Author + information
- Linn M.A. Kennedy, MS* ()
- Kenneth Dickstein, MD, PhD,
- Stefan D. Anker, MD, PhD,
- Krister Kristianson, PhD,
- Ronnie Willenheimer, MD, PhD,
- OPTIMAAL Study Group
To the Editor:
Obesity is related to cardiovascular risk factors and is an independent risk factor for coronary artery disease (CAD) and premature death (1,2). However, the importance of obesity to mortality and morbidity in patients with established CAD is not well defined. The present analysis evaluated the importance of body mass index (BMI) to prognosis after complicated acute myocardial infarction (AMI).
In post-hoc analysis, we examined the impact of baseline BMI on all-cause death, cardiac death (death from AMI, chronic heart failure [CHF], other cardiac causes, and sudden cardiac death), cancer death, and AMI and CHF hospitalization in 5,388 patients with complicated AMI who were included in the OPtimal Trial In Myocardial infarction with the Angiotensin II Antagonist Losartan (OPTIMAAL) (3). Prior to data analysis, patients were categorized into four BMI groups: underweight, <22.00; normal weight, 22.00 to 24.99; overweight, 25.00 to 29.99; obese, ≥30.00 kg/m2. Univariate and multivariate logistic regression analysis and analysis of variance were performed (StatView 5.0.1, SAS Institute, Cary, North Carolina). Multivariate analyses adjusted for 46 variables comprising demographics, patient history, medication, physical examination, and biochemical analyses. Mean study drug dose percent (MSDD%) was calculated as mean percentage of the study drug target dose (captopril/losartan) during each patient's follow-up. We denoted significance to be p < 0.05.
Baseline BMI ranged from 13.2 to 49.4 kg/m2. Table 1depicts baseline characteristics. Median follow-up was 2.8 years (range 1 to 1,471 days). Table 2shows the main results of univariate and multivariate analyses. Additionally, BMI categories had similar adjusted risk of cancer death (n = 83), CHF hospitalization (n = 585), and number of CHF hospitalizations (n = 916) and CHF hospitalization days (n = 8,646). One-week mortality was comparable among BMI categories (1.8% to 2.5%). The BMI category was independently (p = 0.018) related to the number of AMIs (n = 975)—adjusted relative frequency compared with normal-weight: underweight, 0.93; overweight, 0.78; obese, 0.82.
In univariate analysis, there was a significant association between BMI category and MSDD%: underweight, 68.55; normal weight, 70.24; overweight, 74.13; obese, 76.91. In univariate analysis, a higher MSDD% was significantly protective of all end points (p < 0.0001), except cancer death. However, the addition of MSDD% to the multivariate analyses did not attenuate the independent associations shown in Table 2.
After complicated AMI, compared with normal-weight patients, overweight/obese patients did not show a higher risk of mortality or recurrent AMI or CHF hospitalization. On the contrary, in univariate analysis, being overweight/obese was associated with lower risk of mortality and recurrent AMI, and the lower risk of recurrent AMI in obese patients remained significant in multivariate analysis. Underweight patients had 52% higher adjusted mortality risk than normal-weight patients. Being overweight or obese, compared with being underweight, was independently predictive of more than 50% better survival (results not shown).
The results may seem remarkable considering the vast range of comorbidity associated with obesity, including CAD. Obesity may confer excess risk by producing an inflammatory state, promoting thrombosis and activating the sympathetic nervous system and the renin-angiotensin system (4,5). Because obesity increases the risk of CAD and mortality in the population (1,2,6), it might be anticipated that obesity also increases mortality in patients with AMI. Nevertheless, our results are supported by previous studies in patients undergoing percutaneous coronary intervention/cardiac surgery; the risk of mortality and cardiovascular events was decreased in obese patients and/or increased in underweight patients (7,8). However, in patients with CAD, elevated BMI was associated with increased risk of acute coronary events (9). Previous results in patients after AMI are contradictory; obesity was associated with increased risk of recurrent coronary events during three years of follow-up in one study (5), but with lower one-year mortality in another (10). Conversely, obesity independently predicted in-hospital death in older, but not in younger, patients (10).
The reasons for the findings of the present study are probably multifactorial. Overweight/obese patients may have better tolerance to afterload-reducing medication, which may lead to ingestion of higher doses of renin-angiotensin system inhibitors and improved survival (11). In the present study, MSDD% increased with increasing BMI, which may partly explain our findings. Another possible explanation is related to the fact that neurohormonal activation may be important to the increased cardiovascular risk associated with obesity (4). In our study, much of the negative effects of obesity might have been attenuated by the treatment given; all patients received renin-angiotensin system inhibition and 75% had beta-blockade. Consequently, any beneficial effects associated with being overweight/obese might have come into play, such as a higher metabolic reserve, providing better resistance against the metabolic stress associated with heart failure (12). Variations in the use of neurohormonal blockers also might possibly explain different study results concerning the impact of BMI in patients with CAD (5,7–10).
Our observation that underweight patients had the worst prognosis is supported by studies in patients with CAD (7,8) and CHF (12,13). Cachectic patients with CHF demonstrate multiple alterations in physiologic, biochemical, and neurohormonal pathways, leading to hemodynamic, immunologic, endocrine, metabolic, and muscular abnormalities (14). Perhaps, after complicated AMI, lean patients may have similar deleterious abnormalities.
One may argue that our results were skewed by the younger age in overweight/obese patients. Indeed, when excluding age from the multivariate analyses, overweight and obese patients had significantly lower risk of death and recurrent AMI, and overweight patients also had lower risk of cardiac death (data not shown). Thus, age seems to be the major factor explaining why being overweight or obese was not independently protective of mortality compared with being at a normal weight. However, despite adjustment for age and numerous other variables, the adjusted mortality risk of being overweight or obese was still not higher compared with being at a normal weight but was rather lower, although not statistically significant.
Please note: Dr. Kristianson has been an employee at Merck & Co. (now retired). This study was supported by a grant from Merck & Co.
- American College of Cardiology Foundation
- Eckel R.H.
- Potapov E.V.,
- Loebe M.,
- Anker S.,
- et al.
- Wolk R.,
- Berger P.,
- Lennon R.J.,
- et al.
- Horwich T.B.,
- Fonarow G.C.,
- Hamilton M.A.,
- et al.