7 Year Study on Glucose Targets and Post-operative Surgical Site Infection and Mortality

Author list
Shanice Ker1, Roderica Ng2, Sophia Chew1,2,3
1Yong Loo Lin School of Medicine, National University of Singapore
2Department of Anaesthesiology, Singapore General Hospital
3Cardiovascular and Metabolic Disorders Programme, Duke-NUS Medical School

Abstract

Background and Objectives
Hyperglycaemia is associated with adverse outcomes, including increased morbidity and mortality, in cardiac surgical patients. There is overriding evidence that glycaemic control improves morbidity and mortality in such patients. A quality improvement project was initiated to keep target glucose between 4-8mmol/l perioperatively.  Thereafter, this was revised to 10mmol/l in accordance with the Society of Thoracic Surgeons guidelines. However, the association of these changing glucose targets with surgical site infection (SSI) and mortality remains to be elucidated. This retrospective study aims to determine if changing target glucose levels affects the incidence of SSI and mortality amongst Asian patients undergoing coronary artery bypass grafting (CABG).

Methods
All patients having cardiac surgery at our Centre were placed on insulin infusions titrated to achieve perioperative target glucose levels of 4-8mmol/l in 2009-2010 and 10mmol/l from 2011. The first glucose upon arrival in the cardiothoracic intensive care unit (CTSICU) was recorded. Outcomes were incidence of SSI and mortality over the 7 year period. Univariate analysis was done using Pearson chi-square test for categorical variables, and independent sample t-test for quantitative variables. 

Results
4925 patients underwent CABG over this 7-year period.  The mean first glucose level upon arrival in CTSICU was well within target of 4-10mmol/l with no significant variation. Patients undergoing CABG were mainly males, Chinese, hypertensive, with Euroscore of 2.2. Prevalence of diabetic patients was approximately 50%. The average SSI rate was 3.14% in 2008 which fell to 1.65% after initiation of the project, remaining stable thereafter. There was no association between glucose control and mortality.

Conclusion
Glucose targets were maintained throughout the 7-year period with the implementation of the insulin regime. There was a fall in the incidence of SSI at the beginning of the project, but no further fall thereafter.  This may be partially attributed to the Hawthorne effect. The sustained control of SSI suggests that other variables are in play.  We postulate preoperative glycaemic control is important as the prevalence of diabetes is significant in our population.  Furthermore, the target levels may be too liberal given the high incidence of diabetes which would have to be further elucidated. Mortality remained low and was not associated with change in glucose targets.

7 Year Study on Perioperative Glucose Targets and Post-operative Surgical Site Infection and Mortality