On demand sildenafil as a treatment of Raynaud's Phenomenon: a series of N-of-1 trials

Background: Treatment of Raynaud phenomenon (RP) with phosphodiesterase-5 inhibitors has shown moderate efficacy. Adverse effects decrease the risk-benefit profile of these drugs, and patients may not be willing to receive long-term treatment. On-demand single doses before or during exposure to cold may be a good alternative. Objective: To assess the efficacy and safety of on-demand sildenafil in RP. Design: Series of randomized, double-blind, n-of-1 trials. (ClinicalTrials. gov: NCT02050360) Setting: Outpatients at a French university hospital. Participants: Patients with primary or secondary RP. Intervention: Each trial consisted of a multiple crossover study in a single patient. Repeated blocks of 3 periods of on-demand treatment were evaluated: 1 week of placebo, 1 week of sildenafil at 40 mg per dose, and 1 week of sildenafil at 80 mg per dose, with a maximum of 2 doses daily. Measurements: Raynaud Condition Score (RCS) and frequency and daily duration of attacks. Skin blood flow in response to cooling also was assessed with laser speckle contrast imaging. Mixed-effects models were used and parameters were estimated in a Bayesian framework to determine individual and aggregated efficacy. Results: 38 patients completed 2 to 5 treatment blocks. On the basis of aggregated data, the probability that sildenafil at 40 mg or 80 mg was more effective than placebo was greater than 90% for all outcomes (except for RCS with sildenafil, 80 mg). However, the aggregated effect size was not clinically relevant. Yet, substantial heterogeneity in sildenafil's efficacy was observed among participants, with clinically relevant efficacy in some patients. Limitation: The response to sildenafil was substantially heterogeneous among patients. Conclusion: Despite a high probability that sildenafil is superior to placebo, substantial heterogeneity was observed in patient response and aggregated results did not show that on-demand sildenafil has clinically relevant efficacy. In this context, the use of n-of-1 trials may be an original and relevant approach in RP.