Author + information
- Received November 20, 2012
- Revision received May 31, 2013
- Accepted June 3, 2013
- Published online August 20, 2013.
- Michael E. Buschur, MD∗,
- Dean Smith, PhD, MPH∗,
- David Share, MD†,
- William Campbell, MD‡,
- Stephen Mattichak, MD§,
- Manoj Sharma, MD⋮ and
- Hitinder S. Gurm, MD∗∗ ()
- ∗Division of Cardiovascular Medicine, University of Michigan, Ann Arbor, Michigan
- †Blue Cross Blue Shield of Michigan, Detroit, Michigan
- ‡Borgess Medical Center, Kalamazoo, Michigan
- §Bay Regional Heart and Vascular, Bay City, Michigan
- ⋮Division of Cardiovascular Medicine, Covenant Healthcare, Saginaw, Michigan
Reprint requests and correspondence:
Dr. Hitinder S. Gurm, Department of Medicine, Division of Cardiovascular Medicine, University of Michigan Cardiovascular Center, 2A394, 1500 East Medical Center Drive, Ann Arbor, Michigan 48109-5853.
Objectives This study sought to examine the prevalence and clinical implications of morbid obesity among patients undergoing percutaneous coronary intervention (PCI).
Background The prevalence of obesity, and morbid obesity in particular, continues to rise rapidly in the United States. Obese patients are at increased risk for cardiac disease and are more likely to need invasive cardiac procedures. There is a paucity of contemporary data on the prevalence and clinical implications of morbid obesity among patients undergoing PCI.
Methods We examined the prevalence of morbid obesity (body mass index [BMI] ≥40 kg/m2) among 227,044 patients undergoing PCI and enrolled in the Blue Cross Blue Shield of Michigan Cardiovascular Consortium registry from 1998 to 2009.
Results The proportion of morbidly obese patients undergoing PCI increased from 4.38% in 1998 to 8.36% in 2009. Compared with overweight patients (BMI 25 to 30 kg/m2), these patients had significantly increased vascular complications (adjusted odds ratio [OR]: 1.31; 95% CI: 1.17 to 1.47; p < 0.0001), contrast-induced nephropathy (adjusted OR: 1.89; 95% CI: 1.70 to 2.11; p < 0.0001), nephropathy requiring dialysis (adjusted OR: 4.08; 95% CI: 2.98 to 5.59; p < 0.0001), and mortality (adjusted OR: 1.63; 95% CI: 1.33 to 2.00; p < 0.0001).
Conclusions Morbid obesity is increasing in prevalence among patients undergoing PCI and is associated with a higher risk of mortality and morbidity. These epidemiological changes have important implications for technical considerations of cardiac catheterization, design of the catheterization lab to accommodate these patients, and most importantly, for societal effort toward prevention of obesity.
Obesity is a rapidly increasing worldwide epidemic. In 2010, 35.7% of all adults over 18 years of age were obese (body mass index [BMI] ≥30 kg/m2) (1) compared with 12% of all adults in 1991 and 17.9% in 1998 (2). Obese patients are at increased risk for cardiac disease and are more likely to require invasive cardiac procedures (3). However, multiple studies have shown that obesity has a paradoxical protective effect on outcomes after percutaneous coronary intervention (PCI) (4,5).
Although there have been many studies on the outcomes of obese patients undergoing PCI, there are limited data about the prevalence and health outcomes of the morbidly obese (BMI ≥40 kg/m2). One small study estimated that the prevalence of morbidly obese adults has increased from 2.20% in 2001 to 3.07% in 2005 in the general U.S. population (6). It is likely that the increasing prevalence of morbid obesity in society will translate into an increase in the prevalence of morbid obesity among patients with cardiovascular disease. There is a paucity of data on the prevalence and implications of morbid obesity among patients undergoing PCI. This study examined the prevalence of morbid obesity among patients presenting for PCI between 1998 and 2009 and examined complications in morbidly obese patients.
The study population comprised 227,044 patients undergoing PCI and enrolled in the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) registry from 1998 to 2009. The details of the BMC2 registry, data collection, and auditing have been previously documented and are detailed in the Online Appendix(7–9).
Patients were divided into 4 categories according to BMI (kg/m2): lean (BMI <25), overweight (25≤ BMI <30), submorbidly obese (30≤ BMI <40), and morbidly obese (BMI ≥40). The primary outcome of interest was the change in prevalence of morbidly obese patients from 1998 to 2009. We also compared the in-hospital outcomes across the 4 categories of BMI, with the specific outcomes being in-hospital death, emergency coronary artery bypass grafting (CABG), all CABG, contrast-induced nephropathy (CIN), nephropathy requiring dialysis (NRD), post-procedure transfusion, stroke/transient ischemic attack, myocardial infarction (MI), major adverse cardiac events, vascular complications, and gastrointestinal bleeding. In-hospital death was defined as death from cardiac or noncardiac causes during the same admission as that for PCI. Vascular complications were defined as a composite of pseudoaneurysm, arteriovenous fistula, femoral neuropathy, retroperitoneal hematoma, access site hematoma requiring transfusion or associated with prolonged hospital stay, drop in hemoglobin >3.0 g/dl, or any access site complication requiring surgical repair. CIN was defined as a creatinine increase ≥0.5 mg/dl above baseline (10). Major adverse cardiac events was defined as a composite of death, stroke/transient ischemic attack, MI, all CABG, and revascularization involving the index lesion.
Continuous variables were expressed as mean ± SD, and discrete variables were expressed as frequency counts and percentages. Secular changes in the proportion of morbidly obese patients presenting for PCI were assessed with the Cochrane-Armitage test. We developed multiple hierarchical models to assess the impact of BMI categories on different outcomes of interest using overweight patients (the group that has traditionally demonstrated the best post-PCI outcomes) as the reference. The pre-procedural variables demonstrating marginal association (p < 0.1) with the specific outcome of interest were selected through stepwise selection and included demographic information, comorbidities, presenting cardiac diagnosis, laboratory variables, and angiographic findings. Model discrimination was assessed using the C-statistic, and calibration was assessed using the Hosmer Lemeshow test. The variables included in the final models are listed in the Online Appendix. All analysis was performed using SAS version 9.2 (SAS Institute, Cary, North Carolina).
Our study enrolled 227,042 patients, of whom 19.1% (n = 43,346) were in the lean category, 36.3% (n = 82,487) were in the overweight category, 37.2% (n = 84,479) were in the submorbidly obese category, and 7.4% (n = 16,730) were in the morbidly obese group. The baseline characteristics of the population are detailed in Table 1. Among the 32 hospitals, the prevalence of morbid obesity ranged from 5.0% to 11.4% (Online Fig. 1).
The morbidly obese group was overall significantly younger (mean age 59.2 years vs. 64.9 years in the overweight group) than the other groups, with a male/female ratio of approximately 1:1. In addition, they were less likely to be smokers and had a significantly higher incidence of hypertension and diabetes.
The pre- and post-procedure medications are listed in Table 1. The morbidly obese group was more likely to be receiving statins, beta-blockers, and angiotensin-converting enzyme-inhibitors or angiotensin receptor blockers pre-procedure as compared with the other categories.
Over the course of the study, significant changes in the obesity profile were noted, with a small but significant rightward shift in the BMI distribution (Fig. 1). Between 1998 and 2009, the proportion of lean patients decreased from 22.64% to 18.48%, the overweight patients decreased from 39.38% to 35.07%, the submorbidly obese patients increased from 33.59% to 38.08%, and the morbidly obese category increased from 4.38% to 8.36%. There was a 91% increase in the number of morbidly obese patients presenting for PCI.
The unadjusted and adjusted clinical outcomes of patients in different BMI categories are listed in Table 2 and Figure 2. The lean patients had a lower risk of CIN but a higher risk of post-procedure transfusion, vascular complications, and death. The submorbidly obese category had an increased risk of CIN, NRD, and vascular complications compared with the overweight group. Finally, the morbidly obese patients had a lower risk of emergent CABG and all CABG but an increased risk of CIN, NRD, vascular complications, and death.
The key finding of our study was that there has been a significant increase in the proportion of morbidly obese patients undergoing PCI. Second, these patients have a markedly increased adjusted risk of complications, suggesting that the obesity paradox does not appear to be protective in this population. The findings of our study, derived from a large unselected population of 227,044 patients, significantly adds to the prior work evaluating the impact of obesity on PCI outcomes.
A large body of data suggests that obesity is a growing epidemic in the general population, and our study suggests that this has been accompanied by an increase in the obesity profile of the patients undergoing PCI. There has been a steady increase in the proportion of patients who are submorbidly and morbidly obese, with the submorbidly obese population making up the largest cohort of patients in recent years.
Several studies have demonstrated that overweight patients have a lower risk of cardiovascular events and all-cause mortality than normal-weight patients (5,11). These results have not only been confirmed in overweight patients with coronary artery disease but also in overweight patients with heart failure as well (12). There has been an increasing focus on the obesity paradox in both the medical and the lay press (13). However, the term “obesity paradox” obfuscates the adverse outcomes associated with morbid obesity, a group that was at significantly higher risk of mortality and morbidity in our study.
The increase in complications may be due to many reasons. The increase in the risk of CIN and NRD is probably directly related to the increased use of contrast media in this population (14). The poor fluoroscopic visualization associated with obesity often necessitates multiple contrast injections, with increased risk of renal complications. Second, calculation of renal function in morbidly obese patients may be erroneous because the commonly used methods of calculating glomerular filtration rate or creatinine clearance to guide contrast dosing have not been validated in and may not be applicable to patients with extreme body mass.
The increase in vascular complications in the morbidly obese is not surprising because the majority of patients underwent transfemoral PCI, for which both access and vascular closure would be more difficult with increasing body mass. Although radial access was rarely employed in this population, there has been a more recent upsurge in the use of radial access, and it would be interesting to assess if increasing use of radial PCI will negate the impact of BMI on vascular complications.
The most important clinical outcome was the increased risk of death in the morbidly obese patients. This is especially concerning because the obese patients were significantly younger than other patients and were more likely to be undergoing PCI for stable disease.
Further, as compared with overweight patients, the morbidly obese patients had better baseline left ventricular ejection fraction and lower incidence of acute MI, cardiogenic shock, and need for emergent PCI—all of the factors that are traditionally associated with a lower mortality risk in patients undergoing PCI. Our findings corroborate a recent study demonstrating that morbidly obese patients presenting with ST-segment elevation myocardial infarction had a higher mortality despite being younger and having better left ventricular function and less extensive coronary artery disease (15). This increased risk despite the presence of lower-risk features suggests that factors other than the traditional risk factors for post-PCI mortality may play a role in this population. Further, it is possible that these patients may have altered pharmacokinetics of the commonly used periprocedural medications and may demonstrate a differing response to ischemic challenges. Finally, the suboptimal radiographic visualization that is occasionally encountered in morbidly obese patients may preclude detection of less than ideal angiographic results, which may potentially result in a greater likelihood of ischemic complications. Our study was not designed to elucidate the reasons for the increased mortality observed in this population, and these conjectures should be considered hypothesis generating.
It is also possible that the dramatically increased incidence of diabetes in this population contributed to both development of coronary artery disease and the adverse outcome following PCI seen in this population. Further research is warranted to identify the mechanistic underpinning of this association.
Finally, the dramatic increase in the number of morbidly obese patients has important implications for technical considerations of cardiac catheterization and design of the catheterization laboratory to accommodate these patients. The need to deal with increasing body weight will require catheterization tables with greater weight tolerance and systems to safely transfer patients to and from these tables. Further, better radiation shielding and consideration of alternative imaging strategies need to be considered because these patients and their physicians are exposed to increased radiation. Furthermore, the dramatic increase in the proportion of young patients with morbid obesity invokes the need for more upstream interventions for primary prevention and better treatment of obesity.
The major limitations of this study are that it only included nonrandomized patients and offered observational data. There is an unaccounted bias in which certain morbidly obese patients were selected for revascularization and in the revascularization strategies offered to these patients. There may also be selection bias in that morbidly obese patients may have been offered PCI more often than surgical revascularization. Additionally, it is possible that operator experience overall and specifically in performing PCI in morbidly obese patients may be an important predictor of outcome. Our study was unable to account for this variable. Finally, this study examined the prevalence of obesity in a single state, the fifth most obese state in 2011, and results may not be extrapolated to the remainder of the United States (16).
This large study analyzed 227,044 patients and found that there has been a 91% increase in the number of morbidly obese patients presenting for PCI from 1998 to 2009. The prevalence of morbidly obese patients is growing at a faster rate than overweight or obese patients. The morbidly obese patients have an increased risk of CIN, NRD, vascular complications, and death despite being younger and having a more stable presentation than overweight patients. The findings of our study have important implications for contemporary interventional practice and prevention and treatment of obesity.
The Blue Cross Blue Shield of Michigan Cardiovascular Consortium registry is funded by Blue Cross Blue Shield of Michigan. The sponsor had no role in analysis, study design, or decision to publish these results. Dr. Gurm received research funding from Agency for Health Care Research and Quality and the National Institutes of Health. Dr. Share is employed by Blue Cross Blue Shield of Michigan. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- body mass index
- coronary artery bypass grafting
- contrast-induced nephropathy
- myocardial infarction
- nephropathy requiring dialysis
- percutaneous coronary intervention
- Received November 20, 2012.
- Revision received May 31, 2013.
- Accepted June 3, 2013.
- American College of Cardiology Foundation
- Eckel R.H.,
- Krauss R.M.
- Moscucci M.,
- Kline-Rogers E.,
- Montoye E.,
- et al.
- Moscucci M.,
- Kline-Rogers E.,
- Share D.,
- et al.
- Brown H.
- Marenzi G.,
- Lauri G.,
- Assanelli E.,
- et al.
- Das S.R.,
- Alexander K.P.,
- Chen A.Y.,
- et al.
- ↵Overweight and Obesity. Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/obesity/data/adult.html. Accessed June 21, 2012.