Sarcopenia definitions: where to draw the line? Response to Scarlata et al

Abstract

We thank Scarlata et al 1 for their interest in our study 2 in which we demonstrated a 15% prevalence in stable COPD by European Working Group on Sarcopenia in Older People (EWGSOP) criteria. 3 Professor Scarlata and colleagues acknowledge these are the most commonly adopted sarcopenia criteria. Consequently, prevalence estimates in other populations 4 allow us to compare our findings beyond respiratory medicine. Towards the end of recruitment, The Foundation for the National Institutes of Health (FNIH) Sarcopenia Project published their findings. 5 We acknowledge the strength of the FNIH criteria, with cut points derived from a large pool of patient-level data. Nonetheless, the project team had to make difficult decisions around data harmonisation and handling, and by group's own admission the criteria are a 'work in progress' to be evolved and refined. 6Professor Scarlata and colleagues suggest the FNIH employ gait speed as an end point rather than a parameter of muscle function. However, low gait speed (<0.8 m/s) was the principal criterion for mobility disability, from which grip strength and then appendicular lean mass cut points were derived. In this respect, we view gait speed as an intrinsic component of the FNIH operationalisation. 5With regards to lean mass assessment, the FNIH elected only to accept dual-energy X-ray absorptiometry (DXA) measurements and in doing so excluded over half of their data set (14 603/26 625 cases). While DXA offers a more accurate assessment of lean mass than bioelectrical impedance analysis, it may not be readily available in practice, particularly within settings where sarcopenia is relevant (nursing homes, critical care). In contrast, the EWGSOP criteria are more pragmatic and accept use of bioelectrical impedance analysis, a measure we routinely use in practice.7 Finally, when deciding between absolute and normalised cut points, FNIH data were inconsistent across genders; body mass index adjustment did not affect relationships between sarcopenia domains in men, but markedly improved the fit of models in women. 5 Further, for all combinations of FNIH-defined low strength and mass, there were no significant associations with survival among women. 6 This contrasts with recent studies in which EWGSOP-defined sarcopenia carried a twofold mortality risk. 8 We are unable to offer a sarcopenia prevalence by FNIH criteria owing to lack of DXA measures in our cohort. However, 2119 patients within the FNIH data set had a COPD diagnosis and disease-specific publications may emerge. 5 We agree that multiple definitions may hinder efforts to treat sarcopenia. However, we are encouraged that a consistent phenotype is emerging in which a lean mass criterion is combined with one or more markers of physical function. This is important and exciting given the effectiveness of exercise, and nutritional and pharmacological interventions on functional end points. We observed a modest whole-group effect on all EWGSOP sarcopenia domains following pulmonary rehabilitation.2 To this end, patients ...

Sarcopenia and frailty in chronic respiratory disease

“…To highlight this issue, the Foundation for the National Institutes of Health (FNIH) Sarcopenia Group required measures derived from DXA and in doing so had to exclude more than half of their validation data set in whom measures were unavailable. 53 , 54 In contrast, the European Working Group on Sarcopenia in Older Persons (EWGSOP) criteria are more pragmatic and accept the use of BIA, a practical measure routinely used in our day-to-day practice, 55 but this may overestimate muscle mass, particularly in overweight or obese patients, resulting in a ‘hidden’ population with undiagnosed sarcopenic obesity.…”

Correlation between body composition and disease severity in patients with chronic obstructive pulmonary disease

“…In this study, FFMI and PhA were included in the linear regression equation, and the results showed that FEV1%pred was correlated with PhA, but not with FFMI, indicating that PhA was more sensitive than FFMI in predicting lung function in COPD patients. It has also been found that CAT scores and mMRC scores were significantly higher in patients with low PhA compared with patients with normal PhA in COPD, whereas there was no significant difference in CAT and mMRC scores between patients with low FFMI and those with normal FFMI, suggesting that PhA is more sensitive than FFMI in predicting symptoms in patients with COPD (44). A study by Vitalii et al (45) showed that the frequency of acute exacerbations of COPD was negatively correlated with FFMI.…”