Defining Risk Factors for Perioperative Complications Following Shoulder Arthroscopy: An Analysis of the National Surgical Quality Improvement Program Database

Abstract

IntroductionLimited data are available to prioritize quality improvement initiatives in shoulder surgery. This study identified factors associated with increased complication rates in commonly performed arthroscopic surgical procedures about the shoulder to better guide future quality initiatives.MethodsThis study utilized the 2005-2011 National Surgical Quality Improvement Program (NSQIP) database. Current Procedural Terminology (CPT) codes were used to retrospectively identify shoulder arthroscopy cases (29805-29828). The primary endpoint was the occurrence of any complication within 30-days of surgery, and secondary endpoints were major and minor complications. Bivariate analyses (Chi-square and Fisher's exact tests) and multivariate logistic regression analysis were conducted to assess the relationship between patient risk factors (demographic and preoperative clinical characteristics) and the endpoints. Findings from the bivariate analyses (risk factors with p<0.2) and a theoretical framework were used to identify variables for inclusion in the multivariate logistic regression models for each endpoint.ResultsThe study included 6,530 cases of shoulder arthroscopy procedures. Overall, 78 complications were reported in 69 cases. The overall rates of any, major and minor complications in the cohort were 1.06% (n=69), 0.66% (n=43), and 0.40% (n=26) respectively. Return to operating room (35%, n=27), superficial surgical site infections (14%, n=11) and urinary tract infections (13%, n=10) were the three most frequently occurring post-operative complications in patients undergoing shoulder arthroscopy. After adjusting for relevant patient risk factors, the risk of developing complications was higher for patients undergoing arthroscopy on an inpatient basis [[Adjusted Odds Ratio (AOR): 3.15 , 95% Confidence Limits (CL): 1.73 to 5.71, p<0.01] as compared to those undergoing arthroscopy on an outpatient basis, for patients whose arthroscopy lasted for more than 1.5 hours [AOR: 2.17, 95% CL: 1.37 to 3.44, p<0.01] than those whose arthroscopy was less than 1.5 hours in duration, and for patients with disseminated cancer as compared to those without [AOR: 34.65, 95% CL: 3.56 to 337.05, p<0.01].ConclusionThis study helps identify factors associated with perioperative complications to improve quality of care and provide prognostic information to patients undergoing arthroscopic shoulder surgery. To decrease the risk of complications, operative time should always be carefully monitored, and caution should be sought in patients with disseminated malignancy. While outpatient shoulder arthroscopy is associated with decreased risk of complications compared to inpatient, further research should be conducted to explore this relationship. IntroductionLimited data are available to prioritize quality improvement initiatives in shoulder surgery. This study identified factors associated with increased complication rates in commonly performed arthroscopic surgical procedures about the shoulder to better guide future quality initiatives. Limited data are available to prioritize quality improvement initiatives in shoulder surgery. This study identified factors associated with increased complication rates in commonly performed arthroscopic surgical procedures about the shoulder to better guide future quality initiatives. MethodsThis study utilized the 2005-2011 National Surgical Quality Improvement Program (NSQIP) database. Current Procedural Terminology (CPT) codes were used to retrospectively identify shoulder arthroscopy cases (29805-29828). The primary endpoint was the occurrence of any complication within 30-days of surgery, and secondary endpoints were major and minor complications. Bivariate analyses (Chi-square and Fisher's exact tests) and multivariate logistic regression analysis were conducted to assess the relationship between patient risk factors (demographic and preoperative clinical characteristics) and the endpoints. Findings from the bivariate analyses (risk factors with p<0.2) and a theoretical framework were used to identify variables for inclusion in the multivariate logistic regression models for each endpoint. This study utilized the 2005-2011 National Surgical Quality Improvement Program (NSQIP) database. Current Procedural Terminology (CPT) codes were used to retrospectively identify shoulder arthroscopy cases (29805-29828). The primary endpoint was the occurrence of any complication within 30-days of surgery, and secondary endpoints were major and minor complications. Bivariate analyses (Chi-square and Fisher's exact tests) and multivariate logistic regression analysis were conducted to assess the relationship between patient risk factors (demographic and preoperative clinical characteristics) and the endpoints. Findings from the bivariate analyses (risk factors with p<0.2) and a theoretical framework were used to identify variables for inclusion in the multivariate logistic regression models for each endpoint. ResultsThe study included 6,530 cases of shoulder arthroscopy procedures. Overall, 78 complications were reported in 69 cases. The overall rates of any, major and minor complications in the cohort were 1.06% (n=69), 0.66% (n=43), and 0.40% (n=26) respectively. Return to operating room (35%, n=27), superficial surgical site infections (14%, n=11) and urinary tract infections (13%, n=10) were the three most frequently occurring post-operative complications in patients undergoing shoulder arthroscopy. After adjusting for relevant patient risk factors, the risk of developing complications was higher for patients undergoing arthroscopy on an inpatient basis [[Adjusted Odds Ratio (AOR): 3.15 , 95% Confidence Limits (CL): 1.73 to 5.71, p<0.01] as compared to those undergoing arthroscopy on an outpatient basis, for patients whose arthroscopy lasted for more than 1.5 hours [AOR: 2.17, 95% CL: 1.37 to 3.44, p<0.01] than those whose arthroscopy was less than 1.5 hours in duration, and for patients with disseminated cancer as compared to those without [AOR: 34.65, 95% CL: 3.56 to 337.05, p<0.01]. The study included 6,530 cases of shoulder arthroscopy procedures. Overall, 78 complications were reported in 69 cases. The overall rates of any, major and minor complications in the cohort were 1.06% (n=69), 0.66% (n=43), and 0.40% (n=26) respectively. Return to operating room (35%, n=27), superficial surgical site infections (14%, n=11) and urinary tract infections (13%, n=10) were the three most frequently occurring post-operative complications in patients undergoing shoulder arthroscopy. After adjusting for relevant patient risk factors, the risk of developing complications was higher for patients undergoing arthroscopy on an inpatient basis [[Adjusted Odds Ratio (AOR): 3.15 , 95% Confidence Limits (CL): 1.73 to 5.71, p<0.01] as compared to those undergoing arthroscopy on an outpatient basis, for patients whose arthroscopy lasted for more than 1.5 hours [AOR: 2.17, 95% CL: 1.37 to 3.44, p<0.01] than those whose arthroscopy was less than 1.5 hours in duration, and for patients with disseminated cancer as compared to those without [AOR: 34.65, 95% CL: 3.56 to 337.05, p<0.01]. ConclusionThis study helps identify factors associated with perioperative complications to improve quality of care and provide prognostic information to patients undergoing arthroscopic shoulder surgery. To decrease the risk of complications, operative time should always be carefully monitored, and caution should be sought in patients with disseminated malignancy. While outpatient shoulder arthroscopy is associated with decreased risk of complications compared to inpatient, further research should be conducted to explore this relationship. This study helps identify factors associated with perioperative complications to improve quality of care and provide prognostic information to patients undergoing arthroscopic shoulder surgery. To decrease the risk of complications, operative time should always be carefully monitored, and caution should be sought in patients with disseminated malignancy. While outpatient shoulder arthroscopy is associated with decreased risk of complications compared to inpatient, further research should be conducted to explore this relationship.