Cannabis Use and Fungal Infections in a Commercially Insured Population, United States, 2016

Case reports have identified invasive fungal diseases in persons who use cannabis, and fungal contamination of cannabis has been described. In a large health insurance claims database, persons who used cannabis were 3.5 (95% CI 2.6–4.8) times more likely than persons who did not use cannabis to have a fungal infection in 2016.

Among patients with fungal infections, persons who used cannabis were significantly younger than persons who did not use cannabis (median age 41.5 years vs. 56.0 years; p<0.001), more likely to be immunocompromised (43% vs. 21%; p<0.001), more likely to be hospitalized on the fungal infection diagnosis date (40% vs. 13%; p<0.001), and more likely to have tobacco use codes (40% vs. 9%; p<0.001) (Table). Sixty percent (n = 24) of persons who used cannabis and had fungal infections had cannabis abuse or dependence codes, compared with 79% of persons who used cannabis and did not have fungal infections, and 48% (n = 19) of persons who used cannabis and had fungal infections had unspecified cannabis use codes, compared with 29% of persons who used cannabis and did not have fungal infections. Persons who used cannabis and had fungal infections and unspecified cannabis use codes were older (median age 52 years vs. 28 years) and more frequently immunocompromised (63% vs. 25%) than persons who used cannabis and had dependence codes.

Conclusions
In this large commercially insured population in the United States, cannabis use was associated with a higher prevalence of certain fungal infections. Although these infections were uncommon, they can result in substantial illness and even death, particularly in immunocompromised persons.
Several hypotheses could explain our findings. First, on the basis of immunocompromised status and hospitalizations, persons who used cannabis appeared to be sicker than persons who did not use cannabis and were therefore presumably at higher risk for fungal infections in general. Some persons who used cannabis might be using medical cannabis to help manage their underlying conditions. In this analysis, it was not possible to determine the source of infection, although contaminated cannabis has been previously implicated in aspergillosis, mucormycosis, and cryptococcal meningitis (3)(4)(5). We are not aware of any reports of blastomycosis, histoplasmosis, or coccidioidomycosis acquired from contaminated cannabis. However, a small risk likely exists; 1 histoplasmosis outbreak occurred in a cannabis field (6), and fomites, such as hay and vegetables, are involved in rare coccidioidomycosis cases (7). Another possible explanation is that smoking-induced structural and immunological lung damage confers increased susceptibility to infection (8), although the lung effects of cannabis might differ from those of tobacco (9). Confounding by tobacco smoking might be another explanation because tobacco use is typically more common among persons who use cannabis (10). Tobacco can also be contaminated with fungi, possibly to a lesser extent than cannabis (11).
Our results could also reflect medical coding artifacts. In general, cannabis use is likely greatly underrepresented by ICD codes (12), supported by the finding that <0.3% of our study population had cannabis use codes, whereas ≈9% of the US population reported using cannabis in the past month (13). Although ICD-10-CM codes cannot distinguish between medical and recreational cannabis use, the higher frequency of immunocompromising conditions and older age among persons who use cannabis and persons who used cannabis and had unspecified cannabis use codes suggests medical cannabis use among some of these patients. We were also unable to differentiate smoking cannabis from other modes of use (e.g., ingestion), which is relevant because smoking might lead to greater fungal exposure through inhalation. Injection drug use, which also might be more common among persons who use cannabis, is an emerging risk factor for some fungal infections, such as invasive candidiasis, although this mode of acquisition seems less likely for the fungal infections described here, which are typically acquired through inhalation.
Another limitation is that we did not evaluate immunocompromised status associated with medications such as corticosteroids and tumor necrosis factor inhibitors. The similar geographic distribution of fungal infections between persons who use cannabis and persons who did not use cannabis is notable because state laws vary substantially regarding medical and recreational use. There is more legalization overall in the western and northeastern United States.
Despite the limitations inherent in administrative data and our inability to infer causality between cannabis use and fungal infections, our study adds to emerging evidence about this association. This finding is consistent with a recommendation that solid organ transplant recipients avoid smoking cannabis (14). Patients with other immunocompromising conditions should be also aware of the possible link between cannabis smoking and fungal infections and might also consider avoiding this exposure. Physicians should remain aware of the possible link between fungal infections and cannabis use.