Abstract
BACKGROUND: Many serologic markers are routinely tested prior to joint arthroplasty, but only few are commonly used to guide surgeons in determining patients most at risk of periprosthetic joint infection (PJI). The objective of this study was to investigate the association between preoperative bilirubin level and PJI after primary hip and knee arthroplasty. METHODS: A retrospective analysis was performed on patients undergoing revision hip and knee arthroplasty at our hospital from January 2016 to December 2019. Laboratory biomarkers were collected before the primary arthroplasty, as well as general patient information. The association between the above serologic markers and postoperative PJI was analyzed. RESULTS: A total of 72 patients (30 hips/42 knees) were analyzed, including 39 patients with PJI and 33 patients without PJI. Except for total bilirubin (TB) and direct bilirubin (DB), there was no significant difference between the remaining laboratory biomarkers. The preoperative TB and DB in the PJI group were 10.84 ± 0.61 μmol/L and 3.07 ± 0.19 μmol/L, respectively, which were lower than those in the non-PJI group (14.68 ± 0.75 μmol/L and 4.70 ± 0.39 μmol/L, P < 0.001). The area under the curve (AUC) of preoperative TB to predict PJI was 0.755 (P < 0.001, cutoff = 11.55 μmol/L, sensitivity = 66.67%, specificity = 75.76%). Meanwhile, the AUC of preoperative DB was 0.760 (P < 0.001, cutoff = 4.00 μmol/L, sensitivity = 84.62%, specificity = 54.45%). CONCLUSIONS: The serum levels of TB and DB before the primary arthroplasty were lower in PJI patients than in non-PJI patients, and the preoperative values lower than 11.55 μmol/L and 4.00 μmol/L could be considered as a risk factor for postoperative PJI.