As with all things related to functional medicine the causal factors involved are important diagnostically in order to appropriate a potential benefit.
I've seen benefit from using acupuncture alone for advanced cases of RLS, but ultimately RLS is a topical expression of deeper pathology.
When RLS presents alone, and isn't related to underlying disease states supplementing with Mg (400-800 Mg) before bed is often enough to completely alleviate the issue.
When that makes a difference having a separate conversation with that patient on what is causing low tissue levels of Mg becomes important to me.
from a scientific point of view there is only few vidence, the latest cochrane review concludes: "There is insufficient evidence to determine whether acupuncture is an efficacious and safe treatment for RLS. Further well-designed, large-scale clinical trials are needed."
From a clinical point of view I agree with James Philips, however I relaised that treatment succes may be related to the intensity of needling (deqi sensation, neural irritation, etc.).
A case report stated that the treatment succes with acupuncture may be lowered in patients pretreated with dopaminergic drugs (http://aim.bmj.com/content/29/3/240.long).
I am a PhD student of TCM in China. From the symptoms of RLS, we can write out a prescription about acupuncture based on the TCM theories. However, I just went over the papers in cnki which was the most common used Chinese data base. There are only 8 papers refered to treating PLS with acupuncture, while the press of all the 8 papers is not so influence in China.
So, I agree with the cochrane review concludes above. You can treat RLS with acupuncture as a alternative try. Or you can design a RCT about this topic.
There is only one article named" WARMING NEEDLE TECHNIQUE WITH ACUPOINT INJECTION FOR THE TREATMENTOF 42 CASESOF RESTLESS LEG SYNDROME" which was writed in English.
From a clinical perspective the technique of Chinese Scalp Acupuncture (CSA) using the Chorea/Tremor line treats one phenotype of RLS. Always screen for iron deficiency, thyroid disease, diabetes, obstructive sleep apnoea, REM sleep disorder, Dystonias, Parkinson's disease and "Parkinson's plus" syndromes.
Designing an RCT for CSA is difficult, as the technique specifically involves strong stimulation with very close doctor patient contact. Streitberger needles may be suitable as an "active placebo", and a third arm with non-specific scalp massage could act as yet another "active placebo".
A "true" test would be to randomly assign 4 patient groups to Naloxone, Metoclopramide, Naloxone +Metoclopramide or placebo and then all groups to CSA.
Ethics approval and financial backing would be difficult here.