Aim: This pilot study compared the efficacy of superficial dry needling (SDN), deep dry needling (DDN), placebo dry needling (PDN) and a common physiotherapy control group (CON) when used in the treatment of myofascial trigger points (MTrPs) in rotator cuff syndrome (RCS) patients. Methodology: A randomised, single-blind, placebo-controlled pilot study (n=20) was conducted comparing the three needling groups to each other and to a common physiotherapy protocol. Participants were selected patients presenting for treatment in a private practice. The objectives of the study were to compare the groups on three levels: Pre trial-Post trial, within individual treatment session (Intra-treatment), and between treatment sessions (Inter-treatment). All groups were treated with the same basic common protocol but three of them had the addition of one each of the needling interventions. A modified Constant-Murley scale, range of motion and power were used as outcomes measures. Ethical permission was obtained from the University of the Witwatersrand. Results: Results were analysed for the four groups using an ANCOVA. DDN had significant improvement over CON over the trial period (p≤0.05) and SDN (p≤0.02). This was particularly due to highly significant intra-treatment effect on internal range of motion at session 3 (p≤0.01) and the highly significant inter-treament effect between session 3 and 4 (p≤0.03). DDN was significantly less effective than the other groups at session 3 (p≤0.01) and session 4 (p≤0.03). External rotation power was also significantly greater for DDN between sessions 2 and 3 (inter-treatment) (p≤0.05). 49% of the MTrPs identified were found within the infraspinatus muscles. Discussion: Twitch-obtaining dry needling (DDN) appears to show greater clinical benefit on the effects of myofascial trigger points than SDN, CON or PDN. The effect appears to correlate with the greater incidence of MTrPs in the infrapinatus muscles whose functions directly relate to the improved parameters. The clinical effect may be related to the effects of the bleeding elicited by intramuscular needling (humoral effects). This is evidenced by the transiently poor effect of DDN immediate following treatment becoming significanly better by the following treatment. Conclusion: The pilot study showed that DDN may be an effective treatment for RCS when used in conjunction with a conventional physiotherapy programme. The elicitation of a local twitch response and associated bleeding may be significant. In future studies, particular attention should be paid to both the infraspinatus muscle and the timing of the intervention and observation intervals.