Diabetes has been found to be an independent predictor of several postoperative complications, including superficial surgical site infections (SSIs), sepsis and 30-day readmission, in patients who underwent coronary artery bypass graft (CABG) surgery. According to the researchers, further investigation of the results may aid surgical planning and prevention of these adverse outcomes.
“Diabetic patients are common for CABG surgeries and present with unique postoperative risks,” said Ziyad Knio, BS, a biostatistician at Beth Israel Deaconess Medical Center and Harvard Medical School, both in Boston. “After univariate and multivariate analysis, diabetes was independently associated with 30-day readmission, urinary tract infection, sepsis, superficial surgical site infection and blood transfusions.”
Co-author Rabya Saraf, BA, who also presented the findings at the 2016 annual meeting of the Society of Cardiovascular Anesthesiologists (abstract 11), said the link between diabetes and increased risk for cardiovascular disease is well established, but the association between diabetes and complications after cardiac surgery remains less studied.
“One study examining a cohort of diabetic patients who underwent CABG surgery found that patients with type 1 diabetes had a twofold increase in risk of death as compared to nondiabetics,” said Ms. Saraf, who cited a study from the Journal of the American College of Cardiology (2015;65:1644-1652). “However, the association between diabetes and other post-CABG outcomes has not been studied on a large, prospective cohort.”
National Quality Database Used
In the American College of Surgeons National Surgical Quality Improvement Program database, the researchers identified 8,666 CABG patients from 2007 to 2013, by querying Current Procedural Terminology codes. Of those patients, 3,287 (38%) were diabetic. For this retrospective study, a patient was considered diabetic if his or her treatment included either prescribed insulin or non-insulin antidiabetic agents.
The researchers conducted a univariate analysis comparing diabetic with nondiabetic patients to identify outcomes potentially associated with diabetes. Univariate analysis of these outcomes was then used to identify possible additional predictors. Finally, these predictors were entered into a stepwise logistic regression model to find which ones remained significant and could be considered independent predictors of the postsurgical outcomes.
On univariate analysis, diabetes was associated with the following postoperative complications: 30-day readmission, acute renal failure, urinary tract infection, cerebrovascular accident/stroke, sepsis, blood transfusions needed intraoperatively or within 72 hours postoperatively, sepsis and superficial SSIs.
With each outcome, the researchers next performed a univariate analysis to determine additional predictors (e.g., diabetes, body mass index >30 kg/m2, American Society of Anesthesiologists physical status ≥3, smoking, history of congestive heart failure, hypertension requiring medication). These variables were then entered into a stepwise logistic regression model to determine predictive value.
After a risk-adjusted model was created, diabetes was independently associated with the following predictors: 30-day readmission, urinary tract infection, sepsis, superficial SSIs and blood transfusions (Table).
As a follow-up to this study, Mr. Knio and Ms. Saraf intend to investigate the identified outcomes to discover the underlying mechanism of diabetes that might account for these increased risks.
“You will notice that the odds ratios are all over 1,” Mr. Knio said. “Clearly, diabetes is associated with more of these complications, and we plan to investigate those further.”
Intensive Glycemic Control and Mortality
The moderator of the session, Thomas Floyd, MD, CM, FRCPC, professor of anesthesiology at Stony Brook School of Medicine, in New York, said given the results of the NICE-SUGAR trial (N Engl J Med 2009;360:1283-1297), which “demonstrated that aggressive insulin management in patients led to increased mortality,” the research presented by Mr. Knio and Ms. Saraf did not come as a surprise.
“Greater insulin resistance is indicative of severe multisystem organ disease,” Dr. Floyd said. “If you’re diabetic, I’m not sure that correcting your glucose level is going to improve your outcome, but I think that whether or not a patient can get that in line is indicative of systemic disease.
“Determining the impact of intensive glycemic control on mortality … will hopefully lead us to more focused therapy in the future,” he concluded.