We read with interest the article by Lontchi-Yimagou et al. (1), in which they describe diabetes in individuals with low BMI from India. This is a valuable contribution to the literature that suggests the need for more investigation into this phenotypic category. However, it seems premature to suggest that the findings indicate the need to revisit the category of “malnutrition-related diabetes mellitus,” given the lack of any causal link between malnutrition and the findings reported.
Compared with type 2 diabetes (T2D) in White Caucasians, T2D in Asian Indians has unique phenotypic features, including lower BMI. We (2) have shown that there exists a subcategory of T2D known as severe insulin-deficient diabetes, which differs from the “low-BMI diabetes” described by Lontchi-Yimagou et al. (1) only in the degree of leanness, which could reflect differences in the populations studied rather than a specific clinical phenotype. Indeed, the BMI of the representative Asian Indian population that we used for validation of the severe insulin-deficient diabetes subcategory shows considerable overlap with those of the participants with lean diabetes in the current study. Similarly, the metabolic characteristics of lean diabetes (low insulin secretion and less insulin resistance) noted by the authors are only to be expected on the basis of low BMI and need not represent a unique pathophysiology. In this context, it is pertinent to recall that Asian Indians with dysglycemia have, overall, poorer β-cell function than White Caucasians (3). It is therefore possible that the lean individuals studied by Lontchi-Yimagou et al. (1) represent an extreme end of the spectrum of T2D in Asian Indians rather than a specific subcategory.
For individuals with diabetes to be grouped into a specific diagnostic category, several criteria need to be fulfilled: a specific pathogenic mechanism (e.g., autoimmune β-cell destruction in type 1a diabetes) should be identified, and a categorical clinical or genetic marker (e.g., presence of pancreatic calculi in fibrocalculous pancreatic diabetes or specific mutations in monogenic diabetes) should be present. Until these criteria are fulfilled, it would be rational to consider these lean individuals as having a subtype of T2D rather than assigning them to a new category with arbitrary cutoffs based on a continuous variable such as BMI.
Indeed, even though the concept of malnutrition-related diabetes mellitus first gained acceptance in 1985 when the World Health Organization accepted it as a specific subtype of diabetes (4), further studies over the course of the following two decades failed to prove that it differs sufficiently from other forms of diabetes to warrant its status as a separate diagnostic category (5). Unsurprisingly, therefore, the syndrome is not mentioned in subsequent classification systems of diabetes.
Notwithstanding the caveats described above, we are in complete agreement with the authors that lean non–type 1 (ketosis-resistant) diabetes in the Asian Indian population warrants further study. Further characterization of pathophysiology, phenotype, and genotype of these individuals would decide whether lean diabetes can move beyond the realm of semantics and be established as a distinct diagnostic entity.
See accompanying article, p. e159.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
