Ketamine for chronic non-cancer pain: A meta-analysis and trial sequential analysis of randomized controlled trials.

BACKGROUND: Ketamine has been suggested to be efficient in relieving chronic pain. However, there is inconsistency across studies investigating the effect of ketamine for chronic pain management. We aimed to perform a meta-analysis in order to assess the efficacy of this compound during chronic non-cancer pain conditions.

METHODS: The study consisted in a meta-analysis of clinical trials comparing ketamine to a placebo during chronic non-cancer pain. The primary endpoint of this study was pain relief 4 weeks after the beginning of treatment. Secondary outcomes were: pain relief 1, 2, 8 and 12 weeks after the beginning of treatment and incidence of psychedelic manifestations.

RESULTS: Six studies were included in this meta-analysis. Overall, 99 patients received ketamine and 96 received placebo. Ketamine did not decrease pain intensity at 4 weeks (MD (on a 0 to 10 scale) = -1.12 [-2.33, 0.09], GRADE evidence: very low). However, analysing studies with no high-risk bias found ketamine to decrease pain intensity at 4 weeks and increased the level of GRADE evidence to moderate. Trial sequential analysis confirmed the overall result and revealed the lack of power of this meta-analysis. Ketamine also decreased pain intensity at all other evaluated points in time. Ketamine increased the incidence of psychedelic manifestations in comparison to placebo.

CONCLUSION: Results of this meta-analysis found moderate evidence suggesting the efficacy of ketamine during chronic pain. Further studies are warranted to conclude about the effect of ketamine during chronic pain conditions and to determine optimal administration regimes of this agent during this condition.

SIGNIFICANCE: Ketamine has been found interesting for managing chronic pain. We performed a meta-analysis aiming to confirm those results. Ketamine was found efficient in alleviating pain up to 12 weeks after the beginning of treatment. However, overall evidence favouring the use of this compound was very low.