Complications After Pediatric ACL Reconstruction: A Meta-analysis

OBJECTIVES: There are a variety of surgical techniques used for pediatric ACL reconstruction. With a majority of studies being small case series, there is limited data on the outcomes of surgery, frequency of complications, and rate of growth disturbance. The purpose of this meta-analysis is to review clinical outcomes following pediatric ACL reconstruction and the complications of growth disturbance. METHODS: The PubMed and EMBASE databases were searched for clinical studies on ACL ruptures in the skeletally immature from 1985 to 2016. Full-text studies in English and performed on humans were included (n = 5718). Titles were included if they discussed operative intervention on skeletally immature patients with ACL tears (n = 160). The references from these studies were reviewed to ensure no studies were missed. Studies that discussed complications with ACL reconstruction that were specific to the pediatric population were included (growth disturbance). Complications that were not specific to the pediatric population (i.e. knee stiffness, pain) were excluded, as were studies that reported results of surgeon surveys. Data was extracted including demographics, graft type, surgical technique, follow up, growth disturbance, re-rupture, and patient reported outcome scores . This data was analyzed in aggregate. RESULTS: Eighteen (18) studies meeting the inclusion criteria were identified (Table 1). There were a total of 225 patients with 226 knees who underwent pediatric ACL reconstruction. The average age at surgery was 12.5 years, 86% were male, and the mean post-operative follow up was 45.8 months. The majority of the grafts were hamstring autograft (169 of 226, or 75%). One hundred fifty eight patients (70%) underwent transphyseal reconstructions, 64 (28%) partial transphyseal, and 4 (2%) physeal sparing reconstructions. There were eight (3.5%) re-ruptures prior to skeletal maturity. Sixteen patients (7%) developed angular deformity, of which the majority was valgus deformity (n=13). One patient had valgus and flexion deformity. Two patients had recurvatum, and one patient had both varus and recurvatum deformities. Three patients required osteotomy for realignment and eight (72.7%) were observed. There were 10 patients with at least a 1 cm leg length discrepancy. Six patients with leg length discrepancy were treated with epiphysiodesis (66.7%) and 3 were observed. Two patients who had combined deformities (in multiple planes) were treated with osteotomy. Eight studies reported IKDC scores (range 81-100%) and grades (all either A or B). Five studies reported excellent Lysholm scores with mean scores greater than 94. CONCLUSION: Growth disturbance after ACL reconstruction in the skeletally immature is more common than what has been initially reported. This meta-analysis showed that even with hamstring autograft these occurrences do occur, particularly with transphyseal reconstructions. Of the complications noted, leg length discrepancy greater than 1cm was more likely to be surgically treated with epiphysiodesis. Angular deformity (valgus) occurred as well but was generally observed. Although there traditionally has been a concern for growth disturbance in very young patients (under 12 in males, under 10 in females) who undergo reconstruction, patients who are nearing skeletal maturity may actually be at higher risk of clinically significant growth disturbances due decreased ability for the body to respond to physeal insult.