The longitudinal associations of sarcopenia definitions with adverse outcomes: a comparative study

Introduction: Frailty, sarcopenia and fragility fractures are closely related to epidemiologically, biologically and clinical impactions. Fragility fracture is observed more frequently in sarcopenic older adults. Older persons who have had a fragility fracture should be assessed for sarcopenia to better develop prevention and recovery after fractures. For these reasons global consensuses tried to create the definition of sarcopenia that best predicts clinical outcomes in the recent years.

Objectives: There is no global consensus definition of sarcopenia despite the great efforts in the recent years. We aimed to study the longitudinal associations of different sarcopenia definitions with functional outcomes.

Methods: We recruited participants admitted to geriatrics outpatient clinics of a university hospital. The patients that have follow-up evaluation for usual gait speed (UGS), activities of daily living (ADL) and instrumental ADL and frailty included. Body composition was assessed by bioimpedance analysis (TANITA BC532). HGS, UGS, ADL, IADL were assessed by Jamar hydraulic hand dynamometer, gait speed at 4m course, Katz and Lawton scales, respectively. Frailty was screened by FRAIL questionnaire. Sarcopenia was defined by EWGSOP1, EWGSOP2 and the two alternative FNIH definitions. EWGSOP2-sarcopenia was evaluated by universally suggested cut-offs (i.e. 27 and 16 kg). EWGSOP2-probable sarcopenia was assessed also by Turkish cut-offs (i.e. 32 and 22 kg) as an additional parameter alternatively. The patients were assessed for detoriation in UGS, ADL, IADL, FRAIL scores and also for decrease in UGS to <= 0.8 m/s and detoriation to frailty.

Results: Among a total of 1881 patients, 264 patients had follow-up data for functional measures and included in the study. The mean age was 75.3+6.3, 195 (73.9%) patients were female with a mean and median follow-up days of 600 and 511 days (Table 1a, 1b). In the first evaluation, the prevalence of sarcopenia ranged between 0.8%-6.1% with standard definitions while it increased to 37% when EWGSOP2-probable sarcopenia was assessed by Turkish cut-offs (probable sarcopenia EWGSOP2-Turkish) (Table 1). The EWGSOP1 or EWGSOP2-confirmed/probable sarcopenia were not associated with adverse outcomes. The EWGSOP2-probable sarcopenia by Turkish cut-offs was associated with detoriation in IADL and decrease in UGS to <= 0.8 m/s (p=0.049 and pearson chi square=3.9; p=0.044 and pearson chi square=4.1, respectively) (Table 2). The FNIH sarcopenia definition including slow UGS was associated with detoriation in IADL (p=0.045, pearson chi square=5.0) (Table 2). In the regression analysis including age, MNA-SF, number of chronic diseases and drugs, dementia and diabetes none of the regression parameters were associated with adverse outcomes.

Conclusions: In this follow-up study of about 1.5 years, the adverse outcomes associated with sarcopenia were decrease in UGS to <= 0.8 m/s and detoriation in IADL. The sarcopenia definitions associated with adverse functional outcomes were EWGSOP2-probable sarcopenia by Turkish cut-offs and FNIH sarcopenia definition including the slow gait speed. EWGSOP2-probable sarcopenia by Turkish cut-offs came forward as having more extensive association with adverse functional outcomes.