Methods

Study population

We retrospectively evaluated 4996 patients who underwent PCI with drug-eluting stents at Beijing Anzhen Hospital, Beijing, China, between January 2004 and December 2006. We used the hospital’s comprehensive interventional database, in which charts of coronary intervention patients were reviewed by independent investigators unaware of the objectives or purpose of the study. The study design was reviewed and approved by our institutional research ethics committee. Exclusion criteria were acute myocardial infarction with ST-segment elevation; cardiogenic shock; congenital or valvular heart disease; primary cardiomyopathy; age <17 or >80 years; moderate or severe renal insufficiency with estimated glomerular filtration rate <60 ml/min per 1.73 m²; previous coronary stent implantation or coronary artery bypass graft; thrombocytopenia; and ongoing bleeding or history of bleeding diathesis.

Body mass index

The presence of obesity was assessed using BMI, which was defined as weight in kilograms divided by height in metres squared (kg/m²). BMI was categorised according to World Health Organization criteria; BMI <18.5 was classified as underweight, 18.5–24.9 as normal, 25–29.9 as overweight and ⩾30 as obese.17

Stent procedure

All interventions were performed according to current standard guidelines, with the type of stent implanted and the use of glycoprotein IIb/IIIa inhibitors at the discretion of the operator. A loading dose of 300 mg clopidogrel was given to all patients at least 6 hours before PCI, followed by a maintenance dose of 75 mg daily for 9–12 months unless severe bleeding complications appeared. In addition, all patients were given 300 mg aspirin daily, starting on the day of the procedure. Compliance with clopidogrel and aspirin treatment was determined by telephone interview during follow-up. Glycoprotein IIb/IIIa inhibitors were used to treat 28.0% of patients during the procedure.

Endpoints and definitions

The primary endpoint was defined as a composite of cardiovascular thrombotic events, consisting of cardiovascular death and non-fatal myocardial infarction. The second endpoint was defined as definite or probable stent thrombosis. Cardiovascular death over the follow-up period was defined as death from acute myocardial infarction, coronary artery disease or congestive heart failure. Myocardial infarction diagnostic criteria included two or more of the following: high cardiac enzyme levels (twice the normal level of total creatine kinase and an abnormal MB isoenzyme or cardiac troponin I level); development of typical electrocardiographic changes (new pathological Q-wave in two or more contiguous leads, or as new left bundle branch block); and typical chest pain. Stent thrombosis was classified by the Academic Research Consortium definitions as definite, probable, or possible and as early (0–30 days), late (31–360 days), or very late (>360 days).18 Definite stent thrombosis required the presence of an acute coronary syndrome with angiographic or autopsy evidence of thrombus or occlusion. Probable stent thrombosis included unexplained deaths within 30 days after the procedure or acute myocardial infarction involving the target-vessel territory without angiographic confirmation. For those patients who reached at least one of the endpoints, a medical chart review was initiated to determine whether the event met the definitions described. Causes of death were mainly determined by examination of hospital charts, autopsy results, and medical files of the patients’ general practitioners.

Clinical follow-up was performed by either telephone contact or office visit. Investigators blinded to patient BMI performed follow-up interviews and medical event classification. If a patient had more than one procedure during the study period, only the initial procedure was included.

Statistical analysis

Continuous variables are presented as mean (SD), and were compared using the Student t test. Categorical variables are presented as frequencies and percentages. Differences in baseline characteristics between patients in the three study-defined BMI groups were compared with χ² test or a Fisher’s exact test when any expected cell count was less than 5. Event-free survival estimates were created using Kaplan-Meier methods and compared with the log-rank test with respect to categorisation of BMI. Independent predictors of primary and secondary endpoints were identified using Cox proportional hazards regression with stepwise selection using entry and exit criteria of a p value <0.10. The candidate variables entered in the model included age (continuous), gender, smoking status, history of prior myocardial infarction, diagnosis of diabetes mellitus, hypertension, hypercholesterolemia, unstable angina, non-ST-segment elevation myocardial infarction, left ventricular dysfunction with ejection fraction <30%, usage of two-stent techniques in bifurcation lesions, thrombus-presenting lesions, lesion length ⩾20 mm, target vessel reference diameter (continuous) and total stent length (continuous). Hazard ratios and 95% confidence intervals were calculated. All statistical analyses were two-tailed, and a value of p<0.05 was considered statistically significant. Analysis was performed with SPSS software, version 13.0 for Windows.

Results

Patient sample characteristics

Of 7895 patients who underwent PCI during the study period, 4996 patients met the criteria and were available for follow-up analysis. Nearly half of the patients (49.5%) were overweight, and more than 70% were either overweight (n = 2475) or obese (n = 1213), with only 25.7% having normal BMI (n = 1284). Baseline clinical and angiographic characteristics according to BMI category are shown in table 1. Overweight and obese patients were significantly younger than patients of normal weight, and more likely to have diabetes, hypercholesterolaemia and hypertension. There was a higher incidence of triple-vessel disease in obese patients than in normal weight and overweight patients, but vessel diameter and lesion length were very similar among the three groups. Proportion of target vessels, total stent length, use of glycoprotein IIb/IIIa antagonist and use of clopidogrel and aspirin at both 1-year and 3-year follow-ups were not different among the three groups. Concomitant medication use at discharge, including statins, angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers, β-blockers and calcium channel blockers, was similar among the three groups as well.

Only 24 patients (0.5% of the study cohort) were underweight, with a BMI below 18.5. Because of the small number of patients and limited statistical power for this group, they were eliminated from this study. The total number of patients in the study was thus 4972.

Clinical outcomes

Table 2 illustrates the incidence of cardiovascular events during the follow-up period. Composite thrombotic events, including cardiac death and non-fatal myocardial infarction, occurred in 207 (4.2%) of the 4972 patients. Of those, 78 (1.6%) were cardiac death and 129 (2.6%) were non-fatal myocardial infarction. Non-cardiac death occurred in 25 (0.5%) patients. There were 59 (1.2%) episodes of definite or probable stent thrombosis: 18 (0.4%) were early, 19 (0.4%) were late and 22 (0.4%) were very late.

Table 2 and figure 1A illustrate that composite thrombotic events, including cardiac death and non-fatal myocardial infarction, were significantly higher in obese patients (5.9%) than in normal weight (3.2%) and overweight (3.8%) patients during long-term follow-up (p = 0.001). The incidence of definite or probable stent thrombosis during follow-up steadily increased with increasing BMI, with the highest incidence in the obese group (0.9% vs 1.0% vs 1.9% in normal weight, overweight and obese patients, respectively; p = 0.029). Figure 1B shows the Kaplan-Meier curves for event-free survival from stent thrombosis events based on BMI category, with log-rank test results. The event-free survival rate from stent thrombosis was strikingly lower in obese patients (98.6%) than in normal weight (99.4%) and overweight (99.5%) patients at 1-year follow-up (p = 0.013). When the follow-up period was extended to 3 years, the effect of BMI on stent thrombosis persisted (event-free survival rates 99.1% vs 99.0% vs 98.1% in normal weight, overweight and obese patients, respectively; p = 0.023).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Three-year survival curves demonstrate the effect of body mass index on freedom from clinical outcomes after drug-eluting stent implantation. Composite thrombotic events rate (A) and ARC definite/probable stent thrombosis rates (B) in patients with different BMI category.

Multivariate analyses

As presented in table 3, multivariate Cox regression analysis of long-term outcomes showed that BMI ⩾30.0 kg/m² was an independent predictor of cumulative composite thrombotic events during follow-up (hazard ratio 1.86; 95% confidence interval (CI) 1.25 to 2.75; p = 0.003), along with age, diabetes and left ventricular dysfunction. BMI ⩾30.0 kg/m² was also independently associated with increased risk of definite or probable stent thrombosis (hazard ratio 2.17; 95% CI 1.04 to 4.55; p = 0.040), along with other risk factors including diabetes, left ventricular dysfunction and total stent length.

Discussion

The principal findings of this study, which is the first such investigation to evaluate the impact of BMI on clinical outcome after PCI with drug-eluting stents in a Chinese population, are that: (1) nearly three-quarters of Chinese patients undergoing PCI are above ideal body weight; and (2) compared with normal weight and overweight patients, obese patients who received drug-eluting stents demonstrated increased incidences of composite cardiovascular thrombotic events and stent thrombosis during the 3-year follow-up. Obesity, defined as BMI ⩾30.0 kg/m², was an independent predictor of cardiovascular thrombotic events and definite or probable stent thrombosis.

In patients with established coronary artery disease, obesity was reported to be independently associated with increased risk for adverse cardiovascular outcome.19 20 Although the impact of obesity on clinical outcome in patients with coronary artery disease undergoing revascularisation has also been extensively investigated, it still remains a controversial issue. Indeed, several studies have reported better clinical outcomes in overweight and obese subjects after PCI than in normal and underweight subjects, a relation that has been coined the “obesity paradox”.3 4 5 By reporting a worse outcome in obese subjects than in normal or overweight subjects, the results of this study seem to be at odds with these studies. The explanation for this difference is unclear, but one possible reason could be ethnic differences in the relation between BMI and clinical outcome risks. The current World Health Organization definitions of overweight and obesity are based on data from Western populations.17 However, a growing body of literature indicates that cut-off values should be lower among Asian populations.21 22 23 24 25 26 Several epidemiological studies have shown that Asians have higher amounts of body fat at lower BMIs than do Western populations, leading to greater prevalence of cardiovascular risk factors at lower BMIs in Asian populations than in Western populations.22 23 24 25 26 Given these ethnic differences, our results are consistent with the findings that while patients who were overweight and obese had lower mortality and major adverse cardiac event rates than patients of normal weight, the underweight and severely obese subjects (BMI ⩾35.0 kg/m² or ⩾40.0 kg/m²) had significantly higher mortality and major adverse cardiac event rates.7 8 However, the J-shaped association between BMI and clinical outcomes demonstrated by these reports cannot be proved in this study, because the underweight patients were excluded from our analysis. Underweight patients have consistently been shown to have increased in-hospital and long-term adverse outcomes after PCI in previous studies.3 7 8

To our knowledge this is the first study to elucidate the impact of obesity on stent thrombosis after drug-eluting stent implantation. Different aspects may account for the higher risk of thrombotic events in these obese patients. Obesity was found to correlate with higher levels of multiple coagulation factors, such as factor VII, VIII, fibrinogen and plasminogen activator inhibitor-1, which were all associated with increased risk of thrombosis.27 28 Obesity can also be considered a chronic, low-grade inflammatory state, as demonstrated by increased levels of the pro-inflammatory cytokines interleukin-6 and tumour necrosis factor-α, and acute phase proteins such as C-reactive protein.29 This pro-inflammatory state may directly and indirectly cause thrombosis by inducing oxidative stress and endothelial dysfunction.30 Moreover, in several prospective investigations evaluating the relation of BMI and platelet aggregation, overweight and obese patients (BMI ⩾25 kg/m²) were found to more commonly have suboptimal response to loading doses of clopidogrel than normal weight controls.9 10 The influence of BMI on platelet function in patients with long-term clopidogrel treatment (75 mg/day) was also assessed in a recently published trial in which patients who demonstrated effective clopidogrel inhibition had a significantly lower BMI.11 It has also been shown that suboptimal response to clopidogrel is associated with greater incidence of thrombotic events after stent implantation,14 15 leading to the suggestion of weight-adjusted clopidogrel regimens.9 11

The association of obesity and aspirin resistance has also been reported by several studies. In a preliminary study, obese subjects were found to have a blunted response to the platelet-inhibitory effect of a single dose of acetylsalicylic acid.12 A recent investigation also demonstrated that aspirin resistance is significantly associated with haemoglobin A_1c and obesity in diabetic patients.13

In accordance with these previous reports, the present study found that obese patients were more likely to have thrombotic events, especially stent thrombosis, than normal weight and overweight patients after drug-eluting stent implantation. Despite similar clopidogrel therapy among the three groups, the incidence of stent thrombosis was significantly higher in obese patients than in normal weight and overweight patients within 1 year after the procedure. Thereafter, when the mandatory clopidogrel regimen was discontinued and patients took only aspirin, the effect of BMI on stent thrombosis persisted during the 3-year follow-up. These findings, taken in conjunction with previous reports of the effect of BMI on antiplatelet responsiveness, warrant randomised studies to verify whether higher maintenance doses of clopidogrel or aspirin for obese patients could eliminate the adverse effect of obesity on thrombotic events after PCI with drug-eluting stents.

There are inherent limitations to our study. First, this was a single-centre retrospective study and, therefore, the results are not adequate to draw definite wide-ranging conclusions. Second, body weight was recorded only at the time of PCI, such that weight loss and shifts in BMI stratum during follow-up, which may be associated with change in risk, were not taken into account in the present study. Third, BMI may also reflect muscle mass. Other measures of obesity such as waist circumference and waist-to-hip ratio might better reflect body fat content and distribution than BMI. Unfortunately, these measurements for abdominal obesity were not routinely available in this study.

Conclusion

In this observational study, we found that obese patients showed a higher risk for composite cardiovascular thrombotic events and definite or probable stent thrombosis at 3-year follow-up. Further randomised studies to evaluate the efficiency of higher maintenance dose of clopidogrel or aspirin for obese patients after drug-eluting stent implantation are warranted.