Study design: A retrospective comparative study of prospectively collected data.
Objective: To compare clinical, radiographic, and health-related quality of life (Scoliosis Research Society [SRS]-24) outcomes in patients undergoing hemivertebra excision by simultaneous anteroposterior (AP) or posterolateral only approach (PL).
Summary of background data: AP hemivertebra resection for congenital scoliosis has provided reliable and safe long-term outcomes with respect to clinical and radiographic findings. Recently, PL excision has been described with rather high implant failure and revision rates. There are no studies comparing outcomes of AP and PL hemivertebra excision in 1 center. In addition, studies reporting health-related quality of life after surgery for congenital scoliosis are few.
Methods: Between 2001 and 2008, 12 AP and 11 PL hemivertebra excisions were conducted in 21 patients (16 males) with congenital scoliosis or kyphoscoliosis due to unbalanced hemivertebrae in our hospital. The average age at surgery was 4.0 years (range, 1.2-11.4) and the mean follow-up time 2.5 years (range, 1-7 years).
Results: Before surgery, the mean Cobb angle of the main curve was 33° (range, 25°-45°) in the AP group and 43° (range, 26°-87°) in the PL group. At the final follow-up visit, the main curve had been corrected to a mean of 12° (range, 2°-27°) and 15° (range, 6°-28°), respectively (P = NS). The Scoliosis Research Society-24 total scores were similar, 101 (range, 89-109) for the AP and 100 (range, 85-106) for the PL group at the final follow-up visit. Any complication was observed in 8% (1/12) of patients in the AP group and, whereas the PL group had 40% (4/10) complication rate with 2 temporary neurologic deficits, no implant failures were observed. All patients underwent spinal fusion during follow-up.
Conclusion: PL resection is technically more demanding and slightly faster method for hemivertebral resection. It has nearly as good correction rate as the AP-method but more minor complications. Hemivertebrectomy seems to provide a controllable untethering effect in patients with a symptomatic tethered cord.