BackgroundA rotational gastrocnemius flap is often used for soft tissue reconstruction after proximal tibia sarcoma resection. However, little is known about the frequency and severity of complications and the recovery of extensor function after this procedure.Questions/purposesAfter gastrocnemius flap reconstruction with split-thickness skin grafting (STSG) to augment the extensor mechanism repair after proximal tibial resection for sarcoma, we asked: (1) What ROM was achieved (including extensor lag and active flexion)? (2) How often did complications and reoperations occur and what caused them?MethodsBetween 1991 and 2014, one surgeon treated 26 patients with proximal tibial resections for primary bone sarcoma. Of these, 18 were reconstructed with the preferred approach: resecting the proximal tibia leaving the patellar tendon in continuity with the tibialis anterior fascia whenever possible (10), cementing a stemmed proximal tibial endoprosthesis, suturing the patellar tendon to the implant, rotating a medial (16) or lateral (two) gastrocnemius flap over the tendon and prosthesis to augment the repair, and covering the flap with STSG. Alternative methods were used when this was technically impossible (one patient), when there was no advantage to secondary soft tissue coverage (two patients), or when the limb could not be salvaged (five patients). Of the 18 treated with gastrocnemius flaps, two were lost to followup or died of disease before the 24-month minimum and excluded; the median followup of the remaining 16 was 6 years (mean, 9.9 years; range, 2.3-21.7 years); three patients died of disease, and four have not been seen within the last 5 years. We reviewed medical records for passive and active extension, maximum flexion achieved, and complications requiring reoperation. ROM in patients with successful limb salvage was graded as excellent (flexion ≥ 110° and no lag), good (flexion 90°-110° and lag ≤ 10°), fair (one function limited: either flexion < 90° or lag > 10°), or poor (both functions limited: flexion < 90° and lag > 10°).ResultsAt latest followup, three patients had undergone amputation for deep infection. Of those remaining, median active flexion was 110° (mean, 104°; range, 60°-120°) and extensor lag was 0° (mean, 4°; range, 0°-10°). ROM was excellent in nine patients, good in three, fair in one, and poor in none. We observed 18 complications requiring reoperation in 12 patients, including deep infection (four), patellar tendon avulsion/attenuation (three), and flap necrosis (one). Survivorship free from revision or loss of the gastrocnemius flap was 74% (95% confidence interval [CI], 5.6-95.8) at 2, 5, and 10 years. Survivorship free from reoperation for any cause was 74% (95% CI, 52.0-96.0) at 2 years, 52% (95% CI, 25.8-77.8) at 5 years, and 35% (95% CI, 0-61.5) at 10 years using Kaplan-Meier analysis.ConclusionsAlthough most patients regained functional ROM including active extension, 12 required reoperation for complications including infection and early extensor mechanism failures. Despite the observed risks, we believe the gastrocnemius flap with STSG should be considered a suitable approach to provide active extension and soft tissue coverage given the paucity of good surgical options for extensor mechanism reconstruction in this challenging clinical setting.Level of EvidenceLevel IV, therapeutic study.