Friday March 11 2011
Are you shaped like an apple or a pear?
“A medical U-turn has cast doubt on warnings that being overweight and 'apple-shaped' is especially dangerous to the heart”, the Daily Mail has reported.
The news is based on a high-quality review drawing together data on more than 220,000 people to see how well measures of fat, such as body mass index (BMI), waist circumference and waist-to-hip ratio, predict new diagnoses of heart disease or stroke. Despite what some news reports have suggested, these measures were all associated with an increased risk of fatal or non-fatal coronary heart disease, stroke and overall cardiovascular disease. The point missed by many newspapers is that the researchers only found that conventional risk assessments, which already look at established risk factors such as smoking and high blood pressure, were not improved by adding data on these body fat measures. As the researchers have said, their findings do not diminish the importance of controlling body fat to help prevent cardiovascular disease.
This research confirms the health risks associated with being overweight and obese, and simply says that the harmful effects of being overweight mainly act through the other established risk factors for heart disease and stroke. Recommendations to follow a balanced diet, exercise regularly and maintain a healthy weight do not change.
Where did the story come from?
The study was carried out by researchers from the University of Cambridge and was funded by the British Heart Foundation and UK Medical Research Council. It was published in the peer-reviewed medical journal The Lancet.
Some news reports have only partially reflected the nature of this research as they have suggested that a person’s body shape does not necessarily predict heart risk. The study actually found that increases in all three measures of body fat (BMI, waist circumference and waist-to-hip ratio) were independently associated with raised cardiovascular risk. What the study found was that conventional risk prediction models, which take into account traditional risk factors, were not improved by the inclusion of these body fat data. The findings do not diminish the importance of a healthy overall weight in reducing risk of cardiovascular disease.
What kind of research was this?
The current research was inspired by the fact that several guidelines place different emphasis on the value of measures of body fat (adiposity) as predictors of cardiovascular risk. This study, conducted by the Emerging Risk Factors Collaboration, was a systematic review gathering individual patient data collected from 58 study populations. The aim of the review was to study how BMI, waist circumference and waist-to-hip ratio are associated with the development of cardiovascular disease and to explore the relationship between these measures and conventional risk factors.
This large piece of research was well conducted and underlines the importance of individual and collective measures for tackling the major risk factors for heart disease and stroke, primarily smoking, diabetes, blood pressure and ‘bad’ forms of cholesterol.
What did the research involve?
The researchers identified relevant studies through a search of medical databases, hand searching of reference lists and discussion with study authors. They identified a total of 58 studies that fulfilled the following characteristics:
- participants had no known history of cardiovascular disease at study start (confirmed by medical examination)
- information was provided at study start for weight, height, and waist and hip circumference
- the outcomes of cardiovascular disease or cause-specific mortality, or both, were recorded using well defined criteria (the use of validated diagnostic codes and the examination of medical records and death certificates)
- participants had been followed for at least one year
The 58 cohort studies provided records on 221,934 participants from 17 countries. These studies looked at outcomes of either first non-fatal disease event or cause-specific death relating to coronary heart disease, stroke or cardiovascular disease in general (CHD or stroke). The risk of these events was calculated against each one unit increase in three different body fat measurements from the start of the study: each 4.56kg/m² increase in BMI, each 12.6cm increase in waist circumference and each 0.083 increase in waist-to-hip ratio. These measures were equivalent to one standard deviation, which is a statistical term for how far the individual recordings varied from the average.
The researchers adjusted their outcomes for potential confounders of age, sex, smoking status, blood pressure, diabetes, and total and high-density lipoprotein (HDL) cholesterol. Analyses excluded underweight participants with a BMI below 20kg/m². The authors also took into account the nature of statistical differences between the results of the different studies (heterogeneity).
What were the basic results?
The average age of participants at study start was 58 years and just over half were women (56%). The 221,934 participants constituted 1.87 million person-years of follow-up, over which time there were 14,297 new cardiovascular disease events. On average, it took 5.7 years for the first outcome to occur.
In fully adjusted analyses:
- each one standard deviation (SD) rise in BMI increased risk of any cardiovascular disease outcome by 7% (HR 1.07, 95% CI 1.03 to 1.11)
- each one SD rise in waist circumference increased risk of any cardiovascular outcome by 10% (HR 1.10, 95% CI 1.05 to 1.14)
- each one SD rise in waist-to-hip ratio increased risk by 12% (HR 1.12, 95% CI 1.08 to 1.15)
These analyses on any cardiovascular disease event are from data on 144,795 participants with full risk factor information available across 39 studies that reported this outcome. Among these individuals there were 8,347 cardiovascular disease events.
When separate analyses were carried out for the 39 studies reporting the outcomes of coronary heart disease events and the 21 studies reporting stroke as an outcome, similar risk figures were obtained for each one SD rise in BMI, waist circumference and waist-to-hip ratio.
The researchers then added information on BMI, waist circumference or waist-to-hip ratio into a cardiovascular disease risk prediction model that also looked at conventional risk factors (for example, smoking, diabetes, blood pressure and cholesterol). The addition of these adiposity measures did not improve risk discrimination or aid the classification of participants into categories of predicted 10-year risk.
How did the researchers interpret the results?
The researchers conclude that BMI, waist circumference and waist-to-hip ratio, whether assessed singly or in combination, do not significantly improve cardiovascular disease risk prediction when information on conventional risk factors of blood pressure, diabetes and cholesterol is available.
This was well conducted research that has combined data from 58 cohort studies featuring 221,934 people and constituting 1.87 million person-years of follow-up. Each standard unit rise in BMI, waist circumference and waist-to-hip ratio was found to be independently associated with an increased risk of fatal or non-fatal coronary heart disease, stroke or the combined outcome of cardiovascular disease. However, adding these measures into risk prediction models based on conventional risk factors (for example smoking, diabetes, blood pressure and bad cholesterol) did not alter heart disease and stroke risk estimates. This means that none of the measures, individually or in combination, could improve risk prediction when information on other risk factors was available.
A key point to remember is that these findings do not mean that body shape is not important or that BMI, waist circumference and waist-to-hip ratio cannot be used to predict cardiovascular risk. Rather, they mean that their inclusion in conventional clinical risk assessments does not appear to be of benefit. As the researchers have said, their findings “do not diminish the importance of adiposity as a major modifiable determinant of cardiovascular disease” and, indeed, their results demonstrate that a rise in any of these factors does increase cardiovascular risk. Adiposity levels were also found to contribute to the other cardiovascular risk factors of diabetes, cholesterol and high blood pressure.
In this context, the researchers have said that their findings “reliably refute previous recommendations to adopt baseline waist-to-hip ratio instead of BMI as the principal clinical measure of adiposity”. This does not mean that waist-to-hip ratio is of no importance or that it is not associated with cardiovascular disease, rather that it does not appear to be of greater predictive value than BMI, the currently preferred clinical measure. BMI, waist circumference and waist-to-hip ratio were all found to have a similar strength of association with coronary heart disease, stroke and cardiovascular disease overall.
This well conducted systematic review also appears to be reliable, aggregating a large quantity of individual patient data from 58 cohorts. Importantly, all participants (average age 58) were also confirmed to be free of cardiovascular disease at the start of the study, ruling out the possibility that pre-existing cardiovascular disease had clouded the results. Further study is needed, however, particularly in other population groups, as 90% of those included in this study were of European descent.
The findings of this review do not change current recommendations that people should try to eat a balanced diet, take regular exercise and maintain a healthy weight.