Summary

Patients with insulin dependent diabetes were much more likely to have surgical complications following knee arthroscopy, and medical complications following shoulder arthroscopy, while patients with non-insulin dependent diabetes have comparable risks to nondiabetics.

Abstract

Introduction

Diabetes mellitus is a known risk factor for postoperative complications following arthroscopy. Diabetes can further be stratified by method of treatment, including insulin-dependence (IDDM) and non-insulin dependent diabetes (NIDDM). Patients with IDDM have been shown to have much higher rates of postoperative complications, morbidity, and mortality following spine surgery and total joint arthroplasty than nondiabetics or those with NIDDM. The goal of this study is to compare complications following shoulder and knee arthroscopy in patients stratified by their method of glycemic control. Our hypothesis is that patients with IDDM will have a higher risk for postoperative complications than nondiabetics or NIDDM after shoulder and knee arthroscopy.

Methods

A retrospective analysis of the American College of Surgeon’s National Surgical Quality Improvement Program’s (ACS-NSQIP) database for the years 2005-2016 was conducted. Logistic regressions were used to assess the relationship between diabetic status and outcomes. Multivariate models were established to adjust for age, sex, BMI, hypertension, congestive heart failure, chronic obstructive pulmonary disease, smoking status, Charlson comorbidity index and functional status. A post-hoc power analysis was performed based on the observed variances between groups using an alpha equal to 0.05 and a power equal to 0.8.

Results

58,442 patients undergoing shoulder arthroscopy and 86,023 patients undergoing knee arthroscopy were identified. Following knee arthroscopy, patients with IDDM were at a much higher risk for surgical complications, AOR 2.186 (95% CI: 1.226-1.157, p = 0.001), including deep infections (AOR 3.082, 95% CI: 1.753-5.419, p < 0.001) and return to OR (AOR 1.933, 95% CI: 1.280-2.919, p = 0.002), as well as unplanned hospital admission (AOR 1.770, 95% CI: 1.289-2.431, p = 0.002). Following shoulder arthroscopy, patients with IDDM were at a much higher risk for medical complications (AOR 1.524, 95% CI: 1.082-2.147), hospital readmission (AOR 1.581, 95% CI: 1.153-2.169), and 30-day mortality (AOR 3.821, 95% CI: 1.243-11.750). Conversely, patients with NIDDM had comparable risks in all measured outcomes as nondiabetics.

Conclusions

Patients with IDDM were much more likely to have surgical complications following knee arthroscopy, and medical complications following shoulder arthroscopy. Additionally, patients with NIDDM actually have comparable risks to nondiabetics. This data has implications for quality reporting, particularly for surgeons and facilities who care for patients with greater inherent risk of postoperative complications.