BACKGROUND: Repair tension of a torn rotator cuff can affect healing after repair. However, a measurement of the actual tension during arthroscopic rotator cuff repair is not feasible. The relationship between repair tension and healing of a rotator cuff repair remains unclear. PURPOSE/HYPOTHESIS: The purpose of this study was to evaluate the effect of repair tension on healing at the repair site. The hypothesis was that repair tension would be a major factor in determining the anatomic outcome of rotator cuff repair. STUDY DESIGN: Cohort study; Level of evidence, 2. METHODS: Arthroscopic rotator cuff repairs (132 patients) for full-thickness rotator cuff tears were analyzed. An intraoperative model was designed for the estimation of repair tension using a tensiometer. Magnetic resonance imaging (MRI) was performed approximately 1 year (mean [±SD], 12.7 ± 3.2 months) postoperatively for the evaluation of healing at the repair site. Multivariable analysis was performed for tear size, amount of retraction, and fatty degeneration (FD) of rotator cuff muscles. RESULTS: The mean repair tension measured during the arthroscopic procedure was 28.5 ± 23.1 N. There was a statistically significant correlation between tension and tear size (Pearson correlation coefficient [PCC], 0.529; P < .001), amount of retraction (PCC, 0.619; P < .001), and FD of the supraspinatus (Spearman correlation coefficient [SCC], 0.308; P < .001) and infraspinatus (SCC, 0.332; P < .001). At the final follow-up (12.7 ± 3.2 months), healing failure was observed in 18.2% (24/132), and repair tension also showed a significant inverse correlation with healing at the repair site (SCC, 0.195; P = .025). However, when sex, age, tear size, amount of retraction, tendon quality, and FD of rotator cuff muscles were included for multivariable logistic regression analysis, only FD of the infraspinatus showed an association with the anatomic outcome of repair (Exp(B) = 0.596; P = .010). CONCLUSION: Our intraoperative model for the estimation of rotator cuff repair tension showed an inverse correlation of repair tension with healing at the repair site, suggesting that complete healing is less likely with high-tension repairs. A significant association was observed on MRI between a high level of FD of the infraspinatus and repaired tendon integrity.