|Home | About | Journals | Submit | Contact Us | Français|
It has become a somewhat popular academic exercise to question a newly accepted standard of care by comparing it to a less complicated and more historic predecessor. This attempt by Al Humadi of performing a meta-analysis of a technical aspect of laminoplasty, namely, if screw and plate fixation improves outcomes, follows in this vein. The results presented were met with significant questions by our reviewers. In the end our reviewers accepted publication under the condition that several shortcomings of this study were clearly outlined and asked that the Evidence-Based Spine-Care Journal section of Global Spine Journal write an editorial commentary.
The basic idea of the authors seems like a meritorious one—does hardware, usually in the form of small craniofacial-type plate and screws used to span the iatrogenic gap created in a laminoplasty, actually provide value in comparison to the more historic version, which relied on a bone graft alone and usually used some form of a suture anchor to hold the laminoplasty open and allow for the expanded laminar arch to heal in a desired position? The authors’ approach was to mine the copious previous literature on laminoplasties. Following an exhaustive review of 243 studies, the authors concluded that they could not find a meaningful difference of one versus the other technique based on 21 qualitative studies and 12 studies resulting in an analysis of 634 patients (190 patients with plating and 444 nonplated).
Due to the inconsistency of reporting in their source studies, the authors decided to exclusively rely on the Hirabayashi recovery rate, as a least common denominator. This Hirabayashi number represents a percentage figure of the preoperative and the postoperative JOA (Japanese Orthopaedic Association) score and formed the basis of the only quantitative entity for the authors’ quantitative meta-analysis. Common other reported factors, such as neck pain, range of motion, complication rate, reoperation rate, return to functional activities, all were ignored by the authors of this presented study due to inconsistency of presentation in the included source studies. The length of follow-up was also not disclosed and therefore could be open to skewed reporting. These all are not trivial issues for a surgical comparison study, and the authors ask the readers to make a lot of assumptions toward equivalence of reporting. Historically reported problems of nonplated laminoplasty techniques were repeat closure of the laminoplasty site, increased pain, loss of lordosis or kyphosis, new-onset neurologic decline like C5 palsies, cervical instability, and restricted range of motion, especially as patients frequently required postoperative immobilization in a rigid collar to allow for more predictable healing of a graft. Another common problem pertaining to attempts at reporting on a collection of laminoplasty studies that do not use plate fixation lies in the fact that there are almost countless variations of suture fixation techniques. Almost all of these publications come from Asia, with a very narrow patient demographic. In contrast, plating techniques have been introduced in North America and Western Europe, with an inherently likely much larger population diversity. Therefore, another limitation of this meta-analysis could be found in differences of population demographics in form of age, gender, ethnicity, and occupational demands. This potential confounding factor is also not addressed by the authors. As stated earlier, the only variable really addressed by the authors is concentrated on the ultimate goal of a laminoplasty, to improve the myelopathy symptoms of patients using the JOA. In contrast, plated laminoplasties could theoretically incur higher infection rates and hardware-related infection or neural tissue injury rates. To date there have been no direct comparison studies of plating and nonplating, resulting in this attempt of the authors to study this subject with a meta-analysis. The authors also attempted to compare cost of surgery, but they mainly relied on an imputed cost comparison of commonly used devices in their home continent, but did not take into consideration indirect cost factors such as surgical time, surgically related complication rates, for instance, in form of reclosure of a laminoplasty for patients where the laminoplasty is not rigidly fixed with a plate and screw assembly. Similarly, patient outcomes, such as length of stay, resolution of pain, and return to regular unrestricted function, could heavily influence a cost analysis. A formal quality of life year–type assessment was clearly not performed and would be needed to try to make a more substantive value determination.
With all this serious doubt about the validity of this meta-analysis, what then are we left with in this study?
The authors found a consistent improvement of neurologic function in all of the laminoplasty studies that met their selection criteria. While not curative of myelopathy, the value of laminoplasty as an important treatment modality is impressively underscored in this meta-analysis. The study also underscores how inconsistent much of the laminoplasty literature still is not only in techniques used but also in their result reporting. This should be really a thing of the past, as there is enough substance in our studies to find a widely agreeable minimal data set of surgical outcomes along with objectifiable functional recovery and clinical results reporting over a specified time frame.1 After review of this study and its referenced literature, the question posed by the authors, whether plates are really necessary or their expense warranted for laminoplasty, is less relevant than an attempt at answering questions such as determining technique-related complication rates, including the question if French or open door laminoplasty techniques are more effective, and to establish rates of reoperation in longer term follow-up for surgeries that use variations, such as skip laminectomies. Finally, population-based outcomes studies, such as affecting physical laborers and contact sport athletes, would be really meritorious. Cervical compressive myelopathy remains a seriously underdiagnosed condition and raising awareness is a meritorious undertaking.2
The authors’ suggestion that a prospective randomized trial would be needed is very unlikely to ever be done for this issue, since most surgeons in North America and Western Europe did not learn how to do laminoplasties until after the introduction of mini-plate fixation. This was the technical innovation that allowed surgeons to begin to be comfortable with the methods, and certain implants, like ceramic spacers instead of allografts or autografts, were never available in these markets. Ultimately, the authors hopefully will not succeed in creating controversy raised by the potential bias of their study, but will hopefully stimulate further meaningful scientific dialogue about the comparative utility of the 3 operative choices for cervical myelopathy.