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

Given the increasing popularity of the technique of interlaminar stabilization, Schmidt et al clearly state the purpose of their study (i.e., comparing the outcomes of lumbar decompression with and without the use of the Coflex device) but they do not provide their own hypothesis. This randomized controlled trial was essentially an "effectiveness" study involving seven sites with an unknown number of surgeons. The inclusion and exclusion criteria were generally broad enough to include individuals with 1 or 2 levels of moderate to severe stenosis between L3 and L5 with or without a Grade 1 degenerative spondylolisthesis (i.e., ≤ 3 mm); the prevalences of transitional instability documented for both cohorts were reportedly comparable. Interestingly, one of the more unusual inclusion requirements was a VAS score of ≥ 50 mm out of 100 mm for back pain, not leg pain. Another issue is that there was no mention or description of the population from which the study sample was drawn. A total of only 230 patients were randomized from seven sites over the course of 6 years which averages out to approximately five subjects per site per year which is a surprisingly small number given the relative ubiquity of lumbar stenosis. Consequently, without any information about how many individuals were eligible for the trial but were not enrolled, it is difficult to rule out the possibility of selection bias. The baseline comparability between the treatment and control groups appeared to be similar with respect to sex, age, body mass index, preoperative walking distance, and narcotic usage but it was not reported for other factors such as comorbid state or smoking status.

The operative technique is explained in enough detail to ensure that it may be considered to be replicable. With this design, there was no blinding of patients, surgeons, or other assessors of outcomes to the assigned treatment but the radiographs were evaluated by an independent core laboratory. Subject data was collected out to 2 years after surgery and the follow-up rate of 91% is excellent. As part of their methodology, Schmidt et al considered both statistical and clinical significance although they seem to have employed too many hypothesis tests. Multiple widely-accepted patient-reported clinical outcome measures were utilized in this investigation (e.g., VAS for back and leg pain, ODI, Zurich Claudication Questionnaire) which were not statistically significant between the cohorts; however, the authors elected to focus on a composite scale which is not as amenable to clinical interpretation. Consequently, while the rate of CCS was significantly higher for the D+ILS group than that observed for the DA patients, the relevance of this finding remains somewhat unclear.

The analysis appropriately considered the incidences of reoperations and postoperative injections but one confounding factor is that these interventions were presumably performed at the discretion of the unblinded surgeons. Finally, the authors also examined the occurrence of various adverse events such as dural leak which interestingly was observed more frequently with decompression alone than with interlaminar stabilization (13% vs. 4.5%); likewise, the mean operative time was actually longer for the DA procedures than for D+ILS. Nevertheless, there is no discussion of any rationale that might account for these unexpected results.