Background: Hamstring surgical lengthening (HSL) has been frequently performed for the correction of knee flexion deformity in cerebral palsy (CP), although recurrence is described in long-term follow-up. Repeat hamstring surgical lengthening (RHSL) can be an option for recurrent knee flexion deformity; however, the results of this approach are still controversial. The purpose of this study was to compare the results of primary HSL and RHSL in CP.
Methods: Patients with spastic diplegic CP, Gross Motor Function Classification System levels I to III, underwent bilateral medial HSL with complete documentation in the gait laboratory before and after the intervention, were included in the study. A total of 229 subjects met the inclusion criteria and were divided into 2 groups: group A was formed by those who received medial HSL for the first time (185 patients), and group B was composed of individuals who underwent RHSL (44 patients). Clinical and kinematic parameters were evaluated before and after the intervention, and the results compared.
Results: The groups were matched with regard to sex distribution, Gross Motor Function Classification System levels, and follow-up time (>2 y). Popliteal angle was reduced in groups A (60.3 to 51.4 degrees, P<0.001) and B (56.1 to 51.5 degrees, P=0.001) after the intervention. Knee flexion at initial contact was reduced from 40.8 to 28.9 degrees in group A (P<0.001) and from 40.4 to 35.1 degrees in group B (P=0.001). Reduction of minimum knee flexion in the stance phase (24.9 to 17.5 degrees, P<0.001) and improvement of the Gait Deviation Index (52.9 to 60.2, P<0.001) occurred only in group A. Anterior pelvic tilt (APT) increased in groups A (from 17 to 19.5 degrees, P<0.001) and B (from 14.9 to 19.4 degrees, P<0.001) after treatment. Finally, in the comparison between groups, the reduction of knee flexion at initial contact was more significant in group A (P<0.001), whereas the increase of APT was higher in group B.
Conclusions: In the present study, the improvement of knee extension during the stance phase was observed only after the primary medial HSL. Moreover, the increase of APT was more significant when RHSL was performed.
Level of evidence: Level III.