Page 12 - JCTR-11-1

Journal of Clinical and
Translational Research Cannabinoids for cannabis use disorder

diagnosed with CUD; and (3) comparing CB receptor studies, we opted to generate albatross plots encompassing
agonists (i.e., dronabinol and nabilone) or modulators of all included studies. 15
endocannabinoid activity (i.e., nabiximols and CBD) with An albatross plot depicts a scatter plot wherein each
placebo. study’s sample size is plotted against its respective two-
2.2. Search strategy and data extraction sided P-values derived from effect estimates. This visual
representation enables the interpretation of estimated
We systematically searched PubMed, Scopus, and standardized effect sizes (in this case, standardized to
Cochrane Central Register of Controlled Trials from correlation coefficients) within the context of both study
inception to May 2024 with the following search strategy: size and significance level for each individual study, as
(cannabidiol OR CBD OR sativex OR nabiximols OR well as for the collective relationship observed across all
dronabinol OR nabilone OR cesamet OR FAAH OR “fatty studies. 15
acid amide hydrolase”) AND (“cannabis use disorder” OR
“cannabis dependence” OR “cannabis withdrawal”). The To gather the necessary data for constructing these
references from all included studies, previous systematic plots, we retrieved the P-values, sample sizes, and effect
reviews, and meta-analyses were also searched manually estimates from each study. In instances where studies did
for any additional studies. Two authors independently not explicitly provide P-values, we calculated them based
extracted the data following predefined search criteria and on the sample size and effect magnitude (such as Pearson’s
quality assessment. correlation coefficient). 16
The study protocol was registered in the International 3. Results
Prospective Register of Systematic Reviews on May 27,
2024, as CRD42024547431. 3.1. Study selection
The initial search yielded 1,317 results. After the removal of
2.3. Endpoints
duplicate records and ineligible studies, 14 remained and were
We extracted the following data from individual fully reviewed based on inclusion criteria. Of these, a total of
studies: (1) study characteristics: authors, study eight studies met the final eligibility criteria. As two of the
design, intervention, dose, route of administration, eight articles refer to one study, a total of seven studies were
treatment duration, and follow-up days; (2) patient included in the systematic review, comprising 597 patients.
characteristics: gender, mean age, cannabis use in grams The PRISMA flow diagram is illustrated in Figure 1.
per day, days of use in the past 30 days, Marijuana
Craving Questionnaire and Hamilton Depression scale; 3.2. Study characteristics
(3) outcomes: weekly cannabis use, cannabis withdrawal A total of 317 patients (53%) received CB receptor
scale, cannabis cravings, number of adverse effects and agonists (i.e., dronabinol and nabilone) or modulators of
21-day consecutive abstinence. endocannabinoid activity (i.e., nabiximols and CBD), as
monotherapy or in combination with another medication,
2.4. Risk of bias and quality assessment
with concomitant psychotherapy. All the studies included
According to the recommendations from the Cochrane met the final eligibility criteria; however, there was
Handbook for Systematic Reviews of Interventions, we used significant variation in terms of population characteristics,
the revised Cochrane risk-of-bias tool for randomized trials interventions, and outcomes.
13
(RoB-2) to assess the risk of bias in RCTs. Disagreements
were resolved through consensus after discussing reasons All studies were randomized, double-blind, and placebo-
for the discrepancy. The information was presented as a controlled clinical trials with a parallel design. All studies
risk of bias graph and a risk of bias summary figure. included participants with CUD, as defined by the diagnostic
criteria of the DSM-IV, 17-19 DSM-IV-TR, 20,21 DSM-V , and
22
2.5. Statistical analysis the International Statistical Classification of Diseases and
Related Health Problems, 10 edition (ICD-10). 23,24
th
This systematic review was designed following the
PRISMA. The studies included in our analysis exhibited The mean age of the participants ranged from 26.4 to
14
significant heterogeneity, rendering meta-analysis 37.65 years, with a male predominance of 74.2% across
unfeasible. This was due to the use of disparate effect all the studies. Cannabis use at baseline was reported in
measures, including correlation coefficients, odds ratios, different ways: (1) according to the number of days of use,
and regression coefficients, coupled with insufficient data ranging from 28 to 30 days in the past 30 days 17,18,20 or
to standardize effect sizes uniformly across all studies. To 25.7 days in the past 28 days; (2) according to cannabis
23
facilitate a comprehensive interpretation of results across weight, ranging from 0.55 to 3.28 g/day.

Volume 11 Issue 1 (2025) 6 doi: 10.36922/jctr.24.00066