Abstract
Introduction
Sarcopenia is common in patients with hip fracture, but few studies have examined whether assessment of sarcopenia improves prediction of adverse post-operative outcomes. We examined whether sarcopenia, diagnosed using handgrip strength (HGS), could predict outcomes after hip fracture.
Methods
Routinely collected data from the National Hip Fracture Database were combined with locally collected HGS data from a high-volume orthopaedic trauma unit. Patients aged ≥65years with surgically managed, non-pathological hip fracture with grip strength measured on admission were included. The European Working Group on Sarcopenia in Older People (EWGSOP2) thresholds were used to identify patients with or without sarcopenia; those unable to complete grip strength testing were also included in analyses. Outcomes examined were 30-day and 120-day mortality, residential status and mobility, prolonged length of stay (>15 days) and post-operative delirium. Binary logistic regression models were used to examine prognostic value of HGS, and discriminant ability for the Nottingham Hip Fracture Score (NHFS) alone and on adding sarcopenia status were compared using c-statistics.
Results
We analysed data from 282 individuals; mean age 83.2 (SD 9.2) years; 200 (70.9%) were female. 99 (35.1%) patients had sarcopenia and 109 (38.7%) were unable to complete testing. Sarcopenia predicted higher 120-day mortality (OR 13.0, 95%CI 1.7-101.1, p=0.014), but not 30-day mortality (OR 1.5, 95%CI 0.1-16.9, p=0.74). Patients unable to complete HGS testing had higher 30-day mortality (OR 13.5, 95%CI 1.8-103.8, p=0.012) and 120-day mortality (OR 34.5, 95%CI 4.6-258.7, p<0.001). Sarcopenia status did not significantly improve discrimination for mobility but improved prediction of 120-day residential status (c-statistic 0.89 [95%CI 0.85-0.94] for NHFS+sarcopenia vs 0.82 [95%CI 0.76-0.87] for NHFS alone) and post-operative delirium (c-statistic 0.91 [95%CI 0.87-0.94] vs 0.78 [95%CI 0.73-0.84]).
Conclusion
Sarcopenia assessment via HGS testing may provide additional prognostic information to existing risk scores in older patients with hip fracture.