Safety and tolerability of transnasal evaporative cooling for the acute treatment of migraine in an at‐home setting: A randomized, double‐blind, sham‐controlled, decentralized clinical trial

OBJECTIVES/BACKGROUND: Transnasal evaporative cooling (TNEC) is thought to modulate activity in the sphenopalatine ganglion and maxillary division of the trigeminal nerve. The safety, tolerability, and potential for therapeutic benefits of TNEC in adults with migraine have not been assessed in a fully decentralized setting. Our objectives were to determine the safety and tolerability of a TNEC device in adults with migraine and generate hypotheses for testing in studies designed to assess efficacy. METHODS: This prospective, double-blind, sham-controlled, randomized, decentralized clinical trial was conducted in the United States between November 2023 and June 2024. Eligible adults 18-65 years of age who self-reported migraine for ≥1 year were randomized to one of three active treatment groups (4, 6, or 10 L/min of dehumidified air) or to a sham control (2 L/min of ambient air delivered intermittently for ~10% of total time). Tolerability was measured by the percentage of participants who discontinued their treatment session due to discomfort. The assessment of TNEC treatment effects was hypothesis-generating; reported p-values are nominal. The trial was registered at clinicaltrials.gov (NCT06051604). RESULTS: Participants (N = 137) had a mean (SD) age of 39.4 (10.1) years, 79.6% (109/137) were female, and 85.4% (117/137) were White. The most common adverse events were rhinorrhea (2.2% [3/137]) and nasal irritation, ear pressure, runny nose, and sore throat (each 1.5% [2/137]). No participants (0% [0/128]) discontinued treatment early due to discomfort. At 2 h posttreatment, 57.6% ([19/33] 95% confidence interval [CI] = 39.2-74.5) of participants in the sham group had pain relief compared with 48.4% ([15/31] 95% CI = 30.2-66.9) of participants who received 4 L/min TNEC (p = 0.462), 61.8% ([21/34] 95% CI = 43.6-77.8) of those who received 6 L/min (p = 0.727), and 70.0% ([21/30] 95% CI = 50.6-85.3) who received 10 L/min (p = 0.306). The percentage of participants with pain freedom at 2 h posttreatment was higher in the TNEC 10 L/min group than in the sham group (33.3% [10/30] vs. 12.1% [4/33], p = 0.043). CONCLUSIONS: Treatment with TNEC was safe and feasible in a decentralized setting, and it was tolerable at high flow rates. Observed treatment effects in the acute treatment of migraine need to be confirmed in larger, adequately powered clinical trials.