It seems difficult but a simulated condition may be developed in all respects and then may possible, an intelligent exercise is required. 

DOUBLE-BLIND: Too much hard work which is easily criticized after the event and little statistical advantage

It is more elegant have  a placebo control and an one or more intervention groups without blinding. 

Many things cannot be properly blinded so why try? 

It is difficult, but performance bias must be prevented if efficacy of moxibustion is evaluated without blinding.

We have to remember that moxibustion works in two ways: (1) by deep heating a point or area and (2) through the influence of herbal substances contained in moxa. 1. It is not possible to perform a randomized double-blind trial when comparing moxibustion therapy with and without (the same applies to acupuncture). The patient will always see and feel that he has a moxibustion procedure or does not have one. Also, the therapist doing the moxibustion knows whether he is doing a moxibustion treatment or not. On the other hand, it is possible to study only the differences in the effects of heating the area by means of (a) real moxibustion or by means of, for example, radiation: (b) a TDP lamp (moxa lamp) and (C) ordinary infrared radiation (infrared light bulb). However, in the case of moxibustion there will always be a smoke effect perceived by the patient - so, a double blind test will also be debatable. 2. You can try to study the moxibustion effect by changing the herbal composition of moxa by adding to basic herb Artemisia (mugwort) some other herbs - here, however, the patient and the therapist will also feel different smells, so a double blind test will not be possible too.