Severe sepsis is the most common cause of death in critically ill patients (1). The effectiveness of early resuscitation is an important determinant of sepsis survival (2). Specifically, early quantitative resuscitation, the use of a structured set of cardiovascular interventions targeting predefined hemodynamic end points, can have a profound effect on hospital mortality (3). Currently, international consensus treatment guidelines recommend a quantitative resuscitation strategy that includes targeting central venous oxygen saturation (ScvO2) of 70% or greater in the first 6 h of severe sepsis therapy (4). However, the optimal end points of sepsis resuscitation remain controversial (5, 6).
Serum lactate elevation is an important marker of impaired tissue perfusion in patients with sepsis and is often elevated even in the absence of arterial hypotension (7). Numerous studies show that a single early lactate measurement has important prognostic significance and predicts mortality in populations of patients with infection (8, 9). A small number of studies have previously reported that serial measurements of lactate have potential prognostic value during conventional sepsis management (10-12); however, it is not known if lactate clearance is associated with survival in the context of medical centers that have adopted and routinely perform aggressive quantitative resuscitation for severe sepsis. In addition, it is not known if lactate clearance is important independent of other recommended quantitative resuscitation end points, specifically ScvO2.
The objectives of this study were to determine (a) if early lactate clearance is associated with improved survival in emergency department (ED) patients with severe sepsis and (b) the concordance between ScvO2 optimization and lactate clearance during early sepsis resuscitation.
Setting and study design
We analyzed prospectively collected registries of consecutive ED patients diagnosed with severe sepsis at three urban hospitals. The three hospitals are part of a multicenter research collaborative (Emergency Medicine Shock Research Network) (13), where each institution uses ED-based protocol-directed quantitative resuscitation for patients with severe sepsis (14-16). The registries were compiled between 2004 and 2007 and were approved by the institutional review board at each institution. Of the 166 subjects in this report, the subjects from one center (n = 110) were a subset of a total of 1,352 patients in two prior publications (8, 15), and the remainder were previously unpublished. Neither of the previous reports was focused on serial lactate measurement in sepsis.
The inclusion criteria were based on consensus definitions for severe sepsis (17) and included (a) age older than 17 years, (b) suspected infection, (c) two or more criteria of the systemic inflammatory response syndrome, (d) either a systolic blood pressure (SBP) less than 90 mmHg after 20 mL/kg or greater i.v. fluid or an initial serum lactate 4 mmol/L or greater, (e) initial and repeat lactate measurement within 6 h of resuscitation initiation, and (e) intensive care unit admission.
Consistent with international treatment guidelines for patients with severe sepsis and evidence of tissue hypoperfusion (hypotension after i.v. fluids or initial lactate ≥4 mmol/L) (4), each participating center uses a similar quantitative resuscitation algorithm targeting normalization of ScvO2 (≥70%) in the ED (14-16). Figure 1 displays elements of the resuscitation protocol common to each center. The registry used in this study was designed to measure the achievement of each goal within the protocol as a dichotomous variable (yes/no) at specified time intervals (i.e., 6 h). We did not accrue specific data on the use of red blood cell transfusion, dobutamine use, or time of antibiotic administration.
ScvO2 was measured using a central venous catheter inserted into the superior vena cava via internal jugular or subclavian venous access that provides continuous measurement of ScvO2 via reflection spectrophotometry (Edwards Lifesciences, Irvine, Calif). Treatment algorithms at each center mandate the measurement of an initial serum lactate in the ED. Obtaining a repeat lactate level is not a mandated practice at any of the centers and is obtained at the discretion of the treating clinician. All lactate assessments were obtained from venous blood samples and were measured according to local institutional standard methods using a central laboratory at two centers and a point-of-care analyzer (Abbott Point of Care, Abbott Park, Ill) at the third center.
Using a standardized data abstraction template, we queried each of the three registries and collected the following data: demographics, suspected source of infection, vital signs, initial and repeat serum lactate concentrations, laboratory values for organ failure assessment, and outcome information. Illness severity was defined using the presence of individual organ system failures as assessed by the worst recorded values for each organ system while in the ED: cardiovascular = initial SBP less than 90 mmHg; pulmonary = new oxygen requirement or PaO2/FiO2 less than 300 or SpO2/FiO2 less than 221; renal = serum creatinine more than 2.0 mg/dL; hepatic = serum bilirubin greater than 2.0 mg/dL; hematologic = platelets less than 100,000/μL or international normalized ratio more than 1.5 sec (18); and calculation of the Sequential Organ Failure Assessment (SOFA) score (19) as modified by Vincent et al. (20).
To determine the association between lactate clearance and mortality, we stratified patients into two groups defined a priori based on previously published data (12): (a) lactate clearance-repeat lactate decrease by 10% or greater from initial (or both initial and repeat levels ≤2.0 mmol/L), and (b) lactate non-clearance-repeat lactate decrease by less than 10% from initial. The primary outcome was in-hospital mortality.
We wanted to determine if ScvO2 optimization during resuscitation could reliably exclude lactate non-clearance; therefore, we evaluated the concordance between achievement of ScvO2 optimization (≥70%) and lactate clearance goals (≥10%) over the first 6 h of resuscitation using a 2 × 2 table.
We analyzed the difference in proportions of death between lactate clearance and non-clearance groups using the binomial test and the associated 95% confidence intervals (CIs) and P values. All other variables between lactate clearance versus lactate non-clearance groups and between survivors versus nonsurvivors were compared using the binomial test or Student t test when appropriate. We performed multivariate logistic regression analysis using candidate variables that were significantly different (P < 0.05) between survivors and nonsurvivors on bivariate analysis, with in-hospital mortality as the dependent variable. We used SigmaStat (v. 3.5; Systat, Chicago, Ill) for all analyses.
Sample size estimate
We estimated the necessary sample size based on the following assumptions: (a) a predicted mortality rate of 25% overall (based on previously published experience with quantitative resuscitation for severe sepsis at the three centers) (14-16); (b) an event (death) per variable ratio of 10:1 necessary for multivariate modeling (21, 22); and (c) the multivariate model would need to, at a minimum, include the following covariates that have previously been shown to differentiate sepsis survivors from nonsurvivors: arterial hypotension (23), failure to achieve ScvO2 70% or greater (2), and lactate non-clearance (12). To accrue the necessary 30 cases of mortality to test these three covariates in a multivariate model, we estimated that a minimum of 120 total cases would be necessary.
Characteristics of study cohort
There were 166 subjects who met inclusion criteria. The overall mortality rate was 23% (38/166). Table 1 shows characteristics of the entire cohort. Of the 166 subjects, 110 came from center 1, 22 came from center 2, and 34 came from center 3. Using multivariate analysis, there was no apparent center effect on in-hospital mortality.
The quantitative resuscitation algorithm targeting ScvO2 70% or greater was successfully initiated in the ED in 148 (89%) of 166 study subjects. Study patients received an average of 4.4 L (95% CI, 4.1-4.7 L) of i.v. crystalloid fluid during their ED resuscitation. Lactate non-clearance occurred in 15 (9%) of 166, and the remainder (151/166, 91%) cleared lactate.
Lactate clearance versus lactate non-clearance
Table 2 displays data for comparison between lactate clearance and lactate non-clearance groups. Mortality was 60% in the lactate non-clearance group versus 19% in the lactate clearance group (proportion difference, 41% [95% CI, 19%-63%; P < 0.001]). Vasopressor use was not significantly dissimilar between the groups (lactate non-clearance: 11/15 or 73%, and lactate clearance: 87/151 or 58%; P = 0.39). Figure 2 displays the Kaplan-Meier curves for survival fractions over time. The curves diverge significantly by log-rank test (P = 0.003). There was no significant difference in achievement of the ScvO2 goal between the two groups (85% lactate clearance vs. 79% lactate non-clearance; proportional difference, 6% [95% CI, −14% to 26%]; P = 0.84). Table 3 is a 2 × 2 table comparing the concordance of ScvO2 measurement with lactate clearance. There was no evidence of a relationship between lactate clearance and ScvO2 (Fisher exact test, P = 0.457). We found discordance between the two measurements. Specifically, 79% of lactate non-clearance had concomitant ScvO2 70% or greater.
Survivors versus nonsurvivors
Table 4 compares survivors with nonsurvivors. On bivariate analysis, there were four factors that were significantly different (P < 0.05) between survivors and nonsurvivors: (a) initial cardiovascular organ failure (initial SBP <90 mmHg), (b) persistent hypotension (SBP <90 mmHg) despite intravenous fluids, (c) maximum ScvO2 less than 70%, and (d) lactate non-clearance. Table 5 shows the results for the multivariate logistic regression model. Because ScvO2 had a significant association with mortality in the bivariate analysis, we limited the regression model to include only subjects in whom ScvO2 was measured continuously (n = 148). We omitted one of the two hypotension-related covariates (initial SBP <90 mmHg) from inclusion in the model on the grounds that it could be collinear with the persistent hypotension covariate. Lactate non-clearance was found to be a strong independent predictor of in-hospital mortality (odds ratio, 4.9; 95% CI, 1.5-15.9).
In this multicenter sample of ED patients with severe sepsis, we found that early lactate clearance was a strong independent predictor of in-hospital death. In addition, optimization of ScvO2 during resuscitation was not sufficient to exclude lactate non-clearance. This is the first study to evaluate the impact of lactate clearance on survival in the presence of a protocol-directed quantitative resuscitation algorithm for the treatment of patients with severe sepsis, and to date, this is the largest study evaluating lactate clearance in patients with sepsis (10-12).
The cause of an elevated serum lactate in patients with sepsis can be multifactorial. Although lactate elevation may result from acute tissue hypoperfusion and anaerobic metabolism (24), other possible causes may include (a) sepsis-induced impairment of pyruvate-dehydrogenase enzyme activity (25), (b) increased lactate production via catecholamine-driven pathways (26-28), and (c) decreased lactate clearance due to hepatic dysfunction (29, 30). However, regardless of etiology of an elevated serum lactate, lactate elevation in sepsis has been consistently linked to increased mortality (7-12). Although initial reports of lactate clearance in sepsis were published more than 15 years ago (10), evaluating lactate trends during resuscitation is not yet part of the current consensus recommendations for sepsis management (4) and is not routinely performed in practice. While optimizing ScvO2 could in theory be a single comprehensive monitor of the systemic balance between global oxygen delivery and consumption in sepsis patients with circulatory shock (2, 4), our data show that assessment of lactate clearance is important as a predictor of mortality independent of achievement of ScvO2 goals and that tracking ScvO2 does not reliably reflect the effectiveness of lactate clearance during resuscitation. Therefore, these data suggest that serial lactate measurement may provide unique and important information on resuscitation effectiveness.
We acknowledge important limitations in interpreting these results. First, this is a nonexperimental observational study and as such can detect only the association between lactate clearance and mortality but cannot establish cause and effect. Second, because performing serial lactate measurement in patients with sepsis is not a mandatory practice in our centers and measurements are performed at the discretion of the clinician, this could potentially represent a source of selection bias. Third, there was a relatively low incidence (9%) of lactate non-clearance in this cohort. This may be attributable to highly aggressive resuscitation practices in the EDs of the participating centers (14-16). In the only other ED-based study of lactate clearance in severe sepsis, Nguyen et al. (12) reported a lactate non-clearance rate of 29% using the same cutoffs for clearance and non-clearance used in this study. One potential explanation for this disparity is the aggressiveness of crystalloid resuscitation administered in the first 6 h, 4.4 L compared with 3.3 L in the Nguyen et al. study. Therefore, an additional limitation is that the results of the current study may be applicable only to centers that have implemented quantitative resuscitation for sepsis in the ED setting. It is possible that there were deviations from the quantitative resuscitation protocols; however, this study was implemented as an effectiveness trial, so we did not specifically catalogue potential deviations. We identified and enrolled our patients prospectively in the ED, and it is possible that some of the patients diagnosed with sepsis in the ED may have been misclassified and had other etiologies accounting for their presentation. Fourth, our registry did not account for underlying comorbidities that may have had an association with outcome and should be addressed in future studies. Finally, we used all-cause in-hospital mortality as our outcome; it is possible that patients may have died from non-sepsis-related causes.
As severe sepsis carries a persistently high mortality rate and the adequacy of resuscitation is a critical determinant of survival (2, 3), the identification of the optimal markers for restoration of effective tissue perfusion and the ideal end points for a resuscitation algorithm are a high priority for sepsis research. Although ScvO2 optimization has been shown to be important in one clinical trial (2) and was an independent predictor of mortality in the current study, lactate clearance may prove to be an equivalent resuscitation end point for patients with severe sepsis. This hypothesis warrants further investigation in future clinical trials (31).
Early lactate clearance seems to be an important determinant of survival in patients with severe sepsis. In this multicenter study, lactate non-clearance during the resuscitation phase of therapy was a strong independent predictor of in-hospital death. In the context of quantitative resuscitation, assessment of lactate clearance provides unique information on resuscitation effectiveness. ScvO2 optimization does not reliably exclude lactate non-clearance. These data support the rationale for a clinical trial of lactate clearance as a distinct end point of early sepsis resuscitation.
1. Hotchkiss RS, Karl IE: The pathophysiology and treatment of sepsis. N Engl J Med
2. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M: Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med
3. Jones AE, Brown MD, Trzeciak S, Shapiro NI, Garrett JS, Heffner AC, Kline JA: The effect of a quantitative resuscitation strategy on mortality in patients with sepsis: a meta-analysis. Crit Care Med
4. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, Reinhart K, Angus DC, Brun-Buisson C, Beale R, et al.: Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med
5. Ho BC, Bellomo R, McGain F, Jones D, Naka T, Wan L, Braitberg G: The incidence and outcome of septic shock patients in the absence of early-goal directed therapy. Crit Care
6. Peake S, Webb S, Delaney A: Early goal-directed therapy of septic shock: we honestly remain skeptical. Crit Care Med
7. Howell MD, Donnino M, Clardy P, Talmor D, Shapiro NI: Occult hypoperfusion and mortality in patients with suspected infection. Intensive Care Med
8. Shapiro NI, Howell MD, Talmor D, Nathanson LA, Lisbon A, Wolfe RE, Weiss JW: Serum lactate as a predictor of mortality in emergency department patients with infection. Ann Emerg Med
9. Trzeciak S, Dellinger RP, Chansky ME, Arnold RC, Schorr C, Milcarek B, Hollenberg SM, Parrillo JE: Serum lactate as a predictor of mortality in patients with infection. Intensive Care Med
10. Bakker J, Coffernils M, Leon M, Gris P, Vincent JL: Blood lactate levels are superior to oxygen-derived variables in predicting outcome in human septic shock. Chest
11. Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent JL: Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg
12. Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA, Tomlanovich MC: Early lactate clearance is associated with improved outcome in severe sepsis and septic shock. Crit Care Med
13. EMSHOCKNET. Emergency Medicine
Shock Research Network. Available at: www.emshocknet.com
. Accessed: November, 2008.
14. Jones AE, Focht A, Horton JM, Kline JA: Prospective external validation of the clinical effectiveness of an emergency department-based early goal-directed therapy protocol for severe sepsis and septic shock. Chest
15. Shapiro NI, Howell MD, Talmor D, Lahey D, Ngo L, Buras J, Wolfe RE, Weiss JW, Lisbon A: Implementation and outcomes of the Multiple Urgent Sepsis Therapies (MUST) protocol. Crit Care Med
16. Trzeciak S, Dellinger RP, Abate NL, Cowan RM, Stauss M, Kilgannon JH, Zanotti S, Parrillo JE: Translating research to clinical practice: a 1-year experience with implementing early goal-directed therapy for septic shock in the emergency department. Chest
17. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ: 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
18. Surviving Sepsis Campaign: Implementing the Surviving Sepsis Campaign User's Manual (2005)
. Des Plaines, IL: Society of Critical Care Medicine; European Society of Intensive Care Medicine; International Sepsis Forum.
19. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG: The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med
20. Vincent JL, Angus DC, Artigas A, Kalil A, Basson BR, Jamal HH, Johnson G 3rd, Bernard GR: Effects of drotrecogin alfa (activated) on organ dysfunction in the PROWESS trial. Crit Care Med
21. Peduzzi P, Concato J, Feinstein AR, Holford TR: Importance of events per independent variable in proportional hazards regression analysis. II. Accuracy and precision of regression estimates. J Clin Epidemiol
22. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR: A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol
23. Jones AE, Yiannibas V, Johnson C, Kline JA: Emergency department hypotension predicts sudden unexpected in-hospital mortality: a prospective cohort study. Chest
24. Hurtado FJ, Gutierrez AM, Silva N, Fernandez E, Khan AE, Gutierrez G: Role of tissue hypoxia as the mechanism of lactic acidosis during E. coli
endotoxemia. J Appl Physiol
25. Vary TC: Sepsis-induced alterations in pyruvate dehydrogenase complex activity in rat skeletal muscle: effects on plasma lactate. Shock
26. James JH, Luchette FA, McCarter FD, Fischer JE: Lactate is an unreliable indicator of tissue hypoxia in injury or sepsis. Lancet
27. Luchette FA, Jenkins WA, Friend LA, Su C, Fischer JE, James JH: Hypoxia is not the sole cause of lactate production during shock. J Trauma
28. McCarter FD, Nierman SR, James JH, Wang L, King JK, Friend LA, Fischer JE: Role of skeletal muscle Na+
ATPase activity in increased lactate production in sub-acute sepsis. Life Sci
29. Levraut J, Ciebiera JP, Chave S, Rabary O, Jambou P, Carles M, Grimaud D: Mild hyperlactatemia in stable septic patients is due to impaired lactate clearance rather than overproduction. Am J Respir Crit Care Med
30. Pastor CM, Billiar TR, Losser MR, Payen DM: Liver injury during sepsis. J Crit Care
31. Randomized Clinical Trial of Lactate Clearance Versus Central Venous Oxygen Saturation for the Optimal Endpoint of Early Sepsis Resuscitation (NCT00372502)
. Bethesda, MD: National Library of Medicine. Available at: http://www.clinicaltrials.gov
. Accessed January 30, 2009.