Multivariate analysis of blood parameters for predicting mortality in patients with hip fractures

Turan Cihan Dülgeroğlu,Mehmet Kurt,Alaaddin Oktar Üzümcigil,Selçuk Yilmaz,Fatih Karaaslan
DOI: https://doi.org/10.3892/etm.2024.12703
2024-08-30
Abstract:The present retrospective cross-sectional study aimed to evaluate the predictive value of blood parameters and ratios for predicting mortality in patients with hip fractures. In total, 758 patients with hip fractures attending the Department of Orthopedics and Traumatology, Kütahya Health Sciences University Faculty of Medicine (Kütahya, Turkey) between January 2016 and January 2023 were included in the present study. Patients were then divided into two groups, namely the mortality (n=464; 61.2%) and survivor (n=294; 38.8%) groups. Patients in the mortality group were further sub-divided into the following three subgroups: i) Those who succumbed in <1 month (n=117; 25.2%); ii) those who succumbed between 1 and 12 months (n=185; 39.9%); and iii) those who succumbed >12 months later (n=162; 34.9%). In addition, the RDW coefficient of variation, mean platelet volume (MPV), MPV/platelet ratio, neutrophil-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, platelet-to-lymphocyte ratio (PLR), mean platelet volume-to-lymphocyte ratio and monocyte-to-eosinophil ratio means were all found to be significantly higher in the mortality group (P<0.05). MPV (P<0.01), HGB (P<0.05), eosinophil, EOS (P<0.01), HRR (P<0.01), and PLR (P<0.05) were all revealed to exert significant effects on mortality. An age cut-off of 74.50 years had a sensitivity of 81.5% and specificity of 37.1%, whereas an MPV cut-off of 8.85 yielded a sensitivity of 73.5% and specificity of 36.1%. By contrast, an HGB cutoff of 11.05 had a sensitivity of 55.6% and specificity of 35.7%, an eosinophil cut-off of 0.065 had a sensitivity of 47.6% and specificity of 35.4%, whilst a HRR cut-off of 0.7587 had a sensitivity of 55.2% and specificity of 30.3%. Furthermore, a PLR cut-off of 152.620 had a sensitivity of 67.2% and specificity of 41.8% for hip fracture-associated mortality. An age cut-off of 79.50 years had a sensitivity of 70.9% and specificity of 41.5%, while an age cut-off of 83.50 years had a sensitivity of 46.2% and specificity of 64.0% for mortality occurring <1 month after hip fractures. To conclude, results from the present study suggested that HRR has potential predictive value for hip fracture-associated mortality and 30-day mortality, whereas the PLR could only predict hip fracture-associated mortality.
What problem does this paper attempt to address?