Identification of the high risk hip fracture patients

Sammanfattning: Background: Hip fractures are common in older individuals entailing considerable risk of morbidity and mortality. Instruments are needed to enhance risk stratification of hip fracture patients. There is limited knowledge of the role of biomarkers as risk prognosticators for mortality in hip fracture patients.Aims: We assessed how well two well known risk stratification tools performed in a cohort of Swedish hip fracture patients. We investigated whether lactate concentration at admission could help identify hip fracture patients at risk of adverse outcome and performed an exploratory study in which we analysed 92 different biomarkers in plasma samples taken at admission and their possible role as prognosticators. Finally we evaluated the prognostic capabilities of plasma cystatin C, creatinine and shrunken pore syndrome assessed at admission in the same cohort.Methods: This thesis included 997 patients who were admitted to Lund University Hospital with hip fracture between 2011 and 2014. Blood samples were taken from the patients within 3 hours of admission. Follow up was done for one year with the help of the Swedish Hip Fracture Registry. The risk scores we evaluated were the Nottingham Hip Fracture Score (NHFS), the POSSUM and the P-POSSUM score. We analysed plasma lactate, creatinine, cystatin C and 92 biomarkers which are part of the Proseek Multiplex CVD-panel I.Results: During the study period 1845 patients were admitted to Lund University hospital with a hip fracture, 997 of these patients (64%) were included in the cohort and the thirty-day mortality was 6.2%. Lactate at admission had area under the ROC curve as a predictor 30-d mortality of 0.51 (95% CI 0.45–0.57). The area under the ROC curve for prediction of 30-day mortality using the combined POSSUM physiological and operative severity score was 0.66 (95% CI 0.59-0.72) and for the NHFS it was 0.67 (95% CI 0.59-0.74). Of the 92 biomarkers we evaluated, 11 of them remained associated with thirty-day mortality after adjustment with known risk factors. GDF-15 added to the NHFS resulted in net reclassification improvement of 12.1% (95% CI 1.2 - 23.3). Only CA-125 increased significantly the AUC for prediction of 30-day mortality when added to the NHFS. Cystatin C and its derived variables, including the shrunken pore syndrome remained associated with mortality in the adjusted analysis. In contrast, creatinine and eGFRcreatinine were not associated with the outcome in the adjusted analysis.Conclusions: Plasma concentrations of lactate at admission in hip fracture patients did not predict thirty day mortality in our patient cohort. POSSUM, P-POSSUM scores and the NHFS showed equally poor discrimination with regards to 30-day mortality in our cohort. Adding GDF-15 or CA-125 to the Nottingham Hip Fracture Score improved discrimination in our cohort with regard to prediction of 30-day mortality. Plasma concentration of cystatin C, eGFRcystatin and the Shrunken Pore Syndrome are associated with increased mortality when other well established risk factors for mortality are accounted for in patients with hip fracture