In Reply: Guidelines for the Use of Electrophysiological Monitoring for Surgery of the Human Spinal Column and Spinal Cord

To the Editor: The authors of the Guidelines for the Use of Electrophysiological Monitoring for Surgery of the Human Spinal Column and Spinal Cord1 appreciate the complimentary and academic tone of this letter2 and are pleased to respond. We are grateful for the apparent appreciation demonstrated of the large amount of work required for the development of the guidelines, as well as the understanding of the extensive peer review that these guidelines were subjected to prior to eventual publication. The key issues raised by this group of authors regarding the guidelines involve the following: Questions raised by a more recently published systematic review of the literature on intraoperative monitoring, particularly with respect to its use in the context of operation upon intramedullary spinal cord tumors, and the use of a specific “methodological index for nonrandomized studies (MINORS)” as a quality measure3 Different conclusions by different authors on the same topics, providing the example of the use of methylprednisolone in spinal cord injury4,5 Incredulity that the guidelines authors could be suggesting that intramedullary spinal cord tumors can be safely and effectively removed without the use of intraoperative monitoring (IOM) The letter's authors’2 belief that “whenever the best methodology producing guidelines conflicts with the best clinical practice [it] is the methodology which should be changed not the clinical practice” citing the Guidelines for the Management of Severe Traumatic Brain Injury, Fourth Edition6 as an example And, lastly, an exhortation to the guidelines’ authors that “the authors of this important contribution have to realize that they cannot defend an evidence based approach which is clearly in contradiction with other subsequently published review (also evidence based) and with what we call the actual ‘best clinical practice’.” We are happy to undertake responses to all of these points made by the letter's authors,2 recognizing their right to their points of view, but with the knowledge that these differing viewpoints will not necessarily be resolved by our responses. First of all, the systematic review and meta-analysis referred to by the letter's authors7 has already (and very quickly) generated its own Letter to the Editor,8 in which those authors point out the difficulty in making claims or conclusions based upon data that are highly heterogeneous in nature. The letter's authors conclude that they “question the overall validity and conclusions of this meta-analysis in the presence of both clinical and statistical heterogeneity. By pooling dissimilar studies, the authors of this meta-analysis may have been misinformed in their conclusion.” We invite our letter's authors to review this appraisal by these surgical neurophysiologists. In addition, the citing of the use of the MINORS3 criteria for evaluating study quality is an interesting one, especially in light of the failure of this paradigm to detect the noted heterogeneity of the studies under review. Although this is one of several tools recommended by various groups,9 it included the Sala10 reference as a “high-quality comparative study” without recognizing, as our group did, that the study carried “a significant methodological flaw that led to its reclassification as a class III medical evidence study by including nonconsecutive patients over a very lengthy time period in their control group (1988-2000).”1 This application highlights not only the lack of completeness of the criteria, but also the need for topic-specific evaluation to further establish the quality of studies used to determine practice recommendations. The fact that the letter's authors have assumed that this impugns the guidelines process and is an example of “how the application of the [sic] evidence-based medicine ‘alone’ without a sense-based approach can lead to contradictory conclusions starting from the same literature”2 is unfortunate and unwarranted. Regardless of the specific criteria used, paramount to the creation of evidence-based guidelines is a transparent methodology. As with any peer-reviewed publication, the reader is charged with the critical evaluation of that methodology. The methodology used in the creation of the Guidelines for the Use of Electrophysiological Monitoring for Surgery of the Human Spinal Column and Spinal Cord1 has been validated and used extensively in the creation of guidelines for our specialty, particularly with respect to spinal diseases and treatments, and has been discussed in detail elsewhere.11 The strength of the medical evidence depends not only on the purported study design (eg, randomized controlled trial, cohort study, case-control study, etc.), but also on any methodological flaws that might weaken the methodology, results, and conclusions derived from the data in question. To facilitate transparency, we outlined these issues in detail with each study evaluated for inclusion (or not) in the guidelines. By making the process undertaken by guidelines developers as transparent as possible, whichever method of evaluation utilized, the reader is able to determine whether he/she accepts or rejects the recommendations of the guideline document. As with any scientific manuscript, we strove to demonstrate reproducibility of our results with our methodology, stating that, “The choice of any particular paradigm for allocating a designation of ‘strength’ for evidence used in guideline development is somewhat arbitrary, and each professional society must make its own decisions in this matter and make sure that the levels of evidence are clearly defined and transparent.”12 We do not wish to flog a dead horse by undertaking yet another extensive discussion of the differences in interpretation of the literature regarding the use of methylprednisolone in spinal cord injury. Suffice it to say that the “pro” camp is led openly by principal investigators of the methylprednisolone studies, and the “con” group is inhabited by those who did not participate in the studies, who have no vested interest in their claims, and who simply evaluated the data presented in (or withheld from) the publications. It is interesting, however, that the letter's authors cited the “pro” point of view1,11 without also including the equally (if not more) erudite “con” viewpoint.13 With respect to the interpretation that the guidelines’ authors are espousing a belief that intramedullary spinal cord tumors should not be operated upon using IOM, we are astounded. As we have said repeatedly, “The potential of electrophysiological monitoring (IOM) during spinal cord/spinal column surgery is to measure and monitor the integrity of spinal cord function and structure during the surgical procedure, and, when the integrity of spinal cord function and structure is threatened, to modify the operative procedure to reduce/eliminate that threat and thereby avoid subsequent neurological injury and paralysis. While the former has been shown, the latter has not. Further study of the use of IOM as a therapeutic and/or preventative tool during spinal cord/spinal column surgery is needed,”1 “At no point do we suggest ‘the abandonment’ of IOM; we simply recognize that there isn’t the evidence … to argue that a recommendation for its use as a requirement in spinal surgery should have been made,”12 “It is entirely possible that IOM does provide a therapeutic benefit in select patient populations and select procedures. What we reported is the lack of evidence in the published literature that we found of a therapeutic benefit to the use of IOM, while acknowledging evidence for a diagnostic role in spinal surgery,”14 “IOM is well able to show that an injury has/might have occurred (ie, can make the diagnosis), but it has not been demonstrated that the indications of change in neurological status then lead to surgical (or other) responses that provide for an improved outcome (ie, therapeutic benefit) for spinal surgery patients,”15 and, lastly, “therapeutic benefit remains unproven but has not been disproven.”16 These are statements that clearly outline the current literature and its failings, not beliefs on the guidelines authors’ part of the lack of efficacy of IOM. Furthermore, anecdotal and peer-reviewed evidence—which the letter's authors would perhaps consider “sense-based”—suggests that intramedullary spinal cord tumors (or, for that matter, intradural-extramedullary and extradural tumors) can be safely removed without the use of IOM.17 Resection of spinal cord tumors without IOM cannot, therefore, be considered substandard care or as exceptional as the authors of the letter suggest. (Indeed, statements such as these were a major impetus for the creation of these guidelines.) We are confused by the letter's authors’2 claims that somehow there is a discrepancy between “best clinical practice” and guidelines recommendations. Their apparent belief that we have inferred from their remarks—ie, “we all know it works, we just don’t have proof”—is fairly antiquated in its attitude. We think this is also included in their “sense-based approach,” although this is not completely clear. If so, it is the very opposite of what is meant to happen as a result of careful evaluation of the published literature. The letter's authors go on to state in their self-cited letter regarding the most recent severe brain injury guidelines, “The final target of guidelines is, in our opinion, to help neurosurgeons and intensivists to cure their patients and not to keep at any cost an aseptic ‘formally correct’ methodology,”18 as if these are mutually exclusive goals rather than that of supporting “clean” science that guides our efforts on behalf of patients and their desired (and deserved) outcomes. Lastly, the authors of the letter2 conclude with the argument that because Guidelines for the Use of Electrophysiological Monitoring for Surgery of the Human Spinal Column and Spinal Cord1 utilizes an older (one might say well-established) methodology, its results must be inferior to both those subsequently published using “newer” methodologies and those aligning with the self-defined “best clinical practice.” Newer is not necessarily better, particularly when comparing older evidence-based methodology to a newer “sense based approach.” A never-ending discussion is not a negative outcome. Indeed, we would argue that it is the hallmark of the scientific process. We can and should continue to work together to demonstrate which patients definitively benefit from the use of IOM. Until we do, there is insufficient evidence to require, or perhaps even expect, its use as standard of care in any particular operative situation. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.