Back to 2016 Annual Meeting

BACKGROUND: Diabetes mellitus is an increasingly prevalent comorbidity in the United States impacting 9.3% (29.1 million) of Americans, as estimated by the Centers for Disease Control and Prevention. Additionally, approximately 30.0% (86 million) of Americans have prediabetes, and an estimated 15-30% of these prediabetic patients will develop type 2 diabetes mellitus within 5 years. A large body of research has established diabetes as a risk factor for a myriad of chronic illnesses. Comparatively, there is substantially less research examining whether diabetes is associated with poorer surgical outcomes. In plastic surgery particularly, numerous clinical studies have substantiated that current diabetics are at higher risk of wound-related complications. However, the majority of this research has been conducted among patients undergoing breast procedures. Thus, there is a current paucity of literature on the specific ramifications of diabetes on different reconstructive surgeries. Herein, this study aims to identify how diabetes mellitus status is associated with post-operative outcomes and wound-related complications in a wide range of plastic and reconstructive surgery procedures. This is the largest known study examining the link between diabetes mellitus and plastic surgery outcomes, and uses data from the 2007-2012 American College of Surgeons National Surgery Quality Improvement Program (ACS-NSQIP) datasets.
METHODS: Using the American College of Surgeons National Surgical Quality Improvement Program dataset, demographics and outcomes were examined for patients who underwent plastic surgery between 2007 and 2012. The 2012 Current Procedural Terminology (CPT) codes were used to identify and classify these procedures. Multivariable logistic regression models were used to assess the relationship between diabetes status and a spectrum of medical and surgical post-operative outcomes, including: postoperative renal insufficiency, myocardial infarction, deep vein thrombosis, graft or flap loss, wound complications, including wound disruption, deep wound infection, sepsis, and return to operating room. These analyses were adjusted for sex, race, age, operation year, and CPT code. Additionally, patients were evaluated for length of in-patient hospitalization while controlling for multiple demographic factors and type of procedure.
RESULTS: 40,465 plastic surgery patients were identified who had undergone breast, upper and lower extremity, abdominal, and craniofacial procedures. The cohort included 7.91% diabetics. Logistic regression analysis showed that diabetics had a higher likelihood of surgical (OR 1.7, p< 0.001) and medical (OR 1.8, p<0.001) complications than non-diabetics. Diabetic patients had a higher likelihood of both superficial wound occurrence (OR 1.4, p<0.001) and deep incisional surgical site infection occurrence (OR 0.5, p<0.001) than non-diabetic patients. Furthermore, diabetic patients had increased odds for wound complications (OR 1.6, p<0.001) as well as wound disruption (OR 1.5, p=0.013) compared to non-diabetics. A negative binomial model revealed that the average hospital length of stay was 1.8 times longer for a diabetic than for a non-diabetic (p<0.001).
CONCLUSIONS: Diabetes increases a multitude of post-operative complications, and the overall risk profile of patients undergoing plastic surgery, as demonstrated in this largest analysis to date. The negative consequences of diabetes status on outcomes following plastic surgery can be used to counsel patients on the importance of optimizing pre-operative diabetes management, particularly glycemic control.