Aromatase inhibitor (AI) therapy for estrogen receptor-positive breast cancer is known to induce or enhance musculoskeletal problems. We have previously reported that loss of grip strength is more pronounced in AI-users with extremes in BMI. We here report results from a larger prospective study. Postmenopausal early breast cancer patients scheduled to start AI or tamoxifen therapy were recruited. A functional assessment grip strength test was performed at baseline, 3, 6, and 12 months of therapy. BMI was assessed, and a rheumatologic questionnaire was completed at each visit. 188 patients on an AI and 104 patients on tamoxifen were enrolled. 74 % of AI-users reported new/worsened musculoskeletal complaints compared with 37 % in the tamoxifen group. This was translated in a larger grip strength decrease in patients experiencing AI-induced pain opposed to patients without new/worsened complaints (p = 0.0002). 15 % of AI-users discontinued therapy due to musculoskeletal symptoms, who were characterized by a larger grip strength reduction versus adherent patients (p = 0.0107). Young age (p = 0.0135), taxane-based chemotherapy (p = 0.0223), and baseline VAS score >4 (p = 0.0155) were predictors for AI-related musculoskeletal pain. In addition, a quadratic trend of BMI with grip strength change (p = 0.0090) and probability of discontinuation was observed (p = 0.0424). Musculoskeletal events were a substantial problem in AI-treated patients and an important reason for treatment discontinuation. The decrease in grip strength was larger in AI- than in tamoxifen-users, with a more pronounced change in symptomatic patients. The inverse relationship between BMI extremes and grip strength change was confirmed in this large group of AI-patients.