Randomised controlled trial of rhinothermy for treatment of the common cold: a feasibility study

van de Hei, Susanne; McKinstry, Steven; Bardsley, George; Weatherall, Mark; Beasley, Richard; Fingleton, James

[4] Participants in this study 92 were recruited primarily through social media so the results should be generalisable to people in 93 the community with symptoms of the common cold. 94 95 Three main limitations apply to these results. Firstly, this study was designed as a feasibility 96 study which means that although the modified Jackson score was an efficacy outcome measure, 97 we did not formally pre-specify it as the primary outcome measure. Secondly, due to the nature 98 of the intervention we were unable to blind participants and investigators to the allocated 99 intervention and a future study aiming to replicate these results should explore the option of 100 sham rhinothermy. Finally, the small sample size means that although this study showed 101 clinically and statistically significant results it can be considered relatively 'fragile', in that the 102 modified Jackson score would only need to change in a small number of participants who 103 received the rhinothermy intervention to match the placebo group for our conclusions to be 104 different. [34, 35] A formal fragility index cannot be calculated for the change in modified Jackson 105 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 participants used the myAIRVO 2 device for two hours per day on average throughout the 5 day 124 treatment period, and for 2.6 hours on the days of use. As instructed not all participants used the 125 device late in the treatment course once their symptoms had improved. Around a third of 126 participants reduced the flow rate to 25 or 30L/min suggesting that giving participants the option 127 of reducing the flow from the initial 35L/min, or having a lower starting flow is worthwhile. The 128 difference between the randomised groups was observed within 24 hours, so that the efficacy of 129 single dose regimens or sessions over a two to three day period, which may enhance its 130 implementation in clinical practice, will be worth further study. However, while alternative 131 rhinothermy regimens will need to be assessed, our study findings suggest that the regimen used 132 in this feasibility study should also be used in future RCTs. 133 Our study has also been informative in terms of the likely number of participants required in a 135 RCT. We have estimated the change in modified Jackson score which denotes substantial 136 clinical benefit to be a five unit change. Based on the largest SD of 6.6 (from the upper CI of the 137 Jackson score at Day 5), a sample size of 38 per arm would give 90% power, alpha 5%, to detect 138 a difference in symptom score equal to the substantial clinical benefit at day 5. Assuming a 10% 139 drop out rate, derived from the lower confidence interval of our estimate, this would require a total 140 of 126 participants in an RCT investigating three randomised treatments, comprising the current 141 rhinothermy regimen, a sham rhinothermy regimen and vitamin C. 142 143 A specific respiratory virus was identified by PCR in 16/30 (53%) of participants, with the majority 144 being rhinovirus. This respiratory virus detection rate was comparable to previous studies, which 145 report detection rates between 43 and 64%.[36-38] We found no difference in the treatment 146 effect between participants who were nasal swab PCR positive or negative, indicating