The purpose of this study was to compare the effectiveness and safety of percutaneous gracilis tenotomy (PGT) with those of the open precudre (OGT). We conducted this prospective study to describe the effects of PGT and to compare them with those of OGT.

A cross-over randomized controlled trial was conducted on 59 consecutive hips in 31 CP patients scheduled for gracilis tenotomy in the setting of multilevel tendon lengthening release procedures or hip surgery. A pediatric orthopaedic surgeon conducted a PGT. Another surgeon extended the wound to explore what had been cut during the PGT, and completed the tenotomy if necessary. Hip abduction (HA) was assessed by a third surgeon immediately before PGT, after PGT, and then after OGT, using a goniometer, in a standardized reproducible manner. All 3 surgeons were blinded to the others’ findings. Primary end-points included the percentage of muscle portion sectioned percutaneously, and the HA measure. Comparison between HA before and after PGT (1 intervention) was done using a paired t-test with a 95% confidence interval, and comparison between HA after PGT and OGT (2 interventions) was done using a Student's t-test with a 95% confidence interval. The bleeding was assessed and the iatrogenic lesions were identified. The relation between HA after PGT and the percentage of muscle portion sectioned percutaneously was evaluated by calculating the Pearson correlation coefficient.

Mean HA (hips neutral–knees extended) measured 33.71° preoperatively and increased to 45.90° after PGT (P<0.0001). After OGT, HA averaged 48.71° with no statistical gain compared with that observed after PGT (P=0.21). The muscular portion of gracilis origin was cut to an average of 91.95% (completely in only 14 cases, cut to more than 90% in 35 cases, more than 70% in 9 cases, and approximately 60% in 1 case). The gain in HA did not correlate with the extent of the muscular portion sectioned percutaneously (R=–0.043) Partial section of adductor brevis after PGT was encountered in 39 cases, especially the anterior and anterior-inferior aponeurosis fibers (<20% of muscle tendon thickness in the majority of cases). Considerable bleeding with hematoma formation requiring hemostasis during the open control procedure occurred in 30 hips. Partial iatrogenic injury of the anterior branch of the obturator nerve was encountered in one patient bilaterally with very severe adductor contracture, due to an anatomic variant of the branch localization (too medial).

This is the only prospective study concerning the effects of PGT. The authors detail the technique of percutaneous gracilis tenotomy and show that although PGT is a fast, simple and effective procedure, it is not as safe as the open release even when done correctly by an experienced surgeon, mainly because of the increased risk of bleeding.