Evidence-Based Recommendations for Spine Surgery

Alexander R. Vaccaro, MD, PhD, MBA; Charles G. Fisher, MD; Srinivas K. Prasad, MD; Alpesh A. Patel, MD, FACS; John Chi, MD, MPH; Kishore Mulpuri, MBBS, MHSc; Kenneth C. Thomas, MD, MHSc; Peter G. Whang, MD, FACS


Spine. 2020;45(21):E1441-E1448. 

In This Article

Methodological Review

Lee and Ahn essentially conducted an efficacy inquiry which involved a series of procedures performed by a single surgeon on a narrowly defined patient sample and this manuscript represents a continuation of their previously published study.[19] There is certainly sufficient clinical experience with the use of CS for lumbar fusion applications to warrant such a comparative investigation to corroborate some of the purported advantages of this technique.[20] The methodologic design is reasonably sound in that prognostic stratification was employed as part of the randomization process for age, smoking status, and spinal level (i.e., L4/5 or L5/S1); subjects were also blinded to their treatment allocation and were only considered if they had completed all follow-up evaluations. The requisite sample sizes were calculated in advance and it was determined that a minimum of 36 patients in each group with complete 2-year data would be required for this analysis. In addition, the battery of statistical tests that were selected were also appropriate for the purpose of this comparison.

Regardless, there are still some limitations to this study which merit further discussion. While the inclusion and exclusion criteria are clearly described and are largely reproducible, the authors do not relate other critical details about the enrollment process such as how many patients were deemed to be eligible but did not elect to participate in the trial. As part of the postoperative protocol, patients were hospitalized for 14 days following their procedures which is not typical for other regions where the length of stay may be as short as 1 day; furthermore, the incidences of tobacco use in both groups were substantially higher than those observed in other investigations which obviously would be expected to influence the arthrodesis rate. Collectively, these issues may detract from the applicability of these results to the general population as a whole. Perhaps one of the most important points of emphasis is that the relatively modest sample size makes it difficult to draw any definitive conclusions about the utility of CS for this particular indication (i.e., PLIF). Given that there were no significant differences in the primary outcome of fusion success between the CS and PS cohorts, it is certainly conceivable that a beta error may have occurred during the analysis.