Different risk factors for early-onset adjacent segment disease at L3-4 and L5-S1 segments after isolated L4-5 lumbar fusion surgery: 2-year follow-up.

OBJECTIVE

Posterior or transforaminal lumbar interbody fusion (PLIF/TLIF) has become the standard procedure for treating degenerative lumbar diseases. With the increasing number of surgeries, adjacent segment disease (ASD) is becoming widely known as a postoperative complication. However, the association between lumbosacral sagittal alignment and ASD following L4-5 isolated fusion surgery has not been fully elucidated. The aim of this study was to identify the preoperative radiological risk factors for L3-4 ASD and L5-S1 ASD independently after isolated L4-5 PLIF/TLIF.

METHODS

The authors retrospectively reviewed the data of 151 patients with degenerative lumbar diseases who underwent isolated L4-5 PLIF/TLIF at their institution between April 2013 and August 2022. L3-4 ASD and L5-S1 ASD were evaluated separately, and these groups were compared with the non-ASD group. Radiological ASD was defined as disc height (DH) loss (> 3 mm), posterior opening (> 5°) on flexion, or progression of slippage (> 3 mm) as observed on plane radiographs. Symptomatic ASD was defined as the presence of symptoms attributable to the adjacent segment that required revision surgery within 2 years. Preoperative lumbosacral parameters, including slippage at L4-5 (Meyerding grade), pelvic tilt, sacral slope, pelvic incidence (PI), lumbar lordosis (LL), PI-LL, L1 sagittal vertical axis (SVA), and L4 SVA, were measured using standing radiographs. Intraoperative distraction at the L4-5 DH, preoperative disc degeneration (Pfirrman grade), vacuum phenomenon, foraminal stenosis, and additional decompression at adjacent segments were also evaluated.

RESULTS

Of 151 unique patients, 31 (20.5%) had ASD, with 20 (13.2%) included in the L3-4 ASD group and 13 (8.6%) included in the L5-S1 ASD group (2 patients were included in both groups). Multivariate analysis revealed that additional L3-4 decompression and distraction at L4-5 DH were significantly associated with L3-4 ASD, whereas L1 SVA > L4 SVA (L1 plumb line anterior to the L4 plumb line) was associated with L5-S1 ASD. The risk of L5-S1 ASD increased by a factor of 4.13 (p = 0.004) in patients with sagittal imbalance, as indicated by L1 SVA > L4 SVA.

CONCLUSIONS

These findings suggest distinct characteristics between L3-4 ASD and L5-S1 ASD. L3-4 ASD was not associated with lumbosacral sagittal imbalance. By contrast, an anterior shift of the lumbar loading axis, as indicated by L1 SVA > L4 SVA, was associated with the development of L5-S1 ASD. Preoperative L1 SVA > L4 SVA might serve as a convenient predictive parameter for L5-S1 ASD after isolated L4-5 fusion surgery.