The Influence of Body Mass Index on Outcomes and the Benefit of Antiplatelet Therapy following Percutaneous Coronary Interventio
March 2006
Obesity affects over 20% of the U.S. population and most countries in the developing world.1,2 It is an independent risk factor for premature coronary artery disease and is associated with an increase in cardiac morbidity and mortality, similar to the consequences of long-term cigarette smoking, diabetes, hypertension and hyperlipidemia.3,4
In percutaneous coronary intervention (PCI) and acute coronary syndrome (ACS) trials, obesity has been shown to influence prognosis. Surprisingly, obese patients in general have a better short and long-term outcome compared with thinner patients.5–8 This paradox between obesity and a favorable outcome after PCI may reflect a selection bias of “healthier”, younger obese patients in clinical trials compared with thinner older patients who are more likely to have advanced coronary artery disease.4 Alternatively, the more favorable outcomes in obese ACS and PCI patients could be due to a treatment confounder caused by the number or dose of adjunctive medical therapies that these patients typically receive in a weight-based manner.
Finding the right drug dose for each individual patient can be difficult. It is clear that pharmacokinetics and dynamics can vary according to a patient’s age, body size and composition, renal and hepatic function, and concomitant medications. These issues are very important when the therapeutic window for efficacy and safety is narrow, as is the case with some antiplatelet and antithrombotic therapies. In clinical practice, doses of unfractionated, low-molecular weight heparins and intravenous platelet glycoprotein IIb/IIIa receptor antagonists are usually weight-adjusted. In contrast, aspirin and clopidogrel doses are not.
The question of whether the dose of clopidogrel needs to be adjusted for body weight has recently been highlighted in a small study of 48 patients undergoing elective PCI. In this case, the inhibition of platelet aggregation during treatment with clopidogrel and aspirin was found to be significantly less in overweight patients (defined by body mass index [BMI] > 25 kg/m2) both at baseline and at 24 hours when compared with normal weight (BMI 2) patients. The study concluded that obese patients may therefore require a higher clopidogrel dose at the time of PCI.9 An accompanying editorial suggested that a higher loading dose of 600 mg should be considered for overweight patients undergoing elective PCI.10
The aim of our study was to determine the relationship between BMI, clopidogrel therapy and clinical outcomes 1 year following PCI in patients enrolled in the CREDO trial.
Methods
Patient population. BMI, clinical characteristics, and clopidogrel therapy were assessed in 2,116 patients enrolled in the Clopidogrel for the Reduction of Events During Observation (CREDO) trial.11 The CREDO trial randomized moderate-risk coronary artery disease patients with objective evidence of ischemia (e.g., angina symptoms, dynamic ECG changes or positive stress test) who were likely to undergo PCI, to a strategy of a 300 mg loading-dose of clopidogrel (n = 1,053) or placebo (n = 1,063) 3 to 24 hours before PCI. Thereafter, all patients received 75 mg daily of clopidogrel through day 28. From day 29 through 12 months, patients in the loading-dose group received clopidogrel 75 mg daily, and those in the control group received placebo. Both groups received aspirin throughout the study. Overall, at 1 year, there was a 26.9% relative reduction in the combined risk of death, MI or stroke (95% CI = 3.9–44.4%; p = 0.02; absolute risk reduction 3%) in patients randomized to clopidogrel.
Definitions. BMI was categorized by a modified version of the predefined WHO standard classification: low-normal (BMI 2), overweight (BMI 25–29.9 kg/m2), obese (BMI 30–39.9 kg/m2) and severe obesity (greater than or equal to 40 kg/m2).12 The study patients were prospectively defined according to these BMI categories. The primary 1-year outcome was the composite of death, myocardial infarction (MI) or stroke in the intent-to-treat population. The primary endpoint at 28 days was the composite of death, MI or urgent target vessel revascularization (UTVR) in the per-protocol population, which included all randomized patients who underwent PCI. Secondary endpoints focused on safety and included the incidence of any bleeding event and of early discontinuation of study drugs at 28 days and 1 year. Details of these endpoints have been previously described.11 Data on 4 patients were incomplete and were excluded from this study analysis.
Statistical analysis. Data are presented as mean ± standard deviation and percentages. Univariable results are presented with p-values generated from Wilcoxon two-sample tests for continuous variables (as these data were not normally distributed). For the categorical variables, patient demographics between groups were compared using Chi-square tests, or if small expected cell frequencies, exact tests were used. The Breslow-Day test for homogeneity of the odds ratios and the Cochran-Armitage test for trend were both used for testing if a trend existed as BMI increased. Multivariable modeling was used to compare the relationship of increasing BMI with 1-year outcome. This was used so that adjustments could be made for significant baseline and prerandomization variables. Subsequent results of significant findings could then be interpreted on a descriptive basis after accounting for appropriate patient differences. Cox proportional hazard modeling was used to describe all endpoints. The strategy for modeling was to first select from baseline and prerandomization factors. After significant adjustments were made, obesity factors were considered.
Stepwise variable selection techniques were used with a p-value of 0.1 for entry. A p-value of 0.05 was considered statistically significant. Backward and forward selection was also being tested. Treatment is forced into the model. In models where risk factors were significant, interactions for treatment with all other factors and between factors were also considered. Model selection was validated using bootstrap techniques.
Imputation for missing data techniques was used for model selection. but the final data presented are from the original data without imputation.
Results
The baseline demographics of the study population are shown in Table 1. Data to determine BMI were available in all but 4 patients. Of the 2,112 patients included in this analysis, 342 had low-normal BMI, 847 were overweight, 810 were obese and 113 were very obese. In general, increasing BMI was associated with younger age, male gender, non-Caucasian race, a history of hypertension, diabetes and hyperlipidemia (p p = 0.02).
28-day endpoints. The 28-day composite endpoint of death, MI or UTVR was higher (10% vs. 4.4%; p p = 0.09).
Effect of clopidogrel on 28-day endpoints. Univariately, the risk of the 28-day combined endpoint of death, MI or UTVR associated with randomization to clopidogrel was reduced by 31% (OR 0.694; 95% CI = 0.536–0.899; p = 0.006) for every 5-unit increase in BMI (Figure 2). By comparison, there was no significant difference based on BMI in the incidence of 28-day composite endpoints in patients who received placebo therapy.
One-year endpoints. There was a nonsignificant trend favoring increasing BMI in the combined endpoint of death, MI or stroke at 1 year (Figure 3).
Effect of clopidogrel on one-year endpoints. The risk of the 1-year combined endpoint of death, MI or stroke associated with randomization to clopidogrel was reduced by 25% (OR 0.748; 95% CI = 0.601–0.930; p = 0.009) for every 5-unit increase in BMI. By comparison, there was no significant relationship between BMI and the incidence of the 1-year composite endpoint in patients who received placebo therapy (Figure 4).
Any bleeding complications were more frequent at 1 year in patients with a lower BMI (2; p = 0.03) compared with all other BMI groups. There was no difference in major bleeding events between body weight groups (Table 3). Randomization to clopidogrel was not associated with any significant difference in bleeding as BMI increased (41% vs. 34% vs. 35% vs. 25%; p = 0.07).Overall, with every 5-unit increase in BMI, the risk of any bleeding at 1 year decreased by 13% for clopidogrel-treated patients (OR 0.87; 95% CI = 0.775–0.976; p = 0.017).
Regression modeling. Multivariable regression analysis found that as age increased to over 70 years, so too did the risk of death, MI or stroke. When treatment with clopidogrel or placebo and clinical variables were entered into the model, BMI was a less important risk factor for 1-year death, MI or stroke (p > 0.05). While hypertension, diabetes and hyperlipidemia were more prevalent in patients with severe obesity, these risk factors did not significantly influence the 1-year composite endpoint.
There was interaction between BMI and treatment, with clopidogrel decreasing the risk of 1-year combined endpoints as BMI increased, but no effect was seen with placebo.
Discussion
We have shown for the first time that dual antiplatelet therapy with aspirin and clopidogrel is associated with a greater reduction in 1-year death, MI or stroke rates in patients undergoing PCI as BMI increases.
These clinical findings contrast with those of Angiolillo and colleagues, who reported that the effect of clopidogrel on platelet aggregation in patients undergoing PCI decreased with increasing BMI. Platelet aggregation increased with BMI and it was suggested that higher doses of clopidogrel therapy might be required to achieve a clinical benefit. They postulated that without a weight adjustment in clopidogrel therapy, obese patients could be at an increased risk for future adverse clinical events.9 In our post hoc analysis of a prospective, randomized, placebo-controlled trial, we found just the opposite: that dual antiplatelet therapy starting with 300 mg of clopidogrel pretreatment achieves a similar beneficial effect in all BMI groups and that an amplified benefit is actually observed as BMI increases.
The reason for the disparity between these studies (in terms of platelet inhibition and clinical effect with clopidogrel 300 mg) may be explained by the variation reported among ex vivo platelet aggregometry tests and clinical outcomes.13–15 Aggregometry assays of antiplatelet agents do not always correlate with the clinical effect of antiplatelet therapy. In Angiolillo’s study, platelet aggregation was measured by ADP-induced platelet aggregation using standard light transmittance aggregometry.9 Several studies have employed this test to show the effect of clopidogrel on platelet aggregation, and few have correlated the sensitivity and specificity of this with clinical outcomes in patients.16–19 The variation between this surrogate marker of clinical effect and the results from our analysis of the CREDO study highlight a potential risk when using certain measures of platelet inhibition to predict clinical outcome.
A paradox exists between obesity and adverse events after PCI. Previously, studies have found that patients with increasing BMI have better short-term outcomes following PCI compared with thinner patients.5–8 In a recent analysis of the Sibrafiban vs. aspirin to Yield Maximum Protection from ischemic Heart events post-acute cOroNary Syndromes (SYMPHONY) trials, obesity was associated with better 90-day survival after ACS than in normal-weight individuals, but death or MI rates were similar in all BMI groups.20 In our own study, increasing BMI was associated with a trend towards younger male patients and metabolic syndrome-like risk factors. Perhaps these patients’ younger age helped their outcomes or led to earlier presentation to a medical care facility and potentially more aggressive initial management. It may be that increasing BMI represents a surrogate for younger age. In contrast, patients with low-normal BMI tended to be older Caucasian women. It is well known that low BMI may be a surrogate for systemic illness and this may explain part of the increased mortality risk, as may the advancing age of these patients.21
During PCI, patients with lower BMI had a trend towards more bleeding complications compared with higher BMI patients. Similar findings have been reported by Ellis et al. and by Powell et al., both of whose studies have suggested that unadjusted dosing of any antithrombotic medications may have been contributory.5,6 This may in part explain the difference in outcomes between placebo- and clopidogrel-treated patients in the low-normal BMI group. Some studies have reported that obesity favors a hypercoagulable state and that this may lower bleeding risks for obese patients.22
The fact that obesity may be associated with better outcomes after PCI may be due to other factors such as lesion characteristics. Greater culprit vessel lumen size (as evidenced by the larger stent diameters used with increasing BMI) may be associated with lower restenosis rates in obese patients, although restenosis has not been associated with an increased risk of hard endpoints such as death, MI or stroke.23 Interestingly, more obese patients underwent left anterior descending artery interventions, while thinner patients underwent more right coronary artery interventions. However, there was no difference in terms of baseline LV dysfunction or the number of patients with multivessel coronary artery disease across the range of BMI.
The influence of body weight and the effect of maintenance therapy with 75 mg of clopidogrel alone versus aspirin alone have previously been looked at in the Clopidogrel versus Aspirin in Patients at Risk of Ischemic Events (CAPRIE) study. Clopidogrel was associated with no difference in outcomes between the BMI patient groups.24Study limitations. There are a number of limitations with the present study. This was a post hoc observational analysis of the CREDO study. The small numbers of patients in each BMI group means that we were unable to determine the impact of BMI on the time-to-pretreatment effect. Furthermore, the small number of patients may have affected the trend towards slightly higher bleeding rates in the clopidogrel-treated patients than in the placebo group.
Conclusion
In conclusion, despite pharmacodynamic data to suggest a dose- and weight-related antiplatelet effect of clopidogrel, clinical data from over 2,000 PCI patients in the CREDO study suggest the contrary — that as BMI increases, the relative benefit of clopidogrel is actually greater. Further studies are required to assess the mechanism for this and to clarify the correlation between ex vivo platelet aggregometry data and clinical benefit.
1. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among U.S. adults, 1999–2000. JAMA 2002;288:1723–1727.
2. 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;341:1097–1105.
3. Sowers JR. Obesity as a cardiovascular risk factor. Am J Med 2003;115(Suppl 8A):37S–41S.
4. Wilson PW, D'Agostino RB, Sullivan L, et al. Overweight and obesity as determinants of cardiovascular risk: The Framingham experience. Arch Intern Med 2002;162:1867–1872.
5. Ellis SG, Elliott J, Horrigan M, et al. 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 1996;78:642–646.
6. Powell BD, Lennon RJ, Lerman A, et al. Association of body mass index with outcome after percutaneous coronary intervention. Am J Cardiol 2003;91:472–476.
7. 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;39:578–584.
8. 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;93:1229–1232.
9. Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, et al. Platelet aggregation according to body mass index in patients undergoing coronary stenting: Should clopidogrel loading-dose be weight adjusted? J Invasive Cardiol 2004;16:169–174.
10. Patel MR, Peterson ED. Clopidogrel dosing in overweight patients: Does one size fit all? J Invasive Cardiol 2004;16:175–176.
11. Steinhubl SR, Berger PB, Mann JT 3rd, et al. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: A randomized controlled trial. JAMA 2002;288:2411–2420.
12. Aronne LJ. Classification of obesity and assessment of obesity-related health risks. Obesity Research 2002;10:1055–1155.
13. Wiviott SD, Antman EM. Clopidogrel resistance: A new chapter in a fast-moving story. Circulation 2004;109:3064–3067.
14. Varanasi JS, Steinhubl SR. Antiplatelet effect of aspirin in patients with cerebrovascular disease. Stroke 2004;35:E144-E145.
15. Gum PA, Kottke-Marchant K, Poggio ED, et al. Profile and prevalence of aspirin resistance in patients with cardiovascular disease. Am J Cardiol 2001;88:230–235.
16. Jorum P, Lindahl TL, Fransson SG, Richter A. Individual variations of platelet inhibition after loading doses of clopidogrel. J Intern Med 2002;252:233–238.
17. Gurbel PA, Bliden KP, Hiatt BL, O'Connor CM. Clopidogrel for coronary stenting: Response variability, drug resistance, and the effect of pretreatment platelet reactivity. Circulation 2003;107:2908–2913.
18. Gurbel PA, Cummings CC, Bell CR, et al. Onset and extent of platelet inhibition by clopidogrel loading in patients undergoing elective coronary stenting: The Plavix Reduction Of New Thrombus Occurrence (PRONTO) trial. Am Heart J 2003;145:239–247.
19. Taubert D, Kastrati A, Harlfinger S, et al. Pharmacokinetics of clopidogrel after administration of a high loading dose. Thromb Haemost 2004;92:311–316.
20. Eisenstein E, McGuire DK, Bhapkar MV, et al. Elevated body mass index and intermediate-term clinical outcomes after acute coronary syndromes. Am J Med 2005;118:981–990.
21. Garrison RJ, Castelli WP. Weight and thirty-year mortality of men in the Framingham Study. Ann Intern Med 1985;103:1006–1009.
22. Kockx M, Leenen R, Seidell J, et al. Relationship between visceral fat and PAI-1 in overweight men and women before and after weight loss. Thromb Haemost 1999;82:1490–1496.
23. Saucedo JF, Popma JJ, Kennard ED, et al. Relation of coronary artery size to one-year clinical events after new device angioplasty of native coronary arteries (a New Approach to Coronary Intervention [NACI] Registry Report). Am J Cardiol 2000; 85:166–171.
24. The CAPRIE Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischemic events (CAPRIE). Lancet 1996;348:1329–1339.
