Abstract
BACKGROUND: Over the past years, several cardiac risk indices were evaluated and modified, including Goldman, Detsky, and Lee scores. The predictive capacity of these scores in hip fracture patients is lacking. Thus, our objective was to compare the Goldman, Detsky, and Lee scores as predictors of mortality in 6 months after hip fracture. METHODS: We prospectively evaluated 80 consecutive patients with hip fractures, over the age of 65 admitted to an orthopedic ward at Botucatu Medical School. Patient demographic information, Goldman, Detsky and Lee scores were recorded. All patients were followed for 6 months after hip fracture, and mortality was recorded. Multiple logistic regression analyses were performed for mortality prediction. RESULTS: The mortality rate was 23% after a 6-month follow-up period. Patients who died had advanced age and the majority of them were male. They also had lower values of handgrip strength, and higher values of creatinine and urea. In the multiple logistic regression models when adjusted by age, gender, handgrip strength and creatinine, Goldman's score (OR:3.025; 95%CI:1.022-8.953; p:0.046), but not Detsky (OR:2.328; 95%CI:0.422-12.835; p:0.332) and Lee (OR:1.262; 95%CI:0.649-2.454; p:0.494), was associated with mortality 6 months after hip fracture. Each 1 category increase in Goldman score increased the mortality to more than 3-fold. CONCLUSIONS: In conclusion, our data suggest that Goldman score, but not Detsky or Lee indices, predicts mortality associated with hip fracture at up to 6 months post-injury.