Clinical Mimics: An Emergency Medicine–Focused Review of Sepsis Mimics

Brit Long, MD; Alex Koyfman, MD


J Emerg Med. 2017;52(1):34-42. 

In This Article


Why Does SIRS Occur in Patients With Sepsis?

Sepsis ultimately results from a complex interaction of proinflammatory, antiinflammatory, activated complement system, and coagulation mediators that trigger a host response in association with detector and signaling markers. Initiators (e.g., microbes, trauma, hypoxia, ischemia, and toxins) cause local tissue damage, which release local pro- and antiinflammatory markers. Proinflammatory signalers include tumor necrosis factor and interleukins-1 and -6, while anti-inflammatory markers include interleukins-4, -10, and -11 and soluble tumor necrosis factor receptors. These are designed to function at a local level. Other important aspects of sepsis include vascular tone instability through depletion of vasopressin, adrenal insufficiency, and nitric oxide enhancement of vasodilation. If the initiators overwhelm the local response, multiple systems are affected, which may result in end organ dysfunction (Table 2).[2,5,6]

Signs of Sepsis

Hemodynamic effects of septic shock include decreased peripheral resistance with increased cardiac output and tachycardia in the early stages or distributive shock (which may be found in anaphylaxis, pancreatitis, spinal injury, and other conditions). Later stages feature findings similar to hypovolemic shock with increased vascular resistance, lower cardiac output, and cooler peripheral extremities with poor capillary refill. These findings are not specific for sepsis because of the pathophysiology.[2,4–7]

Laboratory data are not specific for sepsis, similar to the hemodynamic effects. Any form of physiologic stress may result in an increased white blood cell (WBC) count with left shift.[7–9] Coagulopathic derangements also occur in this systemic inflammatory state. Fibrin split products, fibrinogen, and a coagulation panel may show derangements in sepsis and its mimics, which does not assist in differentiating mimics from true sepsis. Leukopenia and thrombocytopenia are more suggestive of sepsis, but they are not definitive.[7–10] C-reactive protein levels have been studied for use in sepsis, but they are elevated in any inflammatory state and are not specific for sepsis.[10] Procalcitonin is a propeptide released in response to bacterial infection. It has shown promise in diagnosing bacterial sepsis, but the results of this laboratory study may not return while the patient is in the emergency department (ED).[10–12] Lactic acid elevation and base excess/deficit are commonly used for resuscitation, but they will not pinpoint the cause of shock.[2,6,13,14] Lactate elevation possesses a wide range of etiologies, many of which are not associated with true infection.[14] However, as discussed later, these markers can be used for resuscitation.

Fever is a cardinal sign of infection; however, fever does not necessarily equal infection[2,5,6,15] Cytokine activation induces a febrile response, mediated by the hypothalamus. A review conducted in intensive care unit cohorts stated that the majority of diseases causing true sepsis were associated with temperatures >38.8°C (101.8°F).[15] This is excluding elderly and immunosuppressed patients, where a temperature of 37.0°C (98.6°F) is often a true fever.[16] According to Cunha, a normal host (i.e., otherwise healthy at baseline and not elderly, uremic, or immunosuppressed) with a temperature <38.8°C or >41.1°C (106.0°F) is usually noninfectious.[15] On the other hand, hypothermia suggests bacteremia and true sepsis.[15] More recent studies have questioned the use of fever alone to diagnose true sepsis. A review by Coburn et al. found a temperature ≥37.8°C (100.0°F) displays a likelihood ratio (LR) of 1.5 (95% confidence interval [CI] 1.2–1.9).[9] However, similar to the review by Cunha, this review finds that temperature ≥40.0°C (104.0°F) has a LR of 0.3 (95% CI 0.13–1.0).[8,9]

Decision rules have been created for use in predicting positive bacteremia. However, to the authors' knowledge, no decision aid or rule has been developed for predicting true sepsis versus its mimics. Coburn et al. conducted a meta-analysis to determine which factors predict positive blood cultures.[9] This meta-analysis took into account the Shapiro et al. decision rule for predicting true bacteremia, which states that blood cultures are indicated in the setting of one major or two minor criteria.[8] Major criteria include suspicion of endocarditis, temperature >39.4°C (102.9°F), and indwelling catheter. Minor criteria include temperature 38.3°C–39.3°C (100.9–102.7°F), age ≥65 years, chills, vomiting, systolic blood pressure ≤ 90 mm Hg, WBC ≥ 18,000/μL, or creatinine ≥ 2 mg/dL.[8] Per this meta-analysis, elevated temperature alone does not accurately predict bacteremia (for 38°C: LR 1.9 [95% CI 1.4–2.4]; for 38.5°C: LR 1.4 [95% CI 1.1–2.0]), nor does isolated leukocytosis (LR 1.7). Chill severity may be more useful (shaking chills: LR 4.7 [95% CI 3.0–7.2]).[9] Both SIRS and the Shapiro decision rule are sensitive—but not specific—predictors of bacteremia (SIRS: negative LR 0.09 [95% CI 0.03–0.26]; decision rule: negative LR 0.08 [95% CI 0.04–0.17]).[8,9] None of these can be used in isolation. However, in combination, laboratory markers, history, and physical examination may assist providers in differentiating sepsis from its mimics while evaluating closely for fever and shaking chills.[8,9,15,16]

Sepsis Mimics

Mimics of sepsis are a common cause of misdiagnosis in the ED. All of these clinical conditions produce symptoms and signs that meet SIRS, clouding the clinical picture. These conditions are shown in Table 3, with diagnostic and treatment pearls and pitfalls.

ED Approach

Because of the morbidity and mortality associated with sepsis, specifically in septic shock, rapid identification and resuscitation are vital. ED triage systems are designed to classify patients into different categories based on illness severity with vital signs, chief complaint, and focused examination. This first encounter can provide a great deal of information that may allow providers to recognize sepsis. Because of significant overlap, sepsis and its mimics are often not differentiable upon initial assessment.[2,4–7,56] There are five important factors to consider:

1. The first factor is addressing resuscitation before differentiating sepsis and mimics. Airway assessment, with establishment of definitive airway if needed, in conjunction with evaluation of breathing adequacy and circulatory status are paramount. Obtaining IV access and appropriate diagnostic studies are necessary. These studies include a complete blood cell count, a basic metabolic panel, urinalysis, a chest radiograph, blood cultures, and lactate as appropriate. Initiation of fluid bolus should be conducted to improve perfusion and preload. If sepsis is suspected, broad-spectrum antimicrobials should be initiated. [56]

Lactate is a valuable marker in resuscitation; elevations in this marker are associated with increased mortality, especially in the setting of patients with hypotension.[13,56–58] Even intermediate elevation (i.e., levels 2–4 mmol/L) are associated with a 15% mortality rate.[57]

2. Once resuscitation has been initiated, a focused history and examination can then be completed, evaluating for other conditions and sources. This can allow the provider to target resuscitation and management to the clinical condition. [2,6,56]
3. With completion of the initial assessment and first stages of resuscitation, the next task is to search for a potential source in the setting of SIRS. The LUCCAASS mnemonic will assist the search for source: lung (pneumonia), urine (cystitis/pyelonephritis), cardiac (endocarditis), central nervous system (CNS) (meningitis, encephalitis), abdominal (abscess, cholecystitis), arthritis (septic arthritis), spine (osteomyelitis, abscess), and skin (cellulitis, IV line/peripherally inserted central catheter infection). The information available at this point from diagnostic studies, history, and physical examination can likely assist the provider in differentiating sepsis from its mimics. [2–8]
4. There may be situations in which no source is found, and culture-negative sepsis may account for approximately 40% of sepsis cases. [59] These patient have approximately similar outcomes with culture positive patients. [59] If no source is found, carefully evaluate the patient again while continuing resuscitation, using history and physical examination to evaluate for the above clinical entities that meet SIRS criteria. [2,56] Using a systematic approach will provide the best avenue for diagnosis and treatment, with focused testing. If another entity is discovered, treat that condition as indicated. For example, the patient with urinary tract infection may also have diabetic ketoacidosis, requiring additional fluid rehydration, electrolyte management, and insulin.
5. Frequent evaluation of the patient to response to treatment is necessary. Use of laboratory tests are valuable markers, such as lactate clearance and patient hemodynamic response to resuscitation. Targeting lactate normalization displays ease of use and applicability in the ED for resuscitation goals [56–58,60] A combination of mental status, capillary refill, urine output, blood pressure, heart rate, and laboratory markers are better than one single clinical measure for continued assessment.

Ultimately, the approach of the patient with sepsis or mimic is similar in that resuscitation takes precedence, followed by targeted history and examination. Consideration of sepsis versus its mimics can be completed while resuscitation is underway. Once an etiology is discovered, treatment and management can be targeted.