Katherine J. Deans, MD; Peter C. Minneci, MD
In patients with septic shock treated with hydrocortisone, is intensive insulin therapy (IIT) better than conventional insulin therapy (CIT) for reducing in-hospital mortality? Does fludrocortisone reduce in-hospital mortality?
2 x 2 factorial randomized controlled trial (RCT) (Corticosteroids and Intensive Insulin Therapy for Septic Shock [COIITSS] trial). ClinicalTrials.gov NCT00320099.
11 intensive care units (ICUs) in France.
509 patients (mean age 64 y, 65% men) who had severe sepsis (American College of Chest Physicians/Society of Critical Care Medicine criteria) with multiple organ dysfunction (Sequential Organ Failure Assessment score ≥ 8), needed vasopressor therapy to maintain blood pressure levels (systolic > 90 mm Hg or mean > 60 mm Hg), and were receiving intravenous (IV) hydrocortisone, 50 mg every 6 h. Exclusion criteria included life expectancy < 1 day at ICU admission.
Continuous IV IIT with blood glucose target levels of 80 to 110 mg/dL (n = 126); IIT plus oral 9-α-fludrocortisone, 50-µg tablet daily for 7 days (n = 129); IV or subcutaneous CIT (n = 138); or CIT plus fludrocortisone (n = 116). All patients received IV hydrocortisone, 50-mg bolus every 6 hours, for 7 days.
Primary outcome was in-hospital or 90-day mortality. Other outcomes included 180-day mortality and serious adverse events. The trial had 80% power to detect a 25% relative reduction for in-hospital mortality from 50% (α = 0.05).
95% (intention-to-treat analysis); 100% for primary outcome.
IIT did not reduce in-hospital or 180-day mortality and increased risk for severe hypoglycemia more than CIT; groups did not differ for risk for superinfection (Table). Adding fludrocortisone to hydrocortisone therapy did not reduce mortality but increased risk for superinfection (Table).
In adults with septic shock, intensive insulin therapy did not reduce mortality more than conventional therapy. Adding fludrocortisone to hydrocortisone therapy did not reduce mortality.
Intensive (IIT) vs conventional insulin therapy (CIT) and fludrocortisone (+FLU) vs no fludrocortisone (−FLU) in patients with septic shock†
†Abbreviations defined in Glossary. RRI, RRR, NNH, NNT, and CI calculated from data in article.
‡ Glucose < 40 mg/dL.
Deans KJ, Minneci PC. Intensive insulin therapy did not reduce mortality more than conventional therapy in septic shock treated with corticosteroids. Ann Intern Med. 2010;152:JC5–5. doi: 10.7326/0003-4819-152-10-201005180-02005
Download citation file:
Published: Ann Intern Med. 2010;152(10):JC5-5.
Cardiology, Coronary Risk Factors, Diabetes, Endocrine and Metabolism, Infectious Disease.
Results provided by:
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use