoter polymorphism to acute heart failure after STEMI and to cardiogenic shock. Clin Exp Med. 2015;15(3):405–14.

Liu CH, Kuo SW, Hsu LM, Huang SC, Wang CH, Tsai PR, et al. Peroxiredoxin 1 induces inflammatory cytokine response and predicts outcome of cardiogenic shock patients necessitating extracorporeal membrane oxygenation: an observational cohort study and translational approach. J Transl Med. 2016;14(1):114.

Parenica J, Jarkovsky J, Malaska J, Mebazaa A, Gottwaldova J, Helanova K, et al. Infectious complications and immune/inflammatory response in cardiogenic shock patients: a prospective observational study. Shock. 2017;47(2):165–74.

Takagi K, Blet A, Levy B, Deniau B, Azibani F, Feliot E, et al. Circulating dipeptidyl peptidase 3 and alteration in haemodynamics in cardiogenic shock: results from the OptimaCC trial. Eur J Heart Fail. 2020;22(2):279–86.

Cuinet J, Garbagnati A, Rusca M, Yerly P, Schneider AG, Kirsch M, et al. Cardiogenic shock elicits acute inflammation, delayed eosinophilia, and depletion of immune cells in most severe cases. Sci Rep. 2020;10(1):7639.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Büttner P, Obradovic D, Wunderlich S, Feistritzer HJ, Holzwirth E, Lauten P, et al. Selenoprotein P in myocardial infarction with cardiogenic shock. Shock. 2020;53(1):58–62.

Jentzer JC, Szekely Y, Burstein B, Ballal Y, Kim EY, van Diepen S, et al. Peripheral blood neutrophil-to-lymphocyte ratio is associated with mortality across the spectrum of cardiogenic shock severity. J Crit Care. 2022;68:50–8.

Roth S, M’Pembele R, Stroda A, Jansen C, Lurati Buse G, Boeken U, et al. Neutrophil-lymphoycyte-ratio, platelet-lymphocyte-ratio and procalcitonin for early assessment of prognosis in patients undergoing VA-ECMO. Sci Rep. 2022;12(1):542.

Dudda J, Schupp T, Rusnak J, Weidner K, Abumayyaleh M, Ruka M, et al. C-reactive protein and white blood cell count in cardiogenic shock. JCM. 2023;12(3):965.

Wenzl FA, Bruno F, Kraler S, Klingenberg R, Akhmedov A, Ministrini S, et al. Dipeptidyl peptidase 3 plasma levels predict cardiogenic shock and mortality in acute coronary syndromes. Eur Heart J. 2023;44(38):3859–71. https://doi.org/10.1093/eurheartj/ehad545 .

Kohsaka S, Menon V, Iwata K, Lowe A, Sleeper LA, Hochman JS. Microbiological profile of septic complication in patients with cardiogenic shock following acute myocardial infarction (from the SHOCK study). Am J Cardiol. 2007;99(6):802–4.

Nair RM, Kumar S, Saleem T, Chawla S, Vural A, Abdelghaffar B, et al. Characteristics and impact of bloodstream infections in cardiogenic shock patients on temporary mechanical circulatory support. JACC Cardiovasc Intervent. 2022; https://www.sciencedirect.com/science/article/pii/S1936879822014029

Jentzer JC, Bhat AG, Patlolla SH, Sinha SS, Miller PE, Lawler PR, et al. Concomitant sepsis diagnoses in acute myocardial infarction-cardiogenic shock: 15-year national temporal trends, management, and outcomes. Crit Care Explor. 2022;4(2): e0637.

Schmidt M, Brechot N, Hariri S, Guiguet M, Luyt CE, Makri R, et al. Nosocomial infections in adult cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation. Clin Infect Dis. 2012;55(12):1633–41.

Suffredini AF, Fromm RE, Parker MM, Brenner M, Kovacs JA, Wesley RA, et al. The cardiovascular response of normal humans to the administration of endotoxin. N Engl J Med. 1989;321(5):280–7.

Krack A, Sharma R, Figulla HR, Anker SD. The importance of the gastrointestinal system in the pathogenesis of heart failure. Eur Heart J. 2005;26(22):2368–74.

Attanà P, Lazzeri C, Chiostri M, Picariello C, Gensini GF, Valente S. Endotoxin role in cardiogenic shock: a brief report. Int J Cardiol. 2013;167(6):3031–2.

Ramirez P, Villarreal E, Gordon M, Gómez MD, de Hevia L, Vacacela K, et al. Septic participation in cardiogenic shock: exposure to bacterial endotoxin. Shock. 2017;47(5):588–92.

Lee CT, Wang CH, Chan WS, Tsai YY, Wei TJ, Lai CH, et al. Endotoxin activity in patients with extracorporeal membrane oxygenation life support: an observational pilot study. Front Med. 2021;8: 772413.

Venet F, Textoris J, Blein S, Rol ML, Bodinier M, Canard B, et al. Immune profiling demonstrates a common immune signature of delayed acquired immunodeficiency in patients with various etiologies of severe injury*. Crit Care Med. 2022;50(4):565–75.

Kumar A, Thota V, Dee L, Olson J, Uretz E, Parrillo JE. Tumor necrosis factor alpha and interleukin 1beta are responsible for in vitro myocardial cell depression induced by human septic shock serum. J Exp Med. 1996;183(3):949–58.

Pathan N, Hemingway CA, Alizadeh AA, Stephens AC, Boldrick JC, Oragui EE, et al. Role of interleukin 6 in myocardial dysfunction of meningococcal septic shock. The Lancet. 2004;363(9404):203–9.

Article   CAS   Google Scholar  

Levy RJ, Piel DA, Acton PD, Zhou R, Ferrari VA, Karp JS, et al. Evidence of myocardial hibernation in the septic heart*. Crit Care Med. 2005;33(12):2752–6.

Matkovich SJ, Al Khiami B, Efimov IR, Evans S, Vader J, Jain A, et al. Widespread down-regulation of cardiac mitochondrial and sarcomeric genes in patients with sepsis*. Crit Care Med. 2017;45(3):407–14.

Hollenberg SM, Singer M. Pathophysiology of sepsis-induced cardiomyopathy. Nat Rev Cardiol. 2021;18(6):424–34.

Sato A, Ogita H. Pathophysiological implications of dipeptidyl peptidases. CPPS. 2017;18(8):843–9.

Wattiaux R, Wattiaux-De Coninck S, Thirion J, Gasingirwa MC, Jadot M. Lysosomes and Fas-mediated liver cell death. Biochem J. 2007;403(1):89–95.

Rehfeld L, Funk E, Jha S, Macheroux P, Melander O, Bergmann A. Novel methods for the quantification of dipeptidyl peptidase 3 (DPP3) concentration and activity in human blood samples. J Appl Lab Med. 2019;3(6):943–53.

Lier D, Kox M, Pickkers P. Promotion of vascular integrity in sepsis through modulation of bioactive adrenomedullin and dipeptidyl peptidase 3. J Intern Med. 2021;289(6):792–806.

Baral PK, Jajčanin-Jozić N, Deller S, Macheroux P, Abramić M, Gruber K. The first structure of dipeptidyl-peptidase III provides insight into the catalytic mechanism and mode of substrate binding. J Biol Chem. 2008;283(32):22316–24.

Deniau B, Rehfeld L, Santos K, Dienelt A, Azibani F, Sadoune M, et al. Circulating dipeptidyl peptidase 3 is a myocardial depressant factor: dipeptidyl peptidase 3 inhibition rapidly and sustainably improves haemodynamics. Eur J Heart Fail. 2020;22(2):290–9.

Levy B, Clere-Jehl R, Legras A, Morichau-Beauchant T, Leone M, Frederique G, et al. Epinephrine versus norepinephrine for cardiogenic shock after acute myocardial infarction. J Am Coll Cardiol. 2018;72(2):173–82.

Innelli P, Lopizzo T, Paternò G, Bruno N, Radice RP, Bertini P, et al. Dipeptidyl amino-peptidase 3 (DPP3) as an early marker of severity in a patient population with cardiogenic shock. Diagnostics. 2023;13(7):1350.

Nordin H, Picod A, Hartmann O, Santos K, Azibani F, Bergmann A, et al. Circulating DPP3 is an early predictor of mortality and organ support in patients with cardiogenic shock: post-hoc analyses of the ACCOST-HH trial. Arch Cardiovasc Diseases Suppl. 2023;15(2):208.

Böhme M, Desch S, Rosolowski M, Scholz M, Krohn K, Büttner P, et al. Impact of clonal hematopoiesis in patients with cardiogenic shock complicating acute myocardial infarction. J Am Coll Cardiol. 2022;80(16):1545–56.

Scolari FL, Abelson S, Brahmbhatt DH, Medeiros JJF, Fan CPS, Fung NL, et al. Clonal haematopoiesis is associated with higher mortality in patients with cardiogenic shock. Eur J Heart Fail. 2022;24(9):1573–82.

Jaiswal S, Libby P. Clonal haematopoiesis: connecting ageing and inflammation in cardiovascular disease. Nat Rev Cardiol. 2020;17(3):137–44.

Cook EK, Luo M, Rauh MJ. Clonal hematopoiesis and inflammation: partners in leukemogenesis and comorbidity. Exp Hematol. 2020;83:85–94.

Thiele H, Akin I, Sandri M, Fuernau G, de Waha S, Meyer-Saraei R, et al. PCI strategies in patients with acute myocardial infarction and cardiogenic shock. N Engl J Med. 2017;377(25):2419–32.

Everett BM, Cornel JH, Lainscak M, Anker SD, Abbate A, Thuren T, et al. Anti-inflammatory therapy with canakinumab for the prevention of hospitalization for heart failure. Circulation. 2019;139(10):1289–99.

Bozkurt B, Torre-Amione G, Warren MS, Whitmore J, Soran OZ, Feldman AM, et al. Results of targeted anti-tumor necrosis factor therapy with etanercept (ENBREL) in patients with advanced heart failure. Circulation. 2001;103(8):1044–7.

Chung ES, Packer M, Lo KH, Fasanmade AA, Willerson JT. Randomized, double-blind, placebo-controlled, pilot trial of infliximab, a chimeric monoclonal antibody to tumor necrosis factor-α, in patients with moderate-to-severe heart failure: results of the anti-TNF therapy against congestive heart failure (ATTACH) trial. Circulation. 2003;107(25):3133–40.

Antcliffe DB, Burnham KL, Al-Beidh F, Santhakumaran S, Brett SJ, Hinds CJ, et al. Transcriptomic signatures in sepsis and a differential response to steroids. From the VANISH randomized trial. Am J Respir Crit Care Med. 2019;199(8):980–6.

Sweeney TE, Azad TD, Donato M, Haynes WA, Perumal TM, Henao R, et al. Unsupervised analysis of transcriptomics in bacterial sepsis across multiple datasets reveals three robust clusters. Crit Care Med. 2018;46(6):915.

Davenport EE, Burnham KL, Radhakrishnan J, Humburg P, Hutton P, Mills TC, et al. Genomic landscape of the individual host response and outcomes in sepsis: a prospective cohort study. Lancet Respir Med. 2016;4(4):259–71.

Scicluna BP, van Vught LA, Zwinderman AH, Wiewel MA, Davenport EE, Burnham KL, et al. Classification of patients with sepsis according to blood genomic endotype: a prospective cohort study. Lancet Respir Med. 2017;5(10):816–26.

Tsalik EL, Langley RJ, Dinwiddie DL, Miller NA, Yoo B, van Velkinburgh JC, et al. An integrated transcriptome and expressed variant analysis of sepsis survival and death. Genome Med. 2014;6(11):111.

Baghela A, Pena OM, Lee AH, Baquir B, Falsafi R, An A, et al. Predicting sepsis severity at first clinical presentation: the role of endotypes and mechanistic signatures. EBioMedicine. 2022;75:103776.

Fiorino C, Liu Y, Henao R, Ko ER, Burke TW, Ginsburg GS, et al. Host gene expression to predict sepsis progression. Crit Care Med. 2022;50(12):1748–56.

Cano-Gamez E, Burnham KL, Goh C, Allcock A, Malick ZH, Overend L, et al. An immune dysfunction score for stratification of patients with acute infection based on whole-blood gene expression. Sci Transl Med. 2022;14(669):eabq4433.

Silva KAS, Emter CA. Large animal models of heart failure. JACC Basic Trans Sci. 2020;5(8):840–56.

Tanamati C, Monachini M, Cantarelli M, Khouri P, Amarante G, Martins P, et al. Cardiogenic shock: an experimental animal model. Crit Care. 2005;9(Suppl 2):P10.

Article   PubMed Central   Google Scholar  

Wang Y, Polten F, Jäckle F, Korf-Klingebiel M, Kempf T, Bauersachs J, et al. A mouse model of cardiogenic shock. Cardiovasc Res. 2021;117(12):2414–5.

Rienzo M, Imbault J, El Boustani Y, Beurton A, Carlos Sampedrano C, Pasdois P, et al. A total closed chest sheep model of cardiogenic shock by percutaneous intracoronary ethanol injection. Sci Rep. 2020;10(1):12417.

Braga D, Barcella M, Herpain A, Aletti F, Kistler EB, Bollen Pinto B, et al. A longitudinal study highlights shared aspects of the transcriptomic response to cardiogenic and septic shock. Crit Care. 2019;23(1):414.

Wang Y, Chen Y, Zhang T. Integrated whole-genome gene expression analysis reveals an atlas of dynamic immune landscapes after myocardial infarction. Front Cardiovasc Med. 2023;10:1087721.

Tehrani BN, Drakos SG, Billia F, Batchelor WB, Luk A, Stelling K, et al. The multicenter collaborative to enhance biologic understanding, quality, and outcomes in cardiogenic shock (VANQUISH Shock): rationale and design. Can J Cardiol. 2022;38(8):1286–95.

Heijnen NFL, Hagens LA, Smit MR, Schultz MJ, van der Poll T, Schnabel RM, et al. Biological subphenotypes of acute respiratory distress syndrome may not reflect differences in alveolar inflammation. Physiol Rep. 2021;9(3): e14693.

Moss J, Magenheim J, Neiman D, Zemmour H, Loyfer N, Korach A, et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free DNA in health and disease. Nat Commun. 2018;9(1):5068.

Rhodes A, Cecconi M. Cell-free DNA and outcome in sepsis. Crit Care. 2012;16(6):170.

Jackson Chornenki NL, Coke R, Kwong AC, Dwivedi DJ, Xu MK, McDonald E, et al. Comparison of the source and prognostic utility of cfDNA in trauma and sepsis. Intensive Care Med Exp. 2019;7(1):29.

Garnacho-Montero J, Huici-Moreno MJ, Gutiérrez-Pizarraya A, López I, Márquez-Vácaro JA, Macher H, et al. Prognostic and diagnostic value of eosinopenia, C-reactive protein, procalcitonin, and circulating cell-free DNA in critically ill patients admitted with suspicion of sepsis. Crit Care. 2014;18(3):R116.

Yokokawa T, Misaka T, Kimishima Y, Shimizu T, Kaneshiro T, Takeishi Y. Clinical significance of circulating cardiomyocyte-specific cell-free DNA in patients with heart failure: a proof-of-concept study. Can J Cardiol. 2020;36(6):931–5.

Zemmour H, Planer D, Magenheim J, Moss J, Neiman D, Gilon D, et al. Non-invasive detection of human cardiomyocyte death using methylation patterns of circulating DNA. Nat Commun. 2018;9(1):1443.

Peretz A, Loyfer N, Piyanzin S, Ochana BL, Neiman D, Magenheim J, et al. The DNA methylome of human vascular endothelium and its use in liquid biopsies. Med. 2023;4(4):263-281.e4.

Lim N, Dubois MJ, De Backer D, Vincent JL. Do all nonsurvivors of cardiogenic shock die with a low cardiac index?*. Chest. 2003;124(5):1885–91.

El Sibai R, Bachir R, El Sayed M. ECMO use and mortality in adult patients with cardiogenic shock: a retrospective observational study in U.S. hospitals. BMC Emerg Med. 2018;18(1):20.

Xiao W, Mindrinos MN, Seok J, Cuschieri J, Cuenca AG, Gao H, et al. A genomic storm in critically injured humans. J Exp Med. 2011;208(13):2581–90.

Neyton LPA, Zheng X, Skouras C, Doeschl-Wilson A, Gutmann MU, Uings I, et al. Molecular patterns in acute pancreatitis reflect generalizable endotypes of the host response to systemic injury in humans. Ann Surg. 2022;275(2):e453–62.

Orfanos SE, Kotanidou A, Glynos C, Athanasiou C, Tsigkos S, Dimopoulou I, et al. Angiopoietin-2 is increased in severe sepsis: correlation with inflammatory mediators. Crit Care Med. 2007;35(1):199–206.

Ricciuto DR, Dos Santos CC, Hawkes M, Toltl LJ, Conroy AL, Rajwans N, et al. Angiopoietin-1 and angiopoietin-2 as clinically informative prognostic biomarkers of morbidity and mortality in severe sepsis*. Crit Care Med. 2011;39(4):702–10.

Pöss J, Fuernau G, Denks D, Desch S, Eitel I, De Waha S, et al. Angiopoietin-2 in acute myocardial infarction complicated by cardiogenic shock-a biomarker substudy of the IABP-SHOCK II-Trial: angiopoietin-2 in acute myocardial infarction complicated by cardiogenic shock. Eur J Heart Fail. 2015;17(11):1152–60.

Frydland M, Ostrowski SR, Møller JE, Hadziselimovic E, Holmvang L, Ravn HB, et al. Plasma concentration of biomarkers reflecting endothelial cell- and glycocalyx damage are increased in patients with suspected ST-elevation myocardial infarction complicated by cardiogenic shock. Shock. 2018;50(5):538–44.

Zarbock A, Nadim MK, Pickkers P, Gomez H, Bell S, Joannidis M, et al. Sepsis-associated acute kidney injury: consensus report of the 28th Acute Disease Quality Initiative workgroup. Nat Rev Nephrol. 2023;19(6):401–17.

Merdji H, Levy B, Jung C, Ince C, Siegemund M, Meziani F. Microcirculatory dysfunction in cardiogenic shock. Ann Intensive Care. 2023;13(1):38.

Tolppanen H, Rivas-Lasarte M, Lassus J, Sans-Roselló J, Hartmann O, Lindholm M, et al. Adrenomedullin: a marker of impaired hemodynamics, organ dysfunction, and poor prognosis in cardiogenic shock. Ann Intensive Care. 2017;7(1):6.

Dahabreh IJ, Kazi DS. Toward personalizing care: assessing heterogeneity of treatment effects in randomized trials. JAMA. 2023;329(13):1063.

Goligher EC, Lawler PR, Jensen TP, Talisa V, Berry LR, Lorenzi E, et al. Heterogeneous treatment effects of therapeutic-dose heparin in patients hospitalized for COVID-19. JAMA. 2023;329(13):1066.

Writing Committee for the REMAP-CAP Investigators, Florescu S, Stanciu D, Zaharia M, Kosa A, Codreanu D, et al. Effect of angiotensin-converting enzyme inhibitor and angiotensin receptor blocker initiation on organ support-free days in patients hospitalized with COVID-19: a randomized clinical trial. JAMA. 2023;329(14):1183.

Russell CD, Baillie JK. Treatable traits and therapeutic targets: Goals for systems biology in infectious disease. Curr Opin Syst Biol. 2017;2:140–6.

Stanski NL, Wong HR. Prognostic and predictive enrichment in sepsis. Nat Rev Nephrol. 2020;16(1):20–31.

Shankar-Hari M, Rubenfeld GD. The use of enrichment to reduce statistically indeterminate or negative trials in critical care. Anaesthesia. 2017;72(5):560–5.

Gordon AC, Mason AJ, Thirunavukkarasu N, Perkins GD, Cecconi M, Cepkova M, et al. Effect of early vasopressin vs norepinephrine on kidney failure in patients with septic shock: the VANISH randomized clinical trial. JAMA. 2016;316(5):509–18.

Wong HR, Hart KW, Lindsell CJ, Sweeney TE. External corroboration that corticosteroids may be harmful to septic shock endotype a patients. Crit Care Med. 2021;49(1): e98.

Dellinger RP, Bagshaw SM, Antonelli M, Foster DM, Klein DJ, Marshall JC, et al. Effect of targeted polymyxin B Hemoperfusion on 28-day mortality in patients with septic shock and elevated endotoxin level: the EUPHRATES randomized clinical trial. JAMA. 2018;320(14):1455.

Klein DJ, Foster D, Walker PM, Bagshaw SM, Mekonnen H, Antonelli M. Polymyxin B hemoperfusion in endotoxemic septic shock patients without extreme endotoxemia: a post hoc analysis of the EUPHRATES trial. Intensive Care Med. 2018;44(12):2205–12.

Fleuriet J, Heming N, Meziani F, Reignier J, Declerq PL, Mercier E, et al. Rapid rEcognition of COrticosteRoiD resistant or sensitive Sepsis (RECORDS): study protocol for a multicentre, placebo-controlled, biomarker-guided, adaptive Bayesian design basket trial. BMJ Open. 2023;13(3): e066496.

Santhakumaran S, Gordon A, Prevost AT, O’Kane C, McAuley DF, Shankar-Hari M. Heterogeneity of treatment effect by baseline risk of mortality in critically ill patients: re-analysis of three recent sepsis and ARDS randomised controlled trials. Crit Care. 2019;23(1):156.

Senn S. Individual response to treatment: is it a valid assumption? BMJ. 2004;329(7472):966–8.

Senn S. Statistical pitfalls of personalized medicine. Nature. 2018;563(7733):619–21.

Gewandter JS, McDermott MP, He H, Gao S, Cai X, Farrar JT, et al. Demonstrating heterogeneity of treatment effects among patients: an overlooked but important step toward precision medicine. Clin Pharmacol Ther. 2019;106(1):204–10.

Lawler PR, Hochman JS, Zarychanski R. What are adaptive platform clinical trials and what role may they have in cardiovascular medicine? Circulation. 2022;145(9):629–32.

Download references

Acknowledgements

AGP is funded by a Medical Research Council Clinical Academic Research Partnership Award. Ref: MR/W03011X/1 and the Barts Charity. MB is funded by Barts Charity.

Author information

Marie Buckel, Patrick Maclean have authors contributed equally.

Authors and Affiliations

Department of Perioperative Medicine, Bart’s Heart Centre, St. Bartholomew’s Hospital, London, UK

Marie Buckel & Alastair G. Proudfoot

Wellcome Centre for Human Genetics, University of Oxford, Oxford, UK

Patrick Maclean & Julian C. Knight

Chinese Academy of Medical Sciences Oxford Institute, University of Oxford, Oxford, UK

Julian C. Knight

Peter Munk Cardiac Centre, University Health Network, University of Toronto, Toronto, ON, Canada

Patrick R. Lawler

McGill University Health Centre, McGill University, Montreal, QC, Canada

Queen Mary University of London, London, UK

Alastair G. Proudfoot

You can also search for this author in PubMed   Google Scholar

Contributions

AGP conceived the original review. MB and PM contributed equally to writing the first draft. All authors reviewed subsequent and final drafts.

Corresponding author

Correspondence to Alastair G. Proudfoot .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Availability of data and materials, competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Buckel, M., Maclean, P., Knight, J.C. et al. Extending the ‘host response’ paradigm from sepsis to cardiogenic shock: evidence, limitations and opportunities. Crit Care 27 , 460 (2023). https://doi.org/10.1186/s13054-023-04752-8

Download citation

Received : 18 September 2023

Accepted : 20 November 2023

Published : 27 November 2023

DOI : https://doi.org/10.1186/s13054-023-04752-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Critical illness
• Inflammation
• Host response
• Immune dysregulation

Critical Care

ISSN: 1364-8535

• Submission enquiries: [email protected]

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

Recent advances in understanding and managing sepsis

Daniela berg.

1 Department of Anesthesia, Critical Care Medicine, and Pain Management, Vivantes – Klinikum Neukoelln, Berlin, Berlin, Germany

Herwig Gerlach

The last two to three years provided several “big steps” regarding our understanding and management of sepsis. The increasing insight into pathomechanisms of post-infectious defense led to some new models of host response. Besides hyper-, hypo-, and anti-inflammation as the traditional approaches to sepsis pathophysiology, tolerance and resilience were described as natural ways that organisms react to microbes. In parallel, huge data analyses confirmed these research insights with a new way to define sepsis and septic shock (called “Sepsis-3”), which led to discussions within the scientific community. In addition to these advances in understanding and defining the disease, follow-up protocols of the initial “sepsis bundles” from the Surviving Sepsis Campaign were created; some of them were part of quality management studies by clinicians, and some were in the form of mandatory procedures. As a result, new “bundles” were initiated with the goal of enabling standardized management of sepsis and septic shock, especially in the very early phase. This short commentary provides a brief overview of these two major fields as recent hallmarks of sepsis research.

Introduction

Currently, sepsis and septic shock with subsequent multi-organ failure are the leading causes of death in adult intensive care units (ICUs). Although surgical and pharmacological approaches in sepsis therapy are continually improving, epidemiological studies show an increased incidence of sepsis over the last 20 years 1 . In the past few decades, the high prevalence of sepsis and its high economic impact have led to the development of several projects intended to allow better recognition and more accurate description of the course of the disease 2 .

Sepsis is one of the oldest described illnesses. The term sepsis is derived from the ancient Greek term “σῆψις” (“make rotten”) and was used by Hippocrates around 400 BCE to describe the natural process through which meat decays and swamps release decomposing gases but also through which infected wounds become purulent 3 . After this recognition, it took over 2,000 years to establish the hypothesis that it is not the pathogen itself but rather the host response that is responsible for the symptoms seen in sepsis 4 .

In the last 40 years, one major field of sepsis research was the basic cellular and molecular biology to understand the exact mechanisms behind why the body sometimes reacts with an overwhelming inflammation to infections but sometimes does not. Recent research will be described in the first part of the following text, which gives a possible response to this question.

Clear definitions are of great importance in the medical field, as appropriate treatment of illness demands a correct preceding diagnosis. This is not always simple, and, particularly in emergency and intensive care medicine, fast and reliable diagnosis is needed to treat acute illness. The challenge of fast diagnosis of sepsis is that this syndrome is based on highly complex pathophysiological pathways that may show varying clinical signs and symptoms. Therefore, a brief review of former and new definitions of sepsis will follow; it should be interpreted in the context of the newly described approaches to pathophysiology.

Fast detection and initial treatment of sepsis are of utmost importance. Since 2004, the Surviving Sepsis Campaign (SSC) has developed several guidelines for the management of sepsis and septic shock. From 2005 to 2010, “sepsis bundles” were tested to demonstrate that a protocolized approach in the initial phase of the disease leads to a better outcome. Since this large trial, several similar approaches have been published, and recent articles confirmed the importance of time until treatment as a prognostic factor for patients. These studies as well as the newly described bundles are part of the second section of this brief review.

Old and new approaches to understand the disease “sepsis”

The synonym of sepsis, “blood poisoning”, which has been used for centuries and is still popular among the non-medical population, is an inadequate term for intensive care specialists. A teleological definition was proposed by Hugo Schottmüller in 1914: “Sepsis is present if a focus has developed from which pathogenic bacteria, constantly or periodically, invade the blood stream in such a way that this causes subjective and objective symptoms” 5 . This definition is problematic and increasingly being dismissed, as it is based on subjective clinical observations. In addition, it insinuates an incorrect pathophysiological rationale, as it assumes that bacteria themselves spread. However, today one assumes that the body produces its own transmitters as a response to the infection and that these spread systemically, thus affecting peripheral organs 4 .

In local infections, a normal inflammatory host response controls the focus; a dysregulation of the host response, however, leads to macrocirculatory and microcirculatory failure, by which single or multiple organ failure is induced 6 . The lungs, kidneys, and cardiovascular system are the most affected organs during sepsis and septic shock 7 . This, however, is based on clinical assessment of organ function by routine biomarkers and might not withstand a thorough check by modern cell biology tools. Furthermore, it still does not give an answer to the key question of why some patients (for instance, in cases of severe pneumonia or meningitis due to Streptococcus pneumoniae ) react excessively in terms of hyper-inflammation and “cytokine storm” (see below) whereas others have no symptoms although the same microbe can be detected on their skin or upper airways 8 . The “old” theory is that the latter persons are simply “resistant” (that is, their inflammatory response keeps the contamination under control). However, there are two phenomena which contradict this theory. First, if carriers without symptoms have such a strong resistance, why are they still carriers? Second, if these persons have such a successful “inflammatory response”, why is it not possible to demonstrate this with serum biomarkers?

In a landmark article, Weis et al. describe a biological pathway for how this may be declared 9 : it was found that blood glucose levels influence the mechanisms of “tolerance” against infections. “Tolerance” (or “resilience”) is a form of “defense strategy against infection that preserves host homeostasis without exerting a direct negative impact on pathogens” 9 . In other words, the host organism coexists with the microbes; of course, this may change if this tolerance is disturbed (“dysregulated”) by, for example, other infections, pregnancy, splenectomy (in case of Streptococcus pneumoniae ), or older age. Interestingly, in many of those cases where this disease tolerance fails, the clinical symptoms of sepsis often exert much more dramatic courses than “classical” infections. These recent findings about dysregulation in the pathogenesis of sepsis perfectly fit the new definition (see next section). Finally, the option to differentiate between the individual “type of host response” may foster research to enable the practice of more personalized medicine in patients with sepsis as was suggested for other life-threatening diseases such as acute respiratory distress syndrome 10 .

From understanding to defining sepsis

As already mentioned, the former understanding of sepsis as a hyper-inflammatory response to infection, often accompanied by a fast “cytokine storm” 11 , was the basis for former sepsis definitions: the US critical care specialist Roger Bone organized a consensus conference in 1992 and suggested that the sepsis definition include the aspect of host response. Here, the term systemic inflammatory response syndrome (SIRS), which is still commonly used, was defined 12 . If SIRS occurs without infection (for example, in the case of burns and pancreatitis and in the post-operative setting), the condition is defined only as SIRS; similarly, an infection without SIRS does not equal sepsis. Only when the two criteria are seen in combination can the diagnosis of sepsis be made. Therefore, sepsis was defined as “a systemic inflammatory response syndrome to infection” that may be seen when two or more of the following criteria are fulfilled: heart rate of more than 90 beats/minute, core temperature of more than 38°C or less than 36°C, respiratory rate of more than 20 breaths/minute or partial pressure of carbon dioxide (PaCO 2 ) of less than 32 mmHg, and white blood cell count of more than 12,000/mL or less than 4,000/mL or more than 10% immature neutrophils.

This definition of sepsis (“Sepsis-1”) is still most commonly used. Merely 10 years after the consensus conference hosted by Bone et al . 12 , several experts met in Washington, D.C., to discuss a new definition of sepsis. The experts reached the following conclusions:

• 1. The current concepts of sepsis and septic shock seem, in principle, useful for clinical routine and research.
• 2. These definitions, however, do not allow precise staging of patients or prediction of host response and infection.
• 3. Although SIRS is a useful approach, criteria are too sensitive and non-specific.
• 4. An elaborate list of signs and symptoms of sepsis would better reflect the clinical response to a systemic infection.
• 5. A hypothetical model should be developed that better stages sepsis, better characterizes sepsis on the basis of predisposing factors and comorbidities, better reflects the type of original infection, better describes the host response, and better quantifies the extent of resulting organ dysfunction.

In this manner, a classification system allowing the stratification of patients with sepsis, “PIRO” (today called “Sepsis-2”), was developed at this conference in 2001 13 . “P” stands for predisposition, “I” for type and extent of the primary insult (in the case of sepsis, primary infection), “R” for type and extent of host response, and “O” for extent of organ dysfunction. The benefit of the PIRO model is that it enabled one to separate morbidity due to the infection itself from secondary morbidity that develops through the host response. The introductions of a PIRO model, however, remained theoretical, even though there have been several attempts to introduce a point system that enables scoring of patients with sepsis 14 .

In a roughly two-year process with extended and complex biometric evaluations, a new approach (“Sepsis-3”) was developed that is based on patient data from several validated sources and that was published in the form of three articles in 2016 15 – 17 . What is new in this concept? On the one hand, it is the omission of SIRS as a factor in the definition. The new Sepsis-3 defines sepsis as “a life-threatening organ dysfunction caused by a dysregulated host response to infection” 15 . Therefore, if no organ dysfunction is seen, one may speak only of an infection, not of “sepsis”. The term “severe sepsis” is superfluous, as its criteria (organ dysfunction) are already included in the new definition of sepsis. The term “septic shock” remains; however, it now also includes an elevated lactate level of more than 2 mmol/L as an additional factor.

Part of the new Sepsis-3 definitions is SOFA as a grading score for defining acute organ dysfunction (“Sequential [Sepsis-related] Organ Failure Assessment Score”, or SOFA score) 18 . This score allocates points according to pathological change in six different organ systems; an increase in the total SOFA score by at least two points (with negative patient history, a score of 0 is assumed) indicates acute organ dysfunction, and the diagnosis of sepsis is met if an infection is identified in parallel. If, in addition, hypotension is seen (that is, mean blood pressure of at least 65 mmHg can be reached only using vasopressors, despite adequate fluid management) and the serum lactate levels are more than 2 mmol/L, one speaks of “septic shock”. Sepsis-3 also provides a new tool meant as a simplified screening tool for early recognition of organ dysfunction because of infection: qSOFA (“quickSOFA”) 15 . It is intended primarily for use in emergency departments, peripheral wards, rest homes, and so on and not in ICUs, and it consists of the following three criteria:

• • altered mental status
• • respiratory rate of more than 22 breaths/minute
• • systolic blood pressure of less than 100 mmHg.

When two of these three qSOFA criteria are met, organ dysfunction should be suspected, and the classic SOFA score should be determined by experienced physicians, usually intensive care specialists.

Sepsis without “SIRS”: is it feasible?

In 2015, an Australian study was published that used a large database to determine the influence of SIRS on prognosis 19 . The results may be summarized as follows: first, the presence of SIRS (defined by two or more SIRS criteria) did not influence the overall prognosis; second, about every eighth patient is missed if SIRS is necessary for defining sepsis; third, even though an increasing trend in mortality was seen according to increasing numbers of observed SIRS criteria, no significant difference in mortality rate could be identified, especially comparing patients with zero versus one or one versus two SIRS criteria 19 . These results were confirmed by the Sepsis-3 task force; the basis for this consisted of large data sets of hospitalized patients with suspected infection that were used to assess the validity of several diagnostic and critical care scores. Core results were the low specificity of SIRS (whether in critical care or on peripheral wards), the high prognostic value of the SOFA score in ICUs, and the high prognostic value of a change in cognitive status, respiratory rate, and systolic pressure (qSOFA), particularly in non-ICU patients 16 .

In conclusion, the weakness of the old, SIRS-based Sepsis-1 definition is obvious and was demonstrated by high-level scientific research 19 . Hence, we actually do need a new definition and should no longer use the “SIRS yes or no” criterion for defining sepsis. The new Sepsis-3 definition is based on sound and extended statistical analyses, thus providing a good basis for use in future clinical research 15 – 17 . Furthermore, qSOFA was demonstrated to be a useful tool but needs further validation studies. Finally, the “SIRS” criteria (put in quotation marks, since the authors think that the former entity SIRS should no longer be used) leukocytes, heart rate, and temperature are still important to detect infections but should no longer be used to diagnose sepsis. These two sides of the coin—qSOFA as a screening tool in suspected organ dysfunction plus leukocytes, heart rate, and temperature for surveillance in patients endangered by infections—should be established as a clinical quality standard.

Quality improvement in sepsis management

Around 15 years ago, three international research societies founded the SSC, aiming to reduce sepsis mortality by more than 25% (relative risk reduction) over 5 years. One tool was the creation of international guidelines for the management of sepsis; the first version was published in 2004 and the third revision in 2017 20 . Based on the first version, “SSC Sepsis Bundles” were created, consisting of several measurable interventions (for example, antibiotics, blood cultures, and serum lactate measurement). According to the initial plan, these bundles were tested in more than 30,000 cases of sepsis and septic shock worldwide over a period of 5 years (2005–2010). As a result, it was demonstrated that (1) the planned mortality reduction was reached and (2) not all parts of the bundles had an intrinsic effect on patient outcome 21 .

In parallel, several comparable projects were started all over the world—for example, in Spain (“Edusepsis Group”)—which led to similar results 22 . In 2014, a follow-up study over 7.5 years from the SSC Bundle Project confirmed the data from 2010, and the authors concluded that “These results demonstrate that performance metrics can drive change in clinical behavior, improve quality of care, and may decrease mortality in patients with severe sepsis and septic shock” 23 . Recently, a German group published data, again from a 7.5-year period using a hospital-supported quality improvement program to reduce sepsis mortality 24 . In more than 14,000 included patients, 90-day mortality decreased significantly, from 64.2% to 45.0%, and the length of stay in the hospital decreased from 44 to 36 days.

Recent development in sepsis management

The aforementioned improvement in survival of patients with sepsis and septic shock by standardized protocols and related control instruments (“standard operating procedures”, “check lists”, and so on) led to a broad discussion of whether these protocols should be a mandatory quality indicator. Based on private activities from a New York state (USA) family that was affected by a lethal case of sepsis, the New York State Department of Health in 2013 decided that all state hospitals have to implement evidence-informed protocols for the fast management of sepsis and septic shock (New York Codes, Rules, and Regulations parts 405.2 and 405.4). The way this was carried out could be decided by the hospitals themselves, but the minimum requirement was a 3-hour bundle with the following interventions: (1) blood cultures before administration of antibiotics, (2) serum lactate measurement, and (3) infusion of broad-spectrum antibiotics. Although in the treatment of patients with sepsis there are many more options that provided a beneficial effect (for example, protective mechanical ventilation with low tidal volumes), these regulations concentrated on early treatment, within hours after the detection of sepsis, and therefore included mainly emergency departments.

In 2017, the results from the first 2.25 years after starting these rules were published, presenting data from 149 hospitals, including more than 49,000 patients 25 ; 82.5% of these met the criteria for the 3-hour bundle. Furthermore, it was demonstrated that each 1-hour delay—measured from the initial time of detecting sepsis—increased mortality by 4% (relative risk). Similar results were found for the single interventions of blood culture, antibiotics, and lactate measurement, whereas the effect of early fluid administration was demonstrated only in septic shock patients with a need for vasopressor administration 25 . This latter point supports current discussions that an early fluid challenge might not be favorable in every patient with sepsis (so-called “fluid non-responder”) and that fluid administration should be monitored carefully to avoid a fluid overload with negative effects on patient outcome.

In regard to the effect of early antibiotics, these data did not support the concern that there might be a risk of increasing antibiotic resistance as pointed out by Singer 26 . Furthermore, a recent article by Ferrer et al . demonstrated that an improvement of a more rapid microbiological diagnosis in parallel with early antibiotics facilitates selection of antibiotics and improves outcome 27 . In contrast, new data show that for the source control of infection, the identification of the location, rather than time, might be the most important parameter of improved outcome. Martínez et al . revealed data that outcome may vary according to the source of infection and that urinary tract infection with subsequent sepsis is associated with a lower mortality compared with severe pneumonia 28 . Finally, the microorganism’s virulence and bacterial load may influence the prognosis of sepsis 29 .

Based on these impressive data, the SSC steering group recently published newly defined “SSC Sepsis Bundles 2018”, which are now based on a 1-hour period 30 :

• • Measure lactate level. Re-measure if the initial level is more than 2 mmol/L.
• • Obtain blood cultures prior to the administration of antibiotics.
• • Administer broad-spectrum antibiotics.
• • Begin rapid administration of 30 mL/kg crystalloid for hypotension or lactate of at least 4 mmol/L.
• • Apply vasopressors if the patient is hypotensive during or after fluid administration to maintain mean arterial pressure of at least 65 mmHg.

“Time zero” or “time of presentation” is defined as the time of triage in the emergency department or, if presenting from another care venue, from the earliest chart annotation consistent with all elements of sepsis (formerly severe sepsis) or septic shock ascertained through chart review.

Since these new bundles were published very recently, no validation studies have been performed so far. However, the discussion started within days; many clinicians are very critical of this “mandatory approach”. Two major points (but not the only ones) are that (1) it is presumed that hospitals will perform these bundles more or less “in any case” of sepsis suspicion, even if there may be some time to wait for a more thorough diagnosis, since they fear reduced reimbursement or even legal action, and that (2) the possible “over-therapy”, especially with early administration of antibiotics, may induce side effects for the treated patients as well as a higher rate of resistance against antibiotics over time.

Closing remarks

In patients with sepsis or septic shock, a better understanding of the host response leading to the clinical course, a faster detection of high-risk patients, and an earlier and more standardized approach in managing sepsis are the key challenges in current clinical practice. On the one hand, the recent findings of host defense mechanisms on the cellular level, the new Sepsis-3 definition, and the current developments after investigating the effects of mandatory care of patients with sepsis are significant and promising steps in sepsis research. On the other hand, these steps are tracking new ways which—in some cases—may lead to unknown destinations. At present, it is too early to risk a clear prognosis; we all hope that the sum effect of these new developments will be a positive one. At least it was demonstrated that advances in sepsis research are possible! Perhaps this will foster research engagement by clinicians and scientists in this exciting field of medicine and will bring more attention and support from industrial as well as public institutions.

[version 1; referees: 3 approved]

Funding Statement

The author(s) declared that no grants were involved in supporting this work.

Editorial Note on the Review Process

F1000 Faculty Reviews are commissioned from members of the prestigious F1000 Faculty and are edited as a service to readers. In order to make these reviews as comprehensive and accessible as possible, the referees provide input before publication and only the final, revised version is published. The referees who approved the final version are listed with their names and affiliations but without their reports on earlier versions (any comments will already have been addressed in the published version).

The referees who approved this article are:

• Michael Bauer , Center for Sepsis Control and Care, Jena University Hospital, Jena, Germany No competing interests were disclosed.
• Steven M Opal , Infectious Disease Division, Alpert Medical School of Brown University, Providence, Rhode Island, USA No competing interests were disclosed.
• Antonio Artigas , Critical Care Center, Sabadell Hospital, Corporació Sanitària Universitària Parc Taulí, Universitat Autònoma de Barcelona, Barcelona, Spain No competing interests were disclosed.

Analysis of mild and severe neonatal enterovirus infections in a Chinese neonatal tertiary center: a retrospective case-control study

• Original Article
• Published: 12 April 2024

Cite this article

• Junshuai Li 1 , 2 ,
• Jingjing Xie 1 , 2 ,
• Min Zhang 1 , 2 ,
• Zhuojun Xiao 1 , 2 ,
• Fan Zhang 1 , 2 ,
• Weiqing Huang 1 , 2 ,
• Yong Zhou 1 , 2 ,
• Weiqun Yan 1 , 2 ,
• Rong Zhang 1 , 2 &
• Xiaoming Peng   ORCID: orcid.org/0000-0003-3635-6554 1 , 2  

2 Altmetric

Explore all metrics

To compare the clinical characteristics, virus serotype, and outcome in cases of mild and severe enteroviral infection at a tertiary neonatal intensive care unit in China.

A retrospective analysis of cases hospitalized between June and August 2019. Samples (stool or throat swabs) were examined using reverse transcription polymerase chain reaction. Positive cases were divided into two groups: mild infection and severe infection.

A total of 149 cases were assigned to one of two groups: mild infection ( n  = 104) and severe infection ( n  = 45). There were no significant differences between the groups in terms of sex, gestational age, birth weight, mode of delivery, and onset within 7 days. Clinical symptoms in both groups mostly resembled sepsis (fever, rash, poor feeding, and lethargy); however, there were significant variations in concomitant symptoms such as hepatitis, thrombocytopenia, encephalitis, coagulopathy, and myocarditis. Severe cases were more likely to have abnormal complete blood counts, biochemical parameters, and cerebrospinal fluid markers. The predominant serotypes implicated in neonatal enterovirus infections were echoviruses and Coxsackievirus B. Invasive ventilation, intravenous immunoglobulin, vasoactive medications, and blood product transfusions were often required, with high mortality rates among severe cases.

We found significant differences between mild and severe cases of neonatal enterovirus infection with respect to complications, laboratory findings, and enterovirus serotypes. It is crucial to exercise caution when newborns exhibit symptoms of sepsis, during an enterovirus outbreak. Anemia, thrombocytopenia, abnormal liver function, and coagulation dysfunction should be monitored closely as they could indicate the presence of a severe enteroviral infection.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Data availability

The datasets generated during and/or analysed during the current study are not publicly available due to ethical restrictions but are available from the corresponding author on reasonable request.

Baoming LMF,Alexandra, Valsamakis et al Optimization and evaluation of a novel real-time RT-PCR test for detection of parechovirus in cerebrospinal fluid. [J].J Virol Methods,2019,272:113690. https://doi.org/10.1016/jviromet.2019.113690.6

Baoming, Liu et al (2016) Universal PCR primers are critical for direct sequencing-based enterovirus genotyping. J Clin Microbiol 55(1):339–340. https://doi.org/10.1128/JCM.01801-16

Article   Google Scholar  

Benjamin M, Liu,Sarah B, Mulkey,Joseph M, Campos et al (2023 Dec) Laboratory diagnosis of CNS infections in children due to emerging and re-emerging neurotropic viruses. Pediatr Res 2. https://doi.org/10.1038/s41390-023-02930-6

Elisa Civardi,Chryssoula Tzialla,Fausto Baldanti (2013) Viral outbreaks in neonatal intensive care units: what we do not know. [J] Am J Infect Control 41(10):854–856. https://doi.org/10.1016/j.ajic.2013.01.026

Article   PubMed   Google Scholar  

Zin Naing,Benjamin Rayner,Ananthen Killikulangara Prevalence of viruses in stool of premature neonates at a neonatal intensive care unit. [J] J Paediatr Child Health 2013, 49(3):E221–E226. https://doi.org/10.1111/jpc.12113

Mark J, Abzug et al (2004) Presentation, diagnosis, and management of enterovirus infections in neonates. [J] Paediatr Drugs 6(1):1–10. https://doi.org/10.2165/00148581-200406010-0000

Julia C, Haston,Terry C, Dixon et al Nonpolio enterovirus infections in neonates. [J] Pediatr Ann 2015, 44(5):e103–e107. https://doi.org/10.3928/00904481-20150512-09

Chryssoula Tzialla,Elisa Civardi,Alessandro Borghesi Emerging viral infections in neonatal intensive care unit. [J].J matern fetal neonatal Med,2011,24 suppl 1:156–158. https://doi.org/10.3109/14767058.2011.607588

Nada Harik,Roberta LDB et al (2018) Neonatal nonpolio enterovirus and parechovirus infections. [J] Semin Perinatol 42(3):191–197. https://doi.org/10.1053/j.semperi.2018.02.007

Miren Iturriza-Gómara Brian Megson,Jim Gray, Molecular detection and characterization of human enteroviruses directly from clinical samples using RT-PCR and DNA sequencing. [J].J Med Virol,2006,78(2):243JJLKL-53. https://doi.org/10.1002/jmv.20533

Kaplan MH, Klein SW, McPhee J et al Group B coxsackievirus infections in infants younger than three months of age: a serious childhood illness. [J] Rev Infect Dis 1983, 5(6):1019–1032. https://doi.org/10.1093/clinids/5.6.1019

Modlin JF et al Perinatal echovirus infection: insights from a literature review of 61 cases of serious infection and 16 outbreaks in nurseries. [J] Rev Infect Dis 1986, 8(6):918–926. https://doi.org/10.1093/clinids/8.6.918

Lake,B AM, A Lauer JC, Clark et al Enterovirus infections in neonates. [J] J Pediatr 1976, 89(5):787–791. https://doi.org/10.1016/j.jmii.2019.08.018

Hans de Graaf,Emanuela Pelosi,Andrea Cooper, et al. Severe Enterovirus Infections in Hospitalized Children in the South of England: Clinical Phenotypes and Causative Genotypes. [J].Pediatr Infect Dis J,2016,35(7):723-7. https://doi.org/10.1097/INF.0000000000001093

Kristine R, Rittichier,Paul A, Bryan,Kathlene E, Bassett et al Diagnosis and outcomes of enterovirus infections in young infants. [J] Pediatr Infect Dis J 2005, 24(6):546–550. https://doi.org/10.1097/01.inf.0000164810.60080.ad

Malgorzata A, Verboon-Maciolek,Tannette G, Krediet,Leo J, Gerards et al Clinical and epidemiologic characteristics of viral infections in a neonatal intensive care unit during a 12-year period. [J] Pediatr Infect Dis J 2005, 24(10):901–904. https://doi.org/10.1097/01.inf.0000180471.03702.7f

Malgorzata A, Verboon-Maciolek,Tannette G, Krediet,Leo J, Gerards et al Severe neonatal parechovirus infection and similarity with enterovirus infection. [J] Pediatr Infect Dis J 2008, 27(3):241–245. https://doi.org/10.1097/INF.0b013e31815c1b07

Serena Braccio,Melpo Kapetanstrataki,Mike Sharland Intensive care admissions for children with Enterovirus and Human Parechovirus infections in the United Kingdom and The Republic of Ireland, 2010–2014. [J].Pediatr Infect Dis J,2017,36(3):339–342. https://doi.org/10.1097/INF.0000000000001430

Romero JR, Modlin JF (2015) Coxsackieviruses, echoviruses and numbered enteroviruses (EV-D68). Mandell, Douglas, and Bennett’s principles and practice of infectious diseases. 8thed. Churchill Livingstone, Philadelphia, PA, p 2080e90

Google Scholar  

Abzug MJ (2016) Nonpolio enteroviruses. Nelson’s textbook of pediatrics, 20th edn. Elseiver, Philadelphia, PA, p 1561e1568e1

Hsieh T-YLH-TKS-H et al Neonatal enterovirus infections: emphasis on risk factors of severe and fatal infections. [J] Pediatr Infect Dis J 2003, 22(10):889–894. https://doi.org/10.1097/01.inf.0000091294.63706.f3

Krajden S, Middleton PJ et al Enterovirus infections in the neonate. [J].Clin Pediatr (Phila),1983,22(2):87–92. https://doi.org/10.1177/000992288302200201

American Academy of Pediatrics. Human parechovirus infections. In: Kimberlin DW, Brady MT, Jackson MA, Long SS, et al., eds. American Academy of Pediatrics. Red Book®: 2015 Report of the Committee on Infectious Diseases et al (2015) 592–593

Mark J, Abzug et al (2004) Presentation, diagnosis, and management of enterovirus infections in neonates. [J] Paediatr Drugs 6(1):1–10. https://doi.org/10.2165/00148581-200406010-00001

José R, Romero,Rangaraj, Selvarangan et al (2011) The human parechoviruses: an overview. [J] Adv Pediatr 58(1):65–85. https://doi.org/10.1016/j.yapd.2011.03.008

Huang S-HLY-CHC-G et al Clinical and epidemiologic features of Coxsackievirus A6 infection in children in northern Taiwan between 2004 and 2009. [J].J Microbiol Immunol Infect,2011,44(4):252–257. https://doi.org/10.1016/j.jmii.2011.01.031

Lin Y-PHT-LLT-H et al Molecular and epidemiological study of enterovirus D68 in Taiwan. [J] J Microbiol Immunol Infect 2017, 50(4):411–417. https://doi.org/10.1016/j.jmii.2015.07.015

Hsieh H-YWT-KYJ-Y et al Updates on the molecular epidemiology of Enterovirus D68 after installation of screening test among acute flaccid paralysis patients in Taiwan. [J].J Microbiol Immunol Infect,2018,51(5):688–691. https://doi.org/10.1016/j.jmii.2017.12.001

Hsu T-YYY-PHY-L et al A case of recombinant coxsackievirus A2 infection with neurological complications in Taiwan. [J] J Microbiol Immunol Infect 2017, 50(6):928–930. https://doi.org/10.1016/j.jmii.2016.08.012

Ashley Nino Khetsuriani M, Steven Oberste et al Neonatal enterovirus infections reported to the national enterovirus surveillance system in the United States, 1983–2003. [J].Pediatr Infect Dis J,2006,25(10):889–893. https://doi.org/10.1097/01.inf.0000237798.07462.32

Cherry JD, Krogstad P (2016) Enterovirus, parechovirus, and Saffold virus infections. In: Wilson CB, Victor N, Yvonne M, Remington JS, Klein JO et al (eds) Remington and Klein’s infectious diseases of the Fetus and Newborn Infant, 8th edn. Elsevier Health Sciences, hiladelphia, PA, pp 782–827

American Academy of Pediatrics. Enterovirus (nonpoliovirus). In: Kimberlin DW, Brady MT, Jackson MA, Long SS, et al., eds. American Academy of Pediatrics. Red Book®: 2015 Report of the Committee on Infectious Diseases et al (2015) 333–336

Mark J, Abzug,Marian G, Michaels,Ellen, Wald et al A Randomized, Double-Blind, placebo-controlled trial of Pleconaril for the treatment of neonates with Enterovirus Sepsis. [J].J Pediatric infect dis Soc,2016,5(1):53–62. https://doi.org/10.1093/jpids/piv015

Chen M-HYY-CHM-C et al Effect of intravenous immunoglobulin for neonates with severe enteroviral infections with emphasis on the timing of administration. [J] J Clin Virol ,2015,64:92 – 6. https://doi.org/10.1016/j.jcv.2015.01.013

Download references

This work was supported by Natural Science Foundation of Hunan Province of China(2021JJ40278).

Author information

Authors and affiliations.

Department of Neonatology, Hunan Children’s Hospital, Changsha, China

Junshuai Li, Jingjing Xie, Min Zhang, Zhuojun Xiao, Fan Zhang, Weiqing Huang, Yong Zhou, Weiqun Yan, Rong Zhang & Xiaoming Peng

The School of Pediatrics, Hengyang Medical School, University of South China, Hengyang, Hunan, China

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Junshuai Li, Jingjing Xie, Min Zhang, Zhuojun Xiao, Fan Zhang, Weiqing Huang, Yong Zhou, Weiqun Yan and Rong Zhang. The first draft of the manuscript was written by Junshuai Li and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Xiaoming Peng .

Ethics declarations

Ethics approval.

The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Hunan Children’s Hospital(KY2020-56).

Consent to participate

Written informed consent was obtained from the parents.

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Li, J., Xie, J., Zhang, M. et al. Analysis of mild and severe neonatal enterovirus infections in a Chinese neonatal tertiary center: a retrospective case-control study. Eur J Clin Microbiol Infect Dis (2024). https://doi.org/10.1007/s10096-024-04805-y

Download citation

Received : 25 December 2023

Accepted : 07 March 2024

Published : 12 April 2024

DOI : https://doi.org/10.1007/s10096-024-04805-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Enterovirus infection
• Case-control study
• Virus serotype
• Find a journal
• Publish with us
• Track your research