Anterior single rod instrumentation for thoracolumbar adolescent idiopathic scoliosis with and without the use of structural interbody support

Abstract

Study design: A radiographic and clinical outcomes analysis of 41 patients treated for thoracolumbar adolescent idiopathic scoliosis utilizing a single anterior rigid rod construct.

Objectives: To evaluate the necessity of structural interbody support to improve primary curve correction and preserve or augment lordosis when used in conjunction with a single anterior rigid rod construct, to identify parameters that predict horizontalization of the lowest instrumented vertebra, adjacent disc angulation, and distal uninstrumented vertebrae, and to assess patient satisfaction following surgery.

Background data: Instrumentation-induced kyphosis has been a concern with nonrigid anterior systems used in the past for the treatment of scoliosis. Interbody structural support has been recommended to maintain appropriate sagittal profile when anterior systems are utilized. It has also been suggested that the use of structural interbody support creates a fulcrum to increase curve correction when compression is applied to the convexity of the deformity. However, the necessity of interbody structural support when used in conjunction with a rigid anterior system has not been previously evaluated in patients with adolescent idiopathic scoliosis.

Materials and methods: Forty-one patients mean age 15.9 years (range 12.1-18.6 years) with thoracolumbar adolescent idiopathic scoliosis underwent anterior spinal fusion using a single 6.0 to 6.5 mm solid rod construct between June 1995 and August 1999 performed by the senior author (T.G.L.). Four additional patients with thoracolumbar curves with similar anterior instrumentation over the same time period were lost to follow-up or had incomplete records and were not included in the study. Structural interbody support was used in 21 patients and packed morselized autograft alone was used in 20 patients. The patients in the group with packed morselized bone alone generally underwent surgery earlier in the series before the author began using structural interbody support on a regular basis. Each patient had a minimum follow-up of 3 years. Preoperative, initial, and most recent (>3 years) follow-up radiographs were reviewed to determine in each group Cobb angle measurements, flexibility of primary, secondary, and fractional curves, apical and end vertebral translation, lowest instrumented vertebral and caudal disc angulation, global coronal and sagittal balance, and sagittal Cobb measurements in both instrumented levels as well as lumbar lordosis (T12-S1). In addition, the SRS outcomes instrument was completed by 38 of 41 patients.

Results: The mean preoperative primary curve in patients with structural support was 47 degrees (Group II) and 45 degrees in patients without structural support (Group I). Mean curve correction was to 13 degrees in Groups I and II. One patient in Group II became slightly more unbalanced at final follow-up; otherwise all were improved after surgery. Sagittal measurements over instrumented segments as well as total lumbar lordosis (T12-S1) was maintained between preoperative and final postoperative values in both groups. Similarly, in both groups, when horizontalization of the distal end instrumented vertebra was achieved on the preoperative reverse side-bending radiograph, more normal relationships were achieved between instrumented and distal noninstrumented segments (adjacent disc angulation and fractional lumbar curve) at final follow-up (P <or= 0.01). Patients in both groups were equally pleased with their clinical outcomes based on the SRS outcomes instrument.

Conclusions: The use of interbody structural support does not appear to be necessary to maintain an appropriate sagittal profile or to maximize coronal curve correction when a rigid rod construct with packed morselized bone is used for the treatment of thoracolumbar adolescent idiopathic scoliosis. Parameters predicting horizontalization of the lower instrumented vertebra and uninstrumented segments below the construct were identified, which, if achieved, should predict an optimal long-term outcome. Clinical outcomes were very good in both groups.