Most clinical reports imply that opium has no remarkable impact on blood glucose (Table 1) whether in non-diabetic (Asgary et al., 2008; Roohafza et al., 2013; Javadi et al., 2014; Masoomi et al., 2015; Sanli et al., 2015) or in diabetic individuals (Karam et al., 2004; Hosseini et al., 2011; Rezvanfar et al., 2011; Najafi and Sheikhvatan, 2012a; Bayani et al., 2014; Rahimi et al., 2014). In some of these studies, glycated hemoglobin (HbA1c), as an indicator of poor control of blood glucose, was significantly higher in addicted groups (Karam et al., 2004; Asgary et al., 2008). Another report on the addicted subjects attempting opium abstention showed that fasting blood glucose (FBS) did not change considerably 3 months after opium withdrawal (Mahmoodi et al., 2012). A study on heroin and methadone addicts indicated that FBS was similar with control group, but insulin was significantly higher in the addicted groups. After oral or intravenous glucose load, plasma glucose was significantly higher and insulin response was significantly lower in the addicted groups. Authors suggested that insulin response may be impaired either due to the functional failure of stressed β-cells or by altering of its action on target tissues (Ceriello et al., 1987). In another research, after the subjects had a standard meal, FBS was not different between the heroin addicts and the control group. C-peptide immunoreactivity, however, was lower and insulin immunoreactivity was higher in the addicts. It was concluded that heroin addiction may cause β-cell failure and contemporaneously produce hyperinsulinemia due to the alteration of the rate of hepatic extraction of insulin (Zandomeneghi et al., 1988). In one study in opium addicts, FBS was significantly higher compared to that of the non-addicts (Gozashti et al., 2015). In this report, the increase of FBS was accompanied with a significant decrease in the fasting blood insulin.

In contrast, several reports have indicated that opium has a positive impact on the fall of FBS. In a study on diabetic patients, the opium addiction significantly decreased FBS and 2HPP (glucose concentration 2-h post prandial) compared to the non-addicted patients, but HbA1c did not alter (Azod et al., 2008). The reduction of 2HPP might be due to the decrease of gastric emptying due to opioid μ-receptor activation and thereafter delaying the intestinal glucose absorption (Azod et al., 2008). Several studies on non-diabetic subjects also showed lower levels of FBS in opium addicts compared to the non-addicts (Masoomi et al., 2008; Shirani et al., 2010; Dehghani et al., 2013; Afarinesh et al., 2014; Aghadavoudi et al., 2015); yet, FBS in the heroin addicts was similar to that of the non-addicts (Divsalar et al., 2010).

The findings regarding the prevalence of diabetes are controversial as well (Table 1). Some studies have demonstrated the lower rate of diabetes among opium addicts compared to the non-addicts (Masoomi et al., 2008; Shirani et al., 2010; Najafi and Sheikhvatan, 2012b; Dehghani et al., 2013; Aghadavoudi et al., 2015). One study showed the opposite result (Yousefzadeh et al., 2015), while the data gathered from some other studies have shown no relation between opium addiction and the rate of diabetes mellitus (Davoodi et al., 2005; Javadi et al., 2014; Masoomi et al., 2015). In a more recent population-based study by the authors of this article on 5900 individuals, opium consumption had no significant association with diabetes (Najafipour et al., 2015a), yet the addicted people had a higher rate of uncontrolled diabetes (Najafipour et al., 2015b).

In none of the animal studies, opium or opioids decreased blood glucose (Table 1). In one study on normoglycemic mice, there was no significant difference in the blood glucose between the opium addicted group and the control group (Mohammadi et al., 2012). Moreover, in one study on diabetic rats, regarding serum glucose, there was no significant difference between the opium treated and the control group (Sadeghian et al., 2009). Mami et al. showed that in opium addicted rabbits, serum FBS was significantly higher compared to that of the non-addicted ones (Mami et al., 2011). In one study on conscious dogs, intravenous infusion of morphine had dual effects, low dose (2 mg/h) did not change blood glucose, but high doses (8.16 mg/h) caused hyperglycemia subsequent to the elevation of glucose production in the liver and reduction of glucose clearance in peripheral tissues. These effects were attributed to the increased levels of epinephrine, glucagon, and cortisol in plasma (Radosevich et al., 1984). In a similar study done on conscious rabbits, the administration of 0.3 and 3 mg/kg morphine produced the same results. The authors suggested that hyperglycemia made by high doses might be a response to the secretion of adrenaline (May et al., 1988). In the study of Molina et al on rats, central infusion of morphine significantly raised the blood glucose. These data indicated that hyperglycemic effects of morphine were mediated through central nervous system (CNS), and were led to increased hepatic glucose production secondary to glycogenolysis and in part due to the decrease of insulin release (Molina et al., 1994). Likewise, single intravenous injections of morphine (0.5 mg/kg) resulted in more hyperglycemia in diabetic conscious dogs compared to the normal dogs. This effect of morphine was partly attributed to the increase of glucagon secretion (Ipp et al., 1980). Furthermore, in intra-peritoneally opium injected rats, blood glucose increased gradually between 30 and 120 min after the injection of opium, compared to that of the control group (Asadi et al., 2008). In one study, the orally methadone therapy in the rats for a 35-days period by increasing doses (0.5–1.8 mg/kg) raised blood glucose significantly. In this study, blood glucose was returned back to its normal level 30 days after the withdrawal. In addition, while being exposed to methadone, the animals showed impairment in glucose tolerance test (Sadava et al., 1997).

There is a disparity regarding the impact of opium on lipid indices in humans (Table 2). Several studies on healthy subjects and on individuals who suffered from diabetes mellitus, acute myocardial infarction (AMI) and coronary artery disease (CAD) have indicated no significant relations between opium addiction and triglyceride (TG), total cholesterol (TC), low density lipoprotein (LDL), and high density lipoprotein (HDL) levels (Azod et al., 2008; Shirani et al., 2010; Najafi and Sheikhvatan, 2012a; Roohafza et al., 2013; Afarinesh et al., 2014; Bayani et al., 2014; Javadi et al., 2014; Masoomi et al., 2015; Sanli et al., 2015).

On the other hand, some studies have shown that opium addiction has a harmful effect on one or more lipid parameters. One study on diabetic patients indicated that HDL significantly decreased, or prevalence of low HDL increased in opium addicted compared to non-addicted patients. Yet, other lipids were not different between the two groups (Rahimi et al., 2014). In another study on diabetic patients, HDL was diminished only in men not in women (Karam et al., 2004). Asgary et al. who carried out a study on male cigarette smokers also had the same results. They showed that the ways of opium consumption (smoking or ingestion) had no impact on its outcomes; however, the duration of addiction of more than 2 years significantly augmented LDL (Asgary et al., 2008). Furthermore, in a study on patients who were candidates for coronary artery bypass grafting, the levels of LDL and TG were significantly higher in opium addicts than non-addicts (Aghadavoudi et al., 2015). In another study, it has been shown that the serum levels of TG, TC, and LDL in opium addicted individuals were significantly higher (Salman et al., 2010) or heroin addicts had lower TC and TG, and higher VLDL than the non-addicts (Akif et al., 2013).

Maccaria et al. indicated that in heroin addicts compared to the control group, TG increased while TC and HDL decreased (Maccaria et al., 1991). Furthermore, the frequency of lower cholesterolemia and low HDL were higher in the addicted group than the control. The linear regression analysis showed an inverse correlation between TC and alanine aminotransferase (ALT), an enzyme found mostly in liver damaged cells. Accordingly, the researchers suggested that lower cholesterol might be due to liver disease which is common in heroin addicts (Maccaria et al., 1991).

Unlike the above studies which demonstrated that opium addiction has no beneficial effect on plasma lipids, some data have shown the beneficial effect of opium addiction on such factors. In two studies on diabetic opium users, the level of serum TG was significantly lower than the diabetic non-users, but this was not different with respect to other lipids (Hosseini et al., 2011; Rezvanfar et al., 2011). In three studies on patients with AMI, similar results were obtained (Davoodi et al., 2005; Masoomi et al., 2008; Dehghani et al., 2013). Likewise, Fatemi et al. compared male opium addicts with male non-addicts. They observed no significant difference in TG, LDL, and HDL levels of the two groups, but the total cholesterol level was significantly less in the opium addicts (Fatemi et al., 2008). This was attributed to the lower BMI in the opium addicts. Gozashti et al. showed that TC and HDL significantly decreased in opium addicts, but TG remained unchanged (Gozashti et al., 2015). Divsalar et al. indicated that TC was significantly lower in opium and heroin addicts compared to that of the non-addicts (Divsalar et al., 2010).

The prevalence of dyslipidemia has also been reported differently in various studies, so that in some, no association was found between opium addiction and prevalence of dyslipidemia (Hosseini et al., 2011; Najafi and Sheikhvatan, 2012a; Bayani et al., 2014; Javadi et al., 2014; Najafipour et al., 2015a). In a few researches, the prevalence of dyslipidemia was lower (Maccaria et al., 1991; Dehghani et al., 2013; Rahimi et al., 2014), and in some others, the prevalence was higher in opium addicts (Aghadavoudi et al., 2015). Overall, although opium may reduce blood lipids in some cases, the majority of human studies have demonstrated that opium consumption either is ineffective or has unfavorable impact on serum lipids.

In comparison to clinical studies, more consistency exists among data of animal studies. Accordingly, in most of instances, opiates did not have any useful impact on plasma lipids (Table 2). In a study on STZ-induced diabetic rats, the lipid profiles were similar in the opium treated and the control groups (Sadeghian et al., 2009). In addition, in another study on hamsters, opium administration had no significant influence on plasma lipids; however, opium augmented the effect of alcohol in increasing TC, TG, LDL, VLDL, liver enzymes, and atherogenic index. Opium raised some oxidative stress indices as well (Mohammadi et al., 2013). In another study on rats fed with cholesterol enriched diet, the implantation of a morphine pellet (75 mg) for a 5-days duration raised serum levels of TC, LDL, VLDL, and aortic cholesterol content, and it decreased HDL level. Moreover, in normal diet rats, morphine elevated TC, LDL, and VLDL. All of these effects were prevented by application of opioid antagonist naltrexone (Bryant et al., 1987). In accordance with these data, the oral treatment of mice by morphine during a 28-days period increased liver, aorta and plasma TC levels (Bryant et al., 1988). In another study, hypercholesterolaemic rabbits were exposed to opium smoke for a short period (3 days) and a longer period (28 days). Of these schedules, long term opium smoking diminished the HDL level and caused a trend of increase in LDL (Najafipour et al., 2010). In one study on rats, acute morphine administration resulted in significant increase of the level of plasma TG compared to the control group, but the other lipids were unchanged (Al Sagair, 2005). Mami et al. observed that the oral opium treatment of rabbits for 2 months elevated serum levels of TC, TG, LDL and liver enzymes compared to the control group in spite of the fact that weight gaining and abdominal fat were lower in the opium addicts (Mami et al., 2011). The data of another study showed that opium consumption in rabbits fed with cholesterol-rich diet (for 3 months), significantly intensified the impact of high cholesterol diet on the increase of plasma levels of TG, TC, and LDL. Yet, in the addicted rabbits without cholesterol-rich diets, it decreased the plasma concentrations of TC and LDL (Mohammadi et al., 2009). In another study conducted by Mohammadi et al., it was shown that the mice fed with opium for 1 month had significantly lower total cholesterol compared to the control group, but the two groups had no difference with respect to other lipids (Mohammadi et al., 2012).

In human studies (Table 3), the intravenous injection of morphine via central sympathetic inhibition (sympatholysis) produced vasodilatation (Mansour et al., 1970), but in some studies, there was no significant difference between opium dependents and non-dependents with respect to blood pressure (Roohafza et al., 2013; Aghadavoudi et al., 2015). Furthermore, in a study on diabetic patients undergoing coronary angiography before matching of opium users with non-users for age, sex, and smoking status, the results showed that systolic and diastolic blood pressure were lower in the opium users. Yet after matching, no significant differences was observed between the two groups (Hosseini et al., 2011).

The data on the prevalence of hypertension are different (Table 3). Most studies indicated that opium addiction had no significant relation with the rate of hypertension (Davoodi et al., 2005; Masoomi et al., 2008, 2015; Hosseini et al., 2011; Najafi and Sheikhvatan, 2012a,b; Roohafza et al., 2013; Bayani et al., 2014; Javadi et al., 2014; Rahimi et al., 2014). In addition, in a more recent population-based study by the authors of this article on 5900 adult individuals, opium use had no significant association with hypertension in either occasional or dependent users (Najafipour et al., 2014, 2015a). One study revealed that hypertension was significantly more prevalent in opium users than non-users (Yousefzadeh et al., 2015) while some studies showed the opposite results (Shirani et al., 2010; Dehghani et al., 2013; Aghadavoudi et al., 2015).

In anesthetized rats, intravenous injection of the first dose of morphine (0.025–100 mg/kg) led to fall in their blood pressure, which was transient, and BP immediately returned to near normal (Table 3). This reduction was ascribed both to vagal bradycardia and the decrease in sympathetic tone. After the first and the second injection however, in dose of 100 mg/kg, the acute phase was followed by slow falling of BP (Fennessy and Rattray, 1971). In another study on spontaneously hypertensive rats, the infusion of morphine lessened BP via vasodilatation. The response was reversed by the injection of naloxone methiodide. It was shown that the effectis mediated by peripheral opioid receptors (Bądzyńska et al., 2016). Furthermore, in anesthetized dogs, intravenous administration of two opioid drugs (fentanyl and dextromoramide) resulted in hypotension and bradycardia, which were attributed to a reduced sympathetic tone (Laubie et al., 1974). Furthermore, the subcutaneous injection of morphine in cats or injection in to cisterna magna of their brain caused hypotension and bradycardia due to the inhibition of sympathetic tone (Feldberg and Wei, 1986). However, the rats treated by Intra-cerebroventricular (i.c.v) infusion of morphine or morphine-6-glucuronide showed no change in their arterial blood pressure (Molina et al., 1994). In one study, oral chronic treatment of rats with morphine prevented the elevation of blood pressure subsequent to social deprivation-induced stress. The authors suggested that chronic treatment by morphine caused the desensitization of opioid receptors (Jimenez and Fuentes, 1993). Conversely, in conscious rabbits, the intravenous administration of morphine raised BP via increasing sympathetic activity and serum adrenaline. These effects were blocked by naloxone (May et al., 1988). In a study on rabbits, short (3 days) and long-term (4 weeks) exposure of the animals to opium smoke caused a continuous trend of increase in blood pressure during 4 weeks (Najafipour et al., 2010).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.