From Pages to Practice

By Bhavna Seth, M.D.

Published April 5, 2017


Two years ago, your primary care patient, Mr. Oliver Bickens, a 55-year-old man with hypertension, diabetes, and obesity underwent percutaneous coronary intervention (PCI) for 90% obstruction to his left circumflex coronary artery; his lifetime ASCVD risk is 69%. He has struggled with weight control for years, with BMIs ranging widely from 30–35 kg/m2, depending on his level of stress and anxiety and his ability to follow a prescribed exercise regime. Recently he achieved a 10 kg weight loss. You congratulate him and encourage him to keep up the good work, but wonder if he will rapidly regain all the weight he just lost.

The “obesity paradox” describes the observation that in some cohorts of sick patients, obese people tend to have better outcomes than nonobese people. Uncertainty persists about how weight gain affects cardiovascular disease outcomes, especially among patients with weight loss followed by weight gain (“weight cycling”). A Framingham Heart Study noted that fluctuating body weight contributes to mortality and coronary events in patients without cardiovascular disease at baseline. However, the effect of weight cycling in patients with baseline coronary disease is unknown. A post hoc analysis of data from the Treating to New Targets (TNT) trial, an efficacy and safety trial of intensive versus lower-dose atorvastatin, provides new findings that may be relevant to patients such as Mr. Bickens.

The analysis examined the relation between weight cycling and the risk of adverse cardiovascular outcomes in 9509 participants with clinically evident coronary artery disease and low density lipoprotein (LDL) <130 mg/dL at baseline. Participants had a mean baseline weight of 85±15 kg and were followed for a median of 4.9 years. Intra-individual weight variability was measured as average successive variability (ASV) or the average absolute difference in successive visits at 3, 6, and 9 months, 1 year, and every 6 months thereafter. The primary outcome was any coronary event (composite of coronary heart disease death, nonfatal myocardial infarction, resuscitated cardiac arrest, revascularization, or angina) and secondary outcomes were any cardiovascular event.

Each increase in body-weight variability of 1 standard deviation (1.5–1.9 kg) significantly increased the risk of any coronary event (2091 events; hazard ratio [HR], 1.04; 95% CI 1.01–1.07; P=0.01) Additionally, body-weight variability was significantly associated with an increase in the risk of death (487 events; HR, 1.09; 95% CI 1.07–1.12) and a nonsignificant numerical increase in  myocardial infarction (HR, 1.04; 95% CI 0.98–1.09) and stroke (HR, 1.05; 95% 0.97–1.13).

In adjusted models, compared with patients in the lowest quintile of body-weight variation, patients in the highest quintile of body-weight variation had a 64% increase in any coronary event, 85% increase in any cardiovascular event, 124% increase in death, 117% increase in myocardial infarction, and 136% increase in stroke. Greater degrees of body-weight fluctuation were associated with higher event rates.

This study demonstrates an association, but not a causal relation, between body-weight fluctuation and cardiovascular events in a cohort of patients with coronary disease at baseline and selected to minimize confounding. The authors  note that the weight fluctuation observed might reflect pre-existing illnesses that predispose patients to adverse cardiovascular outcomes.  

Despite certain limitations, the new analysis of the TNT data draws attention to the problem of weight fluctuation and its relation to cardiovascular morbidity and mortality. It may be time to make Mr. Bickens aware that in addition to the importance of weight loss for his cardiovascular well-being, maintaining a stable weight is just as important. 

Browse more From Pages to Practice »

Bhavna Seth, Resident at Boston Medical Center
Bhavna Seth, Resident at Boston Medical Center