Background: The hallmark of sickle cell disease (SCD) is pain, which can be both acute and chronic. Pain in patients with SCD leads to hospital admissions and diminishes quality of life. Pain in patients with SCD is primarily treated with opioid medications, which are often insufficient. In 2017 the Committee on the Health Effects of Marijuana of the National Academies of Science, Engineering, and Medicine reported that there is "conclusive or substantial evidence that cannabis or cannabinoids are effective for the treatment of chronic pain in adults", but the report made no mention of chronic pain in patients with SCD. Currently, 30 states and Washington D.C. have medical marijuana (MM) laws, and 5 states list SCD as a qualifying condition. SCD was added to the list of qualifying conditions in Connecticut in February 2016. Since then we have offered certification to patients in our program who make regular clinic visits and for whom we think MM will be safe. However, not all patients who are certified go on to complete the state regulatory paperwork and obtain MM. We hypothesized that those who had obtained MM would show an improvement in clinical outcomes compared to those patients who had been certified but not obtained MM.
Methods: All patients who requested certification were educated on safety risks of MM. We conducted a review of all certified patients using our electronic medical record and the Connecticut Prescription Monitoring Program, which provides dispensing reports for schedule I-IV medications including MM. Our primary outcome was admissions in the 6 months after compared to the 6 months before obtaining MM, or the date certified for those who did not obtain MM. Our secondary outcomes were acute care services utilization (emergency department and outpatient infusion center visits) for the treatment of acute pain, and opioid use. Baseline hospital admissions, acute services utilization, and daily opioid use were defined as use in the 6 months prior to obtaining MM or certification for those who did not obtain MM. Opioid use was calculated as total oral morphine equivalents (OME) dispensed in a 6-month period and expressed as OME per day. Previous marijuana use was defined as 1 or more urine studies positive for cannabinoids before MM certification. Genotype was divided into more clinically severe (HbSS/Hbβ0) or less severe (HbSC/Hbβ+/Persistent HbF). We compared admissions, acute services, and opioid use for those who obtained MM to those who did not using a difference in differences analysis. Baseline admissions, baseline acute services, baseline opioid use, age, gender, genotype, hydroxyurea use, previous marijuana use, and insurance type of those who obtained MM were compared to those who did not using Student's t tests.
Results: 52 patients requested certification, and 50 patients were certified. 2 patients were not certified due to concerns about inappropriate use based on their past history. Twenty-eight patients who were certified obtained MM; 22 did not. Baseline hospital admissions, acute services utilization, and daily opioid were similar between the two groups. Patients who obtained MM were more likely to be genotype HbSS/HbSβ0. Age, gender, hydroxyurea use, previous marijuana use, and insurance type were similar between the two groups. (Table 1). Patients obtained MM a median of 109.5 days (IQR 54.8 - 188.8) after certification. Two patients concurrently started L-Glutamine within 6 months of MM certification, one who obtain MM and one who did not. Obtaining MM was associated with a decrease in admission rates in the next 6 months compared to those who did not obtain MM (-1.1 admissions 95% CI -0.1 - -2.0, p=0.03). Obtaining MM was not associated with a significant change in acute services utilization (0.3 visits 95% CI -1.4 - 1.9, p=0.8) or daily opioid use (-0.1 OME 95% CI -19.6 - 19.3, p= 1.0).
Conclusion: Though there was no difference in admission rates between the two groups we examined prior to MM certification, and the only clinical difference was that those who obtained MM were more likely to be in the moresevere genotype, those who obtained MM showed a decrease in admissions in the next 6 months compared to those who did not. We suggest that MM allowed patients to improve their pain relief and thereby reduce admission rates. This is consistent with data showing cannabinoid agonists improve pain in murine models of SCD. Larger, randomized and controlled studies of MM for pain in SCD should be conducted.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.