Smoking Water-Pipe, Opium Use and Prevalence of Heart Disease: A Cross-sectional Analysis of Baseline Data from the Pars Cohort Study, Southern Iran

1Student Research Committee, Shiraz University of Medical Sciences, Shiraz, Iran 2Non-Communicable Diseases Research Center, Shiraz University of Medical Sciences, Shiraz, Iran 3School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran 4Liver, Pancreatic, and Biliary Diseases Research Center, Digestive Disease Research Institute, Tehran University of Medical Sciences, Tehran, Iran 5Surveillance and Health Services Research, American Cancer Society, Atlanta, GA, USA


Introduction
Ischemic heart disease (IHD) is one of the leading causes of years of life lost worldwide. 1,2 According to the Global Burden of Disease 2016, total deaths from IHD have increased by 19.0% during the recent years (rising from 7.96 million deaths in 2006 to 9.48 million deaths in 2016). 3 Although IHD deaths rates have been declining during the past three decades in most parts of the world, especially in high-income countries, they remain high in some other regions, including the Middle East. 4 As such, IHD is the first cause of years of life lost in Iran. 5 Hookah, also known as water-pipe, shisha and ghalyan, is a tobacco consumption device 6 that has been historically common in Asia (particularly in the Middle East) and North Africa, but recently, its use has been increasing among youth worldwide. [7][8][9] Hookah users are exposed to many carcinogenic compounds that cigarette smokers are 10 ; thus, similar to cigarette smokers, they may be at higher risk of IHD. 6,11 Opium is one of the most commonly used illicit substances in Iran. 12,13 A few earlier studies have suggested that opium use could be a risk factor for heart disease. 14,15 However, still more research is needed on the nature of the associations between hookah and opium use and IHD. The aim of this cross-sectional study is to investigate the association between these habits and prevalent IHD in Iran.

Study Design and Setting
For this analysis, we used baseline data from the Pars Cohort Study (PCS). The study design and protocol of PCS are described elsewhere. 16 In brief, PCS is a prospective study launched in 2012 to investigate the epidemiology and Open Access http://www.aimjournal.ir risk factors of non-communicable diseases in Valashahr district, a mostly rural setting located in Fars province, southern Iran. A total of 9721 (or almost all) Valashahr inhabitants aged 40-75 were invited and 9264 agreed to participate in PCS. 16 The number of participants included in this analysis was 9248, as we omitted 16 participants due to missing data.

Data Collection
Using a structured questionnaire, trained staff interviewed participants at the PCS center located in Valashahr to collect data on demographic characteristics, several indicator of socioeconomic status (SES), tobacco and opium use, and medical history. Participants had been asked to bring their medication to the PCS center; information on the medications was collected during interviews. Anthropometric indices and blood pressure were measured using calibrated equipment, and fasting blood samples were obtained from all participants.
Self-reported history of IHD was the outcome of interest in this study. The participants were considered to have IHD if they answered yes to the question "have you been diagnosed by a physician to have angina, infarction, or heart failure?" Participants who reported only a history of congenital heart disease were not included in the group of individuals with a history of IHD.
Data on hookah use were collected for "average number of episodes of use per day", "number of days (per week) in which hookah was use", and "overall duration of exposure to hookah (from the first use to the time of interview or the date of cessation; in months/years)". Hookah users were defined as participants who reported using hookah at least once a week for a period of 6 months or more. Cumulative amount of hookah use was calculated in hookah-years by multiplying the average number of episodes of hookah use per day by total years of use.
Methods for reliable measurement of opium use and patterns of use in this populations have been described elsewhere. 17 In this study, we analyzed the overall amount of opium used in nokhods (a local unit for opium use that equals 0.195 g) and overall duration of opium use (from the first use to the date of interview or the date of cessation; in months/years). Cigarette smokers were those who had used cigarette for a period of six months or more. Cumulative use of cigarettes was calculated in pack-years.
Body mass index (BMI, in kg/m 2 ) was calculated and categorized according to the World Health Organization (WHO) recommendation: underweight (<18.5), normal (18.5-24.9), overweight (25-29.9), obese I (30-34.9), obese II (35-39.9), and obese III (>40). Waist to hip ratios (WHR), as an indicator of abdominal obesity, were calculated and categorized into its tertiles. Data on physical activity were collected and metabolic equivalent of task (MET) scores were calculated for each participant (18). Participants were categorized into tertiles of the MET score.
Participants were considered hypertensive if they reported previously diagnosed hypertension by a physician, or were on anti-hypertensive drugs, or had a systolic blood pressure of ≥140 mm Hg or diastolic blood pressure of ≥90 mm Hg at the interview. Participants with diabetes mellitus in this study were those previously diagnosed with diabetes by a physician, or those who were on diabetes mellitus medicines, or had fasting blood sugar levels of ≥126 mg/dL in samples obtained at the interview. The following cutoffs (in mg/dL) were used to define abnormal laboratory tests: triglyceride, ≥200; low-density lipoprotein (LDL), ≥160; high-density lipoprotein (HDL), ≤40; and total cholesterol, ≥240. 19 Using multiple correspondence analysis, we calculated asset-based wealth index, a composite score indicating relative SES of each participant based on self-reported assets and appliances owned by participants including house, car, motorcycle, color or black/white TV, bath in home, vacuum, washing machine, refrigerator, freezer, computer, and microwave device. Participants were categorized into three SES subgroups including low SES, moderate SES, and high SES, according to the tertiles of the wealth index.

Statistical Analysis
We provided mean and standard deviation (SD) for quantitative variables and frequencies and relative frequencies for qualitative variables by IHD status. For reported prevalence, we estimated 95% confidence intervals (CIs) applying binomial distributions. To compare the distribution of the included variables between participants with and without IHD, we used Chi-square and Fisher's exact tests for categorical variables and t-test and Mann-Whitney U test for continuous variables. We used logistic regression to estimate odds ratios (ORs) and corresponding 95% CIs. In multivariable analyses, we included variables with a P value of ≤0.2 in univariate analysis. At first, a saturated model was fitted, and then all candidate variables except main exposures were selected in the model by applying a backward elimination procedure. Accordingly, the model was reduced to the best of available models according to the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) goodness-of-fit indices. We also checked for any statistically significant interaction between hookah use and cigarette smoking, but no significant interaction was observed. Two-sided P-values ≤ 0.05 were considered statistically significant. All statistical analyses were done using Stata software, version 11 (Stata Corporation, College Station, TX, USA).
No association was found between cumulative use of hookah and prevalence of IHD in multivariate analysis (OR 1; 95% CI 0.99-1.001 for hookah year among ever users and OR 0.99; 95% CI 0.98-1.006 for years of hookah use among ever users).

Discussion
We found a statistically significant association between hookah and opium use and prevalence of IHD.
Ever use of hookah in our study was associated with 1.3fold higher odds of having IHD. In a case-control study on the association between hookah use and coronary artery disease confirmed by coronary angiography, the risk of having severe stenosis (>70%) in coronary arteries was 3-times higher in those with a lifetime exposure exceeding 40 hookah-years than non-users. 20 In a crosssectional study in Golestan Province in northern Iran, heavy hookah users (100-180 hookah-year) were at about 4-fold higher odds of having prevalent IHD compared to non-smokers. 21 However, they found no association between ever use of hookah and IHD, probably due to low prevalence of hookah use (1.1% vs. 38.2% in our study) and moderate intensity of use in that population. We did not find any association between cumulative use of hookah and prevalence of IHD in the multivariable analysis. This might be due to various factors. Heavy hookah users may be more likely to die prematurely from respiratory diseases or other conditions related to tobacco smoking, including IHD, and those cases would be unaccounted for in cross-sectional studies. Moreover, heavy hookah users may underreport the amount or duration of hookah use. However, lack of dose-response analysis may also suggest that the association between ever hookah use and IHD in this study is spurious, although given the similarities between exposure to harmful compounds during hookah and cigarette use, the observed association is more likely to be real. The mechanism of association between hookah use and IHD is likely to be similar to that for cigarette smoking, as the toxicant profiles of smoke from hookah and cigarettes have many similarities. 22 Both hookah and cigarette smoke contain polycyclic aromatic hydrocarbons, volatile aldehydes, and heavy metals which have been implicated to play a role in atherosclerosis. 23 They both contain carbon monoxide that can exacerbate angina and reduce arrhythmias threshold through impairment of oxygenation. 23,24 However, hookah may expose users to much more smoke compared to cigarettes, resulting in more toxicant exposure. 10 Burning charcoal (for heating tobacco) in hookah pipe results in much greater generation of the mentioned toxicants. 25 We found that opium use was associated with 1.7fold higher odds of having IHD. Opium use has been associated with increased risk of cancers of the esophagus, stomach, larynx, lung, and urinary bladder. 26 In the Golestan Cohort Study, a prospective study in northern Iran, the risk of death from IHD was about twice higher among opium users than non-users. 13 However, few other well-designed studies have investigated the association between opium use and IHD.
The mechanism of the association between opium use and IHD is unclear. In a case-control study, those with opium addiction had a decreased serum level of adiponectin level, which could result in insulin resistance and increased risk of CVD. 27 In a population-based cross-sectional study, opium addiction was a predictor for hyperhomocysteinemia, which may accelerate the atherosclerosis process. 28 In this population, those with higher SES had higher odds of having prevalent IHD, consistent with the 'third stage' of epidemiologic transition of cardiovascular diseases. 29 Improvement in socioeconomic status and urbanization in certain stages can result in reduced physical activity and an increase in unhealthy diet (e.g., fast foods) and social stress. As all people in our study population were covered by universal health insurance, the higher prevalence of IHD in higher SES groups is unlikely to be entirely related to longer survival due to having better access to care. As expected, IHD was associated with wellestablished IHD risk factors in our study. The observed association between anxiety and depression and IHD may not be causal, as people may be more likely to have stress after being diagnosed with IHD. More research is needed on the nature of these associations. Collection of detailed data on hookah and opium use and other risk factors of IHD, measurement of anthropometric indices and blood pressure by trained staff, and of fasting blood sugar and lipids by laboratory tests are among the strengths of this study. On the other hand, the diagnosis of IHD and data on hookah and opium use were based on self-reports. However, an earlier study showed acceptable reliability and validity for self-reports of opium by comparing self-reports with urine codeine and morphine in a similar setting in northern Iran. 30 Further, crosssectional studies may provide preliminary evidence for a causal association but they cannot prove it. Observing a non-significant dose-response association between cigarette smoking and the IHD prevalence may be also due to the cross-sectional nature of our data. Accordingly, we may interpret that there is no dose-response association between cigarette smoking and being an IHD patient. In other words, it is probable that IHD patients who smoke cigarette more than others did not survive or the patients quit or reduced their cigarette use after being diagnosed with the disease.
In conclusion, hookah and opium use were associated with prevalence of IHD in our study. It seems more research is needed on the nature of these associations, and whether reducing hookah and opium use can potentially improve population health.