The relationship between diabetes and the reoperation rate after lumbar spinal surgery: a nationwide cohort study.

Abstract

Diabetes is present in 5% to 20% of patients undergoing spine surgeries and is a known risk factor for reoperation. Considering the chronicity of diabetes, its influence on the reoperation rate may differ over time. To present the relationship between diabetes and the reoperation rate over time. Retrospective cohort study. A national health insurance database was used to identify a cohort of patients who underwent an initial surgery for lumbar degenerative disease in 2003 (n=34,918). The primary end point was any type of second lumbar surgery after fusion surgery (n=4,792) or decompression surgery (n=30,126) during the early (0-postoperative 90 days), short-term (91-365 days), and midterm (1-6 years) periods. All patients were followed up until December 2008. Cox proportional hazards regression modeling was used to assess the adjusted reoperation rates in the diabetic patients. The incidence of diabetes in the present cohort was 24.5% in the fusion group and 16.9% in the decompression group. Overall, reoperation was performed in 13.2% (631 of 4,792) of the patients after fusion surgery and in 14.0% (4,214 of 30,126) of the patients after decompression surgery. After fusion surgery, diabetes did not make a significant difference in the reoperation rate during the entire follow-up period. After decompression surgery, the reoperation rate was not different during Postoperative Month 3, but diabetic patients showed a 1.2 to 1.4 times higher reoperation rate during postoperative 3 months to 5 years (p<.01). The study did not find a relationship between diabetes at the time of surgery and the reoperation rate during the early postoperative period. Thereafter, the reoperation rate was not higher after fusion surgery in diabetic patients, but it was higher after decompression surgery.