Author + information
- Received December 13, 2012
- Revision received May 17, 2013
- Accepted May 21, 2013
- Published online August 20, 2013.
- Saeed Payvar, MD, MS∗,
- Sunghee Kim, PhD†,
- Sunil V. Rao, MD†,
- Ronald Krone, MD‡,
- Megan Neely, PhD†,
- Nikhil Paladugu∗ and
- Ramesh Daggubati, MD∗∗ ()
- ∗East Carolina University Brody School of Medicine, Greenville, North Carolina
- †Washington University School of Medicine, St. Louis, Missouri
- ‡Duke Clinical Research Institute, Durham, North Carolina
Reprint requests and correspondence:
Dr. Ramesh Daggubati, Brody School of Medicine, East Carolina University, 115 Heart Drive, Greenville, North Carolina 27834.
Objectives The purpose of this study was to compare in-hospital outcomes of percutaneous coronary intervention (PCI) in extreme obesity (EO) (body mass index [BMI] ≥40 kg/m2) with those of normal-weight (NW) patients and to examine the influence of access site on outcomes.
Background Little is known about the outcomes of PCI in EO patients.
Methods We analyzed CathPCI Registry data from patients who underwent radial or femoral PCI and were discharged between July 2009 and June 2011 and compared in-hospital outcomes of EO (N = 83,861) with those of NW patients (BMI 20 to 25 kg/m2; N = 217,616). Outcomes included in-hospital mortality and procedural and bleeding complications. Multivariable logistic regression models were used to assess the independent association of EO with outcomes, using previously validated risk models derived from the CathPCI Registry. The role of access site was specifically examined.
Results Compared with NW patients, EO patients were younger (median age 60 vs. 69 years), more likely female (47% vs. 37%), and more likely African American (12% vs. 7%). EO patients had lower unadjusted mortality (1.2% vs. 2.0%); however, after multivariable adjustment, EO was independently associated with increased risk of in-hospital mortality (odds ratio: 1.22; 95% CI: 1.08 to 1.39) in those presenting with ST-segment elevation myocardial infarction (STEMI). Access site had no effect on bleeding or outcome.
Conclusions EO patients who underwent PCI were younger and had less bleeding compared with NW patients. After multivariable adjustment for risk, EO was independently associated with higher in-hospital mortality overall and particularly in the patients undergoing STEMI.
From 1986 to 2000, the prevalence of extreme obesity (EO), defined as body mass index (BMI) ≥40 kg/m2, quadrupled in the United States, and currently 1 in 50 adult Americans is extremely obese (1). EO has been linked with operative and late mortality after coronary artery bypass grafting (CABG) (2), yet little is known about the outcomes of percutaneous coronary intervention (PCI) in EO patients. BMI ≥45 kg/m2 was an independent risk factor for 30-day mortality in patients undergoing CABG (3).
Conversely, studies have indicated better PCI outcomes in obese patients compared with normal-weight (NW) patients (4), the so-called “obesity paradox,” with better outcomes being attributed to more intensive medical therapy (5), early referral (6), and reduced bleeding and other vascular complications due to the use of radial access and vascular closure devices (7).
We hypothesized that EO patients undergoing PCI have higher post-procedure in-hospital morbidity and mortality and used the CathPCI Registry to test this hypothesis.
This study was a retrospective analysis of in-hospital outcomes of PCI among EO patients compared with those of NW patients included in the CathPCI Registry. This registry is a national reporting system designed for quality improvement of diagnostic cardiac catheterization and PCI procedures, sponsored jointly by the American College of Cardiology (ACC) and the Society for Cardiovascular Angiography and Interventions (SCAI) (8,9). Pre-defined data are collected that include only in-hospital events (10), with neither angiographic nor in-hospital outcomes being adjudicated.
This study included patients ≥18 years of age who had undergone PCI via the femoral or radial artery; were discharged between July 1, 2009, and June 30, 2011; and had BMI ≥40 kg/m2 (EO group) or BMI between 20 and 25 kg/m2 (NW group). Risk factors, comorbid conditions, and hospital characteristics of interest are listed in Table 1. Pre-procedural and procedural characteristics of interest are listed in Table 2. Outcomes of interest included in-hospital death, bleeding, and intraprocedure and post-procedure adverse events. Bleeding was defined as the presence of one or more of the following: <72 h of PCI; any access site, retroperitoneal, intracranial, gastrointestinal, or genitourinary bleeding; cardiac tamponade; or non-bypass surgery–related blood transfusion in patients with a pre-PCI hemoglobin (Hgb) ≤8 g/dl or an absolute decrease from pre-PCI Hgb ≥3 g/dl in patients with a pre-PCI Hgb ≤16 g/dl. Intraprocedure and post-procedure events of interest are listed in Table 3.
Baseline characteristics were compared between NW and EO patients using the Wilcoxon rank-sum test for continuous variables, presented as median (25th, 75th interquartile range), and the chi-square test for categorical variables, presented as counts (proportions). Adjusted associations between EO and outcomes were examined using logistic regression and presented as odds ratios (ORs). Adjustment models for each outcome were based on previously developed CathPCI Registry models for bleeding (11) and mortality, with initial models including 28 covariates for bleeding and 25 for mortality and the final model including 7 covariates as listed in the footnote of Figure 1. Additional stratified analysis was performed by presentation (ST-segment elevation myocardial infarction [STEMI] and non-STEMI). To assess the influence of femoral versus radial access on the relationship between outcomes and EO, a test of interaction between arterial access and EO in the regression models of bleeding was performed. A p value <0.05 was considered significant. All analyses were performed using SAS version 9.2 (SAS Institute, Cary, North Carolina).
Of a total of 1,221,086 admissions included in the registry, 83,861 were EO patients and 217,616 were NW; these 2 groups were included in this study. As compared with NW patients, EO patients had lower median age, higher female-to-male ratio, higher proportion of African Americans, and markedly higher prevalence of diabetes mellitus, hypertension, and hyperlipidemia. A history of PCI was more common in this group, but cerebrovascular and peripheral vascular diseases were more common in the NW patients. The distribution of clinical characteristics in these groups is presented in Table 1.
Emergency PCI was performed in 13% of EO patients versus 20% in NW patients, STEMI in 12% versus 19%, and cardiogenic shock <24 h present in 2% versus 4%. EO patients had less extensive coronary disease, with less multivessel disease, fewer lesions with Thrombolysis In Myocardial Infarction flow grade 0, and fewer high-risk Society of Coronary Angiography and Interventions (SCAI) class lesions (SCAI class >1). Transradial access was used more often in these patients. Although PCI procedures were mostly performed in urban hospitals, the distributions of these 2 groups among hospital locations were not significantly different. Most procedures were performed in the South, yet procedures for EO patients were more prevalent in the Midwest. The distribution of procedural characteristics in these groups is presented in Table 2.
Unadjusted, the EO group had a lower rate of mortality, bleeding, periprocedural MI, cardiogenic shock, heart failure, and tamponade and higher rates of renal failure and access hematoma. The in-hospital outcomes of the EO group and the NW group are shown in Table 3. After multivariable adjustment, EO was independently associated with a higher mortality rate (OR: 1.14; 95% CI: 1.04 to 1.25; p = 0.005) and a lower bleeding rate (OR: 0.80; 95% CI: 0.76 to 0.83; p < 0.001) (Fig. 1). In stratified analysis by presentation in the STEMI and non-STEMI subgroups, EO remained independently associated with higher mortality in the STEMI subgroup, whereas the association did not reach statistical significance in the non-STEMI subgroup (Table 4).
The interaction of femoral versus radial access with outcomes was not statistically significant (p = 0.10 for bleeding and p = 0.70 for mortality). In stratified analysis and in both unadjusted and multivariable-adjusted models, bleeding complication rates were lower for EO patients compared with NW patients in both femoral and radial subgroups. Among femoral patients, the unadjusted OR of bleeding complications for EO versus NW was 0.69 (95% CI: 0.67 to 0.72; p < 0.001) and the multivariable-adjusted OR was 0.79 (95% CI: 0.76 to 0.84; p < 0.001). Among radial cases, the unadjusted OR of bleeding complications for EO versus NW was 0.64 (95% CI: 0.54 to 0.76; p < 0.001) and the multivariable-adjusted OR was 0.76 (95% CI: 0.62 to 0.94; p = 0.011).
We compared the in-hospital outcomes of PCI in EO patients with those of patients with normal BMI and found that EO was an independent risk factor for increased in-hospital mortality in those presenting with STEMI. However, in settings other than STEMI, EO was not independently associated with increased in-hospital mortality. EO patients selected for PCI were younger and had less extensive coronary artery disease (CAD); PCI procedures in this subgroup were performed in less emergent settings with lesser procedural risk and were associated with less bleeding.
With 83,861 EO patients included, this study has a large sample size, making it possible to adjust for multiple covariates, thus providing a unique opportunity to explain the “obesity paradox” hypothesis described in prior studies (4–6). This paradox pertains to the decreased rates of bleeding and in-hospital mortality with increasing obesity and alludes to a “protective effect” of obesity. In a recent study of patients undergoing coronary angiography due to acute coronary syndromes, medical therapy and PCI-treated patients who were overweight (BMI 26.5 to 28 kg/m2) had the lowest risk of mortality in unadjusted analysis; when BMI as a continuous variable was studied in patients with significant CAD, the adjusted risk for mortality decreased with increasing BMI up to approximately 35 kg/m2 and then increased (12). Although our findings from crude analysis were similar, after adjustment for covariates, EO was independently associated with worse outcomes in the STEMI presentation subgroup. Our findings are complementary to the findings from a similar study of the relationship between BMI and in-hospital outcomes of patients with STEMI from the NCDR ACTION (Acute Coronary Treatment and Intervention Outcomes Network) Registry–GWTG (Get With the Guidelines) (13). In that study, 5.1% of the 50,149 STEMI patients were extremely obese, were younger by a decade, had less extensive CAD, and had better left ventricular systolic function. Similar to our study and despite these advantages, EO was found to be independently associated with higher in-hospital mortality.
Unmeasured factors such as pulmonary embolism and difficulties with airway management may explain the increased risk of in-hospital mortality associated with STEMI in these patients. Moreover, it is plausible that difficulties with obtaining venous or arterial access in EO patients may be associated with increased complications; however, there are few data to support it. In a registry analysis, the transradial as compared with transfemoral approach for PCI in EO patients was associated with less major bleeding and less access site injuries (14); however, in our study, the interaction of femoral versus radial access with outcomes was not statistically significant. This could be due to the small number of radial procedures (9%) or relative inexperience in transradial procedures.
In our study, EO was independently associated with less in-hospital bleeding complications, which may reflect underdosing of anticoagulants or higher use of closure devices or transradial access. Although weight-based dosing protocols for unfractionated heparin are not affected by morbid obesity, data on weight-based dosing of low-molecular-weight heparin in this group of patients are limited due to either underrepresentation or even exclusion from trials of these drugs, leading to uncertainty whether the dose should be increased with weight proportionally or kept at a maximal dose (15). Underdosing of anticoagulants could be associated with increased mortality in this subgroup. However, it is notable that in our study, there were 4.4 bleeding complications per mortality in the EO patients versus 3.9 in those with NW.
Many important variables, such as presence of sleep apnea or occurrence of in-hospital complications unrelated to PCI (e.g., aspiration pneumonia, pulmonary embolism), were not included in the registry and hence not accounted for in this study. Moreover, the registry does not include data from all U.S. hospitals that perform PCI, which introduces the possibility of selection bias.
EO patients who underwent PCI were younger and had less bleeding compared with NW patients. Despite these differences and regardless of access route, in the subgroup of patients presenting with STEMI, EO was independently associated with higher in-hospital mortality.
The authors acknowledge the NCDR team, Julie Call, Steve Mills, and Mary Printz, East Carolina Heart Institute at Vidant Medical Center, Greenville, North Carolina.
This research was supported by the American College of Cardiology Foundation's National Cardiovascular Data Registry (NCDR). The views expressed in this paper represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at http://www.ncdr.com. CathPCI Registry is an initiative of the American College of Cardiology Foundation and the Society for Cardiovascular Angiography and Interventions. Dr. Daggubati has received speaker fees from Medtronic, Abbott Vascular, Eli Lilly, and AstraZeneca; and consultant fees from Volcano Corporation. 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
- extreme obesity/extremely obese
- normal weight
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received December 13, 2012.
- Revision received May 17, 2013.
- Accepted May 21, 2013.
- American College of Cardiology Foundation
- Gruberg L.,
- Weissman N.J.,
- Waksman R.,
- et al.
- Brindis R.G.,
- Fitzgerald S.,
- Anderson H.V.,
- Shaw R.E.,
- Weintraub W.S.,
- Williams J.F.
- ↵NCDR. ICD Registry. Available at: https://www.ncdr.com/webncdr/icd/. Accessed July 14, 2012.
- Rao S.V.,
- Kaltenbach L.A.,
- Spertus S.,
- Krone R.J.,
- Singh M.,
- Peterson E.D.
- Angerås O.,
- Albertsson P.,
- Karason K.,
- et al.
- Das S.R.,
- Alexander K.P.,
- Chen A.Y.,
- et al.
- Hibbert B.,
- Simard T.,
- Wilson K.R.,
- et al.