Abstract
Introduction:
Risk prediction tools help guide prognostic conversations and benchmarking in hip fracture care. The Nottingham Hip Fracture Score (NHFS) shows only moderate predictive ability for 30-day mortality. We assessed whether routine markers of inflammation could improve the discriminant ability of the NHFS to predict 30-day mortality following hip fracture surgery.
Methods:
We studied consecutive patients admitted with hip fractures at a large-volume trauma unit between 2015 and 2020. Baseline NHFS and postoperative outcome data were extracted from a local registry and linked to routine laboratory data from patients’ electronic clinical records. We selected measurements taken closest to admission pre-operatively. The biomarkers studied were albumin (negative acute-phase reactant), C-reactive protein (CRP), neutrophil-lymphocyte ratio (NLR) and monocyte-lymphocyte ratio (MLR). Univariate and multivariate logistic regression analyses were performed separately for each combination of NHFS and inflammatory marker. C-statistics were calculated to assess the discriminant ability of the NHFS with and without each inflammatory marker for 30-day mortality.
Results:
We included 1710 patients, mean age 82.5 years (SD 8.2). 1199 (70.1%) were women. 104 (6.1%) patients died within 30 days of admission. In univariate analysis, admission NHFS, albumin, CRP and NLR were significantly different between those alive and dead at 30 days. Higher admission albumin was an independent predictor of 30-day mortality in multivariate analysis (OR=0.86 [95%CI 0.81-0.91], p≤0.001) as was higher CRP (OR=1.93 [95%CI 1.04-1.44], p=0.013). The addition of albumin significantly improved the discriminant ability of the NHFS for 30-day mortality (p≤0.001) (c-statistic 0.742 [95%CI 0.683-0.800] vs 0.681 [95%CI 0.617-0.745] for the NHFS alone). Other inflammatory biomarkers did not significantly improve discrimination of 30-day mortality when added to the NHFS.
Conclusions:
Admission albumin improves the discrimination of 30-day mortality in patients undergoing hip fracture surgery when combined with the NHFS, whereas other markers of inflammation including CRP, MLR and NLR did not.