This study aimed to characterize cannabis use among patients receiving HD, describe patient perspectives related to cannabis use, and evaluate patient experiences related to cannabis and their HD health care team. While other studies have reported on the perspectives of prescribers, nephrologists, and physicians-in-training, this study is one of the first to focus on patient perspectives among those receiving HD.^{21bibr22-20543581241274002bibr23-20543581241274002-24}

Our study sample appeared to be younger than the Canadian average (mean age of patients with ESKD was reported to be 64 years old) ²⁵ and compared with a study done by Collister et al ¹⁹ This may have been at least partially due to our survey methodology (ie, patients were required to complete the survey without assistance), which may have unintentionally selected for younger participants. The average age of patients receiving HD at our center was 60 years old, which is younger than the national average, possibly because of our location (downtown urban center).

Our results revealed 3 themes with important patient care and research implications. First, we found that cannabis use is relatively common among our study sample and is often used for symptom management, highlighting its relevance, and a need for clinicians to incorporate a cannabis assessment in the clinical care of patients. In our study, the frequency of cannabis use among patients receiving HD (21%) was comparable with that reported by Collister et al ¹⁹ in a similar but larger survey of cannabis use in patients with CKD (17.8% smoked cannabis, 8.4% used cannabis edibles, and 5.3% used cannabis oils in the last month). Compared with the general population, the 2020 National Cannabis Survey reported that 20% of those over the age of 15 reported cannabis use in the past 3 months. ²⁶ Notably, we found that 45% were using cannabis for adverse symptom management, but none had formal medical authorization. Collister et al likewise reported that only a small proportion of cannabis users (10.3%) had been prescribed cannabis by a medical practitioner. Furthermore, our study found that only 8% of respondents reported ever being asked about cannabis use by a member of the HD team. This highlights a current lack of clinician involvement in patient self-management of symptoms with cannabis. Potential barriers are insufficient provider knowledge, a lack of routine surveillance of adverse symptoms and cannabis use, and low patient comfort/trust in discussing cannabis in health care settings. A systematic review done by Gardiner et al explored the perspective of health care providers with regards to medical cannabis. They found that although most health care providers showed support for the use of medical cannabis, they reported low self-perceived clinical and legislative knowledge. ²¹ In a survey of Canadian nephrologists by Gitau et al, ²⁴ 59% of respondents felt uncomfortable with their knowledge of medical cannabis literature. Besides addressing this knowledge gap, incorporating questions regarding cannabis use and adverse symptoms in routine patient assessments may facilitate more patient-provider discussions about cannabis use. Provider acknowledgment of cannabis as a therapeutic modality may strengthen patient-clinician relationships by helping patients feel more open to discuss cannabis use.

Next, our study found that the most common symptoms cannabis was being used to manage in our small study sample were insomnia, anxiety, and non-neuropathic pain. However, there is limited evidence to show efficacy or safety of cannabis for these symptoms in patients with kidney disease and/or receiving HD. ⁴ In patients without kidney disease, there is some evidence for using cannabis to treat chemotherapy-induced nausea and vomiting (CINV) in the oncology population,^{10bibr11-20543581241274002-12} for management of neuropathic pain in patients with human immunodeficiency virus (HIV), ¹³ and for treatment of anorexia/cachexia in patients with HIV. ¹⁴ In general, most studies excluded patients with kidney disease, and only one small study demonstrated a benefit of a topical cannabinoid in reducing uremic pruritus and xerosis among patients receiving HD. ¹⁵

In non-ESKD populations, there is limited quality evidence for the use of cannabis and cannabinoids in the treatment of anxiety and other mental health conditions. A systematic review done in 2019 by Black et al summarized evidence regarding the use of cannabinoids for the treatment of depression, anxiety, attention deficit hyperactivity disorder, post-traumatic stress disorder, and psychosis. They found that some studies, although low quality, showed that THC with or without CBD improved anxiety symptoms for those with comorbidities, such as chronic non-cancer pain and multiple sclerosis. They did not find any other significant difference of cannabinoids for other mental health conditions, with the exception of one small study showing that THC increased psychosis symptoms. ²⁷ This lack of data, even in the general population, is an important point to discuss with patients who are interested in using it for improvement of mental health conditions. By identifying the most common symptoms that patients are using cannabis to manage, this may help to focus future research initiatives to these indications in order to inform clinical care.

Finally, we identified high risk cannabis use and opportunities for patient education and harm reduction. The majority of respondents who reported cannabis use either smoked or vaporized cannabis. Similarly, in a survey of patients with CKD, Collister et al ¹⁹ also found that smoked cannabis was the most common route among those who reported using cannabis. This is considered high risk use because cannabis smoke contains many of the same respiratory irritants and toxins as tobacco smoke and is associated with histopathological changes in the respiratory tissue.^8,9 The 2017 Lower Risk Cannabis Use Guidelines state that smoked cannabis is the most harmful route of using cannabis and they suggest other methods, such as edibles or vaping to avoid deep inhalation of cannabis smoke. ²⁸ Interestingly, we found that none of the current cannabis users expressed concern for side effects and none expressed any interest in receiving information about side effects. These findings highlight a knowledge gap and an opportunity for the health care team to educate patients about potential risks, discuss lower risk alternatives, and to facilitate informed decision-making among patients. It also emphasizes the need to develop a stronger therapeutic relationship with patients to facilitate patient receptivity to this information.

Another area for harm reduction is in recognizing and managing potential drug interactions with cannabis. This is significant among those receiving HD, due to the numerous concomitant medications for kidney disease and other comorbidities. There are numerous potential drug interactions with significant clinical consequences. Pharmacodynamic interactions with cannabis include central nervous system depressants (increased drowsiness, ataxia), anticholinergic drugs (increased tachycardia, drowsiness, dry mouth), sympathomimetic drugs (increased tachycardia, hypertension), and stimulants (increased tachycardia). Pharmacokinetically, cannabis is metabolized hepatically through cytochrome P450 enzymes (CYP450). Cannabidiol is metabolized mainly by CYP 2C9, 2C19, 3A4, 1A2, and 2D6 enzymes and THC is metabolized mainly by CYP 2C9, 2C19, and 3A4 enzymes. ⁹ Therefore, pharmacokinetic interactions with cannabis may occur with inhibitors and inducers of these CYP enzymes, for example, ketoconazole (CYP 3A4 inhibitor) and rifampin (CYP 3A4 inducer). Cannabis itself also inhibits CYP2C9, induces CYP 1A2, and upregulates P-glycoprotein. ⁹ Warfarin, a commonly used oral anticoagulant in HD patients, is an example of a drug that has been shown to interact with cannabis. Case reports have described increases in the International Normalized Ratio (INR) among patients taking warfarin and using cannabis (smoked or ingested). ²⁹ Interestingly, drug interactions were identified as the most common concern among cannabis users in our study. This was also the second most common type of information that respondents identified they would be interested in receiving from the health care team. These findings further underscore the opportunity and need for the health care team to incorporate a discussion of cannabis use into routine clinical care, in order to identify, manage, and educate patients on this topic.

Our study had limitations. First, the overall participation rate was lower than anticipated, at approximately 17% of the total population invited. Although we did not collect data on reasons for non-participation, potential reasons include language barrier, lack of interest, decreased willingness to participate due to the COVID-19 pandemic, and/or fears that revealing information about cannabis use may impact their care. A similar study done by Collister et al achieved a higher response rate (27.3% in person, 9.4% online). Some strategies used in their study that we did not employ for reasons of feasibility include the options of completing surveys by mail or online, including patients from multiple sites, and circulating the survey online (social media, CKD organization websites). Providing assistance from the study team to complete the survey was considered, but not pursued because we wanted to ensure anonymity and mitigate any possible privacy concerns that patients may have had in discussing cannabis use.

As a voluntary survey, non-response bias is another possible factor that may further limit the generalizability of our study. This is a limitation cited in other voluntary surveys as well, such as the Canadian Cannabis Survey.

Finally, for feasibility reasons, this survey consisted of only multiple-choice and Likert scale rating questions, therefore limiting the breadth and depth of responses provided. For future studies, patient and provider interviews may provide more detailed insights and identify other themes.