Comments on prevalence and incidence of sarcopenia in the very old: findings from the Newcastle 85+ study

Abstract

We have read the article entitled, ‘Prevalence and incidence of sarcopenia in the very old: findings from the Newcastle 85+ Study’ by Dodds et al.,1 and it is our wish to congratulate the authors for their interesting article and, in turn, to add some contributions on the subject.

The study evaluates handgrip strength, gait speed, and lean mass as a part of the definition of sarcopenia proposed by the European Working Group on Sarcopenia in Older People (EWGSOP).2 The handgrip strength was measured using a dynamometer, a method that is suggested by the mentioned consensus. However, when it comes to cut-off points for weak-grip strength, consensus within the EWGSOP established 20 kg for women and 30 kg for men.2 We notice carefully that the study used the cut-off points suggested by Studenski et al.3 which are 16 kg for women and 26 kg for men. We appreciate the use of these points, since the Studenski publication is more recent and had a larger number of participants (n = 26 625), including participants from different parts of the globe, such as the United States, Italy, Iceland, Puerto Rico, which allows the cut-off points to be more standardized and not directed to a single race or ethnic group.

However, we would like to emphasize that these differences in adopting a certain cut-off point can contribute to the continued misclassification of different participants in the studies and, also, that a globally accepted point, which we can use in Latin American countries, has not been established yet. This kind of misclassification can lead several studies to obtain different results and therefore contradictory conclusions.

Additionally, it is important to mention that this is a retrospective cohort study, in which the baseline prevalence and incidence of sarcopenia was determined. Logistic regression models were fitted to obtain the odds ratio (OR), instead of using the Poisson regression model to calculate the rate ratio or relative risk (RR), which is the most appropriate for a reported incidence of 10.7. Epidemiological studies show that OR may overestimate the real value of the association between variables.4, 5

Finally, we would like to emphasize that Latin America does not have a consensus regarding the cut-off points for the measurement of the different criteria of sarcopenia, which makes it difficult to expand research studies in this specific condition. Previous publications have used the recommendations of EWGSOP. However, controversy in its application can lead to erroneous conclusions, especially when there is clear evidence of differences between the anthropometric measures of the Latin American, European, and Asian population.

The authors have no conflicts of interest.