The Minimally Invasive Lumbar Decompression Procedure: Whimsy or Genuine Invention?

The article by Levy and Deer in this issue of Pain Medicine concerning the safety evaluation of a new technique treating symptomatic lumbar spinal stenosis has many merits, not the least bringing to the reader's attention a new minimally invasive intervention for patients in pain for whom treatment modalities are limited. It also raises several issues symptomatic of the state of clinical research in this country that may negatively affect the development of our specialty, pain medicine. The new technique, known as minimally invasive lumbar decompression (MILD), which reportedly works through a percutaneous remodeling of the ligamentum flavum and the lamina, is done under fluoroscopy. After 510k Food and Drug Administration (FDA) new device approval, several clinical trials have been undertaken in the last 4 years for testing the safety and efficacy of MILD, leading to more than 10 publications, most reporting positive outcomes, with minimal or no complications. Investigators were recruited by the manufacturer, Vertos Medical, and some also became consultants to the company. All followed clinical protocols endorsed by the manufacturer, until this year, when one of them, a professor of neurosurgery at the University of Saskatchewan and a colleague, published his own MILD series in the journal Neurosurgery. This report stated that 60% of the patients enrolled in the trial required a laminectomy within 18 months and, therefore, concluded “that the failure rate (of MILD) is unacceptably high.” I am not aware of the detailed interaction between this Canadian investigator and the manufacturer, but on September 5, 2012, a front-page article appeared in the business section of the New York Times (NYT) titled “A Clash Over a Spine Treatment” . To summarize the content of the article: Dr. Fourney (one of the authors and a Canadian investigator), a respected and well-published professor of neurosurgery and a spine specialist, has been accused by Vertos Medical of “scientific misconduct and violating research ethics” by following up his enrolled patients after the 26-week period of the protocol without agreement from the manufacturer. The publication of such an article in the front page of the business section of one of the most widely read (and respected) newspapers in the world merits our attention. One could argue that the NYT reporter did not read Dr. Fourney's report very thoroughly; otherwise, he would have asked some questions of its authors, such as: why such a small series of patients (10) observed over such a short period of time (26 weeks) without any control group was ever submitted for publication in a major medical journal, Neurosurgery? Why 60% of enrolled patients required a laminectomy in spite of the fact that their outcome data (visual analog scale [VAS] and Oswestry Disability Index [ODI]) at 26 weeks were significantly improved? Etc. But the reporter, who is not a scientist, was interested in something else: why would a medical device manufacturer attempt to control the publication of negative results obtained with its device by intimidating its own investigator? Several comments are in order. The period of a truly independent. A truly independent clinical investigation of new drugs or devices has, for all practical purposes, now ended. Most clinical trials in the United States are now managed, especially for new drugs, by contract research organizations on behalf of the industry, which finances the expenses of the trial, usually involving multiple centers and several countries, in a globalization process, where individual investigators have increasingly fewer patients to test , very little input on the research protocol, and extremely limited access to the resulting data, usually protected for the sponsor by binding legal agreements between investigators and companies. The industry's power (as “owner of the trials”) has created a culture of arrogance concerning the industry's right to control data and ensure only positive results are published. It has also made obligatory the disclosure of conflict of interest when study results are presented by paid consultants or paid investigators. These major ethical problems have been the subject of numerous editorials and debates in the recent medical literature . Again, I do not know the details of the interaction between Vertos Medical and Dr. Fourney, but I can easily guess the damage from the NYT article to the introduction of the MILD procedure in clinical practice. It has raised a cloud of ethical suspicion regarding the device manufacturer and will make any future assessment of this technique much more difficult. As an example, the current article is written by two investigators (RL and TD), who are highly respected in the field of pain medicine, and are also paid consultants for Vertos. I still believe, for reasons discussed earlier, that their contribution is scientifically much more valuable to the clinician than Dr. Fourney's article. Furthermore, as long as the responsibility of clinical trials is left to the industry and governmental National Institutes of Health (NIH)-sponsored research confines itself to helping basic sciences, progress in clinical research will be gravely handicapped. NIH should closely examine the recommendations of another government agency, the Institute of Medicine, which, in its report on pain , has clearly indicated the need for government-sponsored clinical investigations in order to improve pain care in the United States. It would be useful to list the errors that have been compounded in the evaluation of the MILD technique. 1) The head-to-head confrontation of the experimental technique with spine surgery. Not only are the two approaches (minimally invasive technique versus surgery) the proverbial apples and oranges, but the comparison challenges a well-established and organized field of medicine, i.e., spine surgery. This is not necessary. The interventions available to pain medicine, beside surgery, in order to treat symptomatic patients with lumbar spinal stenosis, are very limited. I would not suggest, especially these days, epidural steroid injections, which have repeatedly been shown to be a poor choice for pain relief in lumbar spinal stenosis. Physical therapy has limited efficacy, and opioids are not universally accepted. Even in Fourney's article, I was quite impressed by the fact that 4 patients (out of 10) were not operated upon because they were probably still much improved after the MILD procedure. 2) The absence of standard evaluation protocol, validated by peers and professional organizations and done in collaboration with the FDA, essential to justify any new technique or device being assessed for pain management. 3) Submitting clinical investigators to the control of industry sponsors, creating the opportunity for censorship, pressure, or threat when publishing their results. The reported Vertos reaction to an investigator's opinion is totally inadmissible for scientific, ethical, and even commercial reasons. Because of these errors, MILD's fate may follow that of other pain devices (such as intradiscal electrothermal annulopasty), which were initially introduced to improve the comfort of pain patients using a minimal risk intervention, but which have joined the growing ranks of obsolete pain procedure casualties. It would be a great pity for the MILD procedure to end its evaluation prematurely. A new, minimally invasive treatment modality with minimal risk is badly needed for the treatment of lumbar spinal stenosis, an increasingly prevalent condition due to our aging population and one for which we have very little to offer, except surgery with its limitations. Published results obtained for this new technique, including the present article, are very encouraging, even if these improvements do not last a lifetime. Therefore, let us continue the good work and help pain medicine to raise its standards.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.