Please note the information in this publication may no longer be current but is retained on our website for historical or research purposes.
<< Previous Chapter | Contents | Next Chapter>>
6. The chronic effects of cannabis use on health
6.4 Effects on the respiratory system
The most reliable acute effect of exposure to cannabis smoke is bronchodilation (National Academy of Science, 1982), which has principally been of interest because of its possible therapeutic effect upon asthma (see below pp193-194). Other than bronchodilation, it has proved difficult to demonstrate any effects of acute cannabis smoking on breathing "as measured by conventional pulmonary tests" (National Academy of Science, 1982, p58).
The major concerns about the respiratory effects of cannabis use have been the possible adverse effects of chronic, heavy cannabis smoking (Tashkin, 1993). The two largest issues of concern have been the production of chronic bronchitis as a precursor of irreversible obstructive lung disease, and the possible causation of cancers of the aerodigestive tract (including the lungs, mouth, pharynx, larynx, and trachea) after 20 to 30 years of regular cannabis smoking. These risks are the primary focus of this section of the review.
There is good reason to expect that chronic heavy cannabis smoking may have adverse effects upon the respiratory system (Tashkin, 1993). Cannabis smoke is similar in constitution to tobacco smoke, and contains a substantially higher proportion of particulate matter and of some carcinogens (e.g. benzpyrene) than does tobacco smoke (Leuchtenberger, 1983; National Academy of Science, 1982). Hence, the inhalation of cannabis smoke deposits irritating and potentially carcinogenic particulate matter onto lung surfaces. Cigarette smoking is known to cause diseases of the respiratory system, such as bronchitis, emphysema, and various forms of cancer affecting the lung, oral cavity, trachea, and oesophagus (Holman et al, 1988). Although tobacco smokers smoke many more cigarettes than cannabis smokers, cannabis smoke is typically inhaled more deeply, and the breath held for longer, than tobacco smoke, thereby permitting greater deposition of particulate matter on the lung surface (Hollister, 1986). It therefore seems a reasonable inference that chronic daily cannabis smoking may cause diseases of the respiratory system.
Despite the reasonableness of this hypothesis, it has nonetheless been difficult to investigate the contribution of chronic heavy cannabis smoking to diseases of the respiratory system (Huber et al, 1988; National Academy of Science, 1982). A major problem is that most marijuana smokers also smoke tobacco, which makes it difficult to disentangle the effects of cannabis from those of tobacco smoking. The problems in quantifying current and lifetime exposure to cannabis, because of variations in quality and potency, make it difficult to examine dose-response relationships between cannabis use and the risk of developing various respiratory diseases. There is also likely to be long latency period between exposure and the development of these diseases, especially in the case of cancers of the aerodigestive tract. This period is approximately the length of time since cannabis smoking became widespread in Western societies. There are also technical difficulties in designing studies which are sufficiently sensitive to detect increased risks of diseases arising from relatively rare exposures, such as chronic daily cannabis use.
Top of page
6.4.1 Bronchitis and airways obstruction
There is a small clinical literature containing case reports of acute lung diseases among heavy cannabis smokers in the US military stationed in West Germany during the early 1970s, when hashish was cheap and freely available (Henderson et al, 1972; Tennant et al, 1971). Tennant et al studied 31 soldiers who had smoked 100g or more of hashish monthly for six to 21 months, 21 of whom were also tobacco smokers. Nine complained of bronchitis which had its onset three to four months after they began to smoke hashish. Pulmonary function tests of five cases (two of whom did not smoke tobacco) revealed mild airflow obstruction that partially remitted after a reduction or cessation of hashish use. Tennant (1980) also reported histopathological studies of 23 of these patients in which all patients were found to have atypical cells of the type (squamous metaplasia in 21 cases) associated with chronic bronchitis and carcinoma of the lung.
Henderson et al (1972) reported on 200 servicemen who sought treatment for problems related to hashish use, 90 per cent of whom were also cigarette smokers. Twenty men who smoked large doses of hashish on a weekly basis presented with symptoms of chronic bronchitis, and on testing had vital capacity that was 15-40 per cent below normal. Six had a bronchoscopic examination which showed epithelial abnormalities. The interpretation of these findings was complicated by the fact that the majority of these hashish smokers were also tobacco smokers, as were Tennant et al's subjects, and there was no adequate comparison group.
The field studies of chronic cannabis smokers in Costa Rica (Carter et al, 1980) and Jamaica (Rubin and Comitas, 1975), which included comparison groups, have failed to support the clinical findings of Henderson et al, and Tennant et al. Neither of these studies found any statistically significant differences in lung function, or in the prevalence of respiratory symptoms, between chronic cannabis users and non-cannabis smoking controls. In both studies, however, the measures of respiratory function were relatively unsophisticated, the sample sizes were small, making it difficult to detect all but very large differences, and the comparisons were often confounded by a failure to control for tobacco smoking.
The most convincing evidence that chronic cannabis use may be a contributory cause of impaired lung function and symptoms of respiratory disease comes from a series of controlled studies which have been conducted by Tashkin and his colleagues since the mid-1970s. One of their early studies evaluated the subacute effects of heavy daily marijuana smoking on respiratory function. The subjects were young male marijuana smokers who were studied in a closed hospital ward where they were allowed ad libitum access to marijuana for 47 to 59 days. The results of lung function tests showed a statistically significant decrease in the function of large and medium-sized airways over the course of the study. The degree of impairment was positively correlated with the number of marijuana cigarettes smoked, suggesting that the quantity of inhaled irritants was the important factor, perhaps by producing an inflammatory reaction in the tracheobronchial epithelium. Although the impairment was apparently small and values were still within the normal range, these changes were of clinical significance. If continued over a year, for example, the rate of decline in lung function would be several times greater than the normal rate.
Tashkin and his colleagues (1987) subsequently recruited a volunteer sample of marijuana only smokers (MS, n=144), marijuana and tobacco smokers (MTS, n=135), tobacco only smokers (TS, n=70), and non-smoking controls (NS, n=97). A subset of these subjects were followed to examine changes in lung function, signs and symptoms of respiratory disease, and the occurrence of histopathological changes that may precede the development of carcinoma.
In the baseline observations of their cohort, Tashkin et al (1987) found significant differences in the prevalence of symptoms of bronchitis (such as cough, bronchitic sputum production, wheeze and shortness of breath) between all types of smokers (MS, MTS, TS) and controls. There were no differences between cannabis and tobacco smokers in the prevalence of these symptoms. Lung function tests showed significantly poorer functioning and significantly greater abnormalities in small airways among tobacco smokers (regardless of concomitant cannabis use) while marijuana smokers showed poorer large airways functioning than non-marijuana smokers (regardless of concomitant tobacco use). These findings suggest that "habitual smoking of marijuana or tobacco causes functional alterations at different sites in the respiratory tract, with marijuana affecting mainly the large airways and tobacco predominantly the peripheral airways and alveolated regions of the lung" (Tashkin et al, 1990, p67).
Follow-up studies of a subsample of this cohort have broadly supported the results of the cross-sectional baseline study, while providing more detail on some differences between marijuana and tobacco smoking in their effects on lung function (Tashkin et al, 1990). The first follow-up study was conducted two to three years after the baseline study. Approximately half of these subjects were retested and most remained in the same smoking categories as at baseline, namely, 40 of the 54 MTS, 60 of the 71 MS, 30 of the 32 TS, and 56 of 58 NS, respectively of those who were followed up.
The prevalence of bronchitic symptoms of cough, sputum, and wheeze was higher in all smoking groups than among non-smokers at both time one and time two, and there was no significant change in the respiratory status of any of the smoking groups from time one to time two when those individuals who ceased smoking were excluded. Substantially the same results were obtained when the subjects were followed up three to four years after initial assessment. In addition, there was evidence of an additive adverse effect of marijuana and cigarette smoking, in that the MTS group showed effects of both types of damage attributable to marijuana and tobacco smoking alone.
Tashkin and his colleagues (Fligiel et al, 1988; Gong et al, 1987) undertook histopathological studies of the lungs of a subsample of their cohort. Fligiel et al (1988) compared the bronchial morphology of males aged 25 to 49 years who were heavy smokers of marijuana only (n=30), marijuana and tobacco (n=17), tobacco only (n=15) and non-smoking controls (n=11). Bronchial biopsies were examined by pathologists who were "blind" as to their smoking status, and analyses were made of cellular inflammation. All subjects who smoked (whether cannabis, tobacco or both) showed more prevalent and severe histopathological abnormalities than non-smokers. Many of these abnormalities were more prevalent in marijuana smokers, and they were most marked in those who smoked both marijuana and tobacco.
These findings were especially striking because they were observed in young adults who did not have respiratory symptoms, and they occurred at a younger age on average in marijuana than tobacco smokers, despite the fact that the marijuana smokers smoked less than a quarter as many "joints" as the tobacco smokers smoked cigarettes. Fliegel et al concluded that "marijuana smoking may be as damaging or perhaps even more damaging to the respiratory epithelium than smoking of tobacco" (p46), and there was "a very good possibility ... that marijuana smoking combined with smoking of tobacco, leads to a more significant mucosal alteration than either of these substances smoked alone" (p47).
Evidence of inflammation was sought by examining the presence of alveolar macrophages, lymphocytes, neutrophils and eosinophils in the bronchial lavage of the same subjects. This examination revealed that marijuana and tobacco smoking induced an inflammatory cellular response in the alveoli, and that the combination of marijuana and tobacco smoking produced the largest inflammatory response, "implying an adverse effect of marijuana smoking on the lung that is independent of and additive to that of tobacco" (Tashkin et al, 1990, p74).
Additional research by Tashkin and his colleagues (Tashkin et al, 1988; Wu et al, 1988) suggests that the most likely explanations of the apparently greater toxicity of marijuana smoking are major differences in the topography of marijuana and tobacco smoking. Laboratory studies of the volume of inhaled smoke from tobacco and marijuana, and analyses of its particulate content, indicated that marijuana smokers inhaled a larger volume of smoke (40-54 per cent more), inhaled more deeply, took in more particulate matter per puff, and held their breath about four to five times longer, thereby retaining more particulate matter, and absorbing three times more carbon monoxide, than cigarette smokers (Wu et al, 1988).
Bloom et al (1987) have recently reported findings that broadly confirm those of Tashkin and his colleagues. Bloom et al conducted a cross-sectional study in a general population of the relationship between smoking "non-tobacco" cigarettes and respiratory symptoms and respiratory function. Their study sample was a community sample of 990 individuals aged under 40 years who were being followed as part of a prospective community study of obstructive airways disease. Subjects were asked about symptoms of cough, phlegm, wheeze and shortness of breath, and they were also measured on a number of indicators of respiratory function, including forced expiratory volume and forced vital capacity.
The prevalence of ever having smoked a "non-tobacco" cigarette was 14 per cent (the same as the prevalence of marijuana smoking in general population surveys), with 9 per cent being current smokers and 5 per cent ex-smokers. Non-tobacco smokers were younger and more likely to be male than non-smokers of non-tobacco. The mean frequency of current non-tobacco smoking was seven times per week, and the average duration of use was nine years. Non-tobacco smokers were more likely than non-tobacco non-smokers to have smoked tobacco, and more likely to inhale deeply than tobacco smokers.
Non-tobacco smoking was related to the prevalence of the self-reported respiratory symptoms of cough, phlegm, and wheeze, regardless of whether the person smoked tobacco or not. There were also mean differences in forced expiratory volume and forced vital capacity, with those who had never smoked having the best functioning, followed in decreasing order of function by current cigarette smokers, current non-tobacco smokers, and current smokers of both tobacco and non-tobacco cigarettes. Non-tobacco smoking alone had a larger effect on all flow indices than tobacco smoking alone, and the effect of both types of smoking was additive.
Although there were some inconsistencies between the studies of Tashkin and colleagues and those of Bloom and colleagues, there is reasonable coherence in the available evidence on the respiratory effects of cannabis use. Taken as a whole, it suggests that chronic cannabis smoking increases the prevalence of bronchitic symptoms, reduces respiratory function, and in very heavy smokers produces histopathological changes that may portend the subsequent development of bronchogenic carcinoma, a well known consequence of heavy tobacco smoking. Although, "there is still no conclusive evidence in man of clinically important pulmonary dysfunction produced by smoking marihuana" (Huber et al, 1988; p8), it is nonetheless a reasonable inference that chronic heavy cannabis smoking probably increases the risk of developing respiratory tract cancer, and possibly influences the development of irreversible obstructive pulmonary disease. Persons who wish to reduce their risks of developing these diseases would be wise to desist from cannabis smoking (Tashkin, 1993).
Top of page
6.4.2 Cancers of the aerodigestive tract
Although "not a single case of bronchogenic carcinoma in man has been directly attributable to marijuana" (Tashkin, 1988), it would be unwise to infer from the absence of such cases that there is no such an effect (Huber et al, 1988; National Academy of Science, 1982). There is a 20 to 30-year latency period between the initiation of regular smoking and the development of cancer, and cannabis smoking only became widespread in Western societies in the early 1970s (National Academy of Science, 1982). There has also been a lack of clinical and epidemiological research on this question. Patients with lung or of other types of cancer, for example, have rarely been asked about their cannabis use as part of the clinical history-taking. No cohort or case-control studies of cancers among cannabis smokers have been reported, because the illegality of cannabis has made it difficult to obtain reliable information on habits of the large samples required, while the proportion of cannabis users who become long-term heavy users is likely to be small (Huber et al, 1988).
Despite the absence of such evidence, there are good reasons for suspecting that cannabis may contribute to the development of lung cancer and cancers of the aerodigestive tract (the oropharynx, nasal and sinus epithelium, and the larynx). A major reason is the similarity between the constituents of cannabis and tobacco smoke, an accepted cause of cancers in these organs (Doll and Peto, 1980; International Agency for Research on Cancer, 1990). The major qualitative differences between tobacco and cannabis smoke are the presence of cannabinoids in cannabis smoke and of nicotine in tobacco. There are also some quantitative differences in the amount of various carcinogens with cannabis smoke typically containing higher levels than tobacco smoke (Leuchtenberger, 1983; National Academy of Science, 1982).
The work of Fligiel et al (1988) has indicated that histopathological changes of the type that are believed to be precursors of carcinoma can be observed in the lung tissue of chronic marijuana smokers. These results confirmed the earlier finding of Tennant (1980), who performed bronchoscopies on 30 US servicemen stationed in Europe who had smoked large quantities of hashish and experienced symptoms of bronchitis. He found that 23 of these who also smoked tobacco had one or more pathological changes "identical to those associated with the later development of carcinoma of the lung when it occurs in tobacco smokers" (Tennant, 1983, p78).
The results of these clinical and laboratory studies have recently received suggestive support from case reports of cancers of the upper aerodigestive tract in young adults who have been chronic cannabis smokers. Donald (1991a, b) reported 13 cases of advanced head and neck cancer occurring in young adults under 40 years of age among 3,000 of his cancer patients. Their average age was 26 years (range 19-38 years), compared with an average age of 65 years among his other patients. Eleven of the 13 had been daily cannabis smokers. Interpretation is complicated by the fact that at least five of these patients also smoked tobacco, and at least three were heavy alcohol consumers, both known risk factors for cancers of the upper aerodigestive tract (Holman et al, 1988; Vokes et al, 1993). Donald acknowledged these facts, but emphasised that half of his cases neither smoked tobacco nor consumed alcohol. Moreover, he argued, the implication of marijuana as a cause of cancers of the upper aerodigestive tract was strengthened by the observation that such cancers are rare under the age of 40 years, even among tobacco smokers who consume alcohol.
Similar findings have been reported by Taylor (1988) in a retrospective analysis of cases of upper respiratory tract cancer occurring in adults under the age of 40 years over a four-year period. Because the medical records did not routinely report the patients' use of cannabis, Taylor asked the attending clinicians to make judgments about their patients' cannabis and other drug use. He found 10 cases among the 887 cases of cancer that were treated over the study period. They consisted of six males and four females with an average age of 33.5 years. Nine were cases of squamous cell carcinomas (of the tongue, the larynx, and the lung). Five cases had a documented history of heavy cannabis smoking, two patients were described as "regular" cannabis users, one was classified as a "possible" cannabis user because he was known to abuse other drugs, and two were judged not to be cannabis users. As with Donald's case series, interpretation was complicated by the fact that six out of 10 were heavy alcohol consumers, and six were cigarette smokers (four out of the five heavy cannabis users in each case).
Taylor argued "that the regular use of marijuana is a potent etiologic factor, particularly in the presence of other risk factors, in hastening the development of respiratory tract carcinomas" (p1216). While he allowed that alcohol and tobacco use may have contributed to the development of these cancer, he discounted their importance, arguing like Donald, that the patients were well under 40 years of age, while the peak incidence of such cancers in drinkers and smokers is in the seventh decade of life.
Other investigators (e.g. Caplan and Brigham, 1989; Endicott and Skipper, 1991, cited by Nahas and Latour, 1992) have also reported cases of upper respiratory tract cancers in young adults with histories of heavy cannabis use. Caplan and Brigham's (1989) report of two cases of squamous cell carcinoma of the tongue in men aged 37 and 52 years was especially noteworthy because neither of their cases smoked tobacco or consumed alcohol; a history of long-term daily cannabis use was their only shared risk factor.
These case reports provide limited support for the hypothesis that cannabis use is a cause of upper respiratory tract cancers. They did not compare the prevalence of cannabis use in cases with that in a control sample, and cannabis exposure was not assessed in a standardised way or in ignorance of the case or control status, all of which are standard controls to minimise bias in case-control studies of cancer aetiology. Nonetheless, there is a worrying consistency about the reports that should be addressed by case-control studies which compare the proportions of cannabis smokers among patients with cancers of the upper aerodigestive tract and appropriate controls (National Academy of Science, 1982). Now may be the time to conduct such studies, since chronic cannabis smokers who began their use in early 1970s are now entering the period of risk for such cancers. If carcinomatous changes occur earlier in heavy cannabis smokers, it may be better to restrict attention to early onset cases (e.g. cases occurring in individuals under 50 years of age). Information on cannabis use should also be obtained prospectively in newly diagnosed cases, because of the problems with retrospective assessment of cannabis and other drug use from either clinical records or the relatives of those who have died.
Chronic heavy cannabis smoking probably causes chronic bronchitis, and impairs functioning of the large airways. Given the documented adverse effects of cigarette smoking, it is likely that chronic cannabis use predisposes individuals to develop irreversible obstructive lung diseases. There is suggestive evidence that chronic cannabis smoking produces histopathogical changes in lung tissues that are precursors of lung cancer. Case studies raise a strong suspicion that cannabis may cause cancers of the aerodigestive tract. The conduct of case-control studies of these cancers is a high priority for research into the possible adverse health effects of chronic cannabis smoking.