Effect of attachment site and routing variations in split tendon transfer of tibialis posterior.

Abstract

Spastic equinovarus is a condition that commonly affects the gait of children with cerebral palsy. Split tendon transfers of the tibialis posterior (TP) are often performed to eliminate the excessive hindfoot inversion present in equinovarus. TP muscle moment arms were computed before and after three variations of split TP tendon transfer to assess the effectiveness of each. The three surgeries tested were (1) the original split TP transfer to the distal peroneus brevis tendon routing behind the lateral malleolus; (2) an attachment variation in which the transferred tendon half was attached to the proximal peroneus brevis tendon rather than to the distal site; and (3) a routing variation in which the transferred tendon half was passed through a window in the interosseous membrane and attached to the distal peroneus brevis tendon. Tendon tension was controlled for in these experiments because improper tensioning is often cited as a reason for poor outcomes. All three surgeries significantly reduced the ability of TP to invert the hindfoot, thus eliminating a potential deforming force, but the reduction following interosseous routing was significantly less than that found for the other two transfer procedures. Routing through the interosseous membrane also reduced the ability of the TP to plantarflex the foot, but ankle actions were preserved following the other two surgeries. Routing through the interosseous membrane to an anterior attachment site resulted in a muscle with minimal potential to create or resist action about the ankle joint. Similar moment arms were measured when the attachment site on the peroneus brevis tendon was located either proximally or distally, suggesting that this choice does not appear to significantly affect the mechanical outcome.