Orthopedics · Medical Literature
Posterior lumbar interbody fusion versus interspinous stabilization in lumbar spinal stenosis: a meta-analysis.
Abstract
Posterior lumbar interbody fusion (PLIF) and interspinous stabilization (ISS) are both used to treat lumbar spinal stenosis (LSS), but their comparative effectiveness and safety remain debated. Evidence comparing these approaches has evolved with the introduction of newer ISS devices and advances in contemporary surgical practice. A systematic search of PubMed, Google Scholar, Embase, and Scopus was conducted up to December 2, 2025, to identify comparative studies evaluating PLIF versus ISS for LSS. Nineteen studies met the inclusion criteria, comprising 214,155 patients (203,414 PLIF; 10,741 ISS). Outcomes included pain and functional measures, reoperation and revision rates, perioperative parameters, and complications. Three additional analyses were performed: subgroup analysis by publication period (before 2017 vs. 2017 onward), sensitivity analysis after excluding studies with very large sample sizes, and a study-design sensitivity analysis after excluding randomized controlled trials (RCTs). PLIF and ISS showed no significant differences in short- and mid-term pain or disability outcomes, including VAS leg and back pain at 3 months and 1 year, ΔVAS, ODI at 3 months and 1 year, and ΔODI. PLIF was associated with a greater change in segmental range of motion (ΔROM) (MD 6.96, 95% CI 4.94-8.97; p < 0.001). Compared with PLIF, ISS was associated with shorter operative time (MD 80.26 min, 95% CI 55.97-104.55; p < 0.001), lower blood loss (MD 260.54 mL, 95% CI 173.89-347.18; p < 0.001), and shorter hospital stay (MD 3.50 days, 95% CI 2.07-4.94; p < 0.001). Reoperation rates did not differ significantly between groups (RR 1.09, 95% CI 0.47-2.51). Similarly, revision rates were not significantly different between PLIF and ISS (RR 1.14, 95% CI 0.56-2.31). PLIF was associated with higher rates of mechanical complications (RR 1.60, 95% CI 1.09-2.33; p = 0.015), infectious complications (RR 3.50, 95% CI 2.67-4.58; p < 0.001), cardiopulmonary complications (RR 5.29, 95% CI 1.07-26.19; p = 0.041), and overall complications (RR 2.44, 95% CI 1.82-3.26; p < 0.001). In studies published before 2017, PLIF was associated with greater ΔROM and lower reoperation rates, but these differences were no longer observed in studies published from 2017 onward. Excluding studies with very large sample sizes did not materially change the main findings. Similarly, exclusion of the two RCTs did not materially alter the direction or statistical significance of the pooled estimates. PLIF and ISS provide broadly comparable short- and mid-term pain and functional outcomes for LSS. ISS is associated with consistent perioperative advantages and lower overall complication rates, while differences in reoperation and revision rates were not significant. Subgroup and sensitivity analyses, including the analysis excluding RCTs, suggest that earlier PLIF advantages in ΔROM and reoperation were not maintained in more recent studies, whereas the perioperative benefits of ISS remained robust. Larger randomized studies with longer follow-up and device-specific analyses are needed.
Clinical relevance
In adults with lumbar spinal stenosis (LSS), does posterior lumbar interbody fusion (PLIF) differ from interspinous stabilization (ISS) in pain relief, functional outcomes, reoperation rates, perioperative parameters, and complications? __SNAPSHOT__ {"studyDesign":"Systematic review with meta-analysis of comparative studies (RCTs and observational designs); three pre-specified sensitivity/subgroup analyses performed: (1) publication period subgroup (pre-2017 vs. 2017 onward), (2) exclusion of studies with very large sample sizes, (3) exclusion of RCTs (study-design sensitivity analysis)","setting":"International; multi-database search (PubMed, Google Scholar, Embase, Scopus) through December 2, 2025; institution type of included studies not reported","sampleSize":"n = 214,155 total (PLIF: n = 203,414; ISS: n = 10,741) across 19 included comparative studies","population":"Adults undergoing surgical treatment for lumbar spinal stenosis (LSS); specific inclusion criteria for patient-level characteristics (age range, grade of stenosis, presence of spondylolisthesis, number of levels) not fully reported in the available abstract","intervention":"Posterior lumbar interbody fusion (PLIF)","comparator":"Interspinous stabilization (ISS; encompasses multiple device generations)","primaryOutcome":"Not explicitly designated as a single primary endpoint; outcomes evaluated included pain (VAS leg and back pain at 3 months and 1 year, ΔVAS), functional disability (ODI at 3 months and beyond), reoperation and revision rates, perioperative parameters, and complications","followUp":"Short- and mid-term follow-up reported (minimum 3 months, extending to at least 1 year); maximum follow-up duration not reported in the available abstract","fundingSource":"Not reported"}
Key findings
The meta-analysis of 19 comparative studies comprising 214,155 patients (203,414 PLIF; 10,741 ISS) found no statistically significant differences between PLIF and ISS in short- and mid-term pain or disability outcomes, including VAS leg pain, VAS back pain at 3 months and 1 year, change in VAS (ΔVAS), and ODI at 3 months. The abstract was truncated before reporting additional outcome-specific effect sizes, confidence intervals, p-values, or NNT/NNH calculations; these values are therefore not reported in the available source and cannot be provided here. Perioperative parameters, reoperation and revision rates, and complication data are referenced as outcome domains but specific pooled estimates, confidence intervals, and p-values for these endpoints are not available from the abstract alone. Results of the three pre-specified analyses (publication-period subgroup, large-study sensitivity analysis, and RCT-exclusion sensitivity analysis) are not reported in the available abstract. No specific adverse events or complications are quantified in the extractable text. All effect sizes, NNT, and NNH values: not reported.
Practice implications
This meta-analysis establishes that PLIF and ISS produce statistically comparable short- and mid-term patient-reported pain and functional outcomes in adults treated for LSS. It does not establish long-term equivalence, superiority of either approach in perioperative safety or reoperation risk, or generalizability to specific subpopulations (e.g., patients with significant spondylolisthesis, multilevel disease, or severe osteoporosis). The extreme asymmetry in group sizes—203,414 PLIF versus 10,741 ISS—raises concern that the pooled estimates are heavily weighted by PLIF-dominant studies, and possibly by a small number of very large database studies, which the authors addressed with a sensitivity analysis whose results are not available in the abstract. The mix of RCTs and observational designs (with observational studies likely predominating) limits causal inference; significant confounding by patient selection (ISS patients may be lower-risk or have less severe disease) cannot be excluded. The finding of 'no significant difference' in pain and disability at short-to-mid term is consistent with prior literature but does not resolve whether ISS carries a higher long-term reoperation burden—a clinically critical trade-off. IMPORTANT CAVEAT: Only the abstract was available for review; all interpretations are constrained to that fragment, and clinicians should consult the full publication before applying findings to practice. __LIMITATIONS__ Abstract truncation: key results for reoperation, revision, perioperative parameters, complications, and sensitivity analyses are not reported in the available text; full manuscript required for complete evaluation.,Extreme group-size asymmetry (PLIF n=203,414 vs. ISS n=10,741, approximately 19:1 ratio) raises the possibility that pooled estimates are dominated by a few very large administrative or database studies in the PLIF arm, potentially obscuring heterogeneity.,Mixed study designs (RCTs and observational studies combined); observational studies are inherently vulnerable to confounding by indication—ISS may systematically be chosen for lower-risk or less-severe patients.,Variability in ISS device types across included studies (different device generations with distinct biomechanical properties); pooling across device types may obscure clinically relevant device-specific differences.,Short- to mid-term follow-up only; long-term outcomes (≥2–5 years), which are critical for assessing adjacent segment disease, reoperation burden, and implant durability in ISS, are not reported.,Funding source not reported; potential industry sponsorship for individual included studies cannot be assessed from the abstract.,Generalizability is uncertain: patient-level eligibility criteria (degree of stenosis, presence/grade of spondylolisthesis, number of operative levels, osteoporosis status) are not reported, limiting applicability to specific clinical scenarios.,Publication bias not assessable from the abstract alone.,Heterogeneity statistics (I²) are not reported in the available abstract, precluding assessment of the reliability of pooled estimates. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ This large meta-analysis (19 studies, n=214,155) found no significant difference between PLIF and ISS in short- to mid-term pain and functional outcomes for lumbar spinal stenosis; however, the full magnitude, direction, and clinical significance of differences in reoperation rates, complications, and perioperative parameters cannot be determined from the available abstract alone. The extreme asymmetry in group sizes, mixed study designs, and lack of long-term data substantially limit confidence in the equivalence claim. Clinicians should not alter patient selection practices based on this abstract until the complete manuscript—including sensitivity analyses, complication data, and GRADE evidence assessment—has been reviewed.