International Journal of Cardiology 186 (2015) 266–272
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Clinical outcomes among patients with extreme obesity undergoing elective coronary revascularization: Evaluation of major complications in contemporary practice☆ F. Daniel Ramirez a,1, Benjamin Hibbert a,1, Trevor Simard a, Ronnen Maze a, Ali Pourdjabbar a, Aun-Yeong Chong a, Michel Le May a, Judy Shiau b, Kumanan R. Wilson c, Steven Hawken d, Edward R. O'Brien e, Derek Y. So a,⁎, on behalf of the CArdiovascular Percutaneous TriAL (CAPITAL) Investigators a
Division of Cardiology, University of Ottawa Heart Institute, 40 Ruskin St, Ottawa, ON K1Y 4W7, Canada The Weight Management Clinic, The Ottawa Hospital Civic Campus, 1053 Carling Ave, Ottawa, ON K1Y 4E9, Canada c Department of Medicine, University of Ottawa, The Ottawa Hospital General Campus, 501 Smyth Rd, Ottawa, ON K1H 8L6, Canada d Institute for Clinical Evaluative Sciences, University of Ottawa, The Ottawa Hospital Civic Campus, 1053 Carling Ave, Ottawa, ON K1Y 4E9, Canada e Division of Cardiology, Libin Cardiovascular Institute of Alberta, Foothills Medical Centre, 1403 29 Street NW, Calgary, AB T2N 2T9, Canada b
a r t i c l e
i n f o
Article history: Received 18 June 2014 Received in revised form 21 October 2014 Accepted 17 March 2015 Available online 18 March 2015 Keywords: Body mass index Obesity Coronary artery disease PCI CABG
a b s t r a c t Background/objectives: Individuals with extreme obesity (EO), defined by a body mass index (BMI) ≥40 kg/m2, constitute an increasingly prevalent population at higher risk of procedural complications. The implications of increasing weight burdens among this subset of patients in the setting of elective coronary revascularization have yet to be adequately studied. Methods: We sought to define major complications in this group at one year following contemporary revascularization strategies by retrospectively analysing a cohort of consecutive EO patients undergoing elective percutaneous coronary intervention (PCI) or coronary artery bypass surgery (CABG). The primary endpoint was a composite of peri- and post-procedural complications. Secondary endpoints included a cardiovascular composite and target vessel revascularization (TVR). Results: Adjusted event-free survival curves for the primary endpoint among 133 patients differed significantly with higher BMI (N43.2 kg/m2) associated with greater risk (p = 0.02). The primary endpoint occurred more frequently with CABG compared to PCI (24.2% vs. 5.0%, p b 0.01), which remained significant after adjusting for differences in baseline variables. Rates of the cardiovascular composite and TVR were comparable. Conclusions: Increasing BMI was associated with greater risk for major complications among EO patients undergoing elective coronary revascularization. PCI was associated with fewer complications; however, both revascularization strategies demonstrated equivalent rates of death, MI, and/or stroke. Larger studies may permit a better understanding of the associations between increasing BMI and specific outcomes and to evaluate the role for pre-procedural weight loss in this select population. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Observational studies have reported an ‘obesity paradox’ — a controversial finding that overweight or mildly obese patients with coronary artery disease (CAD) have similar or even improved outcomes, including following myocardial infarction (MI) or percutaneous coronary intervention (PCI), when compared to underweight and/or normal weight counterparts [1–3]. However, it is unclear if this ‘paradoxical’ ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author. E-mail address: [email protected] (D.Y. So). 1 Equally contributing authors.
http://dx.doi.org/10.1016/j.ijcard.2015.03.247 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
protection exists at extremes of body mass index (BMI). Indeed, increased risk of procedural complications, including deep sternal wound infections following coronary artery bypass graft surgery (CABG) [4–6] and access site complications following PCI [7], have been reported with higher levels of obesity. Overall, the implications of increasing weight burdens in such patients in the setting of coronary revascularization have yet to be adequately studied. The World Health Organization stratifies obesity into three classes based on BMI: class I (30.0–34.9), II (35.0–39.9), and III (≥40 kg/m2) [8]. The prevalence of class III obesity, also referred to as morbid or extreme obesity (EO), is increasing at a disproportionately high rate. In the United States, between 1986 and 2000, the overall prevalence of individuals with BMI ≥40 quadrupled and of those with BMI ≥50 quintupled [9]. In the subsequent five years the prevalence of individuals with
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
BMI ≥ 40 increased further by 52% with the subset with BMI ≥50 increasing by 75% [10]. The prevalence of class III obesity in Canada similarly increased by 225% between 1990 and 2003 [11]. It is expected that individuals with EO and its associated risks will continue to form an increasing proportion of patients with CAD requiring revascularization — a trend that may counteract recent improvements in cardiovascular death and hospitalization [12]. Given the dearth of data on revascularization outcomes in the EO population, we sought to characterize major complications associated with contemporary PCI and CABG at our large regional referral centre, including their peri- and post-procedural determinants. 2. Methods 2.1. Study design: retrospective cohort study The University of Ottawa Heart Institute (UOHI) is a tertiary adult cardiac centre serving a population of approximately 1.3 million [13]. Clinical and demographic data, including height and weight, is collected on all patients undergoing coronary angiography. As previously described [7], from January 2007 to August 2010, 21,103 consecutive coronary angiograms with or without PCI were prospectively indexed in the CAPITAL angiography/PCI registry. A cohort of consecutive patients with BMI ≥40 undergoing elective coronary revascularization was identified and included in the analysis. Exclusion criteria included a diagnosis of ST-elevation MI (STEMI) or hemodynamic instability
267
at presentation, incomplete or unavailable electronic medical documentation of procedural/angiographic details, failed PCI, and/or concomitant procedures performed at the time of coronary revascularization (e.g. valvular surgery). 2.2. Baseline patient characteristics Patient records were independently reviewed by three evaluators and differences resolved by consensus. The presence of cardiovascular risk factors (hypertension, diabetes, dyslipidemia, smoking history, family history of CAD), cardiovascular history (prior MI, CABG, PCI, and stroke), baseline renal function, left ventricular systolic function, indication for angiography, number of stenosed coronary arteries, and medications at discharge from hospital were extracted from the registry. Creatinine clearance was calculated using the Cockgroft–Gault equation. Left ventricular dysfunction was defined as an ejection fraction of b50%. Significant obstructive CAD was defined as ≥70% stenosis in any major epicardial coronary artery with the exception of the left main artery in which ≥50% was used. 2.3. Outcomes and definitions The primary endpoint was the incidence of major peri- and postprocedural complications at one year, as defined by a composite of death, MI [14], stroke (as diagnosed by neurological imaging and/or confirmed by a neurologist), deep surgical site infection (deep SSI, as
Fig. 1. Study flow diagram. AVR: aortic valve replacement, BMI: body mass index, CABG: coronary artery bypass surgery, LV: left ventricle, MVR: mitral valve repair, PCI: percutaneous coronary intervention, PFO: patent foramen ovale.
268
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
diagnosed by an infectious disease specialist and/or meeting Center for Disease Control and Prevention criteria [15]), vascular access site complication (VASC) requiring intervention (for the PCI cohort), and Thrombolysis In Myocardial Infarction (TIMI) major or CABG-related bleed (for the PCI and CABG cohorts, respectively) [16,17]. A secondary composite endpoint of death, MI, and/or stroke at one year was also assessed. Other outcomes included the rate of target vessel revascularization (TVR) at one year, length of stay in hospital (LOS), and components of the primary outcome. Length of follow-up and time to death, MI, or stroke was calculated from the date of the index catheterization prompting revascularization. Time to deep SSI, VASC, or major/CABGrelated bleeding was calculated using the date of diagnosis (as per medical records). The LOS was defined as the total number of days in hospital after the revascularization procedure (PCI or CABG).
2.4. Statistical analyses Continuous variables are reported as mean ± standard deviation or median with interquartile range and were compared using a t-test or Mann–Whitney rank sum test, when appropriate. Categorical variables are reported as number (%) and were compared using chi-square or Fisher's exact test. Time to event outcomes were analysed in a Cox proportional hazards regression analysis. Both univariate and multivariable adjusted Cox models were fit. Covariates included BMI, age, sex, hypertension, diabetes, dyslipidemia, smoking history, and positive family history of CAD. A two-sided p-value b 0.05 was the criterion used to declare statistical significance. All analyses were conducted using SAS version 9.2.
3.3. Major complications by mode of revascularization The primary endpoint occurred in 24.2% of CABG patients and 5.0% of PCI cases (HR 5.55; 95% CI 1.81–17.03; p b 0.01). When adjusted for differences in baseline variables, event-free survival for the primary endpoint remained in favour of PCI (HR 5.49; 95% CI 1.50–20.05; p b 0.01) (Fig. 3A). All-cause mortality was 6.1% in CABG cases compared to 0% in the PCI cohort (p = 0.10). No differences were observed in the rates of MI (3.0% vs. 4.0%, p = 0.78) and no strokes occurred in either group at one year. In patients revascularized by CABG, deep SSI occurred in 12.1%, all of which consisted of sternal wound infections. Superficial SSI was diagnosed in 30.3% of CABG cases with both sternal and graft harvest sites included. No cases of VASC requiring surgical intervention were observed in the PCI arm. Major bleeding trended towards an increased frequency with surgery with TIMI CABG-related/major bleeds identified in 9.1% of CABG patients and 1.0% of PCI cases (p = 0.08, Table 2). No significant difference was observed in either the unadjusted or adjusted rates of the cardiovascular composite of death, MI, and stroke (9.1% CABG vs. 4.0% PCI, Table 2, Fig. 3B). Median LOS was significantly longer in the CABG group than in the PCI cohort (11 vs. 1 days,
Table 1 Baseline characteristics, angiographic findings, and medical therapies at discharge of patients undergoing elective coronary revascularization.
Age, years Male BMI, kg/m2 (IQR) Height, m Weight, kg
3. Results 3.1. Patient population A cohort of 602 EO patients were identified from 21,103 coronary angiograms performed over the study period. Of these, 133 underwent elective isolated coronary revascularization: 33 patients undergoing CABG and 100 patients undergoing PCI. Fourteen patients who underwent surgical revascularization were excluded due to a concomitant procedure (most often valvular repair/replacement) (Fig. 1). Baseline characteristics and cardiovascular risk factors were comparable among the CABG and PCI groups with the exception of diabetes, which was more common in the surgery cohort (81.8% vs. 51.0%, p b 0.01). Patients in the CABG group had more triple-vessel disease (57.6% vs. 13.0%, p b 0.001) and left main stenosis (36.4% vs. 4.0%, p b 0.001). Conversely, single-vessel disease was more common in the PCI group (12.1% vs. 58.0%, p b 0.001) as was clopidogrel prescription at discharge (12.1% vs. 98.0%, p b 0.001). No significant differences were found in the use of acetylsalicylic acid, beta blocker, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker, or statin therapy at discharge between cohorts (Table 1). Median follow-up duration was 22 months in both revascularization groups (p = NS, Table 2).
3.2. Influence of incremental BMI The primary endpoint of any major complication at one year occurred in 9.8% of all EO patients undergoing elective revascularization. To investigate the incremental effects of increasing BMI, we divided the total cohort into two groups based on BMI above or below the median value (43.2 kg/m2). Patients with BMI above the median had an increased risk of the primary endpoint (HR 3.70; 95% CI 1.02–13.47; p b 0.05). This effect persisted after adjustment for baseline covariates (HR 5.89; 95% CI 1.33–26.14; p = 0.02) (Fig. 2).
Cardiovascular risk factors Hypertension Diabetes Dyslipidemia Smoking history Positive family history Cardiovascular history Myocardial infarction PCI CABG Stroke Creatinine clearance, mL/min (IQR)a Baseline creatinine, μmol/L Indication Positive non-invasive testing Stable angina NSTEACS Other Stenosed coronary arteries (≥70%) Single-vessel Double-vessel Triple-vessel Left main stenosis (≥50%) LV dysfunction (EF b 50%) Medications at discharge Acetylsalicylic acid (ASA) Clopidogrel Beta blockerb ACE inhibitor/ARBb Statinb
CABG (n = 33)
PCI (n = 100)
p-Value
59.3 ± 9.5 22 (66.7) 42.7 (40.6–46.5) 1.7 (1.6–1.8) 124.3 (112.9–132.2)
57.5 ± 9.7 56 (56.0) 43.7 (41.5–48.1) 1.7 (1.6–1.8) 122.3 (110.3–140.6)
31 (93.9) 27 (81.8) 30 (90.9) 21 (63.6) 5 (15.2)
78 (78.0) 51 (51.0) 80 (80.0) 61 (61.0) 18 (18.0)
9 (27.3) 5 (15.2) 0 (0) 1 (3.0) 124 (82–176)
27 (27.0) 30 (30.0) 8 (8.0) 5 (5.0) 135 (109–165)
98 (72–116)
87 (76–105)
9 (27.3) 9 (27.3) 14 (42.4) 1 (3.0)
20 (20.0) 19 (19.0) 57 (57.0) 4 (4.0)
0.53 0.45 0.21 0.78
4 (12.1) 10 (30.3) 19 (57.6) 12 (36.4) 5 (15.2)
58 (58.0) 28 (28.0) 13 (13.0) 4 (4.0) 6 (6.0)
b0.01 0.98 b0.01 b0.01 0.20
32 (97.0) 4 (12.1) 30 (90.0) 25 (75.8) 31 (93.9)
96 (96.0) 98 (98.0) 60 (87.0) 58 (84.1) 65 (94.2)
0.78 b0.001 0.80 0.46 0.69
0.38 0.38 0.11 0.64 0.82
0.07 b0.01 0.24 0.95 0.91 0.85 0.15 0.21 0.99 0.43
Values in parentheses represent percent of total cases in which the relevant data were available unless otherwise stated. Standard deviation reported for continuous variables. ACE: angiotensin converting enzyme; ARB: angiotensin receptor blocker; BMI: body mass index; CABG: coronary artery bypass graft surgery; EF: ejection fraction; eGFR: estimated glomerular filtration rate; IQR: interquartile range; LV: left ventricular; NSTEACS: non-ST-segment elevation acute coronary syndrome; PCI: percutaneous coronary intervention. a As estimated by the Cockcroft–Gault equation. b Data available for 69 cases in PCI arm.
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
269
Table 2 Clinical endpoints in extremely obese patients undergoing elective coronary revascularization (unadjusted).
Length of follow-up, months (IQR) Primary endpoint Death Myocardial infarction Stroke Deep surgical site infection Vascular access site complication TIMI major/CABG-related bleed Blood transfusion CV composite of death, MI, and stroke Length of stay in hospital, days (IQR) Target vessel revascularization
Total cohort (n = 133)
CABG (n = 33)
PCI (n = 100)
p-Value⁎
22 (3–33) 13 (9.8) 2 (1.5) 5 (3.8) 0 (0)
20 (3–38) 8 (24.2) 2 (6.1) 1 (3.0) 0 (0) 4 (12.1) – 3 (9.1) 8 (24.2) 3 (9.1) 11 (8–22) 1 (3.0)
23 (3–32) 5 (5.0) 0 (0) 4 (4.0) 0 (0) – 0 (0) 1 (1.0) 5 (5.0) 4 (4.0) 1 (0–2) 7 (7.0)
0.76 b0.01 0.10 0.78 –
4 (3.0) 13 (9.8) 7 (5.3) 2 (0–7) 8 (6.0)
0.08 b0.01 0.49 b0.001 0.68
CABG: coronary artery bypass surgery; CV: cardiovascular; IQR: interquartile range; PCI: percutaneous coronary intervention; TIMI: Thrombolysis In Myocardial Infarction. ⁎ p-Value calculated for CABG vs. PCI groups.
p b 0.001). No difference was observed in the rates of TVR between groups (p = 0.68) (Table 2). 4. Discussion The prevalence of obesity, particularly class III obesity, continues to rise — a change in demographic that is shifting the physical and metabolic profile of patients undergoing coronary revascularization [18]. Accordingly, it is important to characterize clinical outcomes in this understudied population. Our study is the first to exclusively focus on elective coronary revascularization in this group. Our evaluation showed an incidence of overall complications of 9.8% among EO patients undergoing revascularization and suggested an incremental risk among those with higher BMI. Furthermore, patients undergoing CABG appeared to have an increased risk of complications compared to those undergoing PCI. Previous studies examining the impact of obesity in the setting of stable CAD, MI, and/or coronary revascularization have largely compared patients with obesity to their non-obese counterparts [2,19,20]. Data from such studies support the existence of a ‘paradoxical’ protective effect among mildly obese cardiac patients. However, these initial
studies lacked focus on EO patients (with BMI ≥ 40). Emerging evidence suggests that this subset is exempt from this protection and may represent a higher-risk population for coronary revascularization [18], in part presumably because the metabolic and physiologic strain conferred by greater weight burdens predispose to peri- and postprocedural complications. Consistent with these findings, our study – comprising contemporary revascularization techniques and outcome data to one year – found high complication rates among this population and suggests that even within this group increasing BMI further augments peri-procedural risk. Interestingly, we additionally observed less procedural morbidity with PCI compared to CABG. Although not randomized, adjustment by baseline covariates did not alter these results. EO has been associated with higher rates of major adverse clinical events both in the setting of CABG [4] and PCI [18,22], including primary PCI for STEMI [21,23], the cause of which is likely multifactorial. In the setting of PCI, EO patients have been shown to have higher rates of VASC [24]; image acquisition during PCI may be suboptimal rendering the procedure technically more difficult; and previous studies have suggested BMI as a risk factor for clopidogrel unresponsiveness [25], which may place patients at risk for thrombotic complications after stenting. Similarly, EO
Fig. 2. Event-free survival curves for the primary endpoint in extremely obese patients undergoing elective coronary revascularization stratified by median of body mass index (BMI).
270
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
patients may face multiple potential risks when undergoing CABG. Technical aspects, such as graft harvesting and implantation as well as control of haemostasis, may be more challenging because of body habitus. Post-operatively, prolonged need for ventilatory support and LOS has been reported [26] while impaired healing may increase the risk for wound infections. In our study, the increased overall complication rate among patients undergoing CABG seemed to be driven by wound infections. The incidence of deep sternal SSIs in our study is higher than in previous reports; however, direct comparisons are limited by differences in study designs and baseline patient characteristics. Moreover, definitions for
SSIs vary considerably among studies. Two registry-based studies found rates of deep sternal infections of 1.4% and 1.8% in EO patients following CABG [4,6]; however, both studies were restricted to inhospital infections and the studies failed to define how sternal SSIs were diagnosed. Similarly, Birkmeyer et al. reported a sternal SSI rate of only 2–3% in a cohort of patients undergoing CABG who were identified as being “severely obese”. However, the study's criteria for diagnosing sternal SSIs differed from the CDC criteria used in our study. Furthermore, “severe obesity” was defined as a BMI N36, only 45.7% of patients in their cohort were diabetic, and only in-hospital outcomes were assessed [27]. The importance of accounting for complications
Fig. 3. Event-free survival curves for the primary (A) and secondary (B) cardiovascular composite endpoints following elective coronary artery bypass surgery (CABG) versus percutaneous coronary intervention (PCI) in patients with extreme obesity.
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
arising after discharge from hospital is underscored by reports that 13.8% of CABG patients require re-hospitalization within 30 days of discharge of which nearly 1 in 5 are for wound infections [27]. Our study also suggested that EO patients who underwent CABG were at increased risk of severe bleeding compared to PCI while PCI patients had a low rate of VASC requiring intervention. We previously reported a low incidence of VASC in this population [7] and a recent large retrospective study similarly found a VASC rate of 2.75% in EO patients undergoing PCI [18]. Higher rates of radial access at our centre as well as the use of bivalirudin (previously reported to be 36% and 17%, respectively, in this patient group) may in part explain the low rate of bleeding events seen in our PCI cohort compared to other reports in the literature [7,17,21]. Additionally, as has been postulated by others, underdosing of anticoagulants (particularly low molecular weight heparins) in this patient population due to uncertainty in optimal dosing at extremes of BMI may result in a favourable bleeding risk profile [21]. Although ischemic complications, as defined by our secondary endpoint, were not statistically increased among patients revascularized with CABG compared to PCI, our study is likely underpowered to definitively ascertain this observation. Overall, these observations raise two important questions for the management of EO patients requiring revascularization: 1) could interventions aimed at weight reduction before coronary revascularization reduce the risk of procedural morbidity, and 2) should less invasive alternatives be preferred in patients needing revascularization? The notion that pre-procedural weight reduction at such extremes of BMI may improve peri- and post-procedural outcomes is an area of considerable research and debate. Pre-operative weight loss has been associated with reduced operative times, operative complications, and LOS — benefits observed with weight reductions of as little as 5% in studies of bariatric surgery patients, for instance [28,29]. Indeed, while still debated [31], such findings have led to institutes and organizations recommending specific weight loss goals before consideration for bariatric surgery [30,31]. The question of whether such an approach could be considered before elective coronary revascularization – for instance, using a low-calorie diet incorporating meal replacements – remains to be determined [32]. The question of whether less invasive alternatives should be preferred in EO patients similarly remains unanswered. To date, despite advances in stent designs and marked improvements in associated clinical outcomes [33] as well as perceived increased operative risk with EO as suggested by the underrepresentation of obese patients with severe ischemic heart disease in CABG registries [34], there is little data to support PCI as the predominant strategy in this population. Indeed, the applicability of studies comparing surgical to percutaneous revascularization to EO patients is questionable given the relatively small number EO patients represented in the study populations. The use of PCI over CABG may afford the benefits of a less invasive approach thereby reducing infections and bleeding; however, this would likely come at the cost of less complete revascularization. Ultimately, larger studies comparing the efficacy of percutaneous to surgical revascularization in EO patients are needed. 4.1. Study limitations Our study has several limitations. The retrospective, non-randomized design of the study may have allowed unaccounted factors to influence the differences between PCI and CABG. However, we adjusted for key covariates that may affect revascularization outcomes in an attempt to account for differences among the two groups. Though coronary anatomy was assessed, detailed assessment of disease complexity was not performed. Our study was also relatively small despite the large volume of procedures performed at our institute as patients with EO accounted for only 3.7% of all angiograms (consistent with the 3.1% prevalence of class III obesity reported in the 2007 to 2009 Canadian Health Measures Survey) [35]. Nevertheless, our study is the largest to date to focus on the association between increasing BMI and clinical outcomes in the EO
271
population following elective coronary revascularization and highlights the need for further research into harm-reduction strategies in these high-risk patients. 5. Conclusions Within the EO population increasing BMI is associated with an increased risk of adverse clinical outcomes following elective coronary revascularization. The risk of SSI, in particular, is an important contributor to the morbidity associated with surgical approaches in this population. Modifiable factors influencing outcomes in these high-risk patients and the potential role for interventions aimed at weight reduction prior to revascularization remain to be fully elucidated. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Acknowledgements None. References [1] A. Romero-Corral, V.M. Montori, V.K. Somers, et al., Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies, Lancet 368 (2006) 666–678. [2] L. Mehta, W. Devlin, P.A. McCullough, et al., Impact of body mass index on outcomes after percutaneous coronary intervention in patients with acute myocardial infarction, Am. J. Cardiol. 99 (2007) 906–910. [3] M. Schmiegelow, C. Torp-Pedersen, G.H. Gislason, et al., Relation of body mass index to risk of stent thrombosis after percutaneous coronary intervention, Am. J. Cardiol. 110 (2012) 1592–1597. [4] G. Prabhakar, C.K. Haan, E.D. Peterson, L.P. Coombs, J.L. Cruzzavala, G.F. Murray, The risks of moderate and extreme obesity for coronary artery bypass grafting outcomes: a study from the Society of Thoracic Surgery's database, Ann. Thorac. Surg. 74 (2002) 1125–1130. [5] V.G. Fowler Jr., S.M. O'Brien, L.H. Muhlbaier, G.R. Corey, T.B. Ferguson, E.D. Peterson, Clinical predictors of major infections after cardiac surgery, Circulation 112 (2005) I358–I365. [6] R. Jin, G.L. Grunkemeier, A.P. Furnary, J.R. Handy Jr., Is obesity a risk factor for mortality in coronary artery bypass surgery? Circulation 111 (2005) 3359–3365. [7] B. Hibbert, T. Simard, K.R. Wilson, et al., Transradial versus transfemoral artery approach for coronary angiography and percutaneous coronary intervention in the extremely obese, JACC Cardiovasc. Interv. 5 (2012) 819–826. [8] W.H.O. Expert Consultation, Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies, Lancet 363 (2004) 157–163. [9] R. Sturm, Increases in clinically severe obesity in the United States, 1986–2000, Arch. Intern. Med. 163 (2003) 2146–2148. [10] R. Sturm, Increases in morbid obesity in the USA: 2000–2005, Public Health 121 (2007) 492–496. [11] P.T. Katzmarzyk, C. Mason, Prevalence of class I, II and III obesity in Canada, CMAJ 174 (2006) 156–157. [12] N.R.C. Campbell, J. Onysko, H. Johansen, R.N. Gao, Changes in cardiovascular deaths and hospitalization in Canada, Can. J. Cardiol. 22 (2006) 425–427. [13] M.R. Le May, D.Y. So, R. Dionne, et al., A citywide protocol for primary PCI in STsegment elevation in myocardial infarction, N. Engl. J. Med. 358 (2008) 231–240. [14] K. Thygesen, J.S. Alpert, A.S. Jaffe, M.L. Simoons, B.R. Chaitman, H.D. White, Third universal definition of myocardial infarction, Circulation 126 (2012) 2020–2035. [15] A.J. Mangram, T.C. Horan, M.L. Pearson, L.C. Silver, W.R. Jarvis, Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee, Infect. Control Hosp. Epidemiol. 20 (1999) 250–278. [16] R. Mehran, S.V. Rao, D.L. Bhatt, et al., Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the bleeding academic research consortium, Circulation 123 (2011) 2736–2747. [17] B. Hibbert, A. MacDougall, M. Labinaz, et al., Bivalirudin for primary percutaneous coronary interventions: outcome assessment in the Ottawa STEMI registry, Circ. Cardiovasc. Interv. 5 (2012) 805–812. [18] M.E. Buschur, D. Smith, D. Share, et al., The burgeoning epidemic of morbid obesity in patients undergoing percutaneous coronary intervention. Insight from the Blue Cross Shield of Michigan Cardiovascular Consortium, J. Am. Coll. Cardiol. 62 (2013) 685–691. [19] S.G. Ellis, J. Elliott, M. Horrigan, R.E. Raymond, G. Howell, Low-normal or excessive body mass index: newly identified and powerful risk factors for death and other complications with percutaneous coronary intervention, Am. J. Cardiol. 78 (1996) 642–646.
272
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
[20] A. Bechlioulis, K. Vakalis, K.K. Naka, et al., Paradoxical protective effect of central obesity in patients with suspected stable coronary artery disease, Obesity 21 (2013) E314–E321. [21] S. Payvar, S. Kim, S.V. Rao, et al., In-hospital outcomes of percutaneous coronary interventions in extremely obese and normal-weight patients: findings from the NCDR (National Cardiovascular Data Registry), J. Am. Coll. Cardiol. 62 (2013) 692–696. [22] R.M. Minutello, E.T. Chou, M.K. Hong, et al., Impact of body mass index on inhospital outcomes following percutaneous coronary intervention (report from the New York State Angioplasty Registry), Am. J. Cardiol. 93 (2004) 1229–1232. [23] S.R. Das, K.P. Alexander, A.Y. Chen, et al., Impact of body weight and extreme obesity on the presentation, treatment, and in-hospital outcomes of 50,149 patients with ST-segment elevation myocardial infarction: results from the NCDR (National Cardiovascular Data Registry), J. Am. Coll. Cardiol. 58 (2011) 2642–2650. [24] N. Cox, F.S. Resnic, J.J. Popma, D.I. Simon, A.C. Eisenhauer, C.R. Rogers, Comparison of the risk of vascular complications associated with femoral and radial access coronary catheterization procedures in obese versus nonobese patients, Am. J. Cardiol. 94 (2004) 1174–1177. [25] D.J. Angiolillo, A. Fernandez-Ortiz, E. Bernardo, et al., Variability in individual responsiveness to clopidogrel: clinical implications, management and future perspectives, J. Am. Coll. Cardiol. 49 (2007) 1505–1516. [26] C.H. Wigfield, J.D. Lindsey, A. Muñoz, P.S. Chopra, N.M. Edwards, R.B. Love, Is extreme obesity a risk factor for cardiac surgery? An analysis of patients with a BMI ≥40, Eur. J. Cardiothorac. Surg. 29 (2006) 434–440.
[27] N.J.O. Birkmeyer, D.C. Charlesworth, F. Hernandez, et al., Obesity and risk of adverse outcomes associated with coronary artery bypass surgery, Circulation 97 (1998) 1689–1694. [28] S. Cassie, C. Menezes, D.W. Birch, X. Shi, S. Karmali, Effect of preoperative weight loss in bariatric surgical patients: a systematic review, Surg. Obes. Relat. Dis. 7 (2011) 760–767. [29] C.D. Still, P. Benotti, G.C. Wood, et al., Outcomes of preoperative weight loss in highrisk patients undergoing gastric bypass surgery, Arch. Surg. 142 (2007) 994–998. [30] R.S. Alami, J.M. Morton, R. Schuster, et al., Is there a benefit to preoperative weight loss in gastric bypass patients? A prospective randomized trial, Surg. Obes. Relat. Dis. 3 (2007) 141–146. [31] M.R. Ali, S. Baucom-Pro, G.A. Broderick-Villa, et al., Weight loss before gastric bypass: feasibility and effect on postoperative weight loss and weight loss maintenance, Surg. Obes. Relat. Dis. 3 (2007) 515–520. [32] M.M. Elahi, G.K. Chetty, A.W. Sosnowski, M.S. Hickey, T.J. Spyt, Morbid obesity increases perioperative morbidity in first-time CABG patients — should resources be redirected to weight reduction, Int. J. Cardiol. 105 (2005) 98–99. [33] T. Simard, B. Hibbert, F.D. Ramirez, M. Froeschl, Y.X. Chen, E.R. O'Brien, The evolution of coronary stents: a brief review, Can. J. Cardiol. 30 (2014) 35–45. [34] B.C. Reeves, R. Ascione, M.H. Chamberlain, G.D. Angelini, Effect of body mass index on early outcomes in patients undergoing coronary artery bypass surgery, J. Am. Coll. Cardiol. 42 (2003) 668–676. [35] M. Shields, M.D. Carroll, C.L. Ogden, Adult obesity prevalence in Canada and the United States, NCHS Data Brief, no. 56, National Center for Health Statistics, Hyatsville, 2011.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Clinical outcomes among patients with extreme obesity undergoing elective coronary revascularization: Evaluation of major complications in contemporary practice☆ F. Daniel Ramirez a,1, Benjamin Hibbert a,1, Trevor Simard a, Ronnen Maze a, Ali Pourdjabbar a, Aun-Yeong Chong a, Michel Le May a, Judy Shiau b, Kumanan R. Wilson c, Steven Hawken d, Edward R. O'Brien e, Derek Y. So a,⁎, on behalf of the CArdiovascular Percutaneous TriAL (CAPITAL) Investigators a
Division of Cardiology, University of Ottawa Heart Institute, 40 Ruskin St, Ottawa, ON K1Y 4W7, Canada The Weight Management Clinic, The Ottawa Hospital Civic Campus, 1053 Carling Ave, Ottawa, ON K1Y 4E9, Canada c Department of Medicine, University of Ottawa, The Ottawa Hospital General Campus, 501 Smyth Rd, Ottawa, ON K1H 8L6, Canada d Institute for Clinical Evaluative Sciences, University of Ottawa, The Ottawa Hospital Civic Campus, 1053 Carling Ave, Ottawa, ON K1Y 4E9, Canada e Division of Cardiology, Libin Cardiovascular Institute of Alberta, Foothills Medical Centre, 1403 29 Street NW, Calgary, AB T2N 2T9, Canada b
a r t i c l e
i n f o
Article history: Received 18 June 2014 Received in revised form 21 October 2014 Accepted 17 March 2015 Available online 18 March 2015 Keywords: Body mass index Obesity Coronary artery disease PCI CABG
a b s t r a c t Background/objectives: Individuals with extreme obesity (EO), defined by a body mass index (BMI) ≥40 kg/m2, constitute an increasingly prevalent population at higher risk of procedural complications. The implications of increasing weight burdens among this subset of patients in the setting of elective coronary revascularization have yet to be adequately studied. Methods: We sought to define major complications in this group at one year following contemporary revascularization strategies by retrospectively analysing a cohort of consecutive EO patients undergoing elective percutaneous coronary intervention (PCI) or coronary artery bypass surgery (CABG). The primary endpoint was a composite of peri- and post-procedural complications. Secondary endpoints included a cardiovascular composite and target vessel revascularization (TVR). Results: Adjusted event-free survival curves for the primary endpoint among 133 patients differed significantly with higher BMI (N43.2 kg/m2) associated with greater risk (p = 0.02). The primary endpoint occurred more frequently with CABG compared to PCI (24.2% vs. 5.0%, p b 0.01), which remained significant after adjusting for differences in baseline variables. Rates of the cardiovascular composite and TVR were comparable. Conclusions: Increasing BMI was associated with greater risk for major complications among EO patients undergoing elective coronary revascularization. PCI was associated with fewer complications; however, both revascularization strategies demonstrated equivalent rates of death, MI, and/or stroke. Larger studies may permit a better understanding of the associations between increasing BMI and specific outcomes and to evaluate the role for pre-procedural weight loss in this select population. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Observational studies have reported an ‘obesity paradox’ — a controversial finding that overweight or mildly obese patients with coronary artery disease (CAD) have similar or even improved outcomes, including following myocardial infarction (MI) or percutaneous coronary intervention (PCI), when compared to underweight and/or normal weight counterparts [1–3]. However, it is unclear if this ‘paradoxical’ ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author. E-mail address: [email protected] (D.Y. So). 1 Equally contributing authors.
http://dx.doi.org/10.1016/j.ijcard.2015.03.247 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
protection exists at extremes of body mass index (BMI). Indeed, increased risk of procedural complications, including deep sternal wound infections following coronary artery bypass graft surgery (CABG) [4–6] and access site complications following PCI [7], have been reported with higher levels of obesity. Overall, the implications of increasing weight burdens in such patients in the setting of coronary revascularization have yet to be adequately studied. The World Health Organization stratifies obesity into three classes based on BMI: class I (30.0–34.9), II (35.0–39.9), and III (≥40 kg/m2) [8]. The prevalence of class III obesity, also referred to as morbid or extreme obesity (EO), is increasing at a disproportionately high rate. In the United States, between 1986 and 2000, the overall prevalence of individuals with BMI ≥40 quadrupled and of those with BMI ≥50 quintupled [9]. In the subsequent five years the prevalence of individuals with
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
BMI ≥ 40 increased further by 52% with the subset with BMI ≥50 increasing by 75% [10]. The prevalence of class III obesity in Canada similarly increased by 225% between 1990 and 2003 [11]. It is expected that individuals with EO and its associated risks will continue to form an increasing proportion of patients with CAD requiring revascularization — a trend that may counteract recent improvements in cardiovascular death and hospitalization [12]. Given the dearth of data on revascularization outcomes in the EO population, we sought to characterize major complications associated with contemporary PCI and CABG at our large regional referral centre, including their peri- and post-procedural determinants. 2. Methods 2.1. Study design: retrospective cohort study The University of Ottawa Heart Institute (UOHI) is a tertiary adult cardiac centre serving a population of approximately 1.3 million [13]. Clinical and demographic data, including height and weight, is collected on all patients undergoing coronary angiography. As previously described [7], from January 2007 to August 2010, 21,103 consecutive coronary angiograms with or without PCI were prospectively indexed in the CAPITAL angiography/PCI registry. A cohort of consecutive patients with BMI ≥40 undergoing elective coronary revascularization was identified and included in the analysis. Exclusion criteria included a diagnosis of ST-elevation MI (STEMI) or hemodynamic instability
267
at presentation, incomplete or unavailable electronic medical documentation of procedural/angiographic details, failed PCI, and/or concomitant procedures performed at the time of coronary revascularization (e.g. valvular surgery). 2.2. Baseline patient characteristics Patient records were independently reviewed by three evaluators and differences resolved by consensus. The presence of cardiovascular risk factors (hypertension, diabetes, dyslipidemia, smoking history, family history of CAD), cardiovascular history (prior MI, CABG, PCI, and stroke), baseline renal function, left ventricular systolic function, indication for angiography, number of stenosed coronary arteries, and medications at discharge from hospital were extracted from the registry. Creatinine clearance was calculated using the Cockgroft–Gault equation. Left ventricular dysfunction was defined as an ejection fraction of b50%. Significant obstructive CAD was defined as ≥70% stenosis in any major epicardial coronary artery with the exception of the left main artery in which ≥50% was used. 2.3. Outcomes and definitions The primary endpoint was the incidence of major peri- and postprocedural complications at one year, as defined by a composite of death, MI [14], stroke (as diagnosed by neurological imaging and/or confirmed by a neurologist), deep surgical site infection (deep SSI, as
Fig. 1. Study flow diagram. AVR: aortic valve replacement, BMI: body mass index, CABG: coronary artery bypass surgery, LV: left ventricle, MVR: mitral valve repair, PCI: percutaneous coronary intervention, PFO: patent foramen ovale.
268
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
diagnosed by an infectious disease specialist and/or meeting Center for Disease Control and Prevention criteria [15]), vascular access site complication (VASC) requiring intervention (for the PCI cohort), and Thrombolysis In Myocardial Infarction (TIMI) major or CABG-related bleed (for the PCI and CABG cohorts, respectively) [16,17]. A secondary composite endpoint of death, MI, and/or stroke at one year was also assessed. Other outcomes included the rate of target vessel revascularization (TVR) at one year, length of stay in hospital (LOS), and components of the primary outcome. Length of follow-up and time to death, MI, or stroke was calculated from the date of the index catheterization prompting revascularization. Time to deep SSI, VASC, or major/CABGrelated bleeding was calculated using the date of diagnosis (as per medical records). The LOS was defined as the total number of days in hospital after the revascularization procedure (PCI or CABG).
2.4. Statistical analyses Continuous variables are reported as mean ± standard deviation or median with interquartile range and were compared using a t-test or Mann–Whitney rank sum test, when appropriate. Categorical variables are reported as number (%) and were compared using chi-square or Fisher's exact test. Time to event outcomes were analysed in a Cox proportional hazards regression analysis. Both univariate and multivariable adjusted Cox models were fit. Covariates included BMI, age, sex, hypertension, diabetes, dyslipidemia, smoking history, and positive family history of CAD. A two-sided p-value b 0.05 was the criterion used to declare statistical significance. All analyses were conducted using SAS version 9.2.
3.3. Major complications by mode of revascularization The primary endpoint occurred in 24.2% of CABG patients and 5.0% of PCI cases (HR 5.55; 95% CI 1.81–17.03; p b 0.01). When adjusted for differences in baseline variables, event-free survival for the primary endpoint remained in favour of PCI (HR 5.49; 95% CI 1.50–20.05; p b 0.01) (Fig. 3A). All-cause mortality was 6.1% in CABG cases compared to 0% in the PCI cohort (p = 0.10). No differences were observed in the rates of MI (3.0% vs. 4.0%, p = 0.78) and no strokes occurred in either group at one year. In patients revascularized by CABG, deep SSI occurred in 12.1%, all of which consisted of sternal wound infections. Superficial SSI was diagnosed in 30.3% of CABG cases with both sternal and graft harvest sites included. No cases of VASC requiring surgical intervention were observed in the PCI arm. Major bleeding trended towards an increased frequency with surgery with TIMI CABG-related/major bleeds identified in 9.1% of CABG patients and 1.0% of PCI cases (p = 0.08, Table 2). No significant difference was observed in either the unadjusted or adjusted rates of the cardiovascular composite of death, MI, and stroke (9.1% CABG vs. 4.0% PCI, Table 2, Fig. 3B). Median LOS was significantly longer in the CABG group than in the PCI cohort (11 vs. 1 days,
Table 1 Baseline characteristics, angiographic findings, and medical therapies at discharge of patients undergoing elective coronary revascularization.
Age, years Male BMI, kg/m2 (IQR) Height, m Weight, kg
3. Results 3.1. Patient population A cohort of 602 EO patients were identified from 21,103 coronary angiograms performed over the study period. Of these, 133 underwent elective isolated coronary revascularization: 33 patients undergoing CABG and 100 patients undergoing PCI. Fourteen patients who underwent surgical revascularization were excluded due to a concomitant procedure (most often valvular repair/replacement) (Fig. 1). Baseline characteristics and cardiovascular risk factors were comparable among the CABG and PCI groups with the exception of diabetes, which was more common in the surgery cohort (81.8% vs. 51.0%, p b 0.01). Patients in the CABG group had more triple-vessel disease (57.6% vs. 13.0%, p b 0.001) and left main stenosis (36.4% vs. 4.0%, p b 0.001). Conversely, single-vessel disease was more common in the PCI group (12.1% vs. 58.0%, p b 0.001) as was clopidogrel prescription at discharge (12.1% vs. 98.0%, p b 0.001). No significant differences were found in the use of acetylsalicylic acid, beta blocker, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker, or statin therapy at discharge between cohorts (Table 1). Median follow-up duration was 22 months in both revascularization groups (p = NS, Table 2).
3.2. Influence of incremental BMI The primary endpoint of any major complication at one year occurred in 9.8% of all EO patients undergoing elective revascularization. To investigate the incremental effects of increasing BMI, we divided the total cohort into two groups based on BMI above or below the median value (43.2 kg/m2). Patients with BMI above the median had an increased risk of the primary endpoint (HR 3.70; 95% CI 1.02–13.47; p b 0.05). This effect persisted after adjustment for baseline covariates (HR 5.89; 95% CI 1.33–26.14; p = 0.02) (Fig. 2).
Cardiovascular risk factors Hypertension Diabetes Dyslipidemia Smoking history Positive family history Cardiovascular history Myocardial infarction PCI CABG Stroke Creatinine clearance, mL/min (IQR)a Baseline creatinine, μmol/L Indication Positive non-invasive testing Stable angina NSTEACS Other Stenosed coronary arteries (≥70%) Single-vessel Double-vessel Triple-vessel Left main stenosis (≥50%) LV dysfunction (EF b 50%) Medications at discharge Acetylsalicylic acid (ASA) Clopidogrel Beta blockerb ACE inhibitor/ARBb Statinb
CABG (n = 33)
PCI (n = 100)
p-Value
59.3 ± 9.5 22 (66.7) 42.7 (40.6–46.5) 1.7 (1.6–1.8) 124.3 (112.9–132.2)
57.5 ± 9.7 56 (56.0) 43.7 (41.5–48.1) 1.7 (1.6–1.8) 122.3 (110.3–140.6)
31 (93.9) 27 (81.8) 30 (90.9) 21 (63.6) 5 (15.2)
78 (78.0) 51 (51.0) 80 (80.0) 61 (61.0) 18 (18.0)
9 (27.3) 5 (15.2) 0 (0) 1 (3.0) 124 (82–176)
27 (27.0) 30 (30.0) 8 (8.0) 5 (5.0) 135 (109–165)
98 (72–116)
87 (76–105)
9 (27.3) 9 (27.3) 14 (42.4) 1 (3.0)
20 (20.0) 19 (19.0) 57 (57.0) 4 (4.0)
0.53 0.45 0.21 0.78
4 (12.1) 10 (30.3) 19 (57.6) 12 (36.4) 5 (15.2)
58 (58.0) 28 (28.0) 13 (13.0) 4 (4.0) 6 (6.0)
b0.01 0.98 b0.01 b0.01 0.20
32 (97.0) 4 (12.1) 30 (90.0) 25 (75.8) 31 (93.9)
96 (96.0) 98 (98.0) 60 (87.0) 58 (84.1) 65 (94.2)
0.78 b0.001 0.80 0.46 0.69
0.38 0.38 0.11 0.64 0.82
0.07 b0.01 0.24 0.95 0.91 0.85 0.15 0.21 0.99 0.43
Values in parentheses represent percent of total cases in which the relevant data were available unless otherwise stated. Standard deviation reported for continuous variables. ACE: angiotensin converting enzyme; ARB: angiotensin receptor blocker; BMI: body mass index; CABG: coronary artery bypass graft surgery; EF: ejection fraction; eGFR: estimated glomerular filtration rate; IQR: interquartile range; LV: left ventricular; NSTEACS: non-ST-segment elevation acute coronary syndrome; PCI: percutaneous coronary intervention. a As estimated by the Cockcroft–Gault equation. b Data available for 69 cases in PCI arm.
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
269
Table 2 Clinical endpoints in extremely obese patients undergoing elective coronary revascularization (unadjusted).
Length of follow-up, months (IQR) Primary endpoint Death Myocardial infarction Stroke Deep surgical site infection Vascular access site complication TIMI major/CABG-related bleed Blood transfusion CV composite of death, MI, and stroke Length of stay in hospital, days (IQR) Target vessel revascularization
Total cohort (n = 133)
CABG (n = 33)
PCI (n = 100)
p-Value⁎
22 (3–33) 13 (9.8) 2 (1.5) 5 (3.8) 0 (0)
20 (3–38) 8 (24.2) 2 (6.1) 1 (3.0) 0 (0) 4 (12.1) – 3 (9.1) 8 (24.2) 3 (9.1) 11 (8–22) 1 (3.0)
23 (3–32) 5 (5.0) 0 (0) 4 (4.0) 0 (0) – 0 (0) 1 (1.0) 5 (5.0) 4 (4.0) 1 (0–2) 7 (7.0)
0.76 b0.01 0.10 0.78 –
4 (3.0) 13 (9.8) 7 (5.3) 2 (0–7) 8 (6.0)
0.08 b0.01 0.49 b0.001 0.68
CABG: coronary artery bypass surgery; CV: cardiovascular; IQR: interquartile range; PCI: percutaneous coronary intervention; TIMI: Thrombolysis In Myocardial Infarction. ⁎ p-Value calculated for CABG vs. PCI groups.
p b 0.001). No difference was observed in the rates of TVR between groups (p = 0.68) (Table 2). 4. Discussion The prevalence of obesity, particularly class III obesity, continues to rise — a change in demographic that is shifting the physical and metabolic profile of patients undergoing coronary revascularization [18]. Accordingly, it is important to characterize clinical outcomes in this understudied population. Our study is the first to exclusively focus on elective coronary revascularization in this group. Our evaluation showed an incidence of overall complications of 9.8% among EO patients undergoing revascularization and suggested an incremental risk among those with higher BMI. Furthermore, patients undergoing CABG appeared to have an increased risk of complications compared to those undergoing PCI. Previous studies examining the impact of obesity in the setting of stable CAD, MI, and/or coronary revascularization have largely compared patients with obesity to their non-obese counterparts [2,19,20]. Data from such studies support the existence of a ‘paradoxical’ protective effect among mildly obese cardiac patients. However, these initial
studies lacked focus on EO patients (with BMI ≥ 40). Emerging evidence suggests that this subset is exempt from this protection and may represent a higher-risk population for coronary revascularization [18], in part presumably because the metabolic and physiologic strain conferred by greater weight burdens predispose to peri- and postprocedural complications. Consistent with these findings, our study – comprising contemporary revascularization techniques and outcome data to one year – found high complication rates among this population and suggests that even within this group increasing BMI further augments peri-procedural risk. Interestingly, we additionally observed less procedural morbidity with PCI compared to CABG. Although not randomized, adjustment by baseline covariates did not alter these results. EO has been associated with higher rates of major adverse clinical events both in the setting of CABG [4] and PCI [18,22], including primary PCI for STEMI [21,23], the cause of which is likely multifactorial. In the setting of PCI, EO patients have been shown to have higher rates of VASC [24]; image acquisition during PCI may be suboptimal rendering the procedure technically more difficult; and previous studies have suggested BMI as a risk factor for clopidogrel unresponsiveness [25], which may place patients at risk for thrombotic complications after stenting. Similarly, EO
Fig. 2. Event-free survival curves for the primary endpoint in extremely obese patients undergoing elective coronary revascularization stratified by median of body mass index (BMI).
270
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
patients may face multiple potential risks when undergoing CABG. Technical aspects, such as graft harvesting and implantation as well as control of haemostasis, may be more challenging because of body habitus. Post-operatively, prolonged need for ventilatory support and LOS has been reported [26] while impaired healing may increase the risk for wound infections. In our study, the increased overall complication rate among patients undergoing CABG seemed to be driven by wound infections. The incidence of deep sternal SSIs in our study is higher than in previous reports; however, direct comparisons are limited by differences in study designs and baseline patient characteristics. Moreover, definitions for
SSIs vary considerably among studies. Two registry-based studies found rates of deep sternal infections of 1.4% and 1.8% in EO patients following CABG [4,6]; however, both studies were restricted to inhospital infections and the studies failed to define how sternal SSIs were diagnosed. Similarly, Birkmeyer et al. reported a sternal SSI rate of only 2–3% in a cohort of patients undergoing CABG who were identified as being “severely obese”. However, the study's criteria for diagnosing sternal SSIs differed from the CDC criteria used in our study. Furthermore, “severe obesity” was defined as a BMI N36, only 45.7% of patients in their cohort were diabetic, and only in-hospital outcomes were assessed [27]. The importance of accounting for complications
Fig. 3. Event-free survival curves for the primary (A) and secondary (B) cardiovascular composite endpoints following elective coronary artery bypass surgery (CABG) versus percutaneous coronary intervention (PCI) in patients with extreme obesity.
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
arising after discharge from hospital is underscored by reports that 13.8% of CABG patients require re-hospitalization within 30 days of discharge of which nearly 1 in 5 are for wound infections [27]. Our study also suggested that EO patients who underwent CABG were at increased risk of severe bleeding compared to PCI while PCI patients had a low rate of VASC requiring intervention. We previously reported a low incidence of VASC in this population [7] and a recent large retrospective study similarly found a VASC rate of 2.75% in EO patients undergoing PCI [18]. Higher rates of radial access at our centre as well as the use of bivalirudin (previously reported to be 36% and 17%, respectively, in this patient group) may in part explain the low rate of bleeding events seen in our PCI cohort compared to other reports in the literature [7,17,21]. Additionally, as has been postulated by others, underdosing of anticoagulants (particularly low molecular weight heparins) in this patient population due to uncertainty in optimal dosing at extremes of BMI may result in a favourable bleeding risk profile [21]. Although ischemic complications, as defined by our secondary endpoint, were not statistically increased among patients revascularized with CABG compared to PCI, our study is likely underpowered to definitively ascertain this observation. Overall, these observations raise two important questions for the management of EO patients requiring revascularization: 1) could interventions aimed at weight reduction before coronary revascularization reduce the risk of procedural morbidity, and 2) should less invasive alternatives be preferred in patients needing revascularization? The notion that pre-procedural weight reduction at such extremes of BMI may improve peri- and post-procedural outcomes is an area of considerable research and debate. Pre-operative weight loss has been associated with reduced operative times, operative complications, and LOS — benefits observed with weight reductions of as little as 5% in studies of bariatric surgery patients, for instance [28,29]. Indeed, while still debated [31], such findings have led to institutes and organizations recommending specific weight loss goals before consideration for bariatric surgery [30,31]. The question of whether such an approach could be considered before elective coronary revascularization – for instance, using a low-calorie diet incorporating meal replacements – remains to be determined [32]. The question of whether less invasive alternatives should be preferred in EO patients similarly remains unanswered. To date, despite advances in stent designs and marked improvements in associated clinical outcomes [33] as well as perceived increased operative risk with EO as suggested by the underrepresentation of obese patients with severe ischemic heart disease in CABG registries [34], there is little data to support PCI as the predominant strategy in this population. Indeed, the applicability of studies comparing surgical to percutaneous revascularization to EO patients is questionable given the relatively small number EO patients represented in the study populations. The use of PCI over CABG may afford the benefits of a less invasive approach thereby reducing infections and bleeding; however, this would likely come at the cost of less complete revascularization. Ultimately, larger studies comparing the efficacy of percutaneous to surgical revascularization in EO patients are needed. 4.1. Study limitations Our study has several limitations. The retrospective, non-randomized design of the study may have allowed unaccounted factors to influence the differences between PCI and CABG. However, we adjusted for key covariates that may affect revascularization outcomes in an attempt to account for differences among the two groups. Though coronary anatomy was assessed, detailed assessment of disease complexity was not performed. Our study was also relatively small despite the large volume of procedures performed at our institute as patients with EO accounted for only 3.7% of all angiograms (consistent with the 3.1% prevalence of class III obesity reported in the 2007 to 2009 Canadian Health Measures Survey) [35]. Nevertheless, our study is the largest to date to focus on the association between increasing BMI and clinical outcomes in the EO
271
population following elective coronary revascularization and highlights the need for further research into harm-reduction strategies in these high-risk patients. 5. Conclusions Within the EO population increasing BMI is associated with an increased risk of adverse clinical outcomes following elective coronary revascularization. The risk of SSI, in particular, is an important contributor to the morbidity associated with surgical approaches in this population. Modifiable factors influencing outcomes in these high-risk patients and the potential role for interventions aimed at weight reduction prior to revascularization remain to be fully elucidated. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Acknowledgements None. References [1] A. Romero-Corral, V.M. Montori, V.K. Somers, et al., Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies, Lancet 368 (2006) 666–678. [2] L. Mehta, W. Devlin, P.A. McCullough, et al., Impact of body mass index on outcomes after percutaneous coronary intervention in patients with acute myocardial infarction, Am. J. Cardiol. 99 (2007) 906–910. [3] M. Schmiegelow, C. Torp-Pedersen, G.H. Gislason, et al., Relation of body mass index to risk of stent thrombosis after percutaneous coronary intervention, Am. J. Cardiol. 110 (2012) 1592–1597. [4] G. Prabhakar, C.K. Haan, E.D. Peterson, L.P. Coombs, J.L. Cruzzavala, G.F. Murray, The risks of moderate and extreme obesity for coronary artery bypass grafting outcomes: a study from the Society of Thoracic Surgery's database, Ann. Thorac. Surg. 74 (2002) 1125–1130. [5] V.G. Fowler Jr., S.M. O'Brien, L.H. Muhlbaier, G.R. Corey, T.B. Ferguson, E.D. Peterson, Clinical predictors of major infections after cardiac surgery, Circulation 112 (2005) I358–I365. [6] R. Jin, G.L. Grunkemeier, A.P. Furnary, J.R. Handy Jr., Is obesity a risk factor for mortality in coronary artery bypass surgery? Circulation 111 (2005) 3359–3365. [7] B. Hibbert, T. Simard, K.R. Wilson, et al., Transradial versus transfemoral artery approach for coronary angiography and percutaneous coronary intervention in the extremely obese, JACC Cardiovasc. Interv. 5 (2012) 819–826. [8] W.H.O. Expert Consultation, Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies, Lancet 363 (2004) 157–163. [9] R. Sturm, Increases in clinically severe obesity in the United States, 1986–2000, Arch. Intern. Med. 163 (2003) 2146–2148. [10] R. Sturm, Increases in morbid obesity in the USA: 2000–2005, Public Health 121 (2007) 492–496. [11] P.T. Katzmarzyk, C. Mason, Prevalence of class I, II and III obesity in Canada, CMAJ 174 (2006) 156–157. [12] N.R.C. Campbell, J. Onysko, H. Johansen, R.N. Gao, Changes in cardiovascular deaths and hospitalization in Canada, Can. J. Cardiol. 22 (2006) 425–427. [13] M.R. Le May, D.Y. So, R. Dionne, et al., A citywide protocol for primary PCI in STsegment elevation in myocardial infarction, N. Engl. J. Med. 358 (2008) 231–240. [14] K. Thygesen, J.S. Alpert, A.S. Jaffe, M.L. Simoons, B.R. Chaitman, H.D. White, Third universal definition of myocardial infarction, Circulation 126 (2012) 2020–2035. [15] A.J. Mangram, T.C. Horan, M.L. Pearson, L.C. Silver, W.R. Jarvis, Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee, Infect. Control Hosp. Epidemiol. 20 (1999) 250–278. [16] R. Mehran, S.V. Rao, D.L. Bhatt, et al., Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the bleeding academic research consortium, Circulation 123 (2011) 2736–2747. [17] B. Hibbert, A. MacDougall, M. Labinaz, et al., Bivalirudin for primary percutaneous coronary interventions: outcome assessment in the Ottawa STEMI registry, Circ. Cardiovasc. Interv. 5 (2012) 805–812. [18] M.E. Buschur, D. Smith, D. Share, et al., The burgeoning epidemic of morbid obesity in patients undergoing percutaneous coronary intervention. Insight from the Blue Cross Shield of Michigan Cardiovascular Consortium, J. Am. Coll. Cardiol. 62 (2013) 685–691. [19] S.G. Ellis, J. Elliott, M. Horrigan, R.E. Raymond, G. Howell, Low-normal or excessive body mass index: newly identified and powerful risk factors for death and other complications with percutaneous coronary intervention, Am. J. Cardiol. 78 (1996) 642–646.
272
F.D. Ramirez et al. / International Journal of Cardiology 186 (2015) 266–272
[20] A. Bechlioulis, K. Vakalis, K.K. Naka, et al., Paradoxical protective effect of central obesity in patients with suspected stable coronary artery disease, Obesity 21 (2013) E314–E321. [21] S. Payvar, S. Kim, S.V. Rao, et al., In-hospital outcomes of percutaneous coronary interventions in extremely obese and normal-weight patients: findings from the NCDR (National Cardiovascular Data Registry), J. Am. Coll. Cardiol. 62 (2013) 692–696. [22] R.M. Minutello, E.T. Chou, M.K. Hong, et al., Impact of body mass index on inhospital outcomes following percutaneous coronary intervention (report from the New York State Angioplasty Registry), Am. J. Cardiol. 93 (2004) 1229–1232. [23] S.R. Das, K.P. Alexander, A.Y. Chen, et al., Impact of body weight and extreme obesity on the presentation, treatment, and in-hospital outcomes of 50,149 patients with ST-segment elevation myocardial infarction: results from the NCDR (National Cardiovascular Data Registry), J. Am. Coll. Cardiol. 58 (2011) 2642–2650. [24] N. Cox, F.S. Resnic, J.J. Popma, D.I. Simon, A.C. Eisenhauer, C.R. Rogers, Comparison of the risk of vascular complications associated with femoral and radial access coronary catheterization procedures in obese versus nonobese patients, Am. J. Cardiol. 94 (2004) 1174–1177. [25] D.J. Angiolillo, A. Fernandez-Ortiz, E. Bernardo, et al., Variability in individual responsiveness to clopidogrel: clinical implications, management and future perspectives, J. Am. Coll. Cardiol. 49 (2007) 1505–1516. [26] C.H. Wigfield, J.D. Lindsey, A. Muñoz, P.S. Chopra, N.M. Edwards, R.B. Love, Is extreme obesity a risk factor for cardiac surgery? An analysis of patients with a BMI ≥40, Eur. J. Cardiothorac. Surg. 29 (2006) 434–440.
[27] N.J.O. Birkmeyer, D.C. Charlesworth, F. Hernandez, et al., Obesity and risk of adverse outcomes associated with coronary artery bypass surgery, Circulation 97 (1998) 1689–1694. [28] S. Cassie, C. Menezes, D.W. Birch, X. Shi, S. Karmali, Effect of preoperative weight loss in bariatric surgical patients: a systematic review, Surg. Obes. Relat. Dis. 7 (2011) 760–767. [29] C.D. Still, P. Benotti, G.C. Wood, et al., Outcomes of preoperative weight loss in highrisk patients undergoing gastric bypass surgery, Arch. Surg. 142 (2007) 994–998. [30] R.S. Alami, J.M. Morton, R. Schuster, et al., Is there a benefit to preoperative weight loss in gastric bypass patients? A prospective randomized trial, Surg. Obes. Relat. Dis. 3 (2007) 141–146. [31] M.R. Ali, S. Baucom-Pro, G.A. Broderick-Villa, et al., Weight loss before gastric bypass: feasibility and effect on postoperative weight loss and weight loss maintenance, Surg. Obes. Relat. Dis. 3 (2007) 515–520. [32] M.M. Elahi, G.K. Chetty, A.W. Sosnowski, M.S. Hickey, T.J. Spyt, Morbid obesity increases perioperative morbidity in first-time CABG patients — should resources be redirected to weight reduction, Int. J. Cardiol. 105 (2005) 98–99. [33] T. Simard, B. Hibbert, F.D. Ramirez, M. Froeschl, Y.X. Chen, E.R. O'Brien, The evolution of coronary stents: a brief review, Can. J. Cardiol. 30 (2014) 35–45. [34] B.C. Reeves, R. Ascione, M.H. Chamberlain, G.D. Angelini, Effect of body mass index on early outcomes in patients undergoing coronary artery bypass surgery, J. Am. Coll. Cardiol. 42 (2003) 668–676. [35] M. Shields, M.D. Carroll, C.L. Ogden, Adult obesity prevalence in Canada and the United States, NCHS Data Brief, no. 56, National Center for Health Statistics, Hyatsville, 2011.
