Management of substance abuse

The health and social effects of nonmedical cannabis use

New WHO publication on cannabis


Chapter 7. Long-term cannabis use and noncommunicable diseases

7.1 What do we know?

7.1.1 Respiratory diseases

7.1.1.1 Chronic bronchitis

Considerable epidemiological and clinical research has assessed whether cannabis smoking is a risk factor for chronic obstructive pulmonary disease (COPD). The major symptoms of COPD are chronic bronchitis, chest tightness, increased cough and increased sputum most days of the year for two or more years. In most of these studies, cannabis-only smokers have been more likely to have reported cough, sputum and wheezing but no more likely to report shortness of breath than controls who do not smoke cannabis (Aldington et al., 2007; Bloom et al., 1987; Moore et al., 2005; Tan et al., 2009; Tashkin et al., 1987; Taylor et al., 2000).

In follow-up studies of habitual cannabis smokers, those who quit show reductions in cough, sputum and wheeze compared to those who continue to smoke cannabis. For example, a detailed analysis of a large, well-characterized cohort of nearly 1000 subjects followed from birth to age 38 years and assessed for respiratory symptoms at 18, 21, 26, 32 and 38 years, found strong associations between current cannabis use and morning cough, sputum production and wheeze over multiple assessments at different ages. Those who quit or very substantially reduced their cannabis smoking showed marked improvements in these symptoms (Hancox et al., 2015). Similar findings had been reported previously (Tashkin, Simmons & Tseng, 2012). Taken together, these and other findings suggest that the chronic bronchitis induced by cannabis smoke is reversible (Hancox et al., 2015; Tashkin, Simmons & Tseng, 2012).

A common finding in video bronchoscopy studies in cannabis-only smokers is swelling and oedema that blocks and partially occludes the bronchi (Roth, 1998). This is consistent with a modest increase in airway resistance that is of unclear clinical significance (Tashkin, 1987; Aldington, 2007; Hancox et al., 2010). Biopsies of bronchial mucosal tissue of cannabis-only smokers showed more replacement of the normal ciliated cells lining the airway with mucus-secreting and other cells than in nonsmokers (Roth et al., 1998; Fligiel et al., 1997). The reduction in ciliated cells, and subsequent increased mucus secretion from the larger number of mucus-secreting cells, probably explain the increased symptoms of chronic bronchitis in regular cannabis smokers (Tashkin, 2015).

7.1.1.2 Chronic obstructive pulmonary disease

COPD is a progressive disease that produces faster-than-normal annual age-related decreases in lung function in tobacco smokers. Studies have not found an increased risk of COPD in cannabis-only smokers. Chronic bronchitis occurs in the absence of COPD in most studies of cannabis smokers (e.g. Hancox et al., 2015; Kempker, Honig & Martin, 2015; Pletcher et al., 2012; Sherrill et al., 1991; Tashkin et al., 1980; Taylor et al., 2000). No impairments in respiratory function were observed in three of the four longitudinal studies of lung function in regular cannabis smokers (Hancox et al., 2010; Pletcher et al., 2012; Tashkin et al., 1997). Although cannabis smokers do not appear to be at greater risk of COPD, they lose lung function more quickly than nonsmokers, even if the rate of decline is slower than that in tobacco smokers.

The only abnormality found in lung function tests was a modest increase in airway resistance (Tashkin et al., 1987; Aldington et al., 2007; Hancox et al., 2010), which is probably due to oedema in the airways of cannabis-only smokers. In some studies, cannabis smokers have had larger lung volumes than non-cannabis smokers, probably because the deep inhalations used in smoking cannabis stretch the lung (Tashkin, 2015).

Tobacco smoke activates alveolar macrophages, the major immune cell located in the lungs. These macrophages release cytokines and chemokines that stimulate the release of tissue-damaging products that can cause chronic bronchitis and emphysema. Cannabis smokers, in contrast to tobacco smokers, show reduced macrophage activity, presumably because of the immunosuppressive activity of THC (Tashkin, 2015).

7.1.1.3 Other respiratory diseases

There have been case reports of bullous lung disease (pathologically enlarged air spaces in the lung parenchyma measuring more than 1 cm) in cannabis smokers who use varying amounts of tobacco (Johnson, 2000; Phan, Lau & Li, 2005; Hii et al., 2008). This disease could compromise lung function and predispose to pneumothorax, but the causal relationship to cannabis is unclear (Tan, Hatam & Treasure, 2006).

Several cases of Aspergillus and other forms of pneumonia have also been reported in immunocompromised cannabis smokers (Tashkin, 2015). Cannabis smoking impairs the function of alveolar macrophages, key immune-effector cells in the lung’s defence against infection (Baldwin et al., 1997). The loss of cilia and increased mucus-secreting cells in the airways of regular cannabis smokers (Fligiel et al., 1997) may impair mucociliary clearance, thereby increasing the risk of pneumonia. Cannabis use could predispose to pneumonia as a consequence of cannabis-related impairment in the lung’s defences against infection. Since cannabis has also been shown to be frequently contaminated with Aspergillus fumigatus (Kagen, 1983) and potentially pathogenic gram-negative bacteria (Ungerleider, 1982), the introduction of these microorganisms into the lung by smoking provides an additional mechanism whereby cannabis could increase the risk of pneumonia. Well-designed epidemiological studies are needed to investigate this risk.

7.1.2 Cardiovascular diseases

One of the most reliable signs of cannabis intoxication is tachycardia or elevated heart rate (Chesher & Hall, 1999; Jones, 2002; Sidney, 2002). The CB1 and CB2 cannabinoid receptors are both found in the cardiovascular system (Montecucco & Di Marzo, 2012). Young daily cannabis users in laboratory studies develop tolerance to these effects within 2-4 weeks (Jones, 2002). Middle-aged men with a history of myocardial infarction who smoke cannabis can experience acute symptoms of angina; such cases were reported in the literature as early as in the 1970s (Gottschalk, Aronow & Prakash, 1977). Furthermore, cannabis has been shown to trigger, earlier than does tobacco, the occurrence of angina pectoris symptoms after physical effort among patients with a history of coronary disease or stable angina pectoris (Aronow & Cassidy, 1974)

There is a limited number of epidemiological studies of CVD in cannabis smokers (Sidney, 2002). Mittleman et al. (2001) found that the risk of myocardial infarction was four times higher in patients with a recent myocardial infarction in the hour after smoking cannabis, and then rapidly declined. The authors note that this risk was much less than that of cocaine (nearly 24 times in the first hour after use) (Mittleman et al., 1999).

A prospective study of 1913 adults found a dose-response relationship between cannabis use and cardiovascular mortality over 3.8 years (Mukamal et al., 2008). The risk was 2.5 times higher in those who used cannabis less than weekly and 4.2 times higher among those who used it weekly or more often. Cannabis use was not significantly associated with long-term mortality in patients from the same cohort after 18 years follow-up, although mortality rates were systematically higher in cannabis users than in non-users (Frost et al., 2013).

Recent case reports and case series suggest that cannabis smoking may increase the risk of CVD in younger cannabis smokers who are otherwise at relatively low risk (Arora et al., 2012; Bailly et al., 2010; Basnet, Mander & Nicolas 2009; Canga et al., 2011; Casier et al., 2014; Deharo, Massoure & Fourcade, 2013; Duchene et al., 2010; Hodcroft, Rossiter & Buch, 2014; Karabulut & Cakmak, 2010; Kocabay et al., 2009; Pratap & Korniyenko, 2012; Renard et al., 2012). Jouanjus et al (2014) reported that there were 35 cases of CVD in French cannabis users reported by health providers to the national Addictovigilance network between 2006 and 2010. These complications occurred in patients with a mean age of 34 years and, on the basis of their clinical history, could be attributed to cannabis use. The authors used capture-recapture measures to estimate the rate of reporting that resulted in 0.4% (Jouanjus et al., 2012). On the basis of this result, the 35 cases reported should be regarded as a considerable underestimate of the number of CVD that could have occurred among young cannabis users in France during the study period. During this study period, it is noteworthy that the portion of cannabis-related cardiovascular complications increased by a factor of three.

A third of cannabis-related hospitalizations in Toulouse, France, were attributed to CVD. These included seven myocardial infarctions, four cerebral strokes and three cases of juvenile thromboarteritis (Jouanjus et al., 2011). In another study of hospitalizations for CVD in young people (aged 15-30 years), cannabis was involved in 18 of 20 cases that involved a psychoactive drug. Many of these myocardial infarctions occurred in young cannabis users with few risk factors for CVD. Overall the data in the literature report that normal arteries were generally found in coronary angiographies, suggesting that vasospasm could be responsible for these events. Besides, cannabis smoking increases CVD risk by increasing carboxyhaemoglobin levels (Wu et al., 1988). As a result, the signal of an increasing risk of serious cannabis-related cardiovascular disorders was identified and confirmed in France. The cardiovascular complications observed in young cannabis users differed from those presented by non-using patients of the same age.

The impact of second-hand cannabis smoke on vascular endothelial function has recently begun to be discussed and examined. This research (Xiaoyin et al., 2014) concluded that cannabis and tobacco smoke impair endothelial function similarly under comparable exposure conditions. It also concluded, as have several other studies, that it is the smoke and not the THC that causes the impairment.

7.1.2.1 Stroke

In recent decades, the incidence of stroke in young adults has increased, as have case reports of stroke in cannabis smokers (Wolff et al., 2013). In 2013, Wolff found only 59 cases of cannabis-associated strokes in the literature. These were ischaemic strokes or transient ischaemic attacks that occurred in persons with a mean age of 33 years. By 2015, around 100 cases of cannabis-related ischaemic stroke had been reported (Wolff et al., 2013; Desbois & Cacoub, 2013; Hackam, 2015; Wolff et al., 2015). Some case-control studies also suggested that cannabis smoking was a risk factor for stroke in young adults (Barber et al., 2013), and at least five cases of ischaemic stroke have been reported in persons using synthetic cannabinoids (Benson-Leung, Leung & Kumar, 2013; Freeman et al., 2013; Takematsu et al., 2014).

Wolff (Wolff et al., 2011), for instance, described a 21-year-old male who had a cerebellar ischaemic stroke after smoking cannabis. Wolff argued that this stroke could be attributed to multifocal intracranial arterial stenosis because: (a) the patient had a normal cranial angiogram six months before he started smoking cannabis; (b) his cerebral arteries were stenosed at the time of the stroke; and (c) the stenoses were reversed after three months of abstinence from cannabis use. Cannabis-associated strokes usually occur in chronic or current cannabis users who smoke tobacco (Wolff et al., 2013). The stroke often occurs while the drug is being smoked, or minutes afterwards. There are several case reports of a recurrence of strokes in patients who did not stop using cannabis (Wolff et al., 2013). The cardiovascular effects of cannabis provide possible mechanisms for these strokes – namely, orthostatic hypotension, altered cerebral vasomotor function, supine hypertension and swings in blood pressure, cardio-embolism, vasculopathy, vasospasm and reversible vasoconstriction cerebral syndrome (Wolff et al., 2013). Furthermore, a French study of young adults (under the age of 45 years) who had an ischaemic stroke over a two-year period found that 13 of 48 were cannabis users. In 10 of the 13, the cause of stroke was multifocal intracranial arterial stenosis (Wolff et al., 2011). There was a reversal of cerebral vasoconstriction (documented by vascular imaging three months after diagnosis) in all patients who stopped using cannabis. This suggests that cannabis use can produce ischaemic stroke in young adults by inducing reversible cerebral vasoconstriction.

A five-year follow-up of cases of reversible cerebral vasoconstriction syndrome (RCVS) in 159 ischaemic strokes in young patients found RCVS to be the cause of 13% of these strokes, most often in men with a mean age of 32 years. In 67% of these cases the precipitant was smoking cannabis resin. The cerebral vasoconstriction resolved within 3-6 months if patients abstained from smoking cannabis (Wolff et al., 2015). The cerebral vasoconstriction induced by cannabis is a possible mechanism for these strokes (Wolff et al., 2015).

7.1.3 Cancer

THC and other cannabinoids are not carcinogens in microbial assays (MacPhee, 1999; Marselos & Karamanakos, 1999) or tests using rats and mice (Chan, 1996). However, cannabis smoke is carcinogenic in these assays (MacPhee, 1999; Marselos & Karamanakos, 1999; Leuchtenberger, 1983). This suggests that cannabis smoking could, like cigarette smoking, be a cause of cancers of the lung, the upper aerodigestive tract (mouth, tongue, oesophagus) and bladder (MacPhee, 1999). This could be true because there is a strong qualitative similarity between the carcinogens found in cannabis and tobacco smoke (Institute of Medicine, 1999; Van Hoozen & Cross, 1997). The existing case reports raise a suspicion, but provide limited support for the hypothesis, that cannabis use can cause upper respiratory tract cancers. The quality of the case reports is insufficient as they do not compare rates of cannabis use in cases and controls; rather, they assess cannabis exposure retrospectively, in the knowledge that the user has cancer, and they do not control for confounding factors such as alcohol and tobacco use (Hall et al., 2002).

7.1.3.1 Upper aerodigestive tract cancers

The evidence on the risks of upper area digestive tract cancers in cannabis smokers is not consistent. Two studies have shown an increased risk (Zhang et al., 1999; Feng et al., 2009), two have shown a decreased risk (Liang, 2009; Zhu et al., 2002), and five have found no association (Aldington et al., 2008a; Hashibe et al., 2006; Llewellyn, Johnson & Warnakulasuriya, 2004; Llewellyn et al., 2004; Rosenblatt et al., 2004). Pooled analyses have not found an overall association for head and neck cancer (Berthiller et al., 2008), but there is a possible increased risk of oropharyngeal cancer and a decreased risk of oral and tongue cancers (Marks et al., 2014). Three studies on the relationship between human papillomavirus (HPV) and cannabis and the risk of head and neck cancer suggest that HPV (which is a strong risk factor for oropharyngeal cancer) may be a modifying risk factor (Gillison et al., 2008; Liang et al., 2009; Marks et al., 2014). Future research on upper aerodigestive tract cancers needs to take into account the effects of concomitant alcohol and tobacco use as well as HPV infection.

7.1.3.2 Respiratory cancers

The Swedish conscript study (Callaghan, Allebeck & Sidorchuk, 2013) found a doubling of lung cancer rates among conscripts who had smoked cannabis 50 or more times by the age of 18 years. However, the study could control only for baseline tobacco use. Case-control studies of lung cancer in North Africa have found consistent associations but in all these studies cannabis smoking has been confounded by cigarette smoking (Mehra et al., 2006). A Tunisian case-control study of 110 cases of hospital-diagnosed lung cancer and 110 community controls found an association with cannabis use (OR = 8.2) that persisted after adjustment for cigarette smoking. A pooled analysis of three Moroccan case-control studies also found an elevated risk of lung cancer among cannabis smokers, but these cannabis users also smoked tobacco (Berthiller et al., 2008). A New Zealand case-control study of lung cancer in 79 adults under the age of 55 years and 324 community controls (Aldington et al., 2008b) found a dose-response relationship between frequency of cannabis use and lung cancer risk. Yet a collaborative pooled analysis of these epidemiological studies found no overall association between cannabis smoking and lung cancer and no evidence of a dose-response relationship (Zhang et al., 2015).

7.1.3.3 Testicular cancer

Three studies have found an association between cannabis smoking and testicular cancer. All are USA studies published since 2009. One of these, Daling et al. (2009), reported a case-control study of cannabis use among 369 men diagnosed with a testicular germ cell tumour and 979 age-matched controls. They found a higher rate of cannabis use among cases (OR = 1.7 [95% CI: 1.1, 2.5]). The risk was higher for a nonseminoma (OR = 2.3 [95% CI: 1.4, 4.0]) and was higher in those who used cannabis before the age of 18 years and used cannabis more often than weekly. These findings have since been replicated in two further case-control studies (Lacson et al., 2012; Trabert et al., 2011). Another case study published by of the University of Southern California in Los Angeles found that men with these tumours were about twice as likely to have a history of using cannabis.

A meta-analysis of these studies (Gurney et al., 2015) found an odds ratio of 1.5 for high frequency cannabis users and an odds ratio of 1.5 for those who had used cannabis for 10 or more years. This is a moderate and consistent relationship and, because tobacco smoking is not a cause of testicular cancer, there is no potential confounding by tobacco smoking. Cannabinoid receptors are found in the testes so a causal connection is not implausible.

7.1.3.4 Other cancers

With cannabis use, there is a trend towards increased prostate cancer (i.e. 3-fold risk), and cervical cancer (i.e. 1.4-fold risk). An elevated risk of prostate cancer was reported among cannabis smokers in an 8.6-year follow-up of 64 855 members of the Kaiser Permanente Medical Care Program (Sidney et al., 1997). Males who smoked cannabis had an increased risk of prostate cancer, as did males who were current cannabis smokers (Sidney et al., 1997). Confounding by other lifestyle factors was a possible explanation of the finding because AIDS-related deaths were higher among cannabis users in this study

Smaller sized studies have implicated cannabis use in the development of bladder cancer and testicular germ cell tumours. The reasons for the great heterogeneity in epidemiological studies correlating cannabis use and cancer may be related to difficulties in quantifying cannabis use, unmeasured confounders in the cases or controls, and variable expression of cannabinoid receptors in target tissues.

Cannabis smoking during pregnancy has been associated with cancers among children. Three case-control studies for different cancers have found an association (Robinson et al., 1989; Grufferman, 1993; Kuijten, 1992). Each study examined cannabis use as one of many other risk factors for these cancers and there was no a priori reason to expect a relationship between cannabis use and the cancers. None of these results have been replicated. The incidence of these cancers did not increase over the period 1979-1995 in the USA when cannabis use was common (Reis et al., 2000; Smith et al., 2000; Gurney et al., 2000).

7.1.4 Areas that require more research

In view of the high prevalence of cannabis use worldwide, the French finding of an increasing risk of serious cannabis-related cardiovascular disorders should be explored in other countries.

Among the possible directions for future research are: a) to identify the mediators involved in the occurrence of the cardiovascular effects of cannabinoids; b) to assess the role of risk factors such as pre-existing CVD; and iii) to evaluate the potential therapeutic value of medical cannabinoids in preventing myocardial infarction.

·       To collect more data on the increasing risk of CVD, health providers should be encouraged to investigate cannabis exposure systematically, including from second-hand smoke, in young patients presenting with cardiovascular disorders and patients receiving treatment with pharmaceutical preparations of cannabinoids.

  • Better estimates of the rate of ischaemic stroke in young cannabis users are needed. Health providers should systematically asked about cannabis use, and especially the use of cannabis resin, when treating young patients with ischaemic stroke.
  • More research is needed on the effects of cannabis on cerebral arteries in patients who develop a stroke. It is imperative that physicians ask young people with ischaemic strokes about their drug use, do arterial imaging to search for intracranial arterial stenosis, and evaluate the reversibility of the vascular abnormalities.
  • Larger cohort and better designed case-control studies are needed to better control for the effects of cigarette smoking in order to clarify lung, head and neck, prostate and testicular cancer risks among long-term regular cannabis smokers (Hashibe et al., 2005).
  • Future research on upper aerodigestive tract cancers needs to take into account the effects of concomitant alcohol and tobacco use as well as HPV infection.
  • Studies are needed to compare the effects of smoking with vaporizing and ingestion of cannabis on the occurrence of cancer in various tissues, and to account for levels of THC and other cannabinoids.

Cannabis smoking impairs the function of alveolar macrophages, key immune-effector cells in the lung’s defence against infection. The loss of cilia and increased mucus-secreting cells in the airways of regular cannabis smokers may impair mucociliary clearance and thereby increase the risk of pneumonia.

  • Well-designed epidemiological studies are needed to investigate this risk and also the impact of cannabis smoking on persons other than the smoker (by environmental smoke).