Bariatric Surgery in Patients With Type 2 Diabetes Mellitus
I read with great interest the article by Panunzi et al1 conducting a meta-analysis to check the usefulness of body mass index (BMI) as an indicator for predicting remission of type 2 diabetes mellitus (T2DM) by bariatric surgery. The authors selected 94 studies and concluded that BMI had no ability for predicting T2DM remission, and the only significant predictor of glycated hemoglobin A1c (HbA1c) reduction was waist circumference (WC) at baseline. Although metaregression analysis showed that lower baseline WC was associated with higher reduction of HbA1c, the authors considered the significance by selecting bias. I have some queries regarding their conclusion.
First, there is a tendency that patients with more severe state of T2DM mellitus are more likely to be offered bariatric surgery. As patients with relatively lower WC are likely to be at higher baseline HbA1c level in their meta-analysis, the authors speculated that a larger reduction in HbA1c after bariatric surgery was observed in patients with lower WC. As WC is related to insulin resistance, which was also observed in their meta-analysis, I think that an inverse relationship between baseline WC and percent delta HbA1c should be used as a useful marker for predicting remission of T2DM by bariatric surgery. In their Figure 4, basal BMI and WC in their meta-analysis were higher than 25 kg/m2 and 100 cm, respectively, and there is a clear trend of WC against percent delta HbA1c. To confirm the causality of the association, a prospective study controlling for baseline severity of diabetes should be set as the next step of research.
Second, the authors recommended monitoring the changes of central and peripheral adiposity. Does this mean that measuring visceral and subcutaneous adipose tissue should be checked as predictors of T2DM remission? There is a report that combination of BMI and WC can be used for the estimation of visceral adiposity.2 Schlecht et al3 conducted a validation study on ultrasonography for the measurement of visceral and subcutaneous adipose tissue and concluded that ultrasonography could be recommended as a valid tool. In contrast, ultrasonography measurement can be used as a valid method in assessing the amount of visceral adipose tissue by well standardized technique and trained operators.4
Anyway, there is a space for discussion on the valid method of detecting central and peripheral adiposity, including computed tomography, magnetic resonance imaging, or other appropriate apparatus.