Spotlight: Overdiagnosis and Delay: Challenges in Sepsis Diagnosis
- Spotlight Case
- Realize the importance of early sepsis diagnosis and treatment.
- Identify challenges in diagnosing sepsis and distinguish common mimickers.
- Review potential etiologies of circulatory shock.
- Recognize the strengths and limitations of different sepsis criteria and clinical decision support tools.
- Understand the role of sepsis alerts and standardized order sets in sepsis management.
A 66-year-old man presented to the emergency department (ED) with generalized weakness and nausea. He was noted to be confused but protecting his airway. His blood pressure was low (80/50 mm Hg), and he was started on intravenous fluids and broad spectrum antibiotics with some initial improvement in his symptoms and blood pressure. Laboratory test results revealed a mildly elevated white count, acute kidney injury, and elevated liver function tests. CT scans of his head and abdomen were ordered, and he was admitted to the medical intensive care unit (ICU) with a presumed diagnosis of septic shock.
The patient's blood pressure continued to trend downward to 60/40 mm Hg. Physicians placed a central line and started him on vasopressors. In reviewing the laboratory tests ordered in the ED, the ICU resident noticed that the patient's troponin level had returned markedly elevated. Initial electrocardiogram (ECG) from the ED revealed some T-wave inversions. A repeat ECG in the ICU showed obvious ST segment elevations suggestive of acute myocardial infarction (AMI). At that point, the ICU physicians came to believe that a large AMI was the likely explanation for the patient's low blood pressure on presentation, not septic shock.
The patient was immediately sent to the cardiac catheterization laboratory. He was found to have complete occlusion of the right coronary artery and a stent was placed to alleviate the blockage. He required ionotropic support to maintain cardiac output and was transferred to the coronary care unit in critical condition. He continued to deteriorate and ultimately experienced a cardiac arrest. He could not be revived and died 12 hours after presenting to the ED.
In the end, clinicians realized that the patient had initially been misdiagnosed with sepsis and that his presenting symptoms and exam findings were most likely caused by AMI and cardiogenic shock. Had the correct diagnosis been made earlier, the outcome might have been different.
Commentary by Ifedayo Kuye, MD, MBA, and Chanu Rhee, MD, MPH
The patient in this case was mistakenly diagnosed with sepsis, the syndrome of life-threatening organ dysfunction caused by a dysregulated host response to infection. Sepsis afflicts approximately 1.7 million adults each year in the United States and may contribute to up to half of all inpatient deaths.(1-3) It is also the most expensive condition treated in hospitals, resulting in an estimated $24 billion in costs each year to the health care system.(4)
Reacting to the high mortality associated with sepsis, as well as several high-profile cases in which patients have experienced preventable morbidity and mortality resulting from a delayed or missed diagnosis of sepsis, state and national regulations have been enacted to facilitate improved inpatient sepsis diagnosis and management. In 2013, in response to a missed diagnosis of sepsis that resulted in the death of Rory Staunton, a 12-year-old boy, New York State established a law requiring hospitals to adopt protocols for the early diagnosis and treatment of sepsis (referred to as Rory's Regulations). Other states have adopted or are considering similar measures. In 2015, the Centers for Medicare and Medicaid Services (CMS) implemented the "SEP-1" quality measure, which requires hospitals to report adherence to sepsis management bundles based on the Surviving Sepsis Campaign guidelines.(5)
These initiatives are supported by existing literature, which suggests that sepsis is a medical emergency in which every hour matters and that compliance with sepsis bundles leads to better outcomes. For example, a retrospective study of patients with septic shock found that each hour of delay in antibiotic administration after the onset of hypotension was associated with a decrease in survival of 7.6%.(6) A more recent analysis of sepsis cases reported under Rory's Regulations in New York demonstrated similar findings; rapid administration of bundled sepsis care was associated with improved survival.(7)
However, as seen in this case, identifying and correctly diagnosing sepsis can be challenging. The clinical signs and symptoms are nonspecific, and there is no gold standard test that confirms the diagnosis. The systemic inflammatory response syndrome (SIRS) criteria developed as part of the 1991 consensus definitions ("Sepsis-1") defined sepsis as infection leading to abnormalities in temperature, heart rate, respiratory rate, and white blood cell count.(8) However, the SIRS criteria have received substantial criticism due to their low specificity. One study found that 15% of patients admitted to a general inpatient ward met SIRS criteria on admission and 47% met SIRS criteria at least once during their hospitalization.(9) The Third International Consensus Definitions for Sepsis and Septic Shock ("Sepsis-3") redefined sepsis as infection leading to organ dysfunction and suggested using greater than or equal to 2 quick Sequential Organ Failure Assessment (qSOFA) score points (1 point each for altered mental status, elevated respiratory rate, and a systolic blood pressure ≤100 mm Hg) as a screening tool based on ease of measurement and strong association with mortality in patients with suspected infection.(1) These criteria are more specific for sepsis but come at the expense of sensitivity, leading to concerns that many sepsis diagnoses could be delayed or missed.(10)
The major challenge, which none of the sepsis definitions address, is the difficulty in determining whether or not patients are infected, particularly in the early stages of presentation. Blood cultures can take 24–48 hours to turn positive, and most patients with sepsis do not have documented bacteremia.(2) Many patients with sepsis, in fact, never have pathogens recovered from any cultures.(11) While fever, tachycardia, and elevated white blood cell count can be seen in numerous other conditions, infections can also present without the cardinal symptoms of fever, leukocytosis, and tachycardia, particularly in patients who are older or immunosuppressed. Although procalcitonin testing can aid in the diagnosis of bacterial infection and sepsis, the test has imperfect sensitivity and specificity.(12)
Central to the pathophysiology of sepsis is the presence of organ dysfunction, such as hypotension, elevated lactate, respiratory failure, altered mental status, renal failure, or coagulopathy, but these are also not specific to sepsis. Although a host of conditions can present like sepsis, one prospective study found that the most common mimickers of septic shock in ICU patients were adverse drug reactions (particularly metformin and antihypertensive drugs), acute mesenteric ischemia, malignancies, inflammatory diseases, adrenal insufficiency, diabetic ketoacidosis, and acute pancreatitis.(13)
In the case described above, the patient presented with nonspecific symptoms of weakness, nausea, and altered mental status. It is easy to say in retrospect that sepsis was the inappropriate diagnosis in this case since the presenting symptoms were not classic for infection. However, nearly 40% of patients with sepsis present with vague symptoms.(14) In addition, the patient described had laboratory evidence of inflammation (elevated white blood cell count) and end-organ dysfunction (elevated liver function tests, acute kidney injury, and shock). Given the current regulatory and clinical push toward aggressive sepsis recognition and management, it is understandable why he was presumed to have sepsis. The real error was prematurely closing off the possibility of alternate diagnoses while concurrently initiating treatment for possible sepsis.
Although an abundance of literature purports the benefits of sepsis performance improvement initiatives and adherence to bundled care, there is scant published literature on the dangers of erroneous diagnoses of sepsis (as occurred in this case). However, one retrospective study of patients admitted to an ICU with a diagnosis of sepsis determined that 43% were unlikely to have actually had an infection.(15) As in this clinical vignette, this study also found that patients who likely did not have infection in retrospect had higher mortality rates, which may have been related to delays in appropriate therapy secondary to an incorrect working diagnosis. For the patient described, the misdiagnosis delayed the correct diagnosis of cardiogenic shock, leading to a concomitant delay in appropriate treatment and likely contributed to his death.
Therefore, when a patient presents with circulatory shock or organ dysfunction, it is important that clinicians do not anchor the diagnosis solely on sepsis even while treating empirically with antibiotics for this possibility. Clinicians must consider the four broad causes of shock: hypovolemic, cardiogenic (e.g., acute myocardial infarction, valvular heart disease, myocarditis, or cardiac arrhythmias), obstructive (e.g., pulmonary embolism, cardiac tamponade, or tension pneumothorax), or distributive (e.g., sepsis, anaphylaxis, or pancreatitis).(16) Performing a full clinical examination for shock, including assessment of skin color, temperature, and jugular venous distention, can aid in diagnosing the correct etiology. In this case, bedside echocardiography may have been helpful in identifying the type of shock as cardiogenic resulting from myocardial infarction rather than sepsis. Echocardiography may have revealed wall motion abnormalities, which would have raised the suspicion for cardiogenic shock. Finally, a 12-lead electrocardiogram should have been performed upon the patient's arrival to the ED, as up to 33% of patients with myocardial infarctions do not present with chest pain.(17)
The challenges in identifying sepsis have spurred the development of various clinical decision support tools, most notably alerts that flag patients based upon real-time information in a patient's electronic health record (EHR). Most of the current alert tools use SIRS criteria, but some also employ predictive models incorporating an array of physiologic and clinical data. However, these alerts generally tend to have low positive predictive values.(18,19) The use of such alerts can thus contribute to alert fatigue as well as inappropriate sepsis treatment for patients who have noninfectious conditions. Studies have also shown that these alerts lead clinicians to more frequently order lactate, chest radiographs, blood cultures, antibiotics, and fluids, but there is no clear evidence that these actions improve outcomes for patients. Finally, clinicians have not reliably embraced these alerts. In one study, only one-third of clinicians felt that sepsis alerts were useful and one-quarter felt that they improved care.(20)
Standardized order sets, on the other hand, have been reported in several cases to be associated with lower mortality rates.(21,22) However, given the before–after study design employed in these studies, along with the likely cointerventions of sepsis education and training, it is difficult to know the extent to which these findings are attributable to the order sets alone. Before–after studies of sepsis outcomes are especially prone to ascertainment bias, because enhanced awareness of sepsis leads to the diagnosis of sepsis among more mildly ill patients that previously may not have been given a diagnosis of sepsis.(23) The inclusion of these less ill patients in the analysis can lead to overly optimistic estimations of improved mortality.
With advances in EHRs and machine learning methods, as well as the increasing availability and breadth of rapid microbiologic testing, the diagnosis of sepsis will likely become more refined and personalized in the near future. Until then, current best practices involve active surveillance and reassessment of at-risk patients by clinicians who must simultaneously be aware of the strengths and limitations of various sepsis criteria. From a systems standpoint, hospitals can focus on implementation of comprehensive performance improvement programs centering on education and process improvement. The greatest improvements in outcomes tend to come from programs that have both components, which can include regular lectures, stakeholder meetings, bedside training and educational materials, audit and feedback to clinicians, order sets, clinical decision support, and the implementation of rapid sepsis response teams.(24) Systematic measurement of serum lactates in patients with possible sepsis, as mandated by the CMS SEP-1 measure, may also prompt more rapid physician interventions and improve sepsis outcomes.(25)
At the same time, clinicians must not anchor on a diagnosis of sepsis too early in a patient's course and should consider other potential life-threatening diagnoses that might present similarly, as in the case of the patient described above. Pending significant progress in the science of sepsis diagnosis, these relatively simple measures might hold the greatest promise in improving outcomes for patients with sepsis while minimizing the unintended consequences of inappropriate sepsis diagnosis among patients with other serious noninfectious conditions.
- Sepsis is a medical emergency, as delays in diagnosis and management are associated with higher mortality rates.
- Identifying sepsis is challenging given that the clinical signs and symptoms are nonspecific and there is no gold standard for diagnosis.
- When patients present with circulatory shock or signs of end-organ dysfunction, empiric antibiotics are reasonable, but clinicians should not anchor solely on a diagnosis of sepsis and must consider other serious diagnoses that could present similarly.
- Clinical decision support tools, including sepsis alerts and order sets, can facilitate early recognition of sepsis and appropriate care.
- However, most sepsis alerts have low positive predictive value and may contribute to alert fatigue and the administration of inappropriate sepsis therapies for noninfected patients.
- The current best practices for sepsis diagnosis involve active surveillance and reassessment of at-risk patients by clinicians who must be aware of the strengths and limitations of various sepsis criteria.
Ifedayo Kuye, MD, MBA
Brigham and Women's Hospital
Chanu Rhee, MD, MPH
Assistant Professor of Medicine
Brigham and Women's Hospital
Faculty Disclosure: Drs. Kuye and Rhee have declared that neither they, nor any immediate member of their families, have a financial arrangement or other relationship with the manufacturers of any commercial products discussed in this continuing medical education activity. In addition, the commentary does not include information regarding investigational or off-label use of pharmaceutical products or medical devices.
1. Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315:801-810. [go to PubMed]
2. Rhee C, Dantes R, Epstein L, et al; CDC Prevention Epicenter Program. Incidence and trends of sepsis in US hospitals using clinical vs claims data, 2009–2014. JAMA. 2017;318:1241-1249. [go to PubMed]
3. Liu V, Escobar GJ, Greene JD, et al. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA. 2014;312:90-92. [go to PubMed]
4. Torio CM, Moore BJ. National Inpatient Hospital Costs: The Most Expensive Conditions by Payer, 2013. HCUP Statistical Brief #204. Rockville, MD: Agency for Healthcare Research and Quality; May 2016. [Available at]
5. Venkatesh AK, Slesinger T, Whittle J, et al. Preliminary performance on the new CMS Sepsis-1 National Quality Measure: early insights from the Emergency Quality Network (E-QUAL). Ann Emerg Med. 2018;71:10-15.e11. [go to PubMed]
6. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34:1589-1596. [go to PubMed]
7. Seymour CW, Gesten F, Prescott HC, et al. Time to treatment and mortality during mandated emergency care for sepsis. N Engl J Med. 2017;376:2235-2244. [go to PubMed]
8. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101:1644-1655. [go to PubMed]
9. Churpek MM, Zadravecz FJ, Winslow C, Howell MD, Edelson DP. Incidence and prognostic value of the systemic inflammatory response syndrome and organ dysfunctions in ward patients. Am J Respir Crit Care Med. 2015;192:958-964. [go to PubMed]
10. Haydar S, Spanier M, Weems P, Wood S, Strout T. Comparison of QSOFA score and SIRS criteria as screening mechanisms for emergency department sepsis. Am J Emerg Med. 2017;35:1730-1733. [go to PubMed]
11. Phua J, Ngerng WJ, See KC, et al. Characteristics and outcomes of culture-negative versus culture-positive severe sepsis. Crit Care. 2013;17:R202. [go to PubMed]
12. Rhee C. Using procalcitonin to guide antibiotic therapy. Open Forum Infect Dis. 2017;4:ofw249. [go to PubMed]
13. Contou D, Roux D, Jochmans S, et al. Septic shock with no diagnosis at 24 hours: a pragmatic multicenter prospective cohort study. Crit Care. 2016;20:360. [go to PubMed]
14. Filbin MR, Lynch J, Gillingham TD, et al. Presenting symptoms independently predict mortality in septic shock: importance of a previously unmeasured confounder. Crit Care Med. 2018 Jun 29; [Epub ahead of print]. [go to PubMed]
15. Klein Klouwenberg PMC, Cremer OL, van Vught LA, et al. Likelihood of infection in patients with presumed sepsis at the time of intensive care unit admission: a cohort study. Crit Care. 2015;19:319. [go to PubMed]
16. Vincent JL, De Backer D. Circulatory shock. N Engl J Med. 2013;369:1726-1734. [go to PubMed]
17. Canto JG, Shlipak MG, Rogers WJ, et al. Prevalence, clinical characteristics, and mortality among patients with myocardial infarction presenting without chest pain. JAMA. 2000;283:3223-3229. [go to PubMed]
18. Alsolamy S, Al Salamah M, Al Thagafi M, et al. Diagnostic accuracy of a screening electronic alert tool for severe sepsis and septic shock in the emergency department. BMC Med Inform Decis Mak. 2014;14:105. [go to PubMed]
19. Umscheid CA, Betesh J, VanZandbergen C, et al. Development, implementation, and impact of an automated early warning and response system for sepsis. J Hosp Med. 2015;10:26-31. [go to PubMed]
20. Guidi JL, Clark K, Upton MT, et al. Clinician perception of the effectiveness of an automated early warning and response system for sepsis in an academic medical center. Ann Am Thorac Soc. 2015;12:1514-1519. [go to PubMed]
21. Thiel SW, Asghar MF, Micek ST, Reichley RM, Doherty JA, Kollef MH. Hospital-wide impact of a standardized order set for the management of bacteremic severe sepsis. Crit Care Med. 2009;37:819-824. [go to PubMed]
22. Micek ST, Roubinian N, Heuring T, et al. Before–after study of a standardized hospital order set for the management of septic shock. Crit Care Med. 2006;34:2707-2713. [go to PubMed]
23. Rhee C, Gohil S, Klompas M. Regulatory mandates for sepsis care—reasons for caution. N Engl J Med. 2014;370:1673-1676. [go to PubMed]
24. Damiani E, Donati A, Serafini G, et al. Effect of performance improvement programs on compliance with sepsis bundles and mortality: a systematic review and meta-analysis of observational studies. PLoS One. 2015;10:e0125827. [go to PubMed]
25. Han X, Edelson DP, Snyder A, et al. Implications of Centers for Medicare & Medicaid Services Severe Sepsis and Septic Shock Early Management Bundle and initial lactate measurement on the management of sepsis. Chest. 2018;1542:302-308. [go to PubMed]