Jourdan M Cancienne1, Matthew J Deasey1, Michelle E Kew1, Brian C Werner2. 1. Department of Orthopaedic Surgery, University of Virginia Health System, Charlottesville, Virginia, U.S.A. 2. Department of Orthopaedic Surgery, University of Virginia Health System, Charlottesville, Virginia, U.S.A.. Electronic address: bcw4x@virginia.edu.
Abstract
PURPOSE: To determine the association between glycemic control and adverse events after arthroscopic rotator cuff repair (RCR). METHODS: Patients with a diagnosis of diabetes mellitus who underwent arthroscopic RCR and had a hemoglobin A1c (HbA1c) level determined within 3 months before or after surgery were identified in a national database and stratified by HbA1c level. The incidence of postoperative infection within 6 months was determined using Current Procedural Terminology and International Classification of Diseases, Ninth Revision codes. A receiver operating characteristic (ROC) curve analysis was performed to determine whether a threshold HbA1c level existed above which the risk of infection, revision rotator cuff surgery, and lysis of adhesions (LOA)-manipulation under anesthesia (MUA) after arthroscopic RCR was significantly increased. This threshold was then tested using a logistic regression analysis. RESULTS: The study included 3,740 patients with an infection rate ranging from a low of 0.29% to a high of 1.14% after RCR. The inflection point of the ROC curve for infection corresponded to an HbA1c level between 7.0 and 8.0 mg/dL (P = .035; area under the curve, 0.648; specificity, 61%; sensitivity, 59%). We then used 8.0 mg/dL as a threshold to test for adverse outcomes. We found a significant difference in infection rates for patients with levels below versus above the threshold (0.30% vs 0.84%; OR, 2.0; 95% confidence interval, 1.2-3.4; P = .014) but no difference in revision rates (P = .240) or LOA-MUA (P = .650) in patients with levels above versus below the threshold. CONCLUSIONS: The risk of infection after RCR in patients with diabetes mellitus increases as the perioperative HbA1c level increases and, although statistically significant, remains low. ROC curve analysis determined that a perioperative HbA1c level above 8.0 mg/dL could serve as a threshold level; however, the area under the curve and low sensitivity reflected the poor utility of this test as an independent predictor. This study did not find an association between increased perioperative HbA1c levels and rates of revision rotator cuff surgery or LOA-MUA after RCR. LEVEL OF EVIDENCE: Level III, retrospective cohort study.
PURPOSE: To determine the association between glycemic control and adverse events after arthroscopic rotator cuff repair (RCR). METHODS:Patients with a diagnosis of diabetes mellitus who underwent arthroscopic RCR and had a hemoglobin A1c (HbA1c) level determined within 3 months before or after surgery were identified in a national database and stratified by HbA1c level. The incidence of postoperative infection within 6 months was determined using Current Procedural Terminology and International Classification of Diseases, Ninth Revision codes. A receiver operating characteristic (ROC) curve analysis was performed to determine whether a threshold HbA1c level existed above which the risk of infection, revision rotator cuff surgery, and lysis of adhesions (LOA)-manipulation under anesthesia (MUA) after arthroscopic RCR was significantly increased. This threshold was then tested using a logistic regression analysis. RESULTS: The study included 3,740 patients with an infection rate ranging from a low of 0.29% to a high of 1.14% after RCR. The inflection point of the ROC curve for infection corresponded to an HbA1c level between 7.0 and 8.0 mg/dL (P = .035; area under the curve, 0.648; specificity, 61%; sensitivity, 59%). We then used 8.0 mg/dL as a threshold to test for adverse outcomes. We found a significant difference in infection rates for patients with levels below versus above the threshold (0.30% vs 0.84%; OR, 2.0; 95% confidence interval, 1.2-3.4; P = .014) but no difference in revision rates (P = .240) or LOA-MUA (P = .650) in patients with levels above versus below the threshold. CONCLUSIONS: The risk of infection after RCR in patients with diabetes mellitus increases as the perioperative HbA1c level increases and, although statistically significant, remains low. ROC curve analysis determined that a perioperative HbA1c level above 8.0 mg/dL could serve as a threshold level; however, the area under the curve and low sensitivity reflected the poor utility of this test as an independent predictor. This study did not find an association between increased perioperative HbA1c levels and rates of revision rotator cuff surgery or LOA-MUA after RCR. LEVEL OF EVIDENCE: Level III, retrospective cohort study.