Network meta-analysis of therapies for cluster headache: Effects of acute therapies for episodic and chronic cluster

Objective We used network meta-analysis (NMA) to characterize the relative effectiveness and harms of acute treatment options for cluster headache. Background There are few evidence-based acute treatments available for cluster headache. As most treatments were compared only against placebos in clinical trials, few head-to-head comparisons of treatments are available. Methods An a priori registered scoping review was performed to identify randomized controlled trials evaluating treatments in adult patients (>18 years old) with cluster headache per accepted diagnostic criteria. Bayesian NMAs were performed to compare treatments in terms of headache relief at 15 or 30 min, and also the occurrence of adverse events. We report odds ratios (ORs) of relative treatment effects along with corresponding 95% credible intervals (CrIs), as well as measures of treatment ranking. Results A total of 13 randomized controlled trials informed NMAs. We found high flow oxygen to be the most effective therapy for headache response at 15 and 30 min (OR 9.0, 95% CrI 5.3 to 15.9 vs. placebo), with injectable sumatriptan demonstrating the next highest effect (OR 6.4, 95% CrI 3.75 to 11.1 vs. placebo). High flow oxygen was also more effective than low flow oxygen (OR 2.55, 95% CrI 1.13 to 5.8), nasal spray zolmitriptan (OR 3.75, 95% CrI 1.72 to 8.4), octreotide (OR 4.5, 95% CrI 1.64 to 12.5), and non-invasive vagal nerve stimulation (nVNS; OR 5.2, 95% CrI 2.29 to 11.9). Sumatriptan injectable was also effective for headache relief and was found to be better than nasal spray zolmitriptan (OR 2.67, 95% CrI 1.21 to 5.9), octreotide (OR 3.20, 95% CrI 1.17 to 8.8), and nVNS (OR 3.69, 95% CrI 1.63 to 8.4). Octreotide (OR 4.1, 95% CrI 1.71 to 10.5) and sumatriptan (OR 2.40, 95% CrI 1.39 to 4.2) were associated with greater risk of adverse events compared to placebo, while other treatments did not demonstrate increased risk. When focusing on patients with episodic cluster headache, nVNS was significantly better than placebo (OR 4.9, 95% CrI 1.89 to 14.1). Conclusions Our findings suggest that high flow oxygen is more efficacious when compared to low flow oxygen for headache relief. When low flow oxygen fails in patients who can tolerate oxygen, increased flow rates should be tried. Additionally, high flow oxygen is likely more effective than zolmitriptan nasal spray, nVNS, and octreotide. Sumatriptan injectable is more likely to be effective when compared to zolmitriptan nasal spray, octreotide, and nVNS.