SESSION TITLE: Outcomes/Quality Control
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Wednesday, October 29, 2014 at 08:45 AM - 10:00 AM
PURPOSE: To determine the rate of adoption and deadoption of tight glycemic control (TGC) in adult ICUs following respective publication of the Leuven (Leuven I, November 2001), and NICE-SUGAR trials (March 2009).
METHODS: We conducted an interrupted time series analysis of adult patients admitted to ICUs between January 1, 2001 and December 31, 2012 that contribute data to the APACHE database. The publication dates for Leuven I, and NICE-SUGAR were the main exposure variables. Blood glucose data within the first 24 hours of ICU admission (GLUC24) was used as the measure of glycemic control (previously validated), and a GLUC24 ≤ 6.1 mmol/L was considered TGC. Data was examined via segmented regression analysis using multi-level linear and logistic regression models with admissions clustered by ICU.
RESULTS: The study population consisted of 368,030 ICU admissions in 102 ICUs. The mean (standard deviation) age and APACHE (III or IV) score were 61.9 (17.0) years and 56.0 (28.0) respectively. Prior to Leuven I, 20.3% (95% CI 19.7 - 20.8%) of ICU admissions had TGC. After publication of Leuven I, the proportion of admissions with TGC increased by 0.50% per quarter (95% CI 0.40 - 0.60%, p < 0.0001). Surgical patients demonstrated the greatest increase in TGC (odds ratio (OR) per quarter 1.04, 95% CI 1.03 - 1.04, p < 0.0001). Publication of NICE-SUGAR decreased the rate of increase in TGC (p = 0.002 for change in quarterly rate), but not the overall proportion of admissions treated with TGC (p = 0.7 for quarterly trend). The proportion of patients with TGC during the final quarter of the study was 36.0% (95% CI 34.9 - 37.1%). The incidence of hypoglycemia (GLUC24 < 4.0mmol/L) increased significantly after Leuven I (p <0.0001 for quarterly trend) and did not decrease after NICE-SUGAR (p = 0.3 for quarterly trend). Hypoglycemia was experienced by 2.7% (95% CI 2.5 - 3.0%) of patients at the start of the study period, compared to 5.1% (95% CI 4.6 - 5.7%) at the end.
CONCLUSIONS: Among adult patients admitted to ICUs in the United States, there was a significant increase in TGC and hypoglycemia after Leuven I with little change following NICE-SUGAR.
CLINICAL IMPLICATIONS: This data suggests that there has not been a reversal of medical practice for TGC and that there is an urgent need to understand and promote the de-adoption of ineffective clinical practices.
DISCLOSURE: The following authors have nothing to disclose: Daniel Niven, Gordon Rubenfeld, Andrew Kramer, Henry Stelfox
No Product/Research Disclosure Information