Cannabinoids Offer Modest Pain Relief but Fall Short of First-Line Status, Review Concludes
Cannabinoids Offer Modest Pain Relief but Fall Short of First-Line Status, Review Concludes
A 2025 narrative review published in Neurology International finds condition-specific analgesic benefit from cannabinoids, notable safety concerns including hepatotoxicity and high discontinuation rates, and major regulatory barriers that collectively limit cannabinoid use in chronic pain to an adjunctive role for treatment-resistant patients.
Why This Matters
Chronic pain affects approximately 1.5 billion people worldwide, and cannabinoid-based products are increasingly accessible to patients seeking alternatives to opioids and conventional analgesics. The pharmacological rationale for cannabinoid analgesia is scientifically grounded in endocannabinoid system biology, lending credibility to patient interest. However, the gap between biological plausibility and clinical proof remains substantial, and the regulatory landscape is fragmenting faster than the evidence base can keep pace. This review arrives at a moment when clinicians urgently need consolidated, honest guidance on what cannabinoids can and cannot do for pain.
Clinical Summary
Chronic pain encompasses a heterogeneous set of conditions, from neuropathic syndromes to musculoskeletal disorders, and the search for safe, effective analgesics beyond opioids has driven considerable interest in cannabinoid therapeutics. This 2025 narrative review, published in Neurology International by MDPI, synthesizes existing literature on cannabinoid pharmacology, clinical efficacy, adverse effects, and regulatory frameworks across multiple jurisdictions. The mechanistic basis for cannabinoid analgesia rests on activity at CB1 and CB2 receptors, TRPV1 channels, and PPAR pathways, which collectively modulate nociceptive signaling and neuroinflammation. The authors examine trial data for THC, CBD, nabiximols (a THC and CBD combination), and smoked or vaporized cannabis across neuropathic pain, multiple sclerosis-related spasticity, fibromyalgia, osteoarthritis, and musculoskeletal pain.
The review reports average pain reductions of 0.5 to 1.0 points on a 10-point numerical rating scale, a magnitude the authors characterize as modest and insufficient for first-line classification. Evidence quality is strongest for neuropathic pain and MS-related spasticity, while results for fibromyalgia, osteoarthritis, and musculoskeletal pain remain inconsistent. Safety signals are clinically meaningful: nabiximols cause dizziness in approximately 25% of patients and carry a 12% discontinuation rate, while high-dose CBD raises hepatotoxicity concerns and shows dose-dependent discontinuation rates of 4.3% at low doses and 12.9% at high doses, both exceeding the 3.5% placebo rate. The authors call for large-scale randomized controlled trials with standardized formulations and harmonized international regulations before cannabinoids can be recommended beyond adjunctive use in treatment-resistant patients.
Dr. Caplan’s Take
This review does a credible job of mapping the mechanistic rationale to the clinical reality, and the clinical reality is humbling. Cannabinoids have a plausible pharmacological story, but the analgesic effect sizes are small, the adverse effect profile is not trivial, and the evidence outside neuropathic pain and MS spasticity is genuinely inconsistent. Patients ask me about cannabis for pain more than almost any other topic, and what they deserve is an honest answer: the science supports a modest role, not the transformative one that marketing and media often imply.
In practice, I consider cannabinoid therapy only after conventional options have been adequately trialed and found insufficient. When I do prescribe or recommend, I start with low doses, favor formulations with established pharmacokinetic profiles, monitor liver function if CBD doses are high, and set explicit expectations about the realistic magnitude of benefit. I also ensure patients understand that “natural” does not mean free of side effects. The evidence supports careful, supervised use in select patients, not broad adoption.
Clinical Perspective
This narrative review occupies a familiar position in the cannabinoid research arc: it confirms what higher-quality systematic reviews and meta-analyses have already suggested, namely that cannabinoids provide real but limited analgesia concentrated in neuropathic and spasticity-related pain. What it adds is a useful cross-jurisdictional regulatory analysis and a side-by-side framing against conventional analgesic NNT values (for example, NNT 4.6 for topical diclofenac, NNT 6 to 8 for duloxetine), which helps clinicians contextualize the modest cannabinoid effect sizes. However, the absence of a systematic search protocol means this review cannot claim to represent all available evidence, and clinicians should treat its quantitative claims as summaries of cited sources rather than independently pooled estimates.
From a safety standpoint, the hepatotoxicity signal with high-dose CBD warrants particular attention in patients already on hepatically metabolized medications, including common analgesics and antiepileptics. CBD’s inhibition of CYP3A4 and CYP2C19 can alter levels of clobazam, warfarin, and certain statins. The 25% dizziness rate with nabiximols is relevant for fall-risk populations, including older adults with neuropathic pain. The single most actionable recommendation from this evidence: if you are considering cannabinoid therapy for a pain patient who has failed conventional options, begin with a low-dose, pharmaceutically standardized product, monitor hepatic function at baseline and follow-up, and establish measurable functional outcome goals rather than relying solely on subjective pain scores.
What Kind of Evidence Is This
This is a narrative review, which occupies the lower tier of the evidence hierarchy relative to systematic reviews, meta-analyses, and randomized controlled trials. Narrative reviews synthesize literature based on author selection rather than a predefined, reproducible search strategy, which means the included studies may not represent the full body of available evidence. The single most important inference constraint this imposes is that readers cannot independently verify whether the review’s conclusions would hold if all relevant studies, including negative or null findings, had been included.
How This Fits With the Broader Literature
The review’s central finding of modest, condition-specific cannabinoid analgesia is broadly consistent with prior systematic reviews and meta-analyses, including the 2018 Cochrane review on cannabis-based medicines for chronic neuropathic pain and the 2015 Whiting et al. systematic review in JAMA, both of which reported small to moderate effect sizes with notable adverse event rates. The emphasis on neuropathic pain and MS spasticity as the strongest evidence domains also aligns with the 2017 National Academies of Sciences report on the health effects of cannabis. Where this review adds incremental value is in its cross-jurisdictional regulatory comparison, which contextualizes why large-scale, multi-site trials remain difficult to execute. It does not materially challenge prior conclusions but does reinforce the persistent gap between mechanistic promise and clinical magnitude.
Common Misreadings
The most likely overinterpretation is reading this review as evidence that cannabinoids “work for chronic pain” in a general sense. The data do not support that broad claim. Effect sizes of 0.5 to 1.0 points on a 10-point scale are statistically detectable but clinically modest, and this magnitude of benefit is concentrated in neuropathic pain and MS spasticity, not chronic pain writ large. Extending these findings to fibromyalgia, osteoarthritis, or low back pain exceeds what the evidence supports. A second common misreading involves treating the pharmacological rationale (ECS receptor mechanisms) as equivalent to clinical proof. Biological plausibility is necessary but not sufficient to justify therapeutic use.
Bottom Line
This narrative review confirms that cannabinoids offer modest, condition-specific analgesia strongest in neuropathic pain and MS-related spasticity, but with clinically meaningful side effects and high discontinuation rates. It does not establish cannabinoids as first-line therapy for any chronic pain condition. For now, the evidence supports supervised, low-dose adjunctive use only in patients who have not responded to conventional treatments, with careful safety monitoring and realistic outcome expectations.
Frequently Asked Questions
How much pain relief can I realistically expect from cannabinoid therapy?
Based on the clinical trial data reviewed, the average pain reduction associated with cannabinoids is 0.5 to 1.0 points on a standard 10-point pain scale. For context, a reduction of 2 points is generally considered the threshold for a clinically meaningful difference. This means that while some patients may notice improvement, the average benefit is modest and unlikely to replace the need for other pain management strategies.
Are cannabinoid medications safe for long-term use?
Long-term safety data for cannabinoid medications remain limited, which is one of the key gaps this review identifies. In the shorter-term data available, notable adverse effects include dizziness in approximately one in four patients taking nabiximols, dose-dependent liver enzyme elevations with high-dose CBD, and discontinuation rates meaningfully higher than placebo. Patients on other medications should be particularly cautious about drug interactions, especially those involving liver-metabolized drugs. Long-term risks, including effects on cognition and mental health, have not been adequately studied in pain populations.
Why does my doctor hesitate to prescribe cannabis for my fibromyalgia or arthritis pain?
The evidence supporting cannabinoid use varies considerably by condition. While there is moderate-quality evidence for neuropathic pain and MS-related spasticity, the data for fibromyalgia, osteoarthritis, and general musculoskeletal pain are inconsistent, with some trials showing benefit
Have thoughts on this? Share it:
