Long Adult Deformity Fusions: Is it Better to Stop at L5 or the Sacrum?

Background: For adult spinal deformity the decision often arises whether to terminate a long fusion at L5 or the sacrum. The decision is especially challenging in the presence of a healthy appearing 5-1 disk. The purpose of this study was to compare the outcomes of long adult deformity fusions to either L5 or the sacrum in the presence of a healthy 5-1 disk. In order to minimize the influence of confounding factors, a matched cohort study design was utilized.

Methods: 95 adult deformity patients that underwent fusion from the thoracic spine to either L5 or the sacrum were sorted according to five preoperative prognostic criteria: 5-1 disk degenerative status, patient age, smoking status, # levels fused and global sagittal balance. Patients with advanced preoperative L5-S1 disk degeneration were excluded. The remaining patients were then matched into two cohorts based on the five prognostic criteria (L5: 27 pts; Sac:12 pts). Patients were evaluated at 5.0-year mean follow-up (range: 2-15 years) in terms of SRS-24 outcomes, radiographic data, and complications.

Results: All procedures were performed between 1986 and 2000 at single institution using modern segmental instrumentation. The two cohorts were well matched in terms of 5-1 disk degenerative grade (L5:0.74; Sac:0.75), mean age (L5:44; Sac:44), % smoking (L5:18%;Sac:20%) and preoperative sagittal C7 plumb relative to the posterior cortex of S1 (L5:+3.3cm;Sac:+3.2cm). The # of levels fused (L5:10.4;Sac:11.5), and the use of circumferential procedures (L5:70%,Sac:75%) were also nearly identical. Scoliosis correction was similar for the two cohorts (¯Cobb: L5: 40%; Sacrum: 43%). Despite a similar frequency of osteotomies, surgical correction of sagittal imbalance was superior for Sacrum patients (D C7 plb:L5 :0.9cm; Sac: 3.2 cm; p=0.03). For L5 patients subsequent advanced L5- S1 disk degeneration was common (67%) and was associated with a loss of 5-1 lordosis (preop:19o;latest:10o,p=0.001) and inferior global sagittal balance at latest follow-up (C7plb-L5:+4.0cm; Sac:+1.2cm;p=0.06). Distal transition syndrome required arthrodesis extension to the sacrum for four L5 cohort patients at a mean of 2.5 years.While the sacrum cohort demonstrated superior maintenance of sagittal parameters, it required a more procedures (L5:1.7;Sac:2.8;p=0.03) and resulted in a greater frequency of major complications (L5:22%;Sac:67%;p=0.02) including nonunion (L5:4%; Sac:36%;p=0.001) and major medical morbidity (L5:0%; Sac:33%; p=0.001). SRS-24 scores reflected a similar patient assessment of outcome at latest follow-up (total score:L5:90; Sac:87,120 max.). Patients were 'satisfied' or very satisfied' in 89% cases for L5 and 82% for the sacrum.

Conclusions: At 5-year mean follow-up matched adult deformity patients undergoing arthrodesis from the thoracic spine to either L5 or the sacrum resulted in similar functional outcomes. The two distal fusion levels (L5 and the sacrum), however, each had unique limitations. Long fusions to the sacrum required more procedures and had a higher frequency of early complications than similar fusions to L5. For L5, subsequent subjacent disk degeneration was common and typically resulted in a forward shift in sagittal balance. Arthrodesis extension to the sacrum has been required in a few cases. The ultimate influence of subsequent L5-S1 degeneration on long-term clinical outcomes, however, remains to be seen.