We report a study evaluating and comparing shoulder-related morbidity associated with delayed breast reconstruction using either the conventional latissimus dorsi (LD) flap or the thoracodorsal artery perforator (TAP) flap.We conducted a retrospective cohort study of women over 18 years of age who had a unilateral, delayed breast reconstruction by either an LD or TAP flap at one center over a 56-month period. Shoulder function was assessed using the Constant Shoulder Score (CSS), which evaluated pain, activity of daily life (ADL), range of motion (ROM), and strength. A number of secondary outcomes were also examined.Forty-nine women were included. Demographic and breast treatment data were comparable between the groups. The mean total CSS score for the reconstructed side of the TAP flap was statistically significantly better than that of the LD flap, with a difference of 10.9 points (95% confidence interval [CI] = 2.6-19.2, p-value 0.01). The mean total CSS score for the nonreconstructed side was not statistically significant between groups, with a difference of 0.1 points (95% CI = -6.1-6.2, p-value 0.98). The subscore analysis revealed that women reconstructed using the TAP flap had a difference of 3.2 points for pain (p-value 0.003) and 5.5 points for ROM (p-value 0.011). The factors ADL and strength were of equal magnitude in both groups.Patients who undergo delayed breast reconstruction by the TAP flap seem less prone to suffer from postoperative pain and restricted ROM, thereby suggesting that this flap should be considered an advantageous alternative to the conventional LD flap. A randomized clinical trial is warranted to provide sufficient evidence to this statement.