Abstract
Aim
Traditionally, serum white blood count (WBC), C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) have been utilized as markers to evaluate septic arthritis (SA). Recently, neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) have been identified as prognostic factors for treatment failure, mortality and morbidity in various clinical settings. To date, these markers have not been utilized for evaluating outcomes after hip and knee SA.
Thus, the purpose of this study was to determine the ability of admission NLR and PLR to predict treatment failure and postoperative 90-day mortality in hip and knee SA.
Method
A retrospective study was performed using our institutional research patient database to identify 235 patients with native hip and knee septic arthritis from 2000–2018. Patient demographics, comorbidities and social factors (alcohol intake, smoking and intravenous drug use) were obtained, and NLR and PLR were calculated based on complete blood count values (absolute neutrophil, lymphocyte and platelet count) on admission. Treatment failure was defined as any reoperation or readmission within 90 days after surgery. Receiver operating curves were analyzed, and optimal thresholds for NLR and PLR were determined using Youden's test. Univariate and multivariate analyses were performed to determine if these ratios were independent predictors of treatment failure and 90-day mortality after surgery. These ratios were compared to serum WBC, CRP, and ESR.
Results
Optimal thresholds for NLR was 9.49 (sensitivity=60%, specificity=84%) and PLR was 303 (sensitivity=54%, specificity=77%). With univariate analysis, NLR>9.49 was associated with failure (odds ratio [OR]=7.64, 95%
Confidence Interval [CI]=4.10–14.21) and 90-day mortality (OR=9.83, 95% CI=2.74–35.25). PLR>303 was associated with increased failure (OR=3.85, 95% CI=2.12–7.00). In multivariate analysis controlling for patient demographics, comorbidities and social factors, elevated NLR remained an independent predictor of failure (OR=7.04, 95% CI=3.78–13.14) and 90-day mortality (OR=5.98, 95% CI=1.60–22.32), whereas PLR remained a predictor of failure (OR=3.58, 95% CI=1.95–6.58). NLR was a better predictor of failure and 90- day mortality compared to serum WBC, CRP, and ESR.
Conclusions
This study demonstrates that NLR is a good estimate of SA and performs better than serum WBC, CRP and
ESR to predict treatment failure and 90-day mortality. Elevated NLR is a reliable novel biomarker that may be utilized when evaluating SA patients, and this accessible test could be utilized in the musculoskeletal infection field.
