There is inadequate information on the association of pre-operative and postoperative peak level of lactate with mortality of surgical ICU patients.
To investigate the association between peri-operative lactate level and 90-day mortality in patients admtted to the surgical ICU.
Retrospective cohort study.
ICUs in single tertiary academic hospital.
Adult patients postoperatively admitted to the ICU between January 2012 and December 2017.
MAIN OUTCOME MEASURES
Hazard ratios of 90-day mortality according to the following serum lactate levels were assessed: pre-operative lactate level; peak lactate levels on postoperative day (POD) 0 to 3; and delta values of the lactate level on POD 0 to 3 from pre-operative lactate level. Multivariable Cox regression and receiver operating characteristic analyses were used.
Overall 9248 patients were included, among whom 2511, 8690 and 1958 had measured pre-operative lactate levels, lactate levels within POD 0 to 3, and lactate levels measured at both timepoints, respectively. When the peak lactate level on POD 0 to 3 and delta lactate level all increased by 1 mmol l−1
, 90-day mortality increased by 15% [hazard ratio: 1.15; 95% confidence interval (CI) 1.11 to 1.19; P
< 0.001] and 14% (hazard ratio: 1.14; 95% CI 1.11 to 1.18; P
< 0.001), respectively; the pre-operative lactate level was not significantly associated with 90-day mortality (P
= 0.069). The area under the curve for peak level of lactate on POD 0 to 3 (0.72, 95% CI 0.70 to 0.74) was higher than that of pre-operative lactate level (0.58, 95% CI 0.56 to 0.60) in the receiver operating characteristic analysis.
In patients admitted postoperatively to the ICU, higher peri-operative lactate levels were associated with increased 90-day mortality. The peak level of lactate during POD 0 to 3 showed the most significant contribution to this association.