Abstract
We thank the authors for their thoughtful commentary on our systematic review and meta-analysis comparing laminoplasty (LP) and laminectomy with fusion (LF) for degenerative cervical myelopathy (DCM), and we appreciate their recognition of the importance of cost and patient-centered outcomes. The authors appropriately highlight the potential influence of indication bias on the observed differences in postoperative pain outcomes. We agree that patient selection for LP and LF often differs in clinical practice, with LF more commonly performed in the setting of cervical kyphosis, instability, or more severe axial pain. 1 The primary literature examined in our systematic review did not consistently report baseline deformity parameters or preoperative pain severity in a manner that would permit stratified analyses or meta-regression. Consequently, while our findings suggest greater pain reduction following LF, these results should be interpreted in the context of inherent confounding by indication within the available evidence base. Future studies that systematically account for baseline alignment and pain severity will be essential to disentangling surgical indication from true treatment effect.
With respect to cost outcomes, we agree that the substantial heterogeneity observed reflects variability in cost definitions and reporting across studies. To enhance comparability, we restricted our pooled analysis to studies reporting total index hospitalization costs in U.S. dollars and excluded studies limited to implant-only costs or non-currency metrics. Nevertheless, even among studies reporting “total hospital costs,” there remains considerable variation in included cost components, institutional pricing structures, and payer environments. These findings underscore the need for standardized health-economic frameworks in spine surgery research, as the authors note, and highlight a broader limitation of the current literature. 2 The authors’ concern regarding pooling pain outcomes across variable follow-up intervals is well taken. The limited number of eligible studies and heterogeneity in follow-up timing precluded more granular temporal analyses. While we reported this variability and interpreted pain outcomes cautiously, the inability to examine early vs late recovery trajectories reflects a broader lack of standardized outcome assessment in comparative studies of LP and LF. Prospective investigations with predefined follow-up intervals would substantially strengthen future syntheses.
The omission of neurologic and functional outcomes such as mJOA, NDI, and EuroQol-5D was a deliberate methodological choice aligned with our study objectives. These outcomes have been extensively evaluated in prior systematic reviews and meta-analyses, which generally demonstrate comparable neurologic and functional recovery between LP and LF.3-8 Our intent was to complement this body of work by focusing on patient-centered and value-based outcomes, return-to-work, opioid use, pain, and cost, that are less consistently reported yet highly relevant to patients, employers, and health systems. Nonetheless, we agree that integration of functional metrics alongside economic and occupational outcomes would provide a more comprehensive framework for future studies.
We concur with the authors’ call for prospective, multicenter research with standardized outcome definitions and longitudinal follow-up. The predominance of retrospective designs, variability in outcome reporting, and limited number of high-quality studies in the existing literature constrain definitive conclusions and highlight the need for more rigorous comparative effectiveness research in this field.
