BackgroundKnotted and knotless single-anchor reconstruction techniques are frequently performed to reconstruct full-thickness tears of the upper portion of subscapularis tendon. However, it is unclear whether one technique is superior to the other.Questions/purposes(1) When comparing knotless and knotted single-anchor reconstruction techniques in full-thickness tears of the upper subscapularis tendon, is there a difference in stiffness under cyclic load? (2) Are there differences in cyclic gapping between knotless and knotted reconstructions? (3) Are there differences in the maximal stiffness, yield load, and ultimate load to failure? (4) What are the modes of failure of knotless and knotted reconstruction techniques?MethodsEight matched pairs of human cadaveric shoulders were dissected, and a full-thickness tear of the subscapularis tendon (Grade 3 according to the Fox and Romeo classification) was created. The cadavers all were male specimens, with a median (range) age of 69 years (61 to 75). Before biomechanical evaluation, the specimens were randomized into two equal reconstruction groups: knotless single anchor and knotted single anchor. All surgical procedures were performed by a single orthopaedic surgeon who subspecializes in sports orthopedics and shoulder surgery. With a customized set up that was integrated in a dynamic material testing machine, the humeri were consecutively loaded from 10 N to 60 N, from 10 N to 100 N, and from 10 N to 180 N for 50 cycles. Furthermore, the gapping behavior of the tear was analyzed using a video tracking system. Finally, the stiffness, gapping, maximal stiffness, yield loads, and maximum failure loads of both reconstruction groups were statistically analyzed. Failure was defined as retearing of the reconstructed gap threshold due to rupture of the tendon and/or failure of the knots or anchors. After biomechanical testing, bone quality was measured at the footprint of the subscapularis using microCT in all specimens. Bone quality was equal between both groups. To detect a minimum 0.15-mm difference in gap formation between the two repair techniques (with a 5% level of significance; α = 0.05), eight matched pairs (n = 16 in total) were calculated as necessary to achieve a power of at least 90%.ResultsThe first study question can be answered as follows: for stiffness under cyclic load, there were no differences with the numbers available between the knotted and knotless groups at load stages of 10 N to 60 N (32.7 ± 3.5 N/mm versus 34.2 ± 5.6 N/mm, mean difference 1.5 N/mm [95% CI -6.43 to 3.33]; p = 0.55), 10 N to 100 N (45.0 ± 4.8 N/mm versus 45.2 ± 6.0 N/mm, mean difference 0.2 N/mm [95% CI -5.74 to 6.04]; p = 0.95), and 10 N to 180 N (58.2 ± 10.6 N/mm versus 55.2 ± 4.7 N/mm, mean difference 3 N/mm [95% CI -5.84 to 11.79]; p = 0.48). In relation to the second research question, the following results emerged: For cyclic gapping, there were no differences between the knotted and knotless groups at any load levels. The present study was able to show the following wi...