Correlation of aspiration results with aseptic loosening in total hip arthroplasty.

In the evaluation of patients with a persistently painful total hip arthroplasty establishing an accurate diagnosis is paramount in the selection of a successful treatment regimen. It is unknown whether synovial analysis might differentiate aseptic loosening from other causes of failure. A physiological basis exists to suggest that aseptic loosening might be a process of non-segmented leukocytes. The objective of this study was to determine if the synovial fluid differential cell count might aid in the diagnosis of aseptic loosening. A retrospective chart review of all patients who had undergone revision hip arthroplasty with pre-operative or intra-operative aspiration results was performed. Aseptic loosening was defined as gross intraoperative movement in the absence of infection. From these results Relative-Operating Characteristic (ROC) curves were created, and sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. A diagnosis of aseptic loosening was established in 76 of the 253 hips. The ROC curves indicated that lymphocyte count does have utility in the diagnosis of aseptic loosening. If an aspirate has a combination of fewer than 2500 WBCs (sensitivity 93%, specificity 42%, NPV 94%, accuracy 57%) or more than 10% lymphocytes (sensitivity 86%, specificity 42%, NPV 87%, accuracy 55%) then the sensitivity for aseptic loosening is 96%, the specificity is 33%, the NPV is 95% and the accuracy is 52%. In patients with painful total hip arthroplasties in whom infection has been excluded, aspiration data can be a useful adjunct in the diagnosis of aseptic loosening. In aspirates with neither a WBC cell count of less than 2500 nor a lymphocyte cell count of greater than 10% aseptic loosening can be effectively "ruled out" as fewer than 5% of these patients will have aseptic loosening. While non-specific, aspirate differential can be useful to "rule-out" aseptic loosening with a sensitivity and negative predictive value well exceeding that of standard radiographs.