BackgroundThe margin convergence (MC) technique is used to repair longitudinal-type tears as direct repair of the apex of the longitudinal-type tear from medial to lateral is challenging. Few studies have compared the postoperative clinical outcomes and retear rates of arthroscopic rotator cuff repair (ARCR) using the MC technique with those of conventional ARCR without using the MC technique. Therefore, this study aimed to investigate the efficacy of MC on the clinical outcome and retear rates of patients with large-sized rotator cuff tears (RCTs). It was hypothesized that ARCR using the MC technique would yield clinical outcome and retear rates similar to those of ARCR without using the MC technique. MethodsThe medical records of consecutive patients who underwent ARCR for large-sized RCTs were retrospectively evaluated. Forty-four and 35 shoulders were repaired using MC (MC group) and not using MC (non-MC group), respectively. The range of motion and the Japanese Orthopaedic Association (JOA) score were assessed preoperatively and after a minimum follow-up period of 12 months postoperatively. Magnetic resonance imaging was performed at least 3 months postoperatively to determine whether the tendons had healed. ResultsThe average postoperative follow-up duration was 26.6 months and 24.3 months in the MC and non-MC groups, respectively. The mean range of motion and JOA score improved significantly in both groups postoperatively; however, the postoperative range of external rotation and the total JOA score was significantly lower in the MC group. The overall retear rate did not differ significantly between the MC (13/44, 29.5 %) and non-MC (7/35, 20.0%) groups, respectively (P=0.332). No significant differences were observed between the cases with retears in the two groups in terms of the postoperative range of motion and the total JOA score. In contrast, the postoperative range of external rotation and the total JOA score of the patients with healed tendons in the MC group were significantly poorer than those of the patients with healed tendons in the non-MC group. ConclusionsARCR using MC of large-sized longitudinal-type tears does not lead to better postoperative range of external rotation and clinical outcome compared with those of conventional repair.