Preoperative Diabetes Mellitus and Postoperative Acute Kidney Injury in Colectomies: A Possible Association
Abstract
Introduction and Objective. Colectomies put patients at risk for postoperative acute kidney injury (PO-AKI). This study sought to determine if pre-existing diabetes mellitus led to higher incidence of PO-AKI in colectomy patients and if this association was modified by use of an open vs laparoscopic surgical technique. Methods. A historical cohort study of 35,643 adult colectomy patients was conducted utilizing the 2016 American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database. PO-AKI within 30 days of surgery in patients with diabetes mellitus was assessed as the primary outcome through bivariate analysis while controlling for demographics and comorbidities. The influence of open vs laparoscopic surgical technique was analyzed with an effect modification analysis. Results. Of the 35,643 colectomy patients who met criteria, 5,504 (15.4%) had diabetes mellitus and 268 (0.75%) experienced PO-AKI. Among the 5,504 colectomy patients with diabetes, there were 65 (1.2%) that had PO-AKI, while among the 30,139 patients without diabetes, there were 203 (0.7%) that had POAKI (p < 0.0001). Before adjustment, patients with diabetes had a significant increase in odds of PO-AKI (OR 1.76 [95% CI 1.33-2.33] p < 0.001). However, after adjusting for all control variables, there was no longer a significant odds increase in the PO-AKI odds for patients with diabetes (OR 0.94 [95% CI 0.69-1.29] p-value 0.716). Additionally, those with a BMI>or=30, a GFR < 45, a surgical site infection, a postop transfusion, sepsis, CHF, or an ASA classification status of 3 or greater had higher odds of developing PO-AKI. An effect modification analysis of open vs laparoscopic surgical technique revealed no significant differences in outcome. Conclusions-Implications. While there was no association between prior diabetes mellitus and PO-AKI in colectomy patients, some of the factors adjusted for might have played the role of confounders and partial mediators. Therefore, despite not observing an association, the potential effect of diabetes on PO-AKI cannot be ruled out, it is possible that some variables were acting as both confounders and mediators which may mask an associative relationship. This may be served by a mediation analysis in the future. Therefore, further studies need to focus on elucidating the true nature of the association while incorporating more detailed information about type, onset, and severity of both diabetes mellitus and PO-AKI in colectomy patients, which may be better presented in other databases.
Keywords
Post Operative Acute Kidney Injury, Colectomy, Diabetes Mellitus
Presentation Type
Poster Presentation
Preoperative Diabetes Mellitus and Postoperative Acute Kidney Injury in Colectomies: A Possible Association
Introduction and Objective. Colectomies put patients at risk for postoperative acute kidney injury (PO-AKI). This study sought to determine if pre-existing diabetes mellitus led to higher incidence of PO-AKI in colectomy patients and if this association was modified by use of an open vs laparoscopic surgical technique. Methods. A historical cohort study of 35,643 adult colectomy patients was conducted utilizing the 2016 American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database. PO-AKI within 30 days of surgery in patients with diabetes mellitus was assessed as the primary outcome through bivariate analysis while controlling for demographics and comorbidities. The influence of open vs laparoscopic surgical technique was analyzed with an effect modification analysis. Results. Of the 35,643 colectomy patients who met criteria, 5,504 (15.4%) had diabetes mellitus and 268 (0.75%) experienced PO-AKI. Among the 5,504 colectomy patients with diabetes, there were 65 (1.2%) that had PO-AKI, while among the 30,139 patients without diabetes, there were 203 (0.7%) that had POAKI (p < 0.0001). Before adjustment, patients with diabetes had a significant increase in odds of PO-AKI (OR 1.76 [95% CI 1.33-2.33] p < 0.001). However, after adjusting for all control variables, there was no longer a significant odds increase in the PO-AKI odds for patients with diabetes (OR 0.94 [95% CI 0.69-1.29] p-value 0.716). Additionally, those with a BMI>or=30, a GFR < 45, a surgical site infection, a postop transfusion, sepsis, CHF, or an ASA classification status of 3 or greater had higher odds of developing PO-AKI. An effect modification analysis of open vs laparoscopic surgical technique revealed no significant differences in outcome. Conclusions-Implications. While there was no association between prior diabetes mellitus and PO-AKI in colectomy patients, some of the factors adjusted for might have played the role of confounders and partial mediators. Therefore, despite not observing an association, the potential effect of diabetes on PO-AKI cannot be ruled out, it is possible that some variables were acting as both confounders and mediators which may mask an associative relationship. This may be served by a mediation analysis in the future. Therefore, further studies need to focus on elucidating the true nature of the association while incorporating more detailed information about type, onset, and severity of both diabetes mellitus and PO-AKI in colectomy patients, which may be better presented in other databases.