Clinical and Laboratory Predictors of One-Year Mortality After Proximal Femoral Nailing for Geriatric Intertrochanteric Fractures: Do Albumin and Total Protein Provide Additional Prognostic Value?

Background and Objectives: Serum albumin has been widely studied as a prognostic marker after geriatric hip fracture, whereas less is known about total protein and the additional prognostic information provided by these markers beyond routine clinical factors. This study examined the associations of preoperative albumin and total protein with one-year mortality in older patients with intertrochanteric femur fractures treated with proximal femoral nailing and assessed whether these markers improved mortality prediction when added to routinely available clinical and laboratory variables. Materials and Methods: This retrospective cohort study included 317 patients aged ≥65 years who underwent proximal femoral nailing for intertrochanteric femur fractures. Clinical variables and admission laboratory parameters, including hemoglobin, creatinine, neutrophil and lymphocyte counts, neutrophil-to-lymphocyte ratio, albumin, and total protein, were evaluated. The primary outcome was all-cause mortality within one year after surgery. Multivariable logistic regression was used to identify factors independently associated with mortality. The additional prognostic value of albumin and total protein was assessed by changes in the area under the receiver operating characteristic curve (AUC) and likelihood-ratio tests. Results: Among 317 patients, 133 (42.0%) died within one year. Age (adjusted odds ratio [aOR] 1.14 per year, 95% CI 1.09–1.19; p < 0.001), ASA class III (aOR 4.71, 95% CI 1.27–17.47; p = 0.021), time to surgery (aOR 1.27 per day, 95% CI 1.12–1.44; p < 0.001), and log-transformed creatinine (aOR 3.56 per e-fold increase, 95% CI 1.89–6.69; p < 0.001) were independently associated with one-year mortality. Albumin was lower among patients who died (median 3.60 vs. 3.80 g/dL; p = 0.009), but its adjusted association with mortality was not statistically significant and was imprecisely estimated (aOR 0.78, 95% CI 0.37–1.64; p = 0.506). Total protein was also not independently associated with mortality (aOR 1.11, 95% CI 0.68–1.82; p = 0.681). The base clinical model had an AUC of 0.809 (95% CI 0.761–0.854), compared with 0.810 (95% CI 0.764–0.855) after addition of albumin and total protein (ΔAUC = 0.0013, 95% CI −0.0029 to 0.0058; likelihood-ratio p = 0.801). Conclusions: Age, ASA class, time to surgery, and renal function were independently associated with one-year mortality. Although lower preoperative albumin was associated with mortality in unadjusted analysis, its adjusted association was not statistically significant and was imprecisely estimated. In this cohort, there was no clear evidence that adding albumin and total protein provided incremental prognostic value beyond the base clinical model.