Marijuana Doesn't Curb Agitation, Aggression in Dementia

Pauline Anderson

May 21, 2015

Low-dose oral tetrahydrocannabinol (THC) does not reduce behavioral disturbances in patients with dementia, but it is well tolerated, a trial shows.

The fact that 4.5 mg of THC daily did not produce adverse events suggests that the dose was too low, said study author Marcel Olde Rikkert, MD, PhD, professor, geriatrics, Radboud University Medical Center, Nijmegen, the Netherlands.

"With a higher dose that has more biological effects, I'm pretty sure that we would have found some of the adverse effects that we found earlier using higher doses in younger volunteers," he told Medscape Medical News.

The "trick," he said, is finding a balance between a dose that's high enough for a biological effect but that's low enough to avoid significant adverse effects.

The results suggest that THC can be used as a pharmacologic treatment even in patients with dementia who have a "vulnerable" brain, said Dr Olde Rikkert. "There's now evidence that this group can be further explored."

The study, the largest to date to evaluate the efficacy and safety of oral THC for behavioral disturbances in patients with dementia, was published online May 13 in Neurology.

THC, the main constituent of cannabis, has both psychoactive and analgesic properties. The effect of THC on the endocannabinoid system is mediated by two cannabinoid receptors, CB1 and CB2.

The randomized, double-blind, controlled, multicenter, phase 2 trial included 50 patients (mean age, 78.4 years) with Alzheimer's disease (AD) or vascular or mixed dementia. The patients had clinically relevant neuropsychiatric symptoms (NPS), defined as a Neuropsychiatric Inventory (NPI) score of at least 10, with evidence of agitation, aggression, or aberrant motor behavior at least 1 month before screening.

The study did not include patients with severe aggressive behavior because safety assessments couldn't be carried out in such patients.

All patients were white and 50% were women. Almost half (48%) lived in the community, and about a quarter lived in a specialized dementia care unit, and another quarter lived in a nursing home.

Pain Requirement Dropped

Although the study initially recruited patients with both NPS and pain, the criterion of pain was dropped because not enough such patients came forward. But results from a recent cross-sectional study of 1000 patients with dementia and NPS convinced Dr Olde Rikkert that pain doesn't necessarily go hand and hand with NPI — "at least, that's the case in our nursing homes and in our memory clinics," he said.

Patients in the current study were randomly assigned to receive placebo or 1.5 mg of oral THC 3 times a day for 3 weeks. They also received 1000 mg of acetaminophen 3 times a day in case of reports of pain or, in noncommunicative patients, suspected pain.

The primary outcome was change in the NPS as measured by the NPI, a questionnaire that evaluates 12 behavioral domains, including agitation/aggression, and aberrant motor behavior. A total score for frequency and severity of NPS ranged from 0 (least impairment) to 144 (greatest impairment).

Secondary outcomes included activities of daily living and quality of life as assessed by caregivers using the Caregiver Clinical Global Impression of Change (CCGIC), a 7-point scale.

Overall, 94% of study participants completed the study. Median compliance based on pill counts was 98% in the THC group and 100% in the placebo group.

The NPI total score had decreased in both groups at 14 days (THC, P = .002; placebo, P = .002) and at 21 days (THC, P = .003; placebo, P = .001).

There were no differences in agitation, aberrant motor behavior, or other NPI subdomains except that the "eating disorders" domain favored placebo. Dr Olde Rikkert doesn't think this was "a real relevant effect," and was likely due to chance.

There was no benefit of THC in community-dwelling patients or inpatients.

And there were no between-group differences in quality of life or activities of daily living. CCGIC scores at 21 days showed that 8 patients (36.4%) in the THC group had minimal to marked improvement from baseline, which was not significantly different from the 12 (50.0%) in the placebo group (P = .35).

Adverse events were similar and were only mild. In the THC group, 16 patients (66.7%) experienced at least one adverse event compared with 14 (53.8%) in the placebo group (P = .36). THC-mediated adverse events, such as dizziness, somnolence, and falls were more frequently reported in the placebo group.

Changes in heart rate, blood pressure, and weight were similar for both groups. Episodic memory in 18 patients with mild dementia showed no significant difference in THC vs control patients, as assessed by the Paired Associate Learning Wechsler Memory Scale-Revised.

No "High" Feeling

No patients reported feeling "high," which was surprising to investigators. "That was a remarkable thing because we did expect 'high' feelings beforehand," said Dr Olde Rikkert. "So there's still room for increasing the dose."

When asked whether patients with dementia can even report being high, he said "absolutely."

"People often assume that patients with dementia can't say anything that's reliable, but their main problem is with memory," he said. "They can tell what they feel now, for example, if they're in pain or if they're feeling something very different, like being high. What they can't do is tell you if they were in pain or high 2 days ago."

The authors called the improvement in the placebo group "striking." The improvement might have been due to the attention and support they received during the study or to the placebo effect, said Dr Olde Rikkert.

"There's the expectation in many patients that this intervention might be very helpful. It's difficult to find the difference between this placebo effect and a real effect; it depends on the critical balance of effectiveness and adverse effects."

To counteract this placebo response, the authors suggested a randomized crossover design for future studies.

Finding a treatment for NPS is important because the current treatments, including antipsychotics and neuroleptics, carry significant adverse effects, such as increased risk for falls and for cerebrovascular and other morbidities, said Dr Olde Rikkert. Although haloperidol is not used for this reason anymore in North America, it is still being used in Europe and the Netherlands.

The most probable mechanism explaining the possible effects of THC on NPS is through relaxation and better coping ability. Dr Olde Rikkert believes that neuropsychiatric symptoms are a result of "losing a grip on your surroundings" combined with a personality trait that makes it difficult to cope with cognitive loss.

Although THC may exert neurobiological effects that change amyloid in the brain, "this amyloid linked effect is less likely to mechanistically affect neuropsychiatric symptoms," said Dr Olde Rikkert.

The researchers had aimed to recruit 130 patients for the study, but the underenrollment didn't affect the outcome. According to calculations and chance simulations the authors performed after the 50 study patients were recruited, it was "highly improbable" that exposure of more participants to the study interventions would have influenced the effect and the conclusion, he said.

Researchers did uncover some first signs of change in gait and balance (which are controlled by the cerebellum and basal ganglia) using very sensitive electronic measurements. They are now preparing to publish these results.

However, when reached for comment, Rachelle Doody, MD, PhD, chair, Alzheimer's Disease Research, and director, Alzheimer's Disease and Memory Disorders Center, Department of Neurology, Baylor College of Medicine, Houston, Texas, said that in her view, the study was too small to shed any new light on the usefulness of medical marijuana on dementia symptoms.

"This is far from being interpretable one way or the other," Dr Doody told Medscape Medical News. "It was a study that was supposed to have 130 people in it and ended up with 50. Any study of 50 people with any intervention in Alzheimer's disease is too low and not definitive."

She stressed that the study was not powered to look at the differences it was aiming to find. "It used one dose in a very small number of people and we can find out just a little bit about the safety of that dose by this study but really nothing about whether this is an avenue to pursue or not."

The researchers also plan to do another study using a higher dose of THC, although the exact dose has not yet been determined.

Dr Olde Rikkert has disclosed no relevant financial relationships.

Neurology. Published online May 13, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.