Response to Letters Regarding Article, “The Impact of Body Mass Index and the Metabolic Syndrome on the Risk of Cardiovascular Disease and Death in Middle-Aged Men”

We thank Dr Stefan and colleagues for their insightful comments about our article.¹ Given the missing data on waist circumference and given that no oral glucose tolerance test was performed at baseline, they suggest that our findings may be explained by a potential misclassification of the metabolic syndrome (MetS), insulin resistance, and diabetes mellitus. We agree that we cannot exclude that some participants were misclassified at baseline. However, in contrast to Stefan et al, we believe that this had no major influence on our results. First, all obese participants without the MetS fulfilled the modified National Cholesterol Education Program’s waist circumference criteria (Body Mass Index >29.4 kg/m²).¹ Thus, no additional individuals in this group would have been reclassified as having the MetS even if data on waist circumference would have been available. Second, even if we assume that all overweight participants fulfilled the original waist circumference criteria (>102 cm), 323 participants of our study sample would still be classified as overweight without the MetS. Importantly, these participants were at increased risk for cardiovascular events compared with normal-weight individuals without the MetS (multivariable hazard ratio 1.43, 95% confidence interval 1.16 to 1.76, P<0.001). Third, the fact that our results were similar in participants still free from diabetes mellitus after 20 years follow-up would argue against the potential inclusion of individuals with unidentified diabetes mellitus at baseline as an explanation of our findings. Thus, in our view, the hypothesis of Stefan et al is not supported by our data. Our findings challenge the notion of a healthy obese phenotype as defined by the absence of the MetS or insulin resistance; however, we do agree that other definitions of “healthy” obesity merit further longitudinal studies. Moreover, studies in women, other ethnicities, and other age groups are warranted.

We also appreciate Dr McAuley and colleagues’ relevant comments. Because higher physical activity and cardiorespiratory fitness have beneficial effects on insulin resistance and all MetS components,² it is likely that the overweight/obese without the MetS or insulin resistance in our study were more fit than those with the MetS or insulin resistance. Interestingly, when data on leisure time physical activity³ were added to the multivariable model, the risk for cardiovascular events remained similar (hazard ratio for overweight without MetS: 1.48, 95% confidence interval 1.24 to 1.76, P<0.001; for obese without MetS: 1.89, 95% confidence interval 1.08 to 3.30, P=0.02). However, we acknowledge the limitation that no data on cardiorespiratory fitness were available at baseline. As highlighted by Dr McAuley et al, several important questions remain unanswered about the complex interplay between obesity and cardiovascular risk.