Production of Lactate

Under basal conditions, lactate is produced by muscle, skin, brain, red blood cells, and intestine. In critical illness, additional sources include the lungs, white blood cells, and splanchnic organs. Importantly, activated white blood cells have relatively few mitochondria; thus, they favor anaerobic metabolism. When oxygen is present, glucose is most often metabolized to pyruvate and enters the citric acid cycle to produce adenosine triphosphate, essentially bypassing lactate production. When oxygen supply to tissue is limited, lactate is produced and shunted to the liver as a substrate for gluconeogenesis. Under some circumstances, such as exercise and some states of critical illness, pyruvate may accumulate despite abundant oxygen availability and can be shunted to lactate production.

Other measures of acid-base balance in the critically ill patient may be misleading. Anion gap and base excess are associated with lactate production, but they do not always accurately predict lactate concentration. Both are generally regarded as inferior screening tools for tissue hypoperfusion because they do not account for unmeasured ions and hypoalbuminemia, a common occurrence. In addition, “cryptic” or “occult” shock may be present in the critically ill and may be identified only by lactate elevations in the absence of other indicators generally associated with this state.

Measurement of Lactate

Only the l -lactate isomer is clinically measured. d -Lactate is a bacterial product only rarely relevant to human acid-base balance.

Arterial and venous lactate are generally regarded as equivalent, although recent evidence suggests that venous and arterial peripheral lactates may vary widely in patients with an initially elevated lactate. Future research on this topic is required because it is in these patients that discrimination is most important.

Elevated serum lactate can be a function of increased production, decreased clearance, or both. Myocardial depression, relative hypovolemia (i.e., vascular dilation and fluid loss from capillary leak), mitochondrial dysfunction, excessive adrenergic stimulation, and microcirculatory insufficiency all contribute to decreased oxygen delivery or utilization in shock. Research into hyperlactatemia basically focuses on the use of lactate as a “biomarker” or screening tool to stratify risk and characterize the severity of injury in critical illness and trauma. Additional studies have attempted to characterize lactate as an endpoint of resuscitation in many forms of critical illness.

Prehopsital Measurement of Lactate

Capillary lactate in the prehospital setting is associated with injury severity in trauma patients. Coates and colleagues demonstrated that elevated capiillary lactate levels, in the prehospital setting, equate with injury severity in trauma patients. The investigators also determined that lactate was helpful in the triage of patients with normal vital signs despite evidence of tissue hypoperfusion. Additional evaluation in the prehospital setting has demonstrated that lactate is associated with hospital mortality even in patients with initially normal vital signs. In addition, lactate is significantly different in shocked versus nonshocked patients, and elevated lactate is associated with increased hospital length of stay (LOS), intensive care unit (ICU) LOS, and increased mortality (12.2% vs. 44.3%), especially in patients with normal vital signs on admission (mortality of 35% vs. 7%, P < .001). Thus the prehospital use of lactate as a screening and triage tool may be useful to uncover subtle organ hypoperfusion.

Measurement of Lactate in the Emergency Department

Many studies have been performed assessing presenting lactate levels in the emergency department (ED). Shapiro and colleagues examined 1278 consecutive ED patients with an infection-related diagnosis and uncovered a linear relationship between mortality and lactate. In addition, they found that initial lactate was 36% sensitive and 92% specific for any death within 28 days, whereas lactate was 55% sensitive and 91% specific for early death (within 72 hours of presentation). The same group found a 15% mortality rate in septic patients with a lactate greater than 4.0 mmol/L and normal admission systolic blood pressure. In a multivariate analysis, the odds ratio (OR) of death was 2.1 in patients with lactate between 2.5 and 4 mmol/L and was 7.1 when lactate was greater than 4 mmol/L. The initial lactate level is associated with increased mortality in patients with normal vital signs (“occult” shock) as well as those with overt shock.

Lactate Measurement in the Intensive Care Unit

Investigations into the prognostic value of lactate levels in the ICU have yielded variable results because of unevenness in patient populations, diagnoses, time course of treatment, and the complexity of evolving critical illness. In a well-designed retrospective analysis of 134 mixed ICU patients, receiver-operator analysis was used to investigate the relationship between elevated lactate and Sequential Organ Failure Assessment scores. The risk of organ failure or death increased in patients with a prolonged elevation of lactate. This effect was most profound early in the ICU stay, suggesting that early resuscitation improves mortality in critical illness. In a prospective observational study of 394 consecutive patients, the same group found that mortality decreased in septic patients whose lactate decreased within 12 hours of admission. This difference was independent of hemodynamic status.

Goal-Directed Therapy and Lactate Clearance

It has been proposed that measurement of lactate levels provides a simple point-of-care test to determine an endpoint of resuscitation in critical illness. Recent large prospective randomized trials have confirmed that early goal-directed therapy (EGDT) provides no survival benefit over standard care. The ProCESS (Protocolized Care for Early Septic Shock) investigators randomized 1341 patients with septic shock at 31 centers to protocol-based EGDT, protocol-based standard therapy, or usual care. Although there was significant variability in the use of hemodynamic monitors, as well as blood, fluid, and vasopressor use, there was no difference in 60-day or 1-year mortality and no difference in the need for organ support. Supplemental data from this study suggested that EGDT may provide a survival benefit in patients with an initial lactate level greater than 5 mmol/L. Likewise, the ARISE (Australasian Resuscitation in Sepsis Evaluation) investigators found no difference in survival time, in-hospital mortality, or duration of organ support in 1600 randomized patients with septic shock to either EGDT versus standard care. Neither study used lactate level as a goal of resuscitation; rather, they used it as a marker for tissue hypoperfusion and as a means of inclusion in the study. With this information, it cannot be suggested that lactate be used as a target of resuscitation for shocked patients. The results of another large prospective evaluation of EGDT, the ProMISe (The Protocolised Management in Sepsis Trial) study from the United Kingdom, were similar. Again the investigators did not use lactate as a goal of resuscitation.

Other investigators have found that the early use of physiologic targets in resuscitation is associated with improved survival and decreased organ failure in several clinical shock scenarios, including trauma and sepsis. This effect has been most thoroughly studied in septic shock. To prevent morbidity and overcome obstacles to placement of invasive monitors in septic shock, the use of lactate and its clearance has been proposed as a less invasive measure of the progress of resuscitation.

In a randomized trial by Jansen and colleagues, 348 septic patients were allocated to a lactate-driven protocol versus a lactate-blinded treatment. The study was underpowered, and mortality was not statistically different between the groups (33.9% in the lactate group vs. 43.5% in the control group [ P = .067]); however, a post hoc multivariate analysis demonstrated an observed 9.6% reduction in hospital mortality. In addition, multiple retrospective observational studies have suggested lactate as a monitor of the therapeutic response during EGDT. A randomized prospective trial of septic patients randomly chosen to receive therapy guided by central venous oxygen or lactate showed no difference in outcome, although mortality was lower in the lactate group (23% vs. 17%).

Assuming proper hepatic and renal function, serum lactate should be rapidly cleared. Clearance of lactate has been associated with increased survival in various clinical settings, including trauma, mixed populations of critically ill patients, sepsis, and after myocardial infarction.

Recent experience suggests that lactate clearance may function as a more robust resuscitative endpoint than central venous oxygen saturation or other traditional oxygen-derived variables. The ability to achieve only a central venous oxygen saturation goal was associated with 41% mortality in septic shock, whereas lactate clearance was associated with only 8% mortality. A recent meta-analysis that reviewed 15 original studies examining lactate clearance in critically ill patients found that lactate clearance predicted mortality with 75% sensitivity and 72% specificity.

In summary, the promise of lactate as a goal of resuscitation remains unanswered. It cannot be recommended with current data that lactate be used as the sole endpoint of resuscitation, but it appears to be a reasonable marker for success of resuscitation in shock, particularly in patients with infections and after trauma. If lactate clearance has a role as a goal of resuscitation, then more study is needed.

Clinical Approach to Elevated Lactate in Various Clinical Scenarios

Shock—cardiogenic: Several studies have demonstrated the utility of lactate measurement in the identification of acute coronary syndromes and to predict the development of shock after acute coronary syndrome. In addition, high lactate levels have been associated with 30-day mortality after percutaneous interventions for myocardial infarction and death in the ICU after admission for myocardial infarction.

Attana studied 51 consecutive patients in cardiogenic shock after ST-elevation myocardial infarction and determined that lactate clearance was higher in survivors. Lactate clearance less than 10% was associated with particularly poor survival. Park and colleagues found a mortality of 53% in 96 consecutive patients with cardiogenic shock requiring percutaneous cardiopulmonary support. Multivariate analysis found that a lactate clearance less than 70% at 48 hours was associated with death.

Shock—after arrest: Mullner et al. studied 167 out-of-hospital witnessed cardiac arrest patients and found that lactate levels greater than 16.3 mmol/L were 100% specific for death or poor neurologic recovery. In a review of 394 cardiac arrest patients, multivariate analysis revealed that lactate was an independent predictor of mortality (OR 1.49 per 1 mmol/L increase). Lactate levels greater than 2 mmol/L at 48 hours predicted mortality with a specificity of 86% and predicted poor neurologic outcome with a specificity of 87%. Lactate correlates with good neurologic outcome in out-of-hospital cardiac arrest patients treated with therapeutic hypothermia.

Trauma: Initial serum lactate has been shown in several studies to predict outcome in trauma patients. In a recent study, 1941 patients were retrospectively reviewed. Initial serum lactate (drawn within 35 minutes of admission) was lower in survivors (21 vs. 32 mg/dL, P < .001). In multivariate analysis, initial lactate was a significant predictor of mortality and of the need for operative intervention.

In a broad cohort of trauma patients, initial lactate and clearance of lactate at 6 hours was associated with decreased mortality. In patients with an initial lactate level greater than 4 mmol/L, lactate clearance of greater than 60% was associated with a mortality of 7.5% whereas a lactate clearance of less than 30% was associated with a mortality rate of 28.1% ( P = .001).

Burns: In a study of 166 burn patients, a high initial lactate and an inability to clear lactate at 24 hours were predictive of death (68% survival in lactate clearers vs. 32% in patients whose lactate remained above normal after 24 hours).

Drugs/alcohol: Despite being metabolized via pathways that can produce lactate, alcohol and drugs do not appear to influence the significance of lactate measurement or alter its clinical utility.

Postoperative: Li and colleagues found that elevated lactate after major abdominal surgery was associated with complications. They studied 114 consecutive patients undergoing elective surgery with a Physiologic and Operative Severity Score for the enumeration of Mortality and Morbidity (POSSUM) score of 4 or greater. The degree of lactate elevation and the time-weighted average of lactate over the first 24 hours after surgery correlated with the severity of complications. Lactate clearance at 24 hours was significantly associated with better outcome. They suggested the use of lactate as a therapeutic target in early resuscitation after major abdominal surgery.

In one study of postoperative patients, mortality was increased from 3.9% if lactate cleared in the first 24 hours after surgery to 13.3% if lactate cleared in 48 hours, 42.5% if cleared in the first 96 hours, and 100% if not cleared by 96 hours. Other studies have highlighted the phenomenon of lactate as a marker for poor outcome in surgical ICU patients, in whom mortality was 10% if lactate normalized in the first day after admission versus 67% mortality if lactate did not normalize.