Weighing in on Body Mass Index and Infection After Total Joint Arthroplasty: Is There Evidence for a Body Mass Index Threshold?

BACKGROUND: Although morbid obesity is considered a modifiable risk factor for periprosthetic joint infection (PJI), there is no consensus regarding an appropriate threshold for body mass index (BMI) above which a high risk for infection may outweigh the benefits of surgery. QUESTIONS/PURPOSES: (1) Is there a BMI cutoff threshold that is associated with increased risk for PJI? (2) Is the risk of PJI increased in higher obesity classes?
METHODS: A retrospective study was conducted of all primary THAs and TKAs performed at one institution between 2006 and 2015. Overall 19,226 patients were eligible to be included in the study; 1053 patients were excluded as a result of incomplete data, resulting in a final cohort of 18,173 patients (8757 TKAs and 9416 THAs). PJI was defined using the International Consensus Meeting criteria. To ensure accurate followup, and because there is evidence to support the association between obesity and early infection, we identified PJI within 90 days of the index surgery. This relationship was examined separately for BMI as a continuous variable and for each BMI category as defined by the Centers for Disease Control and Prevention (underweight ≤ 18.49 kg/m; normal 18.5-24.9 kg/m; overweight 25-29.9 kg/m; obese class I 30-34.9 kg/m; obese class II 35-39.9 kg/m; obese class III ≥ 40 kg/m). Analyses were performed with logistic regression, accounting for both patient and surgical risk factors. A BMI threshold was evaluated with a receiver operating characteristic (ROC) curve and the Youden index.
RESULTS: The area under the ROC curve for BMI and risk of PJI within 90 days was only 0.58 (confidence interval [CI], 0.52-0.63) suggesting such a cutoff was not much better than random chance. Among the BMI classes, patients with class III obesity (≥ 40 kg/m) were the only ones showing a higher risk for PJI within 90 days (odds ratio [OR], 3.09 [1.46-6.54]; p = 0.003). The risk of developing PJI was not greater for overweight (OR, 0.72; 95% CI, 0.38-1.4), class I obese (OR, 1.06; 95% CI, 0.57-2.0), or class II obese (OR, 1.08; 95% CI, 0.52-2.2) patients. Underweight patients also demonstrated no increased risk for PJI (OR, 1.80; 95% CI, 0.23-13.9).
CONCLUSIONS: The risk for infection increases gradually throughout the full range of BMI, but no threshold exists. Weight reduction before surgery may mitigate risk for infection for all patients with a BMI above normal. Of note, patients with a BMI > 40 kg/m carried a threefold higher risk for PJI and for these patients, the risks of surgery must be carefully weighed against its benefits. LEVEL OF EVIDENCE: Level III, therapeutic study.