Sepsis is a common condition managed in the emergency department. Current diagnosis relies on physiologic criteria and suspicion of a source of infection using history, physical examination, laboratory studies, and imaging studies. The infection triggers a host response with the aim to destroy the pathogen, and this response can be measured. A reliable biomarker for sepsis should assist with earlier diagnosis, improve risk stratification, or improve clinical decision making. Current biomarkers for sepsis include lactate, troponin, and procalcitonin. This article discusses the use of lactate, procalcitonin, troponin, and novel biomarkers for use in sepsis.
Key points
- •
Clinical biomarkers should be used in association with clinical gestalt, but they cannot replace the bedside clinician.
- •
Lactate has many uses in sepsis including assessment of severity, screening for disease, and as a marker for resuscitation but it is not always elevated in sepsis and can be elevated for other reasons.
- •
Procalcitonin is a marker used to distinguish bacterial from viral infection and has been studied in de-escalation of antibiotics in intensive care unit populations; however, its use in the emergency department for antibiotic use and resuscitation monitoring requires further study.
- •
Besides myocardial infarction, troponin can be elevated in many other conditions and is associated with worse prognosis in sepsis.
- •
New biomarkers include endothelial activators, acute-phase reactants, B-type natriuretic peptide/N-terminal B-type natriuretic peptide, and proadrenomedullin.
Introduction
Sepsis is a common cause of death in patients presenting to the emergency department (ED), and the condition results from the host response to the presence of infection. Current diagnosis relies on physiologic criteria and suspicion of a source of infection using history, physical examination, laboratory studies, and imaging studies. Diagnostic uncertainty often results with the patient who presents with systemic inflammatory response syndrome (SIRS) criteria and suspected sepsis, but a source of infection has not been discovered.
Biomarkers are laboratory assessments used to detect and characterize diseases and improve clinical decision making. Numerous laboratory markers have been used to assist decision making, including complete blood cell count, troponin, creatine kinase, lactate, C-reactive protein, and myoglobin. Some have argued the use of these biomarkers shows a lack of history and examination skills, whereas others have argued these tests have the potential to supplant physical examination and history taking. A reliable biomarker for sepsis should assist with earlier diagnosis, improve risk stratification, or improve decision making for care in sepsis patients.
Introduction
Sepsis is a common cause of death in patients presenting to the emergency department (ED), and the condition results from the host response to the presence of infection. Current diagnosis relies on physiologic criteria and suspicion of a source of infection using history, physical examination, laboratory studies, and imaging studies. Diagnostic uncertainty often results with the patient who presents with systemic inflammatory response syndrome (SIRS) criteria and suspected sepsis, but a source of infection has not been discovered.
Biomarkers are laboratory assessments used to detect and characterize diseases and improve clinical decision making. Numerous laboratory markers have been used to assist decision making, including complete blood cell count, troponin, creatine kinase, lactate, C-reactive protein, and myoglobin. Some have argued the use of these biomarkers shows a lack of history and examination skills, whereas others have argued these tests have the potential to supplant physical examination and history taking. A reliable biomarker for sepsis should assist with earlier diagnosis, improve risk stratification, or improve decision making for care in sepsis patients.
Lactate
Causes of Elevated Lactate Level
Lactate has numerous uses in sepsis, particularly in resuscitation and categorization of illness severity. Lactate is produced from all body tissue with the metabolism of pyruvate and Nicotinamide adenine dinucleotide (NADH), with normal production of 20 mmol/kg/d. Traditionally, lactate production with acidosis was thought to be caused by anaerobic metabolism or impaired hepatic metabolism. Elevated lactate can be broken into several categories, shown in Table 1 . Of note, lactate elevation may be caused by endogenous epinephrine-stimulating β-2 receptors, which produce excess pyruvate during aerobic glycolysis and circulation of inflammatory mediators and liver disease.
| Type A | Type B1 Associated with Disease | Type B2 Drugs and Toxins | Type B3 Associated with Inborn Errors of Metabolism |
|---|---|---|---|
| Tissue Hypoperfusion Anaerobic muscular activity Reduced tissue oxygen delivery | Leukemia Lymphoma Thiamine deficiency Pancreatitis Hepatic or renal failure Short bowel syndrome | Phenformin Metformin Epinephrine Norepinephrine Xylitol Sorbitol Lactate-based dialysate fluid Cyanide β-agonist Alcohols: methanol, ethylene glycol Salicylates Nitroprusside Isoniazid Fructose Paracetamol Biguanides Antiretroviral agents | Pyruvate carboxylase deficiency Glucose-6-phosphatase deficiency Fructose-1,6-bisphosphatase deficiencies Oxidative phosphorylation enzyme defects |
Screening
Serum lactate measurement is recommended as a screen for severe sepsis by the Surviving Sepsis Campaign. Many studies evaluating lactate in sepsis support its use to evaluate and prognosticate for sepsis, with an initial elevated lactate concentration associated with suspected infection and severity of illness. Point of care lactate is useful for sepsis screening, with a specificity of 82% for patients with confirmed sepsis for lactate levels ≥2 mmol/L. However, sensitivities of approximately 30% do warrant caution, and emergency providers should take the clinical picture into account rather than relying on one laboratory value. As lactate levels increase, illness severity, intensive care unit (ICU) admission, and vasopressor requirements increase.
This screening can be conducted through peripheral venous blood and does not necessitate arterial blood draw. Point of care and laboratory levels are equivalent if samples are run on a blood gas machine, and arterial and venous levels strongly correlate. Studies have evaluated the effect of tourniquet and temperature on lactate levels. No effect from tourniquet time or room temperature has been found if analysis occurs within 15 minutes of obtaining the sample. Samples obtained after this period, especially after 30 minutes, should likely be redrawn to minimize error. Other laboratory evaluations, including an electrolyte panel, should not be used as a substitute for lactate levels. Bicarbonate and anion gap levels do not correlate with lactate, as a normal bicarbonate level is found in 22.2% and normal anion gap in 25% of patients with lactate level greater than 4.0 mmol/L.
Prognostication
An association between lactate level and mortality has been established in several studies. Puskarich and colleagues found that with lactate levels of 2.1 mmol/L, the mortality rate was 14.4%, but at levels approaching 20 mmol/L, the mortality rate was 39%. Irrespective of other variables such as blood pressure and illness severity, lactate is an independent predictor of mortality. Levels greater than 2 mmol/L are correlated with increased mortality and meet criteria for severe sepsis, whereas levels greater than 4 mmol/L correlate with mortalities of septic shock even if the patient is normotensive.
Lactate levels greater than 4 mmol/L are strongly associated with increased mortality, no matter the etiology. Patients lacking other criteria for sepsis with a lactate level ≥4.0 should be regarded with caution and given careful consideration for sepsis. Evaluation for other etiologies of elevated lactate levels, including gastrointestinal bleeding, any shock state (eg, cardiogenic, anaphylactic), mesenteric ischemia, and toxicologic etiologies (eg, salicylate overdose), should be conducted. These patients should be admitted for trending of lactate levels to ensure normalization, as these levels are associated with significant mortality regardless of ultimate etiology.
Cryptic Shock
Lactate can also be used to screen for sepsis in the patient with normal vital signs, otherwise known as cryptic septic shock. The hemodynamically stable patient with elevated lactate level (especially ≥4.0 mmol/L) is at risk for increased mortality. As the body begins to undergo inflammation and increased glycolysis, lactate production increases before clinically apparent end-organ damage and patient decompensation. Thus, lactate serves well as an early marker of sepsis and severe sepsis, with elevated levels associated with increased mortality.
Intermediate Levels
Intermediate lactate levels of 2.0 to 3.9 mmol/L present a quandary, particularly in the setting of hemodynamic stability. Lactate levels between 2 and 4 mmol/L meet Centers for Medicare and Medicaid Services criteria for severe sepsis following Surviving Sepsis Campaign guidelines. Recent literature supports increased morbidity and mortality with lactate levels in this range, with mortality rates ranging from 3.2% to 16.4% in patients with no hypotension. Close to one-quarter of these patients progress to having septic shock. Thus, lactate levels in this range warrant close monitoring and aggressive treatment with early fluid administration and antimicrobials. Table 2 describes resuscitation measures based on lactate level.
| Lactate Level | Centers for Medicare and Medicaid Services Measure | Resuscitation Recommendation |
|---|---|---|
| <2 mmol/L | None | Lactate levels may be negative in more than half of patients with sepsis. Clinical gestalt takes precedence over markers. |
| 2–4 mmol/L | Severe sepsis | Resuscitation with intravenous fluids and antimicrobials and reassessment of lactate within 60 min |
| ≥4 mmol/L | Septic shock | Aggressive resuscitation warranted regardless of vital signs. |
Lactate Clearance
Resuscitation of the septic patient can be evaluated with lactate clearance, with a target of 10% clearance in lactate from the initial level. Delayed clearance of lactate in patients with septic shock is correlated with poor outcome. Early lactate clearance is associated with improved outcomes, with targeting lactate normalization as a resuscitation goal. Arnold and colleagues found a lactate clearance of 10% to be a strong predictor of improved survival, with 60% of those in the nonclearance arm suffering death.
Lactate parameters may be a better reflection of body homeostasis compared with oxygen-derived variables, which may not reflect actual clinical status of patients. The advantage of using lactate clearance is that no specialized, invasive equipment is needed, such as in continuous central venous oxygen monitoring (ScvO2), which was used in the original Early Goal-Directed Therapy trial. Jones and colleagues used lactate clearance in place of ScvO2 in the final endpoint of resuscitation and found lactate clearance of 10% to be noninferior to the measurement of ScvO2 as a resuscitation measure for mortality.
Lactate Pitfalls
As discussed earlier, lactate levels may not always be elevated in patients with septic shock. One study found that 45% of patients with vasopressor-dependent septic shock had a lactate level of less than 2.4 mmol/L but a mortality rate of 20%. Patients with elevated lactate had higher rates of prior liver disease, acute liver injury, and acute bacteremia. Hernandez and colleagues found that 34% of patients with septic shock did not have elevated lactate levels, but these patients had a low mortality rate of 7.7% when compared with those with elevated lactate levels (42.9%).
There are also states in which lactate may be elevated but sepsis is not present. These states include hepatic disease, shock states (eg, cardiogenic, obstructive), trauma, seizure, medications and toxins (eg, acetaminophen, metformin, β agonists, epinephrine, propofol, alcohol, cocaine, carbon monoxide, linezolid, cyanide), excessive muscle activity, smoke inhalation, burns, regional ischemia such as mesenteric ischemia, thiamine deficiency, diabetic ketoacidosis, malignancy, and inborn errors of metabolism. Lactate should not be used in isolation for resuscitation but rather in association with other resuscitation measures.
Procalcitonin
Procalcitonin (PCT) is a propeptide of calcitonin produced by endocrine tissue in the thyroid, gastrointestinal tract, and lungs, normally in low concentrations. In bacterial infections, production is upregulated by toxins and proinflammatory mediators, resulting in PCT production. Viral infections increase interferon-γ, which inhibits PCT. Initial levels begin to increase within 3 to 6 hours and peak at 6 to 22 hours with bacterial infection. With infection resolution, levels typically decrease by 50% per day, as opposed to other biomarkers such as white blood cell count and C-reactive protein.
Procalcitonin has several advantages, including specificity for bacterial infection, rapid increase with bacterial infection, rapid decrease with treatment of infection, and no impairment in the presence of neutropenia or immunosuppressive states. Other inflammatory states may cause an increase, however. These states include surgery, paraneoplastic states, autoimmune diseases, prolonged shock states, chronic parasitic diseases (such as malaria), certain immunomodulatory medications, and major trauma.
Antibiotic Stewardship
Most evidence for PCT has been published in ICU patients with lower respiratory tract infections and sepsis. These studies, including meta-analyses, found that PCT-guided algorithms can reduce antibiotic exposure and costs of treatment and hospitalization with no effect on patient outcomes.
In lower respiratory tract infections, especially chronic obstructive pulmonary disease and bronchitis, the clinical picture is not always clear as to whether the patient is experiencing a viral infection or bacterial pneumonia. The ProResp trial randomly assigned patients to standard antibiotic therapy versus PCT-guided therapy whereby if the PCT was less than 0.25 μg/L, antibiotics were discouraged. If PCT was greater than 0.25 μg/L, antibiotics were used. No difference in mortality or length of stay was found, although a significant decrease in antibiotic use was observed in the PCT-guided group (83% vs 44%). A second trial, ProHOSP, evaluated patients with lower respiratory tract infections in an ED using a PCT-guided algorithm with similar cutoff levels. Similar results were found with a reduction in antibiotic use ( Table 3 ).
| Study Antibiotic Use | PCT Level | ||||
|---|---|---|---|---|---|
| <0.1 μg/L | 0.1–0.25 μg/L | 0.25–0.5 μg/L | 0.5–1 μg/L | >1.0 μg/L | |
| ProHOSP antibiotic use (respiratory infection only) | No | No | Yes | Yes | Yes |
| PRORATA antibiotic use (sepsis patients in ICU) | No | No | No | Yes | Yes |
Diagnosis
PCT levels can also be useful in diagnosing sepsis. However, the clinical context and patient scenario including possible source of infection, severity of illness, and likelihood of bacterial infection should take precedence over PCT. A laboratory sample drawn for PCT testing will likely not return while the patient is in the ED. Thus, if severe sepsis or septic shock is suspected, it is imperative the emergency provider treat the patient with broad-spectrum coverage for bacterial etiologies. A PCT can benefit the ICU team, but its use in the ED is controversial.
PCT has potential in identification of culture-positive sepsis, as levels correlate with bacterial load and may have prognostic implications. Differentiating culture-negative sepsis and noninfectious SIRS can be assisted with PCT, with 92% sensitivity for culture-negative sepsis. PCT levels of less than 0.25 μg/L are unlikely to have bacterial infection (<1%). Several meta-analyses have been conducted on PCT diagnostic accuracy in sepsis, with one finding a sensitivity and specificity of 77% and 79%, respectively. Most of these studies state PCT is helpful in diagnosis of documented infection.
The prospective PRORATA trial evaluated septic patients admitted to the ICU in which one study arm had antibiotic initiation dependent on initial PCT level. A PCT level of 0.5 μg/L was used to trigger antibiotic administration. No difference in mortality or length of stay was found between patients in the PCT-guided group versus standard group. A decrease in antimicrobial use was found in the group using PCT to initiate antibiotics (see Table 3 ). A 2015 meta-analysis found the most optimal cutoff value to be 0.5 μg/L for ruling out bacteremia.
Although the literature on PCT-guided antibiotic therapy is encouraging, at this time emergency providers should not be using PCT to direct antibiotic therapy for patients with severe sepsis and septic shock. Currently, standard care in treating severe sepsis and septic shock in the ED requires timely antibiotic administration. In the near future, PCT may have a role in sepsis evaluation in the ED; however, that role is still being defined.
Troponin
Troponin has been most commonly used to diagnose myocardial infarction, with an elevation greater than the 99th percentile in a healthy population meeting criteria for acute coronary syndrome. In patients with concern for acute coronary syndrome, troponin has been used to risk stratify patients as well, as seen in the HEART pathway. Troponin testing has also now changed with the introduction of higher-sensitivity assays. Cardiac troponin consists of troponin I and T, which are cardiac regulatory proteins that control the calcium-mediated interaction of actin and myosin. This interaction allows for myocardial contraction. Injury of the myocardium causes release of these proteins into the bloodstream. Troponin elevation has many etiologies that can be divided into cardiac and noncardiac, shown in Table 4 .
