Faculty Peer Reviewed
The New York Times was on our case again this week reminding us (meaning Americans) of our poor lifestyle habits. They reported that middle-aged Americans are eating fewer fruits and vegetables, drinking more alcohol, exercising less, and, as we all know, becoming more obese, according to a study published in the American Journal of Medicine.Admittedly…mea culpa. With the long hours and late nights of residency, it’s pretty difficult to eat well and exercise as much as we preach to our patients (so I tell myself). But now that I am a parent of two young children, I figure it’s time to get back to eating better and exercising more often. You know, to try to set a good example for them. So what should I eat? For many years we have heard of the benefits of the Atkins diet. Studies have demonstrated that patients on the Atkins diet lose weight, reduce insulin resistance, and lower triglycerides while increasing HDL. One of the many reasons it is difficult to recommend this diet is that it has a tendency to increase LDL levels. The Archives of Internal Medicine published an article on a version of the Atkins diet–the Eco-Atkins diet–which exchanges animal proteins for high quantities of the plant-derived proteins found in gluten, soy, nuts, fruits, vegetables, and vegetable oils. It was a small study comparing 22 overweight and hyperlipidemic patients on an Eco-Atkins diet to 22 controls eating a high-carbohydrate, lacto-ovo vegetarian diet. Although weight loss was comparable over the 4 weeks of the study, there was a statistically significant 8.1% absolute reduction in LDL in the Eco-Atkins group compared to controls. The authors further argue that, compared to the control diet, the Eco-Atkins diet lowered LDL without lowering HDL. Of course, with the small sample size and short duration of the study, these results need to be taken with a grain of salt. Just to be safe… maybe you should stay away from the salt.
Just when I was getting all excited about my lifestyle modification shaping my children’s habits, I came across another article in the New York Times that took away some of my hope. They reported on a study published in the journal Social Science and Medicine www.elsevier.com/locate/socscimed comparinig the eating habits of over 2200 parents and 2600 children. As you might have guessed, there was no correlation in total energy, carbohydrate, saturated fat, or polyunsaturated fat intake between the two groups. It appears that eating habits, particularly in adolescents, are more likely to be influenced by peers than by parents. Now that’s food for thought… On a more positive note, the study showed that if a child does end up mimicking his or her parents, it is more likely to be the mother’s than the father’s eating habits. Good thing too, because my wife loves salads.
As physicians, we have an important role in influencing our patients’ lifestyle choices. We often scold our overweight patients about bad habits and then try to goad them into compliance with a good diet-and-exercise plan by telling them they are at increased risk of developing diabetes, coronary disease, etc. I have always thought…wouldn’t it be great to be able to assess a person’s risk of developing diabetes in the next ten years the way we do with coronary disease using the Framingham Risk Assessment? Well, a group sponsored by the CDC published an article in the Annals of Internal Medicine last week that attempted to do just that. They used 14.9 years of longitudinal data from the Atherosclerosis Risk in Communities (ARIC) Study to develop a diabetes risk scoring system. The ARIC study followed over 15,000 adults between the ages of 45 and 64 in four U.S. communities beginning in 1987. From the data, the group developed and validated two scoring systems. The first was a basic scoring system which used easy-to-assess patient characteristics; the second was an “enhanced” system which incorporated laboratory values into the basic system. In the basic scoring system, points were added for waist circumference (10 to 35 points), maternal diabetes (13 points), and hypertension (11 points). Some points were also given for paternal diabetes, short stature, black race, age >55, increased weight, elevated resting pulse (>68 in men, >70 in women), smoking history, and being a non-drinker (all <8 points each). The enhanced system added additional points for glucose (6 to 28 points), triglycerides, LDL, and elevated uric acid (all <7 points each). The scores were then divided into quintiles and the risk of progression to diabetes was validated in the next ten years. The quintile cut-off points of the basic system were 21, 33, 43, and 55, which were associated with an ascending risk of diabetes of 5.3%, 8.7%, 15.5%, 24.5%, and 33.0%, respectively. The enhanced system cutoff points were 17, 27, 37, and 50, which predicted a 10-year incidence of diabetes of 3.5%, 6.4%, 11.5%, 19.3%, and 46.1%. The simplicity of an integer scoring system makes this model easy to use. You will be able to tell the patient whose score falls into the highest quintile that they have almost a 50-50 chance of developing diabetes. Interestingly, not drinking alcohol gave you some points. [Shameless plug alert!...] Stay tuned for my next “Myths and Realities” piece on alcohol and its “cardioprotective” role.
Since we are talking about diabetes and coronary artery disease, I thought I would mention an article in this week’s New England Journal of Medicine. The study was a comparison of different treatment strategies in patients with both type 2 diabetes and stable coronary artery disease. The Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial was a prospective, randomized, multicenter study that enrolled 2368 patients with both type 2 diabetes and known heart disease (as seen on angiography) to undergo either revascularization (PCI or CABG) with intensive medical therapy or intensive medical therapy alone. They also randomized the patients to receive either insulin-sensitization (metformin, thiazolidinediones, etc.) versus insulin-provision therapy to achieve a hemoglobin A1c of less than 7.0%. The primary endpoint was death from any cause; a secondary endpoint was a composite of death, myocardial infarction, or stroke. After an average follow-up period of 5.3 years, there was no statistically significant difference in the primary and secondary endpoints between the revascularization group and the medical therapy group or between the insulin-sensitization group and those given insulin to treat their diabetes. However, in the subset of patients who underwent CABG, there was a significantly lower secondary composite endpoint compared to the medical therapy control group. The bottom line is that unless you had extensive multivessel disease, there was no indication for revascularization over medical therapy alone in this patient population.
Wish me luck on my lifestyle modification plans and molding my children’s behavior…and I want to take this opportunity to wish all of my colleagues who are moving on in July the best of luck in all your endeavors!
Dr. Matoo is a 3rd year resident in internal medicine at NYU Medical Center.
Peer reviewed by Michael Tanner MD, Section Editor, Clinical Correlations