Lowering of Lactate Threshold for Severe Sepsis Did Not Impact Mortality

Abstract

Listen

Translate article

SESSION TITLE: Education, Research, and Quality Improvement II SESSION TYPE: Original Investigation Poster PRESENTED ON: Wednesday, October 26, 2016 at 01:30 PM - 02:30 PM PURPOSE: In 2015, NYU Lutheran Medical Center, a 450-bed academic teaching hospital sought to reduce mortality from sepsis. In October 2015 the criteria for severe sepsis were modified and the lactate level threshold was decreased from 2.3 mmol/L to 2.0 mmol/L. Furthermore, the diagnosis of sepsis was broadened to include organ dysfunction irrespective of lactate level. Our hospital also identified interdisciplinary teams for intensive education, including simulation training and interactive online modules, to promptly recognize sepsis. Nonetheless, the decreased lactate threshold remained the primary intervention to improve sepsis outcomes. We present a retrospective analysis that demonstrates no significant impact on mortality and and, unexpectedly, worse compliance with the recommended 30ml/kg initial fluids bolus. METHODS: Patients with DRG codes for severe sepsis and septic shock were identified through our EMR. We compared cumulative three-month mortality and compliance with guidelines to provide 30ml/kg initial fluid resuscitation prior and after the change. We used a chi square analysis to test the equality between proportions with the null hypothesis that there was no difference in mortality nor compliance with the fluid resuscitation guidelines for the 3 months prior to and after October 2015 RESULTS: Cumulative mortality from June to September 2015 was 43.9% with a total of 39 patients and mortality for October-December 2015 was 31% with a total of 49 patients. The change was statistically insignificant with a p-value of 0.19. Compliance with fluid resuscitation was 25% for the earlier quarter versus 10% for the later quarter. This change was also statistically insignificant with a p-value of 0.06. CONCLUSIONS: Our experience with lowering of the lactate threshold has not shown any mortality benefit. Although not statistically significant we see a trend of less compliance with the initial fluid resuscitation guidelines. Poor compliance with automated fluid administration is likely due to the lack of applicability of standard volume of fluid to disparate patient populations with variable hemodynamic profiles. Secondly, this analysis may be too premature to detect an impact on fluid compliance or mortality as only three months data reviewed after the change in lactate levels and larger data sets will need to be collected. CLINICAL IMPLICATIONS: Based on our experience, we advocate for individualized and ultrasound guided sepsis management as lowering the lactate threshold and fluid resuscitation compliance guidelines did not alter outcomes. DISCLOSURE: The following authors have nothing to disclose: Farshid Rafatnia, Ravindra Rajmane No Product/Research Disclosure Information