Cyclic testing of flexor tendon repairs: an in vitro biomechanical study.

Forty cadaveric flexor digitorum profundus tendons were repaired using the Tajima, Halsted, Silfverskiold, or Savage techniques. The tendons were cyclically loaded in sets of 4,000 cycles beginning at 25 N and increasing by 10-N increments until failure occurred. Gap formation was continuously monitored with an extensiometer. Only the Savage repair group consistently withstood 4,000 cycles of 25-N loading, with respect to 2-mm gap formation and repair rupture. Tendon repairs subjected to cyclic loading demonstrated gap formation and repair rupture at lower loads than with static testing. Of the methods of flexor tendon repair tested, only the Savage technique could withstand simulated early active motion.