Impact of body mass index on outcomes after primary angioplasty in acute myocardial infarction

Impact of body mass index on outcomes after primary angioplasty in acute myocardial infarction Eugenia Nikolsky, MD, PhD,a,b Gregg W. Stone, MD,a,b Cindy L. Grines, MD,c David A. Cox, MD,d Eulogio Garcia, MD,e James E. Tcheng, MD,f John J. Griffin, MD,g Giulio Guagliumi, MD,h Thomas Stuckey, MD,i Mark Turco, MD,j Manuela Negoita, MD,a,b Alexandra J. Lansky, MD,a,b and Roxana Mehran, MDa,b New York, NY; Royal Oak, MI; Charlotte, Durham, and Greensboro, NC; Madrid, Spain; Virginia Beach, VA; Bergamo, Italy; and Tacoma Park, MD

Background The prognostic importance of obesity after primary percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI) is unknown. We therefore sought to investigate the impact of body mass index (BMI) in patients with AMI undergoing primary PCI. Methods

In the CADILLAC trial, 2082 patients of any age with AMI within 12 hours onset undergoing primary PCI were randomized to balloon angioplasty versus stenting, each Fabciximab. Outcomes were stratified by baseline BMI.

Results Baseline BMI was measured in 2035 (98%) randomized patients; 552 (27%) patients have normal weight (BMI b25 kg/m2), 915 (45%) were overweight (z25 to b30 kg/m2), and 568 (28%) were obese (z30 kg/m2). Compared with normal-weight patients, obese patients were younger and more frequently had diabetes, hyperlipidemia, hypertension, non–anterior myocardial infarction, and higher creatinine clearance. Obese patients were less likely to develop thrombocytopenia (1.8% vs 4.2%), moderate hemorrhagic complications (1.4% vs 3.3%), or required blood product transfusions (3.2% vs 6.3%) (all P V .04). Obese compared with normal-weight patients had lower inhospital mortality (0.9% vs 2.7%, P = .03) at 30 days (1.1% vs 3.8%, P = .02) and 1 year (1.8% vs 7.5%, P b .0001). Independent predictors of 30-day and 1-year mortality included lower ejection fraction, advanced age, 3-vessel disease, anterior AMI, and lower creatinine clearance, but not BMI. Conclusions Obese patients with AMI have an improved prognosis after primary PCI compared with normal-weight patients, a finding attributable to AMI onset at younger age, with better renal function and less anterior infarction. (Am Heart J 2006;151:168-75.) Epidemiologic studies provide evidence that obesity is associated with increased rates of myocardial infarction (MI) and death from cardiovascular diseases.1,2 However, several large clinical trials demonstrated that obese patients have a better prognosis than patients with normal weight after either percutaneous or surgical revascularization.3-7 Data on the impact of body mass index (BMI) on the outcome of patients with acute myocardial infarction

From the aColumbia University Medical Center, New York, NY, bCardiovascular Research Foundation, New York, NY, cWilliam Beaumont Hospital, Royal Oak, MI, d Mid Carolina Cardiology, Charlotte, NC, eHospital Gregorio Maranon, Madrid, Spain, fDuke Clinical Research Institute, Durham, NC, gVirginia Beach General Hospital, Virginia Beach, VA, hOspedali Riuniti di Bergamo, Bergamo, Italy, iMoses Cone Memorial Hospital, Greensboro, NC, and jWashington Adventist Hospital, Tacoma Park, MD. Submitted June 14, 2004; accepted March 15, 2005. Reprint requests: Gregg W. Stone, MD, Cardiovascular Research Foundation, 55 East 59th Street, 6th Floor, New York, NY 10022. E-mail: [email protected] 0002-8703/$ - see front matter n 2005, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2005.03.024

(AMI) are limited. In a cohort of 1760 patients with AMI studied before the advent of reperfusion therapy, 1-year mortality was not related to body weight.8 Among 41 021 patients enrolled in the GUSTO-I randomized trial of 4 thrombolytic strategies, 30-day mortality was weakly associated with both weight and height, with lighter and shorter patients exhibiting a higher risk.9 Higher BMI was also predictive of improvement in left ventricular function between 90 minutes and 5 to 7 days.10 No data are available on the impact of obesity on the outcomes of patients with AMI undergoing primary percutaneous coronary intervention (PCI). We therefore examined the database from the large, multicenter randomized CADILLAC trial to determine the effect of BMI on early and late prognosis of patients with AMI treated with primary PCI.

Methods Full details of the CADILLAC trial methodology have been previously reported.11 Briefly, 2082 patients of any age with AMI within 12 hours onset undergoing primary PCI were

American Heart Journal Volume 151, Number 1

Nikolsky et al 169

Table I. Baseline clinical characteristics and angiographic features BMI (kg/m2)

Clinical characteristics BMI (kg/m2) Age (y) Male sex Diabetes mellitus Hypertension Hyperlipidemia Current smoking Prior MI Prior PCI Prior stroke or transient ischemic attack History of peripheral vascular disease Killip class 2 or 3 Baseline hemoglobin level (g/dL) Baseline creatinine clearance (mL/min) Symptom onset to balloon inflation (h) Angiographic features Single-vessel disease Double-vessel disease Triple-vessel disease Left ventricle ejection fraction (%) Infarct related artery Left anterior descending artery Left circumflex artery Right coronary artery

bbbb 25.0, n = 552

_z _ 25.0 to bbbb 30.0, z z n = 915

23.1 (21.8-24.2) 65 (54-73) 66.7% 11.1% 41.8% 32.2% 44.2% 11.4% 9.6% 2.9% 4.0% 11.1% 14.0 (13.2-15.3) 66.5 (51.7-88.9) 3.9 (2.9-6.1)

27.2 (26.0-28.4) 59 (51-67) 77.8% 15.1% 47.7% 40.4% 44.0% 13.2% 11.1% 3.9% 2.6% 12.0% 14.8 (13.8-15.7) 88.5 (70.1-105.7) 3.9 (2.8-5.7)

51.1% 31.7% 17.2% 57 (48-64) 41.5% 16.1% 42.0%

51.7% 33.9% 14.4% 58 (50-66) 35.4% 16.3% 47.9%

P1

_z _ 30.0, n = 568 z z

P2

P3

b.0001 b.0001 b.0001 .03 .03 .002 .95 .32 .35 .44 .15 .60 b.0001 b.0001 .31

32.9 (31.2-35.9) 56 (49 to 64) 71.7% 24.3% 54.8% 41.5% 41.0% 17.1% 13.2% 1.2% 1.6% 9.0% 14.9 (13.9-15.8) 112.1 (86.2-140.1) 4.1 (2.9-6.9)

b.0001 b.0001 .07 b.0001 b.0001 .001 .28 .007 .06 .05 .01 .25 b.0001 b.0001 .15

b.0001 .0002 .007 b.0001 .008 .67 .25 .04 .23 .003 .19 .07 .62 b.0001 .008

.82 .39 .35 .07

50.5% 33.5% 16.0% 58 (48-65)

.85 .53 .59 .18

.66 .86 .40 .72

.02 .93 .03

34.9% 20.1% 44.7%

.02 .09 .36

.83 .06 .31

P 1, P 2, and P 3 denote P values for the comparisons between BMI b25.0 versus z25.0 to b30.0 (P 1); b25.0 versus z30.0 (P 2); and z25.0 to b30.0 versus z30.0 kg/m2 (P 3).

randomized in a 2  2 factorial design to 1 of 4 mechanical reperfusion strategies: balloon angioplasty F abciximab versus stenting with the Multilink stent F abciximab. Major exclusion criteria included cardiogenic shock, history of gastrointestinal or genitourinary bleeding within 6 months, cerebrovascular event within 2 years or any permanent residual neurologic defect, history of thrombocytopenia, leucopenia, hepatic or renal dysfunction, current thrombolytic administration, and noncardiac illness with life expectancy b1 year. All patients received 324 mg of chewable aspirin, ticlopidine 500 mg or clopidogrel 300 mg, a 5000-U heparin bolus, and intravenous h-blockade in the absence of contraindications before catheterization. Abciximab was administered as a bolus of 0.25 mg/kg, followed by a 12-hour infusion at 0.125 Ag/kg per minute (10 Ag/min maximum). Heparin dosing was guided by a nomogram to achieve an activated clotting time z350 seconds in the absence of abciximab and 200 to 300 seconds if randomized to abciximab. After PCI, medical therapy included aspirin 325 mg daily, as well as h-blockers and angiotensin-converting enzyme inhibitors if not contraindicated. Patients receiving stents were treated with ticlopidine or clopidogrel for 4 weeks. Clinical follow-up was performed at 1, 6, and 12 months, and angiographic follow-up was carried out at 7 months in a prespecified cohort of 656 patients.11 All baseline and follow-up films were analyzed at an independent angiographic core laboratory blinded to clinical events.

The primary composite end point included death from any cause, reinfarction (recurrent ischemic symptoms or electrocardiographic changes accompanied by a creatine kinase level more than twice the upper limit of the normal range or N50% higher than the previous value), target vessel revascularization (TVR) as a result of ischemia if there was evidence of ischemia during functional testing or of angina, and disabling stroke (acute neurologic deficit that lasted N24 hours and affected the ability to perform daily activities or resulted in death), as previously defined.11 Body mass index was calculated at base line by the patient’s measured weight in kilograms divided by his/her reported height in meters.2 By standard convention, patients were divided into 3 groups based on the admission BMI: normal weight (b25 kg/m2), overweight (z25 to b30 kg/m2), and obese (z30 kg/m2).12 Estimated creatinine clearance (CrCl) was calculated by the use of the Cockcroft-Gault formula: CrCl (mL/min) = [(140 age)  weight (kg)]/[SCr (mg/dL)  72], corrected in women by a factor of 0.85.13 Moderate bleeding was defined as one requiring transfusion, and severe bleeding was defined as one causing hemodynamic compromise or hemorrhagic stroke.

Statistical analysis Categorical variables were compared using the Fisher exact test for 2-way comparisons and m2 statistics for 3-way comparisons. Continuous variables are presented as medians with interquartile ranges and were compared using the

American Heart Journal January 2006

170 Nikolsky et al

Table II. Procedural results and inhospital outcomes BMI (kg/m2)

Final reference diameter (mm) Final minimal luminal diameter (mm) Final diameter stenosis (%) Final TIMI flow Grade 0 or 1 Grade 2 Grade 3 Any stent implanted Abciximab administered Contrast media volume (mL) Procedure duration (min) Peak activated clotting time Bleeding Moderate Severe Blood product transfusions Thrombocytopenia Mild Severe Profound

bbbb 25.0, n = 552

_z _ 25 to bbbb 30.0, n = 915 z z

P1

_z _ 30.0, n = 568 z z

P2

P3

2.89 (2.57-3.24) 2.40 (2.03-2.75) 17.7 (8.0-28.3)

3.01 (2.64-3.40) 2.48 (2.11-2.83) 17.8 (9.1-26.8)

.0002 .008 .91

3.05 (2.70-3.44) 2.50 (2.12-2.85) 18.7 (10.0-26.9)

b.0001 .003 .21

.04 .51 .19

1.8% 2.4% 95.8% 49.5% 51.8% 289 (210-375) 63 (49-84) 321 (259-385)

0.7% 3.0% 96.3% 49.0% 54.3% 280 (200-370) 60 (45-80) 316 (255-381)

.04 .50 .58 .95 .35 .62 .02 .51

2.1% 3.4% 94.5% 51.1% 51.8% 300 (225-383) 61 (49-82) 310 (250-370)

.73 .33 .33 .68 .98 .05 .57 .03

.01 .68 .09 .69 .33 .007 .06 .08

3.3% 0.4% 6.3%

2.6% 0.3% 4.2%

.48 .91 .06

1.4% 0.4% 3.2%

.04 .98 .01

.12 .94 .33

4.2% 1.1% 0.5%

4.2% 1.1% 0.1%

.99 .99 .12

1.8% 0.0% 0.0%

.02 .01 .08

.01 .01 .43

P values as in Table I.

nonparametric Kruskal-Wallis test. Survival was estimated by the Kaplan-Meier method and compared by log-rank test. Multivariate analysis of predictors of death and stroke at several periods were performed using Cox proportional hazards regression, and multivariate regression analysis of predictors of restenosis was performed using logistic regression with stepwise selection and entry and exit criteria of P b .1. The candidate variables entered in the model included age, sex, diabetes mellitus, hypertension, hypercholesterolemia, current smoking, history of prior MI or bypass graft surgery, Killip class 2 or 3, creatinine clearance, left anterior descending artery infarct vessel, triple-vessel disease, randomization to abciximab or stent, time from the symptom onset to first balloon inflation, reference vessel diameter, ejection fraction, and BMI value (separately as a categorical and as a continuous variable). All analyses were 2-sided, and significance was established at the .05 level.

Results Baseline BMI was measured in 2035 (98%) randomized patients; 552 (27%) patients have normal weight, 915 (45%) were overweight, and 568 (28%) were obese. As seen in Table I, overweight compared with normal-weight patients were younger, less frequently female, and more likely to have diabetes, hyperlipidemia, hypertension, and nonanterior infarction. The same differences existed between normal-weight and obese patients except for sex. Compared with overweight patients, obese patients were younger still, more frequently female, more commonly had diabetes and hypertension, and had longer reperfusion times. Estimated creatinine clearance and baseline

hemoglobin level also differed significantly as a function of BMI, being highest in obese and lowest in normal-weight patients.

Procedural and inhospital results Post-PCI reference vessel diameter was significantly greater in obese compared with nonobese patients (Table II), whereas left ventricular ejection fraction was independent of BMI (Table I). Obese compared with normal-weight patients had slightly lower peak activated clotting time values but a higher volume of contrast media administered. Procedural success rates and final thrombolysis in myocardial infarction (TIMI) flow grades were independent of BMI (91.7%, 92.2%, and 90.1% in normal-weight, overweight, and obese patients, respectively), as were peak creatine kinase levels (median 1467 vs 1461 vs 1264 U/L, respectively). Inhospital mortality was significantly lower ( P = .03) in obese (0.9%) compared with normal-weight (2.7%) and overweight (1.3%) patients. Compared with normalweight patients, obese patients less frequently developed moderate inhospital hemorrhagic complications and less commonly required blood product transfusions. Analysis of the bleeding source demonstrated that gastrointestinal bleeding was significantly less frequent in obese (0.7%) patients compared with normal-weight (2.4%, P = .02) and overweight (2.0%, P = .05) patients. Bleeding from other sources was not related to BMI. Mild and severe thrombocytopenia (baseline nadir platelet count b100  109/L and b50  109/L, respectively) developed significantly less frequently in

American Heart Journal Volume 151, Number 1

Figure 1

Nikolsky et al 171

Figure 3

Thirty-day and 1-year mortality stratified by sex and BMI.

Table III. Multivariate predictors of mortalityT Hazard ratio (95% CI)

Thirty-day (A) and 1-year (B) outcomes stratified by BMI. P 1 and P 2 denote P values for the comparisons between BMI b 25.0 versus z25.0 to b 30.0 ( P 1), and b 25.0 versus z30.0 ( P 2). Composite adverse events include death from any cause, reinfarction, ischemiadriven TVR, or disabling stroke.

Figure 2

30-d mortality Ejection fraction Creatinine clearance Left anterior descending infarct artery Postprocedure TIMI grade flow 0-1 Hypertension BMI 1-y mortality Ejection fraction Older age Triple-vessel disease Left anterior descending infarct artery BMI

P

0.91 0.96 3.88 9.51 3.16 0.91

(0.88-0.94) (0.94-0.99) (1.16-13.03) (1.11-81.51) (1.07-9.38) (0.80-1.04)

b.0001 .002 .03 .04 .04 .18

0.94 1.06 2.48 2.48 0.95

(0.92-0.96) (1.03-1.09) (1.30-4.71) (1.24-4.99) (0.88-1.03)

b.0001 .0001 .006 .01 .19

TIn order of impact on 30-day and 1-year mortality.

Kaplan-Meier survival curves after primary angioplasty in AMI, stratified by admission BMI.

obese compared with normal-weight and overweight patients; however, there was no significant difference in the rate of profound thrombocytopenia (b20  109/L) among the 3 groups.

Clinical follow-up As seen in Figures 1 and 2, mortality rates at 30 days and 1 year were significantly higher in normal-weight patients compared with both overweight and obese patients. Lower mortality with increasing BMI was observed in both sexes and was especially prominent in women, although statistical significance was reached only in males (Figure 3). Independent predictors of both 30-day and 1-year mortality included lower ejection fraction, older age, 3-vessel disease, and left anterior descending coronary artery infarct vessel, but not BMI neither as a categorical nor as a continuous variable (Table III). The 1-year rate of disabling stroke was greater in normal-weight patients compared with overweight or obese patients (Figure 1). The rate of any stroke was also highest in normal-weight compared with overweight and obese patients. By multivariate analysis, BMI

American Heart Journal January 2006

172 Nikolsky et al

Table IV. Angiographic follow-up at 7 months BMI (kg/m2)

Reference diameter pre (mm) Minimal luminal diameter pre (mm) Minimal luminal diameter post (mm) Minimal luminal diameter at follow-up (mm) Diameter stenosis pre (%) Diameter stenosis post (%) Diameter stenosis at follow-up (%) In-lesion/in-stent restenosis (diameter stenosis N50%) Severe restenosis (diameter stenosis N70%) In-stent restenosis (diameter stenosis N50%) Infarct artery reocclusion

bbbb 25.0, n = 181

_z _ 25.0 to bbbb 30.0, z z n = 272

P1

_z _ 30.0, z z n = 169

P2

P3

2.82 (2.52-3.25) 0 (0.00-0.67) 2.34 (1.97-2.74) 1.86 (1.17-2.18) 100.0 (76.1-100.0) 19.4 (8.7-28.1) 34.0 (24.4-51.2) 26.6% 10.4% 10.5% 8.2%

2.99 (2.62-3.32) 0.00 (0.00-0.67) 2.47 (2.06-2.78) 1.74 (1.29-2.29) 100.0 (76.7-100.0) 19.2 (9.97-27.9) 35.6 (23.5-51.9) 28.0% 11.7% 15.3% 10.0%

.002 .87 .04 .79 .89 .74 .72 .74 .68 .29 .53

3.08 (2.73-3.42) 0.00 (0.00-0.80) 2.49 (2.15-2.85) 1.90 (1.43-2.26) 100.0 (73.6-100.0) 19.6 (10.5-27.0) 35.7(24.7-54.7) 31.3% 10.2% 28.2% 7.0%

b.0001 .31 .002 .15 .24 .57 .57 .33 .96 .002 .65

.01 .22 .17 .23 .28 .71 .82 .46 .65 .01 .28

P values as in Table I.

Table V. Selected outcomes in patients randomized to balloon angioplasty versus stenting as a function of BMI BMI (kg/m2) _z _ 25 to bbbb 30.0, n = 915 z z

bbbb 25.0, n = 552

Peak activated clotting time Moderate or severe bleeding Blood transfusion Thrombocytopenia Mortality 30-d 1-y Disabling stroke 30-d 1-y Reinfarction 30-d 1-y TVR 30-d 1-y Composite adverse events 30-d 1-y Restenosis Infarct artery reocclusion

Balloon angioplasty

Stent

P

315 (255-392) 3.6%

322 (265-377) 3.7%

.68

7.2% 2.5%

_z _ 30.0, n = 568 z z

Balloon angioplasty

Stent

P

314 (250-383) 3.1%

.45

1.00

318 (260-379) 2.8%

5.1% 5.9%

.38 .06

3.4% 4.7%

3.4% 3.6%

1.00 .41

2.2% 1.1%

3.1% 2.4%

.60 .34

2.9% 6.9%

5.2% 8.5%

.17 .45

1.5% 4.4%

1.6% 2.7%

.94 .19

1.1% 1.5%

1.0% 2.5%

.96 .40

0.0% 0.8%

0.4% 2.0%

.31 .23

0.0% 0.0%

0.0% 0.2%

– .31

0.4% 0.7%

0.0% 0.0%

.31 .15

1.5% 3.0%

0.0% 0.8%

.05 .06

0.7% 2.2%

1.4% 2.5%

.28 .75

0.4% 2.6%

1.0% 2.8%

.33 .86

5.8% 21.9%

0.4% 7.6%

.0003 b.0001

4.7% 21.0%

2.9% 12.8%

.15 .0009

4.3% 22.7%

3.8% 11.7%

.76 .0008

8.3% 22.8% 43.8% 10.1%

5.5% 14.8% 19.6% 6.5%

.19 .02 .0007 .43

6.2% 20.7% 41.2% 13.7%

4.5% 13.7% 22.1% 6.4%

.24 .004 .001 .07

5.4% 21.6% 38.8% 9.4%

5.2% 13.0% 25.3% 4.6%

.91 .008 .07 .24

b25.0 kg/m2 was independently associated with increased risk of any stroke at 1 year (hazard ratio 3.99, 95% CI 1.49-10.68, P = .006). No relationship was observed between risk of any stroke at 1 year and BMI when the latter was introduced into the multivariate model as a continuous variable. There were no differences in the rates of reinfarction, TVR, or composite adverse events as a function of BMI.

.84

Balloon angioplasty

Stent

P

305 (246-371) 1.8%

321 (254-370) 1.7%

.34 1.00

Angiographic follow-up As seen in Table IV, the rate of binary angiographic restenosis was highest in obese patients, intermediate in overweight patients, and lowest in normal-weight patients. By multivariate analysis, higher BMI, as a continuous variable, was an independent predictor of in-stent restenosis (odds ratio 1.08, 95% CI 1.01-1.16, P = .02). Obesity was also a significant independent

American Heart Journal Volume 151, Number 1

predictor of in-stent restenosis in the multivariate model when BMI was entered as a categorical variable (odds ratio 3.85, 95% CI 1.28-11.11, P = .02). However, neither severe restenosis (follow-up diameter stenosis N70%) nor infarct artery reocclusion rates varied as a function of BMI.

Impact of randomized treatment modality Angiographic restenosis, 30-day and 1-year TVR rates, and the composite rate of major adverse cardiac events at 1 year were significantly lower in patients treated with stents compared with balloon angioplasty, regardless of BMI (Table V). Overweight patients randomized to abciximab compared with those treated with heparin alone had reduced rates of TVR (2.6% vs 5.1%, P = .04) and composite major adverse cardiac events (3.9% vs 6.9%, P = .04) at 30 days but not at 1 year. Clinical outcomes at 30 days and 1 year were otherwise unrelated to treatment with abciximab in either normalweight or obese patients. Medication use Consistent with the higher prevalence of hypertension and diabetes, obese compared with normal-weight and overweight patients were more likely to have been treated before admission with h-blockers (17.8% vs 10.9% and 15.2%, respectively, P = .02), angiotensin-converting enzyme inhibitors or receptor blockers (13.7% vs 6.2% and 8.6%, P b .0001), diuretics (8.3% vs 4.7% and 5.1%, P = .018), insulin (3.7% vs 1.4% and 1.7%), and oral hypoglycemic agents (9.2% vs 4.5% and 4.7%, P = .0005). At discharge, obese compared with normal-weight and overweight patients were more frequently treated with aspirin (97.3% vs 95.0% and 97.3%, P = .03), thienopyridines (71.8% vs 63.3% and 67.9%, P = .01), h-blockers (82.1% vs 74.7% and 79.3%, P = .01), statins (33.7% vs 22.9% and 31.7%, P = .0001), insulin (3.0% vs 0.4% and 1.1%, P = .0006), and oral hypoglycemic medications (5.5% vs 2.6% and 2.2%, P = .001).

Discussion The principal findings of this study, the first such investigation of the impact of obesity on the outcomes of mechanical reperfusion therapy in AMI, are that (1) by current standards, nearly three quarters of patients undergoing primary PCI for AMI are above ideal body weight; (2) increased BMI was paradoxically associated with lower mortality than in normal-weight patients; (3) obese patients also had paradoxically lower rates of hemorrhagic complications, thrombocytopenia, blood product transfusions , and stroke; (4) rates of TVR and reinfarction, as well as infarct artery restenosis/reocclusion after primary PCI, were independent of BMI, although binary angiographic restenosis rates were increased in obese patients; and (5) by reducing restenosis, treatment with stents enhanced long-term event-free survival, independent of BMI.

Nikolsky et al 173

Obesity is a frequent clinical condition in the industrialized world. It is estimated that more than the half of the adult population in the United States is either overweight or obese.14 Plaque rupture and acute coronary syndromes are more frequent in obese patients than in their normal-weight counterparts,1,2,15 consistent with the observation that overweight and obese patients in CADILLAC comprised 45% and 28% of the total study population, respectively. Nonetheless, the impact of body weight on the outcomes of patients undergoing primary PCI for AMI has not been previously reported. In the present study, despite a greater incidence of diabetes, hypertension, hyperlipidemia and prior MI, longer time from symptom onset to first balloon inflation, and similar procedural success rates, inhospital mortality rates, at 30 days and 1 year in both men and women, were paradoxically lower in obese compared with normal-weight patients. Inhospital, 30-day, and 1-year mortality rates in obese patients were just 0.9%, 1.1%, and 1.8%, respectively, strikingly lower than in normal-weight patients. However, obese patients were also, on average, 9 years younger than normal-weight patients, consistent with prior studies.4,5,16 In addition to younger age, obese patients also had fewer infarctions involving the left anterior descending coronary artery and relatively preserved renal function, both conditions known to affect prognosis after reperfusion therapy in AMI.9,17 By multivariate analysis, BMI itself was not an independent predictor of survival; the greater survival in the obese patient with AMI after primary PCI was attributable to the association of this condition with younger age, a higher incidence of inferior rather than anterior infarction, and better renal function. A recent study based on the New York State Coronary Angioplasty Reporting System also demonstrated lower inhospital mortality and major adverse cardiac events in obese compared with normal-weight patients.7 However, in this report, patients with extreme obesity (BMI N40 kg/m2) had significantly higher inhospital mortality compared with less obese patients (BMI 30-40 kg/m2).7 In the present study, no significant differences in mortality were observed between obese versus extremely obese patients, inhospital (1.0% vs 0%, P = 1.0), at 30 days (1.2% vs 0%, P = .46), or at 1-year follow-up (1.9% vs 2.2%, P = .93). The relatively favorable prognosis in the obese patient with AMI after primary PCI in the present study may also be due to follow-up being truncated at 1 year. Two population-based studies of 2541 and 2677 survivors of first MI found that higher BMI was associated with an increased risk of recurrent coronary events (reinfarction or cardiac death) during a 3-year follow-up period.18,19 In our study, both any stroke and disabling stroke were significantly less likely to occur during the 1-year follow-up period in obese compared with normalweight patients, again, likely a function of younger

American Heart Journal January 2006

174 Nikolsky et al

age. Prior studies have reported conflicting results regarding the association between BMI and stroke, demonstrating either a J- or U-shaped relationship or lack of correlation.20,21 Considering the differences in pharmacologic treatment in patients with different BMI, the better outcomes in obese patients may be explained in part by the higher rates of aspirin, thienopyridine, h-blocker, and statin use at discharge. Compared with normal-weight and overweight patients, obese patients were found to have lower rates of hemorrhagic complications, blood product transfusions, and thrombocytopenia. Similar results have been reported in patients undergoing PCI in the elective setting and in patients treated with coronary bypass surgery.3-6 Exact reasons for the lower propensity to hemorrhagic complications in obese compared with normal-weight patients are unknown. One of the possible explanations includes a lower level of procedural anticoagulation in obese patients as measured by activated clotting time in our study. This might be a result of obesity itself coupled with the higher incidence of diabetes mellitus among obese patients, conditions both associated with hypercoagulable states.22,23 Of note, lower rates of bleeding in obese compared with normal-weight patients were not related to abciximab. The 3 principal determinants of restenosis are diabetes, reference vessel diameter, and lesion length.24,25 In the present study, patients with greater BMI had a higher incidence of diabetes but also larger vessels, possibly explaining why the rates of reinfarction and TVR, as well as severe angiographic restenosis (diameter stenosis N70%) and infarct artery reocclusion, were independent of BMI. However, binary restenosis (diameter stenosis N50%) rates were increased in obese patients, possibly reflecting the overriding effect of diabetes.24

Limitations This post hoc retrospective analysis should be viewed as hypothesis-generating rather than definitive. Markedly obese patients might be less likely to be enrolled in a primary angioplasty trial, and thus, the extent to which these results are generalizable are unknown. The information on actual blood pressure on admission was not captured. The use of BMI as a measure of obesity may be insufficient to completely characterize the impact of body size. Data on waist circumference and waist-to-hip ratio were unavailable; these alternative criteria of obesity may be better markers for coronary artery disease than BMI.26 Finally, changes in BMI during follow-up were not collected, which, if significant, may impact late prognosis. Clinical implications From the present study, it is clear that the apparent bobesity paradox,Q that is, a more favorable prognosis

after primary PCI in obese compared with thin patients with AMI may be explained by the fact that nonobese patients presenting with AMI are older, have worse renal function, and more frequently have anterior infarction than their obese counterparts. The adverse metabolic consequences of obesity, including hypertension, hyperlipidemia, and diabetes, manifest in the occurrence of AMI 9 years earlier than in patients with ideal body weight, with a risk-adjusted survival similar to that seen in patients with ideal body weight. Thus, the high-risk profile of overweight and obese patients mandates aggressive targeting of coronary risk factors for the primary prevention of coronary artery disease.

References 1. Calle EE, Thun MJ, Petrelli JM, et al. Body-mass index and mortality in a prospective cohort of U.S. adults. N Engl J Med 1999;341: 1097 - 105. 2. Wilson PW, D’Agostino RB, Sullivan L, et al. Overweight and obesity as determinants of cardiovascular risk: the Framingham experience. Arch Intern Med 2002;162:1867 - 72. 3. Reeves BC, Ascione R, Chamberlain MH, et al. Effect of body mass index on early outcomes in patients undergoing coronary artery bypass surgery. J Am Coll Cardiol 2003;42:668 - 76. 4. Gruberg L, Weissman NJ, Waksman R, et al. The impact of obesity on the short-term and long-term outcomes after percutaneous coronary intervention: the obesity paradox? J Am Coll Cardiol 2002;39:578 - 84. 5. Gurm HS, Brennan DM, Booth J, et al. Impact of body mass index on outcome after percutaneous coronary intervention (the obesity paradox). Am J Cardiol 2002;90:42 - 5. 6. Powell BD, Lennon RJ, Lerman A, et al. Association of body mass index with outcome after percutaneous coronary intervention. Am J Cardiol 2003;91:472 - 6. 7. Minutello RM, Chou ET, Hong MK, et al. Impact of body mass index on in-hospital outcomes following percutaneous coronary intervention (report from the New York State Angioplasty Registry). Am J Cardiol 2004;93:1229 - 32. 8. Hoit BD, Gilpin EA, Maisel AA, et al. Influence of obesity on morbidity and mortality after acute myocardial infarction. Am Heart J 1987; 114:1334 - 41. 9. Lee KL, Woodlief LH, Topol EJ, et al. Predictors of 30-day mortality in the era of reperfusion for acute myocardial infarction. Results from an international trial of 41,021 patients. GUSTO-I Investigators. Circulation 1995;91:1659 - 68. 10. Lundergan CF, Ross AM, McCarthy WF, et al. GUSTO-I Angiographic Investigators. Predictors of left ventricular function after acute myocardial infarction: effects of time to treatment, patency, and body mass index: the GUSTO-I angiographic experience. Am Heart J 2001;142:43 - 50. 11. Stone GW, Grines CL, Cox DA, et al. Comparison of angioplasty with stenting, with or without abciximab, in acute myocardial infarction. N Engl J Med 2002;346:957 - 66. 12. St Jeor ST, Brownell KD, Atkinson RL, et al. Obesity. Workshop III. AHA Prevention Conference III. Behavior change and compliance: keys to improving cardiovascular health. Circulation 1993;88: 1391 - 6. 13. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31 - 41.

American Heart Journal Volume 151, Number 1

Nikolsky et al 175

14. Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and trends in obesity among US adults, 1999-2000. JAMA 2002;288: 1723 - 7. 15. Wolk R, Berger P, Lennon RJ, et al. Body mass index: a risk factor for unstable angina and myocardial infarction in patients with angiographically confirmed coronary artery disease. Circulation 2003; 108:2206 - 11. 16. Suwaidi JA, Wright RS, Grill JP, et al. Obesity is associated with premature occurrence of acute myocardial infarction. Clin Cardiol 2001;24:542 - 7. 17. Sadeghi HM, Stone GW, Grines CL, et al. Impact of renal insufficiency in patients undergoing primary angioplasty for acute myocardial infarction. Circulation 2003;108:2769 - 75. 18. Rea TD, Heckbert SR, Kaplan RC, et al. Body mass index and the risk of recurrent coronary events following acute myocardial infarction. Am J Cardiol 2001;88:467 - 72. 19. Kaplan RC, Heckbert SR, Furberg CD, et al. Predictors of subsequent coronary events, stroke, and death among survivors of first hospitalized myocardial infarction. J Clin Epidemiol 2002;55: 654 - 64.

20. Wassertheil-Smoller S, Fann C, Allman RM, et al. Relation of low body mass to death and stroke in the systolic hypertension in the elderly program. Arch Intern Med 2000;160:494 - 500. 21. Gillum RF, Mussolino ME, Madans JH. Body fat distribution, obesity, overweight and stroke incidence in women and men-the NHANES I Epidemiologic Follow-up Study. Int J Obes Relat Metab Disord 2001;25:628 - 38. 22. Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15:44 - 54. 23. De Pergola G, Pannacciulli N. Coagulation and fibrinolysis abnormalities in obesity. J Endocrinol Invest 2002;25:899 - 904. 24. Abizaid A, Kornowski R, Mintz GS, et al. The influence of diabetes mellitus on acute and late clinical outcomes following coronary stent implantation. J Am Coll Cardiol 1998;32:584 - 9. 25. Bauters C, Hubert E, Prat A, et al. Predictors of restenosis after coronary stent implantation. J Am Coll Cardiol 1998;31:1291 - 8. 26. Sonmez K, Akcakoyun M, Akcay A, et al. Which method should be used to determine the obesity, in patients with coronary artery disease? (body mass index, waist circumference or waist-hip ratio). Int J Obes Relat Metab Disord 2003;27:341 - 6.

The following article is an AHJ Online Exclusive. Full text of this article is available at no charge at our website: www.ahjonline.com

A randomized pilot study of dalteparin versus unfractionated heparin during percutaneous coronary interventions Madhu K. Natarajan, MD,a James L. Velianou, MD,a Alexander G.G. Turpie, MD,a Shamir R. Mehta, MD,a Dominic Raco, MD,a David M. Goodhart, MD,b Rizwan Afzal, MSc,a and Jeffrey S. Ginsberg, MDa Hamilton, Ontario, and Calgary, Alberta, Canada

Background

Direct comparison of low–molecular-weight

1.7 IU/mL, P = .005)) and at 4 hours (UFH 0.27 IU/mL vs dalteparin

heparin, dalteparin, with unfractionated heparin (UFH) during percutaneous

0.69 IU/mL, P b .0001). Angiographic success was N90% in both

coronary intervention (PCI) is limited. This study examined the relative

groups, and angiographic complications were similar (UFH 2.5% vs

effects of dalteparin and UFH on coagulation and angiographic and

dalteparin 3.8%). The composite of death, myocardial infarction, target

clinical indices during PCI.

vessel revascularization, or bailout glycoprotein IIb/IIIa at hospital discharge

Methods

was 13.7% in the UFH group and 13.1% in the dalteparin group

This was a double-blind randomized study, stratified

by planned glycoprotein IIb/IIIa inhibitor use. Both UFH and dalteparin

( P = not significant). There were 2 major bleedings requiring transfusion,

were administered as an intra-arterial bolus immediately before PCI.

both occurring in the UFH group.

Results

Conclusions

All randomized patients received the assigned study drug

This study suggests that a single intra-arterial bolus

and underwent PCI. Mean activated clotting time levels were 344 seconds

of low–molecular-weight heparin without monitoring is feasible and

for UFH and 234 seconds for dalteparin ( P b .0001). Anti–factor Xa levels

warrants further investigation as an alternative to UFH during PCI.

were higher for dalteparin at 30 minutes (UFH 1.3 IU/mL vs dalteparin

(Am Heart J 2006;151:175.e1-175.e6.)