Does Wallis implant reduce adjacent segment degeneration above lumbosacral instrumented fusion?

Delayed complications following lumbar spine fusion may occur amongst which is adjacent segment degeneration (ASD). Although interspinous implants have been successfully used in spinal stenosis to authors' knowledge such implants have not been previously used to reduce ASD in instrumented lumbar fusion. This prospective controlled study was designed to investigate if the implantation of an interspinous implant cephalad to short lumbar and lumbosacral instrumented fusion could eliminate the incidence of ASD and subsequently the related re-operation rate. Groups W and C enrolled initially each 25 consecutive selected patients. Group W included patients, who received the Wallis interspinous implant in the unfused vertebral segment cephalad to instrumentation and the group C selected age-, diagnosis-, level-, and instrumentation-matched to W group patients without interspinous implant (controls). The inclusion criterion for Wallis implantation was UCLA arthritic grade <II, while the exclusion criteria were previous lumbar surgery, severe osteoporosis or degeneration >UCLA grade II in the adjacent two segments cephalad to instrumentation. All patients suffered from symptomatic spinal stenosis and underwent decompression and 2-4 levels stabilization with rigid pedicle screw fixation and posterolateral fusion by a single surgeon. Lumbar lordosis, disc height (DH), segmental range of motion (ROM), and percent olisthesis in the adjacent two cephalad to instrumentation segments were measured preoperatively, and postoperatively until the final evaluation. VAS, SF-36, and Oswestry Disability Index (ODI) were used. One patient of group W developed pseudarthrosis: two patients of group C deep infection and one patient of group C ASD in the segment below instrumentation and were excluded from the final evaluation. Thus, 24 patients of group W and 21 in group C aged 65+ 13 and 64+ 11 years, respectively were included in the final analysis. The follow-up averaged 60 +/- 6 months. The instrumented levels averaged 2.5 + 1 vertebra for both groups. All 45 spines showed radiological fusion 8-12 months postoperatively. Lumbar lordosis did not change postoperatively. Postoperatively at the first cephalad adjacent segment: DH increased in the group W (P = 0.042); ROM significantly increased only in group C (ANOVA, P < 0.02); olisthesis decreased both in flexion (P = 0.0024) and extension (P = 0.012) in group W. The degeneration or deterioration of already existed ASD in the two cephalad segments was shown in 1 (4.1%) and 6 (28.6%) spines in W and C groups, respectively. Physical function (SF-36) and ODI improved postoperatively (P < 0.001), but in favour of the patients of group W (P < 0.05) at the final evaluation. Symptomatic ASD required surgical intervention was in 3 (14%) patients of group C and none in group W. ASD remains a significant problem and accounts for a big portion of revision surgery following instrumented lumbar fusion. In this series, the Wallis interspinous implant changed the natural history of ASD and saved the two cephalad adjacent unfused vertebra from fusion, while it lowered the radiographic ASD incidence until to 5 years postoperatively. Longer prospective randomized studies are necessary to prove the beneficial effect of the interspinous implant cephalad and caudal to instrumented fusion. We recommend Wallis device for UCLA degeneration I and II.