Uninstrumented in situ fusion for high-grade childhood and adolescent isthmic spondylolisthesis: long-term outcome.

Background: Intermediate-term radiographic studies have shown that anterior and circumferential techniques result in high fusion rates in patients with high-grade spondylolisthesis, whereas posterolateral fusion is less successful. We are not aware of any long-term comparative studies in which these three methods have been evaluated with regard to functional outcome, including systematic spinal mobility and trunk strength measurements.
Methods: Sixty-nine of eighty-three consecutive patients with high-grade isthmic spondylolisthesis who underwent posterolateral (twenty-one), anterior (twenty-three), or circumferential (twenty-five) uninstrumented spondylodesis between 1977 and 1991 participated in the study. The average duration of follow-up was 17.2 years. Radiographs that were made preoperatively and at the time of the most recent follow-up were assessed with regard to fusion quality and degenerative changes. Outcome was assessed at the time of the most recent follow-up by independent observers on the basis of a physical examination, spinal mobility and nondynamometric trunk strength measurements, and Oswestry Disability Index scores.
Results: The mean preoperative vertebral slip was 61% in the posterolateral fusion group, 63% in the anterior fusion group, and 71% in the circumferential fusion group. The final fusion rate was 86% (eighteen of twenty-one) in the posterolateral fusion group, 100% (twenty-three of twenty-three) in the anterior fusion group, and 96% (twenty-four of twenty-five) in the circumferential fusion group. A decrease in lumbar intervertebral disc height at the first mobile level superior to the fusion was noted in five patients in the posterolateral fusion group, seven patients in the anterior fusion group, and one patient in the circumferential fusion group (p = 0.037). The mean Oswestry Disability Index score was 9.7 for the posterolateral fusion group, 8.9 for the anterior fusion group, and 3.0 for the circumferential fusion group (p = 0.035). Nondynamometric trunk strength measurements corresponded with referential values. Abnormally low lumbar flexion affected the posterolateral and circumferential fusion groups more often than the anterior fusion group (p = 0.0015). The percentage of slip showed inverse correlations with lumbar flexion, lumbar extension, and trunk side-bending.
Conclusions: As assessed on the basis of patient-based outcomes, circumferential in situ fusion provided slightly better long-term results than did posterolateral or anterior in situ fusion. When the radiographic and functional results were combined with the patient-based outcomes, the overall differences between the three groups were small.