Low serum albumin has been shown in the primary joint arthroplasty setting to increase the rate of perioperative complications. Our present work examined a large national inpatient administrative dataset to study the relationship between serum albumin level and key outcome measures after revision total knee arthroplasty (RTKA). Our hypothesis was that lower serum albumin would be an independent risk factor for poor outcomes after RTKA. We analyzed the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database from 2006 to 2014, specifically evaluating patients undergoing RTKA. Patients were grouped as having hypoalbuminemia (serum albumin < 3.5 mg/dL) or normal albumin (serum albumin ≥ 3.5). We analyzed data on 22 complications as reported in the NSQIP database and developed composite complication variables (any infections, cardiac/pulmonary complications, and any major complications). For each complication, multivariable logistic regression analysis was used to evaluate its association. The cohort included 4,551 patients undergoing RTKA. Patients in the low serum albumin group were statistically more likely to develop deep surgical site infection, organ space surgical site infection, pneumonia, urinary traction infection, and sepsis. The hypoalbuminemic group was more likely to require unplanned intubation, blood transfusion intraoperatively or postoperatively, remain on a ventilator > 48 hours, and develop acute renal failure. There was also a higher risk of mortality and coma. Across the three composite complication variables, any complication (with or without transfusion), any major complication, and any infection (systemic, wound) were more prevalent among the patients with low serum albumin. This study confirms the relationship between suboptimal nutritional status and complications following RTKA. Hypoalbuminemia may be used as a potential preoperative predictor of outcomes. Understanding the effects of malnutrition on perioperative complications informs the choice of appropriate candidates for surgical intervention, timing of surgery, resource allocation, and risk counseling preoperatively.
Thieme Medical Publishers 333 Seventh Avenue, New York, NY 10001, USA.