Abdominal obesity and the “hypertriglyceridaemic waist” phenotype

Generalised obesity, measured by body mass index (weight (kg)/(height (m)²)), is one of the major causes of ill health in western society. However, abdominal obesity—which is closely associated with intra-abdominal fat and measured either by waist circumference or waist:hip ratio—predicts subsequent coronary artery disease better than body mass index.¹ Furthermore, obesity, particularly abdominal obesity, is associated with insulin resistance, and predicts the development of type 2 diabetes.²,³ On p 716 Després et al argue that waist circumference should be routinely measured in primary care and used to identify people with abdominal obesity, on whom efforts to reduce obesity should be targeted.⁴

High waist measurement may be useful for screening since (a) height contributes little to the variance, (b) it accurately predicts obesity and high waist:hip ratio,⁵ and (c) it predicts traditional coronary artery disease risk factors.⁵ High waist and fasting triglyceride measurements—the hypertriglyceridaemic waist—is a marker for the “metabolic syndrome,” which is associated with the traditional risk factors of hypertension, hyperglycaemia, low high density lipoprotein (HDL) cholesterol and the non-traditional risk factors of insulin resistance, hyperinsulinaemia, raised apolipoprotein B, and small dense low density lipoprotein (LDL) cholesterol particles. A triad of non-traditional risk factors (high apolipoprotein B, hyperinsulinaemia, small dense LDL cholesterol) strongly predicted coronary artery disease in a prospective cohort (odds ratio 5.2) even after traditional risk factors were controlled for.⁶ Furthermore, waist circumference is associated with hyperinsulinaemia and high apolipoprotein B, and hypertriglyceridaemia is associated with dense LDL cholesterol particles.⁷ The combination—that is, the hypertriglyceridaemic waist phenotype—is also associated with coronary artery disease (odds ratio 3.6),⁷ hence Després et al's argument that waist measurement is a vital sign and should be routinely documented.⁴ However, before accepting exhortations to change routine practice some important questions need answering.

How much extra information do non-traditional risk factors provide? The prospective cohort study had few cases (85) and hence wide confidence intervals.⁶ The effect of other important risk factors, including left ventricular hypertrophy, family history, and social class, were not clarified.⁶ Furthermore, the risk of coronary artery disease from the hypertriglyceridaemic waist phenotype (odds ratio 3.6)⁷ is similar to the estimate for traditional risk factors.⁶ Thus additional large prospective studies are needed to clarify the utility of these non-traditional risk factors.

Is triglyceride concentration an independent risk factor? Triglyceride is strongly and inversely related to HDL cholesterol, and the traditional ratio of cholesterol:HDL cholesterol predicts LDL particle size similarly to triglyceride concentrations (respectively r=−0.59; r=−0.54).⁸ Nevertheless, a meta-analysis of prospective studies suggests triglyceride concentration probably is an independent risk factor for coronary artery disease.⁹ A fasting trigylceride of >2.3 mmol/l roughly doubles the risk of myocardial infarction.¹⁰ However, routinely obtaining fasting triglyceride concentrations—which by contrast are not necessary for HDL and LDL cholesterol—may be difficult in primary care. The importance of triglyceride concentration also requires clarification in young men, women, and groups such as South Asians among whom abdominal obesity is prevalent.

How useful is waist measurement as a screening tool? The positive predictive value of risk factors for predicting cardiovascular disease is less than 60%—that is, over 40% of people will not have a risk factor for cardiovascular disease,⁵ although many will subsequently develop type 2 diabetes.²,³ The reliability of waist measurement in primary care, including the optimal position of the tape measure, is also unclear.

Is waist measurement to provide patients with information about their health risks or to guide cost effective treatment? Patients do want information about their health risks and ways of reducing these risks if this is done sensitively. However, most patients who are overweight know that they are and that this carries risks.¹¹ Whether having these risks confirmed and receiving simple advice makes any difference is unclear, but evidence for such simple approaches is sparse.¹² Is more intensive treatment of obesity likely to be effective? A systematic review of diverse treatments suggested that surgical treatment for persistent morbid obesity (body mass index >40) is likely to be effective, as are behavioural treatments, diet and exercise regimens, and drug treatments.¹³ However, most studies were not based in primary care, used volunteers, and concentrated on weight alone and not on risk factors for coronary artery disease. There are also concerns about methodological issues, cost effectiveness, and the training and resource implications for primary care. Finally, most studies document weight regain in the longer term.¹³ A similar range of considerations apply to recent trials of drug treatments.¹⁴ Clearly, better evidence relevant to primary care is needed before we firmly advocate treatment regimes for obesity.

This leaves the primary healthcare team in a dilemma. Yes, both body mass index and abdominal obesity are important, the measurement of fasting triglyceride concentrations may improve estimation of risk, and high waist circumference may identify those at highest risk of coronary artery disease and type 2 diabetes. Thus it would seem prudent to provide information and advice about weight reduction as part of overall management of risk factors to patients with large waists and multiple risk factors for coronary artery disease. Primary healthcare teams should also be aware of the potential to treat patients with high triglyceride and low HDL cholesterol concentrations: in such patients fibrates provide effective secondary prevention (number needed to treat for five years=20), even when LDL cholesterol is low.¹⁵ However, until there is better evidence from primary care it is difficult to support the routine documentation of waist circumference in all patients.

Clinical review p 716