Description
Objective: Malnutrition is paradoxically highly prevalent among patients with obesity undergoing total hip (THA) and knee (TKA) arthroplasty. Clinical guidelines recommend preoperative weight loss, but its benefits may be limited in malnourished patients by exacerbating nutritional deficiencies. The goals of this study were to determine if preoperative malnutrition and insulin resistance (IR) laboratory markers predict postoperative outcomes and limit the effect of preoperative weight loss.
Methods: Among 21,038 primary THAs and 23,726 TKAs performed between 2002 and 2019, we identified 6128 patients with preoperative BMIs>30 kg/m2 measured 1-24 months before surgery, a weight measured at surgery, and preoperative laboratory makers of malnutrition and IR. Malnutrition was defined as total lymphocyte count (TLC) <1,500 cells/mm3, albumin <3.5 g/dL, or transferrin <200 mg/dL, and IR as hemoglobin A1c >6.5%. The mean age was 67 years with 55% female. The mean BMI was 36 kg/m2. Logistic and Cox regressions evaluated prosthetic joint infections (PJI), complications, revisions, and reoperations. Mean follow-up was 5 years.
Results: Preoperative labs met criteria for malnutrition and IR in 38% (2131/5547) and 26% (773/2943) of patients, respectively. Malnutrition overall was not significantly associated with PJI, complications, revisions, or reoperations, however, hypoalbuminemia increased the risk of reoperation (HR=1.5, p=0.035). IR increased the risk of PJI (HR=2.45, p=0.003) and complications (HR=1.4, p=0.041) but was not significantly associated with revisions or reoperations. In multivariable analyses comparing losing ≥10 pounds to maintaining preoperative weight, malnutrition was not a significant effect modifier of PJI (HR=1.03, p=0.89), complication (HR=1.2, p=0.11), revision (HR=0.95, p=0.72), or reoperation (HR=1.1, p=0.25).
Conclusions: Malnutrition (38%) and IR (26%) are relatively common among patients with obesity undergoing THA and TKA. Preoperative screening should include albumin and hemoglobin A1c, while TLC and transferrin do not predict postoperative risks. Malnutrition does not significantly impact the effect of preoperative weight loss on postoperative outcomes.
