OBJECTIVESTo establish the animal model of acute rotator cuff tear in rabbits, and study the effect of timing of surgical repair on healing of tendon-bone interface, formation and distribution of collagens in the supraspinatus tendon insertion and biomechanical properties of supraspinatus.

METHODSSupraspinatus tenotomy was performed in the right shoulder of 90 skeletally matured male New Zealand white rabbits to establish the animal model of acute rotator cuff tear. The rabbits were randomly divided into 3 groups : group of early repair, repaired at 1 week after tenotomy; group of late repair, repaired at 4 weeks after tenotomy; and group without repair, used as control. At 2 weeks, 4 weeks and 8 weeks after repair, healing of tendon-bone interface was observed by HE staining. Collagens were observed by Sirius Red F 3B (SR) in saturated carbazotic acid staining. The areas of type I and III collagens were measured by using imaging analysis software and the ratio of type I and III collagens were calculated. Failure loads of supraspinatus on both sides were measured. The percentage of failure loads of the surgical side was calculated and contralateral supraspinatus were uninjured.

RESULTSThere was no obvious fatty infiltration and muscle atrophy in supraspinatus in all groups. At 8 weeks, the formation of a new enthesis of supraspinatus in groups of early and late repair were observed. In groups of early and late repair, the ratio of areas of type I and III collagens at 8 weeks (2.02 ± 0.77 and 2.06 ± 0.58) was larger than that at 2 weeks (1.10 ± 0.24 and 1.14 ± 0.50, t = 3.082, 3.655, P < 0.01). At 2, 4 and 8 weeks, the percentages of failure loads of the surgical side and uninjured contralateral supraspinatus in group of early repair(38% ± 11%, 66% ± 7%, 89% ± 4%) and group of late repair (41% ± 16%, 63% ± 7%, 89% ± 9%) were both higher than that in group without repair (14% ± 6%, 32% ± 4%, 56% ± 12%); the differences were all statistically significant (group of early repair: t = 3.311, 8.549, 5.719; group of late repair: t = 3.713, 8.063, 6.044; P < 0.01). The percentage of failure loads of the surgical side and uninjured contralateral supraspinatus at 8 weeks was higher than those at 4 weeks (t = 3.878 - 4.613, P < 0.01) and 2 weeks (t = 7.158 - 10.024, P < 0.01) in all groups.

CONCLUSIONSSurgical repair within 4 weeks of acute rotator cuff tear lead to formation of a new enthesis of supraspinatus, improvement of both ratio of type I collagen in the supraspinatus tendon insertion and biomechanical properties of supraspinatus.