METHODS: We searched Medline, Embase, and Google Scholar, as well as the clinicaltrials.gov website from inception through December 16, 2017 for randomized control trials comparing IV ketamine to placebo infusions for chronic pain that reported outcomes for ≥48 hours after the intervention. Three authors independently screened the studies, pooled the data, and appraised risk of bias. Random-effects model was used to calculate weighted mean differences for pain scores and secondary outcomes. Our primary outcome was the lowest recorded pain score ≥48 hours after cessation of treatment. Secondary outcomes included responder rate and adverse effects.
RESULTS: Among 696 studies assessed for eligibility, 7 met inclusion criteria. All studies except one were at high risk of bias. These studies randomly assigned 211 patients with neuropathic (n = 2), mixed (n = 2), and nonneuropathic (nociplastic or nociceptive) (n = 3) pain. Three studies reported significant analgesic benefit favoring ketamine, with the meta-analysis revealing a small effect up to 2 weeks after the infusion (mean difference in pain scores, −1.83 points on a 0–10 numerical rating scale; 95% CI, −2.35 to −1.31 points; P < .0001). In the 3 studies that reported responder rates, the proportion with a positive outcome was greater in the ketamine than in the placebo group (51.3% vs 19.4%; relative risk, 2.43; 95% CI, 1.10–5.40; P = .029; I2 = 0.0%). No differences were noted based on pain classification or condition. Compared to low-dose ketamine studies and investigations that evaluated non–complex regional pain syndrome conditions, a small but nonsignificant greater reduction in pain scores was found among studies that either utilized high-dose ketamine therapy (P = .213) or enrolled complex regional pain syndrome patients (P = .079).
CONCLUSIONS: Evidence suggests that IV ketamine provides significant short-term analgesic benefit in patients with refractory chronic pain, with some evidence of a dose–response relationship. Larger, multicenter studies with longer follow-ups are needed to better select patients and determine the optimal treatment protocol.