Page 145 - Reconstructive Techniques in Musculoskeletal Tumor Surgery

Results
Retrospective study
At review, 26 patients (68%) had no evidence of disease and 12 patients (32%) had died. Median follow-up was 19.5 years (95% CI 13.0 to 26.1). A total of 39 patients (79%) experienced one or more complications. Two patients (5%) required further surgery for joint instability (Henderson type 1 complication); both reconstructions of the distal radius were converted to an arthrodesis. Of 24 patients with a reconstruction around the knee, 15 (63%) had worn a brace for at least one year, or until failure of their graft, because of instability of the reconstructed joint.
Nonunion (Henderson type 2 complication) occurred in six reconstructions (16%). Re-operations took place after a median of 13 months (IQR 9 to 18), in two proximal tibial (14%), two distal femoral (20%) and two proximal humeral reconstructions (17%). One allograft (3%) was removed because of nonunion.
Fractures (Henderson type 3 complication) occurred in ten patients (26%), after
a median of 49 months (IQR 27 to 74). Fractures occurred in reconstructions of the
proximal tibia (n = 5, 36%), proximal humerus (n = 3, 25%) and distal femur (n = 2,
20%). All fractures occurred in reconstructions ≥ 10 cm. 7
Infections (Henderson type 4 complication) occurred in ve patients (13%), after two, 14, 17, 37 and 40 months. Two patients had a primary infection (5%), others occurred after operative intervention for other complications. Four infections occurred in the proximal tibia (29%) and one in the distal femur (10%). Allografts involving the tibia were associated with an increased risk of infection (OR 9.2, 95% CI 0.9 to 93.0, p = 0.06). Tibial grafts with a muscle ap appeared to have a lower infection risk (two of nine, 22%) than those without (two of ve, 40%).
Local recurrences (Henderson type 5 complication) occurred in two patients (5%, one osteosarcoma with wide margins and one parosteal osteosarcoma with an intralesional excision), after six and 13 months, respectively. Metastases developed in eight patients (21%).
In total, 20 reconstructions failed (53%): eight proximal tibial (57%), ve distal femoral (50%) and two distal radial (100%) allografts. Reasons for failure were fracture (n = 7, 18%), degenerative changes or subchondral collapse (n = 5, 13%, gure 1), infection (n = 3, 9%), instability (n = 2, 6%), tumor recurrence (n = 2, 6%, gure 2) and nonunion (n = 1, 3%) (table 2).
Osteoarticular allografts
143