Association Between Early Lactate Levels and 30-Day Mortality in Clinically Suspected Sepsis in Children

Abstract

Importance: Improving emergency care of pediatric sepsis is a public health priority, but optimal early diagnostic approaches are unclear. Measurement of lactate levels is associated with improved outcomes in adult septic shock, but pediatric guidelines do not endorse its use, in part because the association between early lactate levels and mortality is unknown in pediatric sepsis.

Objective: To determine whether the initial serum lactate level is associated with 30-day mortality in children with suspected sepsis.

Design, setting, and participants: This observational cohort study of a clinical registry of pediatric patients with suspected sepsis in the emergency department of a tertiary children's hospital from April 1, 2012, to December 31, 2015, tested the hypothesis that a serum lactate level of greater than 36 mg/dL is associated with increased mortality compared with a serum lactate level of 36 mg/dL or less. Consecutive patients with sepsis were identified and included in the registry following consensus guidelines for clinical recognition (infection and decreased mental status or perfusion). Among 2520 registry visits, 1221 were excluded for transfer from another medical center, no measurement of lactate levels, and patients younger than 61 days or 18 years or older, leaving 1299 visits available for analysis. Lactate testing is prepopulated in the institutional sepsis order set but may be canceled at clinical discretion.

Exposures: Venous lactate level of greater than 36 mg/dL on the first measurement within the first 8 hours after arrival.

Main outcomes and measures: Thirty-day in-hospital mortality was the primary outcome. Odds ratios were calculated using logistic regression to account for potential confounders.

Results: Of the 1299 patients included in the analysis (753 boys [58.0%] and 546 girls [42.0%]; mean [SD] age, 7.3 [5.3] years), 899 (69.2%) had chronic medical conditions and 367 (28.3%) had acute organ dysfunction. Thirty-day mortality occurred in 5 of 103 patients (4.8%) with lactate levels greater than 36 mg/dL and 20 of 1196 patients (1.7%) with lactate levels of 36 mg/dL or less. Initial lactate levels of greater than 36 mg/dL were significantly associated with 30-day mortality in unadjusted (odds ratio, 3.00; 95% CI, 1.10-8.17) and adjusted (odds ratio, 3.26; 95% CI, 1.16- 9.16) analyses. The sensitivity of lactate levels greater than 36 mg/dL for 30-day mortality was 20.0% (95% CI, 8.9%-39.1%), and specificity was 92.3% (90.7%-93.7%).

Conclusions and relevance: In children treated for sepsis in the emergency department, lactate levels greater than 36 mg/dL were associated with mortality but had a low sensitivity. Measurement of lactate levels may have utility in early risk stratification of pediatric sepsis.