To study survival in isolated coronary artery bypass graft (CABG) patients and to evaluate the impact of preoperative chronic opium consumption on long-term outcome.

Cohort of 566 isolated CABG patients as Tehran Heart Center cardiac output measurement was conducted. Daily evaluation until discharge as well as 4- and 12-mo and 6.5-year follow-up information for survival status were fulfilled for all patients. Long-term 6.5-year overall and opium-stratified survival, adjusted survival curves based on opium consumption as well as possible predictors of all-cause mortality using multiple cox regression were determined by statistical analysis.

Six point five-year overall survival was 91.8%; 86.6% in opium consumers and 92.7% in non-opium consumers (P = 0.035). Patients with positive history of opium consumption significantly tended to have lower ejection fraction (EF), higher creatinine level and higher prevalence of myocardial infarction. Multiple predictors of all-cause mortality included age, body mass index, EF, diabetes mellitus and cerebrovascular accident. The hazard ratio (HR) of 2.09 for the risk of mortality in opium addicted patients with a borderline P value (P = 0.052) was calculated in this model. Further adjustment with stratification based on smoking and opium addiction reduced the HR to 1.20 (P = 0.355).

Simultaneous impact of smoking as a confounding variable in most of the patients prevents from definitive judgment on the role of opium as an independent contributing factor in worse long-term survival of CABG patients in addition to advanced age, low EF, diabetes mellitus and cerebrovascular accident. Meanwhile, our findings do not confirm any cardio protective role for opium to improve outcome in coronary patients with the history of smoking. Further studies are needed to clarify pure effect of opium and warrant the aforementioned findings.

Growing number of patients undergoing coronary artery bypass graft (CABG) surgery all over the world[1], justifies studying on possible predicting factors of clinical outcomes such as short and long-term survival. There are published reports of some predictive factors responsible for short-term mortality such as advanced age, previous history of cardiac surgery and myocardial infarction, non-cardiac comorbidities, New York Heart Association functional class (FC) III or IV and serum creatinine (Cr) level[2,3]. Also some additional factors have been suggested for intermediate-term mortality including left ventricular ejection fraction (EF) and history of percutaneous coronary stenting[2,4]. Similarly, long-term survival could be influenced by diabetes mellitus[5], hypoalbuminemia[6], female gender, smoking, cardiogenic shock[7] and severe preoperative renal dysfunction[3,8].

Coronary artery disease (CAD) has a high prevalence in Iranian population, affecting 22.2% of men and 37.5% of women[9]. In addition, opium is the major abused substance in Iran[10] and the prevalence of opium addiction is believed to be higher in CAD patients undergoing revascularization[11]. The predictive role of opium in short-term outcomes of patients undergoing CABG is controversial. Some investigations have suggested cardiac protective role for opium during ischemic events[12,13]. In a propensity-matched study, Sadeghian et al[14] found no association between opium dependence and post CABG in-hospital complications. On the other hand, in Safaii et al[15]’s study, six months post-CABG readmission was significantly more frequent in opium users. However, there are few evidences to clarify the definite relationship between chronic opium abuse and long-term survival in cardiac surgery patients.

So not only there is limited knowledge regarding the adverse impact of opium consumption on long-term survival of CABG patients, but also the available studies on short-term outcomes have shown controversial results. Therefore in this study we aimed to assess the impact of chronic opium consumption on long-term survival of patients who underwent isolated CABG in a cardiac tertiary center.

Based on findings, among 566 patients, 53 (9.4%) deaths occurred during the 6.5 years of follow-up among which 40.9% was cardiac and 59.1% was of non-cardiac cause. Median follow-up time for all of the study population was 78.7 mo (95%CI: 78.5-78.9). Median (25^th-75^th interquartile range) follow-up time of 235 patients with incomplete follow-up was 74.6 (25^th-75^th: 73.5-75.7) mo.

Patients’ demographics characteristics and risk factors: Table 1 demonstrates patients’ baseline demographic and clinical characteristics by survival status. According to univariate analysis, age, blood urea nitrogen and high-density lipoprotein levels were significantly lower and EF, albumin and triglyceride levels were significantly higher among survivors. Moreover, opium consumption, diabetes mellitus, cerebrovascular accident and peripheral vascular disease were significantly more frequent among the expired cases.

The frequency of moderate, high and very high risk EuroSCORE was significantly higher in non-survivors as compared to survivors. However, the number of diseased vessels and performed bypass grafts was similar between the both groups.

Table 2 shows patients’ baseline demographic and clinical characteristics generally and based on opium consumption. Mean ± SD age of patients was 59.08 ± 8.9 years and 75.1% were men. History of opium consumption was present in 14.5%. Forty-one percent had diabetes mellitus and 3.9% had history of cerebrovascular accident. Body mass index and EF mean ± SD was 27.3 ± 4.08 kg/m² and 48.5% ± 10.3%, respectively. Functional class III was documented in 14.7% of patients. Opium users were significantly more often men, younger, smoker and also alcohol consumer.

Patients with positive history of opium consumption also significantly tended to have lower EF (44.9 ± 9.4 vs 49.1 ± 10.3), higher Cr (1.3 ± 0.3 vs 1.2 ± 0.2) and higher prevalence of MI (71.6% vs 47.5%). On the other hand, the level of BMI and fasting blood sugar and prevalence of hyperlipidemia (HLP), hypertension were significantly lower in these patients.

Opium-stratified survival

Based on follow-up information, 6.5 years (78.7 mo) overall survival was 91.8% (95%CI: 89.5%-94.2%). When analyzed based on habitual history of opium consumption, 6.5-year (78.7 mo) overall survival was found to be 86.6% (95%CI: 79.1%-94.7%) in opium users and 92.7% (95%CI: 90.3%-95.1%) in non-opium users.

After adjustments for confounding variables such as age, BMI, EF, diabetes, alcohol, HLP, MI, Cr, BUN and EuroSCORE, we found an evidence of predicting mortality for opium with a borderline P value (HR = 2.16; 95%CI: 0.96-4.84; P = 0.06) (Figure 1). As appears from curves, opium users have a trend to worse long-term survival as compared to non-opium consumers.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Adjusted survival curves based on opium consumption.

The current study represents the overall and opium-based stratified survival of patients undergoing CABG surgery who were followed-up for a median time of 78.7 mo. The prevalence of opium addiction was found to be 14.5% in our study. To our knowledge, studies are lacking to clarify the long-term effects of opium consumption on survival of patients undergoing CABG surgery. One of the notable points of our study was relatively long duration of follow-up.

Considering overall survival of patients undergoing CABG surgery, we found an overall survival of 91.8% in 6.5 years (78.7 mo). Five-year survival has been reported 72.1% in Yoo et al[21]’s study and Dunning et al[22] identified a ten-year survival of 66%. Meanwhile unadjusted 5 and 10-year survivals have been reported 83.8% and 65% in Filardo et al[1]’s research.

We identified that age, BMI, EF, diabetes mellitus and cerebrovascular accident could independently and significantly predict all-cause mortality of patients undergoing CABG surgery. Findings of Leavitt et al[5], Marcheix et al[23] and Barsness et al[24] support our results in terms of adverse effect of diabetes mellitus on long-term survival. Interestingly cerebrovascular accident had the greatest HR of 3.45 in predicting mortality by multiple Cox regression analysis suggesting that non-cardiac comorbidities may play an important role in patients’ outcomes as well as unfavorable cardiac performance.

We found a significant HR of 0.89 for BMI in predicting overall mortality. Our results were in accordance with those reported in Gruberg et al[25]’s work. They found that overweight or obese patients under CABG have significantly better survival outcomes at 3-year follow-up than those with normal BMI. The phenomenon obesity-mortality paradox which is generally accepted in short term outcome studies is described by better outcome in patients with higher BMI compared to the others. However, there is no consensus in long term investigations as Del Prete et al[26] noted that long-term survival was not significantly different between obese and non-obese patients after making adjustment model (HR = 1.2, P = 0.2). This is partially explained by increasing rate of complications due to major cardiovascular risk factors and substantial re stenosis in grafted coronary arteries by time[27,28]. Though we followed our patients for long period of time we found that BMI was still a predictor of mortality. Low mortality rate in our cohort compared to other studies with similar time intervals that reflect lower risks and complications in our patients could be an explanation. The other reason may be the finding of missed to follow up in a group of our patients which decreases the median time of overall follow-up.

Opium can potentially cause coronary atherosclerosis and increase cardiovascular mortality in different ways. Some metabolic changes by opium that could have deleterious effects on cardiovascular system include: Decreasing plasma testosterone and estrogen, and increasing plasma prolactin, increasing insulin resistance, increasing inflammation as well as oxidative stress, increasing fibrinogen and factor VII, and decreasing apolipoprotein A, increasing the release of nitric oxide and inhibiting production and release of hydrogen peroxide. Moreover, opium can decrease myocardial oxygenation from different ways that could extend infarct size and increase probability of death[29,30].

The role of opium in CAD remains controversial. Masoomi et al[13] showed that opium was an independent risk factor of CAD in non-smoker patients. But findings of Sadeghian et al[14]’s research on 4398 isolated CABG patients and opium dependence rate of 15.6%, found no relationship between post CABG in hospital complications and opium addiction. But on the other hand, in a study conducted by Safaii et al[15] on 6-mo outcomes of CABG patients, opium usage leaded to more readmission following CABG operation.

We found a HR of 2.1 for mortality in opium consumers as compared to patients who did not use opium with a borderline P value of 0.06. Since the rate of smoking is higher among opium-consumers, when we further adjusted the results by smoking, we observed that the role of smoking would be more prominent in predicting mortality so that HR for opium decreased to 1.2. Though our findings are not in favor of any protective role for opium in smoker cardiac patients, there is still no evidence to consider opium as a risk marker for long term survival in this group too.

The main problem with clarifying pure effect of opium on long term outcome in cardiac patients is high co incidence of smoking and opium addiction (Table 2). In deed there were only 15 patients who were opium users and non-smokers in our cohort. We need to perform further investigation to clarify pure effect of opium, because ignoring the adverse effects of opium and attributing any poor clinical outcomes to the smoking alone would be potentially associated with worse consequences.

There are other obstacles to detect the effects of opium on outcome in coronary patients that has been discussed elsewhere[30,31]. Briefly there are variations in self-reported dosage, route of usage, and purity of consumed opium. The other important issue probably would be the reason for beginning opium consumption: recreational or for pain relief[29,30].

In conclusion, in the present study, we found advanced age, low EF, DM and CVA as predictors for long term mortality. However, due to the simultaneous impact of smoking as a confounding variable neither the cardio protective role of opium in ischemic phase suggested in some studies nor its role as a predictor for long-term survival of CABG patients could be justified. Further large sample size studies are needed to clarify pure opium role and verify the aforementioned findings.