Development of ankle and knee range of motion after isolated gastrocsoleus lengthening in children with cerebral palsy: a register-based longitudinal cohort study.

Background and purpose:  Outcome after gastrocsoleus lengthening in cerebral palsy (CP) is reported to be influenced by type of lengthening, age, CP subtype, and preoperative range of motion (ROM). We examined the development of ankle and knee ROM after 3 types of isolated gastrocsoleus lengthening.

Methods:  This is a register-based longitudinal cohort study based on data from the Swedish Cerebral Palsy follow Up Program, of children born 2000-2011 who underwent isolated gastrocsoleus lengthening. ROM development was analyzed using mixed-effects modeling. Event limits were defined as ankle ROM ≤ 0° or ≥ 20° and knee extension deficit ≤ -10° and described in Kaplan-Meier curves and Cox regression analyses. The study protocol was published at clinicaltrials.gov.

Results:  184 children were included. The mean differences in ankle ROM 10 years postoperatively between open tendo Achilles lengthening (OTAL) and percutaneous tendo Achilles lengthening (PTAL) was -2.3° (95% confidence interval [CI] -7.4 to 2.7), and between gastrocnemius lengthening (GCL) and PTAL -4.4° (CI -10.4 to 1.5). The adjusted hazard ratio (aHR), adjusted for baseline ROM, Gross Motor Function Classification System level, and CP subtype, comparing ankle event rates between OTAL and PTAL was 2.5 (CI 1.1-5.7). GCL was also associated with a higher event rate compared with PTAL, aHR 2.0 (CI 0.85-4.6). The adjusted mean difference in knee ROM at 10 years between OTAL and PTAL was 5.1° (CI 0.4-9.8), and between GCL and PTAL 1.9° (CI -3.6 to 7.6). Comparing event rates for the knee yielded uncertain results.

Conclusion:  PTAL appears at least as effective as OTAL and GCL for favorable ankle and knee ROM development in children with CP.