Rapid antidepressant effects of the psychedelic ayahuasca in treatment-resistant depression: a randomized placebo-controlled trial

Abstract

Background: Recent open-label trials show that psychedelics, such as ayahuasca, hold promise as fast-onset antidepressants in treatment-resistant depression.

Methods: To test the antidepressant effects of ayahuasca, we conducted a parallel-arm, double-blind randomized placebo-controlled trial in 29 patients with treatment-resistant depression. Patients received a single dose of either ayahuasca or placebo. We assessed changes in depression severity with the Montgomery-Åsberg Depression Rating Scale (MADRS) and the Hamilton Depression Rating scale at baseline, and at 1 (D1), 2 (D2), and 7 (D7) days after dosing.

Results: We observed significant antidepressant effects of ayahuasca when compared with placebo at all-time points. MADRS scores were significantly lower in the ayahuasca group compared with placebo at D1 and D2 (p = 0.04), and at D7 (p < 0.0001). Between-group effect sizes increased from D1 to D7 (D1: Cohen's d = 0.84; D2: Cohen's d = 0.84; D7: Cohen's d = 1.49). Response rates were high for both groups at D1 and D2, and significantly higher in the ayahuasca group at D7 (64% v. 27%; p = 0.04). Remission rate showed a trend toward significance at D7 (36% v. 7%, p = 0.054).

Conclusions: To our knowledge, this is the first controlled trial to test a psychedelic substance in treatment-resistant depression. Overall, this study brings new evidence supporting the safety and therapeutic value of ayahuasca, dosed within an appropriate setting, to help treat depression. This study is registered at http://clinicaltrials.gov (NCT02914769).

Keywords: Ayahuasca; HRS; MEQ; depression; psychedelics; randomized controlled trial (RCT).

Fig. 1.

Trial profile.

Fig. 2.

HAM-D scores at baseline and seven days after dosing. Statistical analysis shows a significant difference between ayahuasca (squares) and placebo (circles) seven days after dosing (p = 0.019). Between-group effect size is high (Cohen's d = 0.98). Values are (mean ± s.e.m.). HAM-D scores: mild depression (8–16), moderate (17–23), severe (⩾24).

Fig. 3.

MADRS scores as a function of time. Significant differences are observed between ayahuasca (squares) and placebo (circles) at D1 (p = 0.04), D2 (p = 0.04) and D7 (p < 0.0001). Between groups effect sizes are high at all time points after dosing: D1 (Cohen's d = 0.84), D2 (Cohen's d = 0.84), and D7 (Cohen's d = 1.49). Values are (mean ± s.e.m.). MADRS scores: mild depression (11–19), moderate (20–34), severe (⩾35). *p < 0.05; ***p < 0.0001.

Fig. 4.

Response and remission rates as a function of time. Response (a) and remission (b) rates were high for both groups at D1 and D2. At D7, response rate was significantly higher for ayahuasca [OR 4.95 (95% CI 1.11–21.02); p = 0.04; NNT = 2.66], while remission rate showed a trend toward significance [OR 7.78 (95% CI 0.81–77.48); p = 0.054; NNT = 3.44].

Fig. 5.

HRS subscales and MEQ30 factors during the dosing session. (a) Significantly higher scores in the ayahuasca group in five HRS subscales: perception (p < 0.0001), somaesthesia (p < 0.0001), cognition (p < 0.0001), intensity (p < 0.0001), and volition (p = 0.0003). Only affect was not significantly different between groups (p = 0.38). (b) Significantly higher MEQ30 scores in the ayahuasca group in the total MEQ30 score (p = 0.004), and three of its factors: mystical (p = 0.049), transcendence of time and space (p = 0.0008), and ineffability (p = 0.003), except for the positive mood (p = 0.32). Values are expressed as a percentage of maximum possible score.