Which Pre-Operative, Modifiable Risk Factors are Most Predictive of Complications in orthopedic Upper Extremity Surgery?

Identification of modifiable comorbid conditions in the preoperative period is important in optimizing outcomes. We evaluate the association between such risk factors and postoperative outcomes after upper extremity surgery using a national database. The National Surgical Quality Improvement Program (NSQIP) 2006-2016 database was used to identify patients undergoing an upper extremity principle surgical procedure using CPT codes. Modifiable risk factors were defined as smoking status, use of alcohol, obesity, recent loss of >10% body weight, malnutrition, and anemia. Outcomes included discharge destination, major complications, bleeding complications, unplanned re-operation, sepsis, and prolonged length of stay. Chi square and multivariable logistic regressions were used to identify significant predictors of outcomes. Significance was defined as P<0.01. After applying exclusion criteria, 53,780 patients were included in the final analysis. Preoperative malnutrition was significantly associated with non-routine discharge (OR=4.75), major complications (OR=7.27), bleeding complications (OR=7.43), unplanned re-operation (OR=2.44), sepsis (OR=10.22), and prolonged length of stay (OR=5.27). Anemia was associated with non-routine discharge (OR=2.67), bleeding complications (OR=13.27), and prolonged length of stay (OR=3.26). In patients who had a weight loss of greater than 10%, there was an increase of non-routine discharge (OR=2.77), major complications (OR=2.93), and sepsis (OR=3.7). Smoking, alcohol use, and obesity were not associated with these complications. Behavioral risk factors (smoking, alcohol use, and obesity) were not associated with increased complication rates. Malnutrition, weight loss, and anemia were associated with an increase in postoperative complication rates in patients undergoing upper limb orthopaedic procedures and should be addressed prior to surgery, suggesting nutrition labs should be part of the initial blood work. Level of evidence: III