CED Clinic
CBD for Recurrent Pericarditis: What the Phase II Trial Really Shows
A 27-person, open-label Phase II study reported lower pain scores, C-reactive protein improvement, and fewer recurrences during medication tapering. Because there was no control group, the first eight weeks added CBD to unchanged background therapy, and all analyses were descriptive, the paper supports a clinical signal rather than proven efficacy.
What This Study Teaches Us
The MAvERIC-Pilot trial is a prospective, multicenter, single-arm study of a standardized oral CBD formulation in adults with active recurrent pericarditis. Mean maximum pain fell by 3.7 points at eight weeks, 8 of 10 participants with elevated baseline C-reactive protein normalized that marker, and 17 of 24 participants entering the extension had no recurrence while background medications were tapered.
The central limitation is attribution. Without randomization or a comparator, and with NSAIDs, colchicine, and corticosteroids held stable during the primary treatment period, the study cannot determine how much improvement came from CBD, concurrent therapy, expectation, regression toward the mean, or the natural fluctuating course of pericarditis.
Why This Matters
Recurrent pericarditis can mean repeated chest pain, medication escalation, steroid exposure, and persistent fear of another flare. An oral, non-intoxicating pharmaceutical cannabinoid is understandably appealing, but this trial does not justify substituting store-bought CBD for prescribed treatment or changing a taper plan without specialist supervision.
The combination of subjective pain improvement and a small objective CRP signal is enough to make the program scientifically credible. It is not enough to change routine management, especially because the tested dose was high, adverse events were common, and the primary endpoint was measured while standard anti-inflammatory therapy remained unchanged.
This paper shows what serious cannabinoid drug development looks like: a defined molecule, verified manufacturing, weight-based dosing, prospective monitoring, and progression toward a randomized trial. It also shows why industry-funded pilot data require disciplined interpretation before regulatory status, mechanistic plausibility, or favorable early results are translated into treatment claims.
Study Snapshot
| Study Type | Phase II, open-label, multicenter, single-arm pilot trial conducted at eight U.S. centers. |
| Population | 27 adults with at least a second symptomatic pericarditis episode, pain of at least 4 on an 11-point scale, and either CRP elevation or cardiac MRI evidence of pericardial inflammation. Mean age was 52.7 years; 18 participants were women; all were White and non-Hispanic or Latino. |
| Intervention | CardiolRx, an oral formulation containing CBD 100 mg/mL with no detectable THC below 5 ppm, titrated to a maximum tolerated dose of up to 10 mg/kg twice daily. |
| Treatment Window | Eight weeks with background pericarditis medications kept stable, followed by an optional 18-week extension during which NSAIDs, corticosteroids, and colchicine were tapered. |
| Comparator | None. Outcomes were compared with each participant’s baseline and prior reported recurrence history. |
| Primary Outcome | Change in the highest patient-reported pericarditis pain score during the seven days before baseline versus the seven days before Week 8. |
| Secondary Outcomes | CRP normalization among participants with elevated baseline CRP, Week 26 pain, recurrence during the extension, medication tapering, and safety measures. |
| Statistical Approach | Descriptive analyses only. The investigators performed no statistical inference and no missing-value imputation. |
| Journal and Year | Journal of the American Heart Association, 2026;15:e047605. |
| DOI | 10.1161/JAHA.125.047605 |
| Funding and Conflicts | Funded by Cardiol Therapeutics, the manufacturer of CardiolRx. Three authors were company employees, sponsor-funded contractors performed data management and statistical analyses, and several investigators reported consulting relationships with Cardiol Therapeutics or other pericarditis-drug companies. |
CardiolRx generated a coherent early signal across pain, CRP, and medication-taper outcomes, but this study cannot determine how much benefit was caused by CBD. It supports completing the randomized Phase III trial. It does not support replacing established pericarditis treatment, using retail CBD as an equivalent, or describing safety and efficacy as settled.
What CBD for Recurrent Pericarditis Looked Like in This Trial
The investigators asked whether a pharmaceutical oral CBD formulation could be added safely to ongoing treatment and produce a sufficiently credible signal in active recurrent pericarditis to justify definitive testing. Participants were already taking stable doses of colchicine, NSAIDs, corticosteroids, or combinations of these medications. Eleven were taking corticosteroids, and 19 were receiving combination therapy.
During the initial eight weeks, those background therapies were intentionally left unchanged while CardiolRx was titrated to each participant’s maximum tolerated dose. This means the primary outcome assessed add-on treatment, not CBD monotherapy. The optional extension then shifted the question: could participants remain stable while clinicians tapered standard medications and continued CBD?
The trial excluded several clinically important groups, including patients currently receiving rilonacept, anakinra, other immunosuppressive therapies, or treatment for several secondary causes of pericarditis. Serial echocardiography and cardiac MRI were not mandated, so the study did not systematically measure whether structural pericardial inflammation changed over time.
What the Paper Found
Mean maximum pain during the prior seven days fell from 5.8 ± 1.7 at baseline to 2.1 ± 1.8 at Week 8, a mean decrease of 3.7 points. The median time to a score of 2 or lower was five days. Mean pain at Week 26 was 1.5 ± 2.2.
Only 10 of 27 participants had CRP of at least 1.0 mg/dL at baseline. Eight of those 10 reached the prespecified normal range by Week 8. Across the full group, mean CRP fell from 2.0 ± 4.9 mg/dL to 0.7 ± 2.5 mg/dL at Week 8 and 0.6 ± 1.3 mg/dL at Week 26.
Twenty-four participants entered the extension, and 17 of 24, or 70.8%, had no new recurrence while background treatment was tapered. Seven recurred, including five who had been taking corticosteroids at baseline. Recurrences were managed by escalating prior background therapy; one participant ultimately transitioned to rilonacept.
Diarrhea occurred in 15 participants, or 55.6%; rash in 8, or 29.6%; nausea in 5, or 18.5%; and fatigue in 5, or 18.5%. Trial medication was discontinued in four participants who experienced at least one adverse event and reduced or temporarily interrupted in eight.
Two participants had serious adverse events. One developed probable drug reaction with eosinophilia and systemic symptoms, or DRESS, which the investigator considered possibly related to the trial drug, although celecoxib had been started concurrently and can also cause DRESS. The other participant’s serious events were judged unrelated. No clinically important QTc change or suicidal behavior was observed.
How Strong Is the Evidence for CBD for Recurrent Pericarditis?
This is nonrandomized interventional evidence designed for signal detection. It is stronger than a retrospective chart review, uncontrolled anecdote, or animal study because investigators prospectively defined eligibility, dosing, outcomes, and follow-up across multiple centers. It is much weaker than a blinded randomized trial because every participant and investigator knew CBD was being given and there was no group showing what would have happened without it.
The primary endpoint was patient-reported pain, an important clinical outcome but one especially vulnerable to expectation effects in an open-label study. CRP is more objective, but only 10 participants had elevated baseline values, the distribution was highly variable, and the authors did not perform inferential testing or provide estimates that quantify statistical uncertainty.
The correct evidentiary label is promising, high-caution Phase II signal. The findings justify the ongoing randomized trial named in the paper. They do not establish treatment efficacy, comparative effectiveness, or a reliable safety profile for broader practice.
Where This Paper Deserves Skepticism
Pain and CRP were measured while participants continued unchanged NSAIDs, colchicine, corticosteroids, or combination therapy. Improvement therefore cannot be attributed to CBD alone.
Pain can improve because of expectation, intensified follow-up, standard treatment, regression toward the mean, or ordinary disease fluctuation. A single-arm design cannot separate those explanations.
The paper reports means and percentages but no formal hypothesis testing, confidence intervals, or inferential analysis. Readers cannot tell how stable these estimates would be in another small sample.
The recurrence denominator was 24, not all 27 enrolled participants. Two participants stopped before Week 8 because of adverse events and one stopped for lack of response, so the extension preferentially retained people able and willing to continue.
The reported decline from 5.8 events per year before enrollment to 0.9 during trial participation compares different observation windows and care conditions. It is vulnerable to recall differences, regression toward the mean, and changes in monitoring or treatment.
Only 10 participants contributed to the CRP-normalization result, serial cardiac imaging was not required, and the trial did not measure NLRP3, NF-κB, IL-1β, or related biomarkers. The proposed mechanism was not demonstrated in these patients.
All participants were White and non-Hispanic or Latino, the sample was highly selected, and patients receiving IL-1 blockers or several immunosuppressive treatments were excluded.
The manufacturer funded the trial, employed three authors, and funded data management and statistical work. The paper’s suggestion that this could represent a “paradigm shift” is a development hypothesis, not a conclusion supported by an uncontrolled pilot.
What This Paper Does Not Show
It does not show superiority to established treatment. CBD was not compared head-to-head with colchicine, corticosteroids, rilonacept, anakinra, or any other active therapy.
It does not show that CBD monotherapy caused the Week 8 improvements. Standard pericarditis medications were kept stable during the primary period.
It does not show that retail CBD is equivalent. The trial used a standardized, THC-free pharmaceutical solution at weight-based doses up to 10 mg/kg twice daily, not a dispensary tincture, gummy, or supplement.
It does not prove the proposed NLRP3 mechanism operated in humans. No direct inflammasome or cytokine biomarker testing was performed.
It does not establish long-term benefit or safety. Follow-up ended at 26 weeks, adverse events were common, and the sample was too small to characterize uncommon risks.
It does not justify self-directed treatment changes. The study’s tapering occurred under protocol-driven clinical supervision in a selected research population.
How This Fits With the Broader Clinical Conversation
The paper enters a field in which recurrent pericarditis is increasingly understood as an autoinflammatory disease involving NLRP3 and interleukin-1 signaling. Controlled trials support IL-1 blockade in appropriately selected refractory disease, while corticosteroids remain clinically useful but can complicate recurrence and tapering. CardiolRx is best understood as a possible future addition to this treatment landscape, not a demonstrated replacement for it.
The authors compare their descriptive pain, CRP, and recurrence outcomes with prior IL-1 inhibitor studies. Those comparisons are useful for generating hypotheses but cannot establish comparable efficacy because the populations, designs, treatment periods, and outcome methods differed.
The study also illustrates an important distinction across cannabinoid medicine. A pharmaceutical CBD drug developed for a defined inflammatory target is not interchangeable with whole-plant cannabis or consumer CBD. Readers interested in that boundary may find CED Clinic’s review of cannabis and cardiovascular risk, its analysis of CBD-related liver enzyme elevation, and its expanded endocannabinoid system guide useful context.
For now, the scientifically appropriate next step is the randomized, double-blind, placebo-controlled Phase III MAVERIC trial identified in the paper. Its results will need to show that benefit persists when expectation, background care, and natural disease variation are controlled.
What catches my attention is not the pain score by itself. Pain can improve for many reasons in an unblinded study. The more useful finding is that symptom improvement moved in the same direction as CRP in the small subgroup that began with measurable systemic inflammation. Even there, ten patients and no control group leave substantial uncertainty.
In practice, I would not use this paper to recommend a retail CBD product, to replace colchicine or biologic treatment, or to accelerate a medication taper. The studied formulation reached high weight-based doses, and tolerability deserves more attention than a simple “generally well tolerated” label suggests. Diarrhea and rash were common, several participants needed dose changes or discontinuation, and one serious DRESS syndrome was considered possibly related to the trial medication, with concurrent celecoxib making causation uncertain.
At this stage, I would describe the work as a promising signal, not a paradigm shift. It earns the randomized trial that is now being conducted. A positive controlled result would create a much more consequential clinical conversation about oral inflammasome modulation, steroid avoidance, biologic de-escalation, and the legitimate place of pharmaceutical cannabinoids in cardiovascular medicine.
What a Careful Reader Should Take Away
This trial makes pharmaceutical CBD for recurrent pericarditis scientifically plausible enough to test rigorously. It does not show that CBD caused the observed improvements, that consumer products will behave similarly, or that current treatment should change. The objective CRP finding is encouraging, the tolerability profile is clinically relevant, and the randomized Phase III result is the evidence that should determine whether this approach advances.
A broader evidence review that helps separate cardiovascular findings about cannabis exposure from this narrow pharmaceutical CBD trial.
Read articleA clinically focused look at laboratory safety signals that matter when CBD is used at pharmaceutical rather than consumer-level doses.
Explore evidenceA mechanistic guide to cannabinoid signaling and related inflammatory pathways, useful background without treating mechanism as proof of benefit.
Read the guideJoin the Conversation
Have a question about how to interpret cannabinoid research? Ask Dr. Caplan or continue the discussion in the CED Clinic forums.
Frequently Asked Questions
It tested CardiolRx, a pharmaceutically manufactured oral CBD solution containing 100 mg/mL and no detectable THC below 5 ppm, in adults with active recurrent pericarditis. It did not test whole-plant cannabis or ordinary retail CBD products.
No. Every participant received CBD, and both participants and investigators knew it. The absence of a control group is the main reason the trial cannot establish that CBD caused the improvements.
Both outcomes moved in a favorable direction. Pain fell across the full sample, and CRP normalized by Week 8 in 8 of 10 participants who began with elevated CRP. The CRP subgroup was small, and no placebo comparison was available.
This study does not support that conclusion. It did not compare CBD directly with those treatments, and the first eight weeks used CBD in addition to stable background therapy.
No. CardiolRx is a standardized investigational drug with verified concentration, manufacturing, and THC limits. Consumer CBD products vary in dose, purity, labeling accuracy, and formulation and were not evaluated here.
The solution was titrated to a maximum tolerated dose of up to 10 mg/kg twice daily. Most participants reached that maximum. This weight-based pharmaceutical exposure is not directly comparable with a typical over-the-counter CBD serving.
Diarrhea occurred in 55.6% and rash in 29.6%. The drug was discontinued in four participants who had experienced at least one adverse event, and eight required dose reduction or interruption. One serious DRESS syndrome was considered possibly related to the trial drug, although concurrent celecoxib was another plausible cause.
Cautiously. The trial enrolled only 27 highly selected adults, all of whom were White and non-Hispanic or Latino, and excluded patients receiving several immunosuppressive or IL-1 therapies.
Cardiol Therapeutics, the manufacturer of CardiolRx, funded the trial. Three authors were company employees, and sponsor-funded contractors performed data management and statistical analyses. These relationships do not invalidate the findings, but they increase the importance of independent, blinded confirmation.
The paper identifies an ongoing randomized, double-blind, placebo-controlled Phase III MAVERIC trial. That study is designed to determine whether the Phase II signal persists when expectation and untreated comparison are addressed.
