PubMed1997
Table 1 is a synopsis of the major findings from an extensive literature on the association between human body habitus and coronary heart disease. Whilst some studies have used quite sophisticated laboratory procedures to quantify body fat most have relied upon anthropometric measurements to determine some component of body habitus. Of these, body weight and height are the simplest measurements and are, therefore, well-suited to large-scale prospective studies. Height and weight are highly reproducible measurements, although in the short term, weight can have considerable physiological variation associated with gastric emptying and state of hydration. Less reliable measurements than height and weight are skinfolds and body circumferences, both of which have been used extensively in cross-sectional and prospective analyses. For skinfolds, both the inter and intra-observer variability is affected by the measurement technique, location of the skinfold site, the skinfold caliper used and skinfold compressibility. As measurement error has been shown to be a function of skinfold thickness, accurate and repeatable skinfold measurements are particularly difficult to make in the obese. In these subjects, it is not always possible to locate a specific anatomical bony landmark or to pull a parallel skinfold away from the underlying tissue. Furthermore, in the extremely obese it is sometimes possible for a skinfold to be thicker than the jaws of the currently available commercial calipers. Alternately, body circumferences are obtainable in all subjects and have greater reproducibility than skinfolds. They are, therefore, the preferred method in obese subjects. However, there is considerable work to be done to establish their association with body fatness. The evidence examined in this review suggests that body weight is a poor predictor of coronary heart disease. Some studies have reported no difference in the body weight of coronary heart disease patients compared to subjects free of the disease, others found the body weight of subjects with coronary heart disease to be slightly greater, and one found the body weight of cardiac patients to be less than controls. Height, however, is associated with coronary heart disease in prospective studies with long-term and shorter-term follow-up periods and case-control designs. Fetal, infant and childhood under-nutrition may link shorter adult height and susceptibility to cardiovascular disease. Many researchers have studied the relationship between overweight and coronary heart disease by using a surrogate measurement of body fatness such as relative weight or a weight-for-height index. In general, results produced by these studies suggest weight-for-height indices, particularly the often used body mass index, are not strong predictors of coronary heart disease. Indeed case-control designs have consistently failed to show a relationship between body mass index and coronary heart disease. Inconsistent results from prospective studies, however, are difficult to interpret. To further confuse the situation, the body mass index has been examined in relation to different coronary heart disease end-points and adjusted for different confounding variables. Explaining the inconsistent results on the basis of length of follow-up is also not straightforward. When follow-up periods exceed 20 years, and sample size is small, however, this closer association has not been found, even with a long follow-up period. Whilst some studies have found no association after 15, 13 and 12 years others have reported a relationship after 8.5, 10, 12, 10 and 7 years. The 22 year follow-up evidence from the Framingham Study shows the strongest 'independent' association between body mass index and coronary heart disease. (ABSTRACT TRUNCATED)