We divided 21 flexor digitorum profundus tendons in the index, middle and ring fingers in seven cadaver hands into three groups. The tendons were cut in zone 2 and repaired using a 4-strand cruciate core suture repair with one of the following three materials in each group: (1) a knotless repair with a 2-0 bidirectional-barbed suture, which has similar tensile strength as a 4-0 non-barbed suture used in the other two groups; (2) a knotted locking repair with a non-barbed 4-0 conventional suture; and (3) a non-locking repair with a non-barbed 4-0 knotless suture. The repaired fingers were cyclically loaded through a simulated active range of motion to a 5 N load. We monitored and recorded the gap sizes at regular intervals during the test. The 2-0 bidirectional-barbed suture group and non-barbed suture groups developed gaps of 2.2 mm after 10 cycles and 2.4 mm after 20 cycles, respectively. Over 1000 cycles, the mean gaps were 3.2 mm in the 4-0 conventional suture group and 9.1 mm in the 2-0 bidirectional-barbed group. The tendons in the 2-0 bidirectional-barbed group gapped earlier, with statistically significant differences compared with those in the locking repair with a non-barbed 4-0 knotless suture group. The repair strength of the barbed suture technique was inferior to the cruciate repairs using a conventional 4-0 non-barbed suture tested in this cyclic-loading model.

Level of evidence: Level V