Diagnosis of Sarcopenia in Long-Term Care Homes for the Elderly: The Sensitivity and Specificity of Two Simplified Algorithms with Respect to the EWGSOP Consensus

Objective

To analyze the prevalence of sarcopenia in elderly care homes using the algorithm of the European Working Group on Sarcopenia in Older People (EWGSOP) and to evaluate the applicability, sensitivity, and specificity of two simplified algorithms that do not include gait speed measurement.

Design

Cross-sectional study.

Setting

Long-term care homes.

Participants

249 residents (187 females and 62 males) with a mean age of 84.9 ± 6.7 years (range, 70–106 yrs).

Measurements

Cognitive impairment was evaluated with the Pfeiffer test, functional capacity with the Barthel index, walking ability by the Functional Ambulation Classification, muscle mass by bioelectrical impedance analysis, muscle strength by handgrip dynamometry and, gait speed over a distance of 4 m. Sarcopenia was assessed using the EWGSOP-recommended algorithm and two simplified algorithms (A and B). Algorithms A and B were validated with respect to the EWGSOP algorithm using the MacNemar test and obtaining the sensitivity and specificity, kappa coefficient, and area under the receiver operating characteristic curve (AUC).

Results

Total or severe dependence was recorded in 67% of the 249 residents, severe or moderate cognitive impairment in 64%, inability to walk in 49%, and capacity to walk only with the aid of an ambulator or two people in 21%. The prevalence of sarcopenia was 63.0% by the EWGSOP algorithm, 62.9% by algorithm A and 63.2% by algorithm B (P = 1.000); 143 participants were diagnosed with sarcopenia by all three methods (P =1.000, K = 1.000).

Conclusion

Further studies that include residents with functional and/or cognitive impairment are required to validate these results, comparing the diagnostic performance of the EGWSOP algorithm with that of the simplified algorithms validated in the present report.