INTRODUCTION

Lumbar adult spinal deformity (ASD) represents a challenge to the physician as there is currently no established decision-making pathway to determine the optimal surgical treatment. The ideal procedure aims to limit the extensiveness of surgery without compromising the outcome, though in practice many uncertainties exist. The aim of this study was to compare the surgical, radiographic and patient-rated outcomes in patients with ASD of the lumbar spine undergoing either short fusion (SF; max 2 levels) or long fusion (LF; 3 or more levels), while controlling for potential confounders.

METHODS

An international, multicentre database (7 European sites, 4 countries) containing the prospectively collected data from patients with adult deformity was searched for patients with deformity of either a degenerative or idiopathic origin, undergoing fusion surgery that included the lumbar spine (for SF, upper instrumented vertebrae L1 or below; for LF, T10 or below), at least 1yr ago. We identified 57 SF and 137 LF patients. Propensity score (PS) adjustment was used to evaluate the difference in outcome between the treatment groups, controlling for relevant baseline variables (age, ASA classification, SRS-22 subtotal score, relative lumbar lordosis, coronal Cobb angle, NRS back pain (% compared with leg pain as well as average pain intensity), sagittal and coronal balance, spondylolisthesis, central stenosis, and osteoporosis/osteopenia). Surgical outcomes (blood-loss, duration of surgery, length of stay) as well as 1-yr patient-rated outcomes (SRS-22) and coronal and sagittal curves were evaluated.

RESULTS

Compared with long fusion, short fusion was associated with less blood-loss during surgery (by 725ml, 95%CI 613-838; p<0.001), a shorter duration of surgery (by 86 min, 95%CI 66-107; p=0.001), and a shorter length of stay (by 3.5 days, 95%CI 2.5-4.5, p=0.001). However, at 1-yr follow-up, it was associated with a higher TL/L coronal curve (by 12.1 deg, 95% CI 9.5-14.6; p<0.001), a higher LS/S coronal curve (by 4.3 deg, 95% CI 2.9-5.8; p<0.001) and a lower relative lumbar lordosis (by 6.4 deg, 95% CI 3.5-9.4; p<0.001). SRS-subtotal scores at 1yr (completed by 74% of all patients) were slightly, but not significantly better in the short fusion group (by 0.11 points, 95%CI, -0.03-0.25; p=0.13); similarly, sagittal balance showed no significant difference between the groups (difference, 0.37, 95%CI, -8.5-9.3; p=0.93).

DISCUSSION

In analyses that controlled for potential confounders, short fusion was shown to be significantly less complex than long fusion — as evidenced by the shorter duration of surgery, lesser blood loss and lesser curve correction — and appeared to have no significant influence on patient-rated outcomes. This likely reflects appropriate and careful patient selection in this cohort. Further analyses of the selection procedure are warranted to refine indications for longer fusion and determine whether there exists, for example, a maximum level of deformity that can still be treated successfully with short fusion.