Sixty-two women with breast cancer scheduled for a mastectomy and implant-based reconstruction were randomized to receive either unrestricted range of motion or limited range of motion exercises from post-operative day (POD) 15 to POD 30. The primary outcomes of interest were the incidence of dehiscence, seroma, infection and necrosis. Secondary outcomes of interest included shoulder range of motion, pain on a verbal rating scale, and upper limb function measured by the Disability of the Arm, Shoulder and Hand (DASH) score. Outcomes were assessed up to 90 days post-operation. No statistically significant differences were observed between the free range and limited range groups in all primary outcomes (p>0.05 for all). The improvement over time in verbal pain scores, DASH scores, and shoulder range of motion in flexion, abduction, and external rotation was statistically significantly in favour of the free range group compared to the limited range group (p<0.05 for all).