Minimally invasive lumbar decompression (the MILD procedure) (Vertos, CA, USA) was introduced over 4 years ago as a new method to treat symptomatic lumbar stenosis (LSS). Guided by contrast epidurography and specially designed instrumentation, percutaneous removal of lamina and ligament in the stenotic area has now been reported in multiple publications to provide symptomatic benefit in patients with LSS. The use of this technique began when the device manufacturer, Vertos Medical, received the Food and Drug Administration (FDA) 510k approval in 2006 as “substantially equivalent” to other instrumentation used in spinal surgery. As a neurosurgeon who performs both traditional and minimally invasive spinal surgery, as well as the MILD procedure, I do not consider the MILD procedure substantially equivalent to any current extant technique of spinal surgery or pain management. The combination of instrumentation and surgical technique (oblique fluoroscopy, epidurography) is, in my opinion, a novel synthesis of technique and instrumentation. Thus, some may consider the 510k approval of MILD a true “curse in disguise” as it allowed the technique to enter clinical practice without prospective, controlled, FDA-supervised studies. This has permitted (dare I say) inappropriate utilization of this technique by many practitioners without appropriate clinical (claudication pain vs radiculopathy) and radiographic (central vs foraminal stenosis) selection criteria—a diagnosis of symptomatic lumbar stenosis of any type, combined with failure of previous interventions, thus being the primary indication for the procedure.

In the absence of Class I evidence, such as that generated by randomized controlled trials, how does the pain management community proceed in studying a new therapy in a prudent and analytical fashion, knowing fully well that the current health care environment is one that revels in “exposing” new procedures as being more expensive but not dramatically better than the current standard of care?

The two articles in this issue of Pain Medicine provide well-needed additions to our expanding knowledge base. While both are, at their core, small case series, each article provides unique additional contributions and should serve as models for further such studies.

Wang et al. analyze the utilization of health care resources in patients undergoing the MILD procedure. In their retrospective analysis, 55% of patients were discharged from chronic pain clinic following MILD. The number of chronic pain procedures decreased from 3.68 per patient pre-procedure to 0.95 following the procedure. Additionally, there was a mean reduction of 2.27 in the visual analog scale (VAS) score per patient, with 68.2% of patients reporting satisfaction with the procedure. The reduction in health care utilization is, of course, a direct correlate of the successful outcomes in over half the patient cohort, with those patients needing no further intervention. In the 10 MILD failures, a total of 23 follow-up procedures were performed, with 3.29 total interventions performed per patient, similar to the pre-MILD mean of 3.68. A major strength of this study is its analysis of a Veterans Administration (VA) patient population, which, as the authors note, obtains all their care through the VA system and is less likely to turn to other centers for chronic pain care, allowing for long-term follow-up. Moreover, as VA physicians are not paid on a fee-for-service basis, this further strengthens the argument that these procedures were not done to merely generate more revenue for the practitioner.

Durkin and colleagues report on a prospectively gathered, multidisciplinary, quality assurance patient registry assessing multiple clinical variables and utilizing a variety of outcome measures. Outcomes were similar to Wang et al., with a significant reduction in VAS in 52% of patients. While the presence of radiographically severe stenosis appeared to be a risk factor for no improvement post-MILD using the National Institutes of Health's Patient-Reported Outcomes Measurement Information System pain interference score, no other preoperative variable appeared to predict successful outcome. Although they acknowledge that further studies are needed to perfect patient selection criteria, their methodology and infrastructure serve as a suitable template for further studies assessing this procedure.

Going forward, the burden will remain on the medical community (both the physician community and the device industry) to demonstrate that new technology is at least as safe, as (or more) effective, and ultimately more cost-effective than the current management options. As pointed out in a previous Pain Medicine editorial [1], there is no true single procedure that can be compared in a head-to-head fashion with the MILD. One study comparing MILD vs epidural steroid injections showed benefit in both groups, but statistically better improvement with MILD [2]. In current clinical practice, however, MILD candidates are those who, almost by definition, have failed previous epidural steroid injections, and thus the real-world relevance of this study may be limited. Demonstrating the superiority of MILD over conventional medical management, such as what was done with spinal cord stimulation in the PROCESS study [3], may be an appropriate avenue of approach. Finally, one could conceive of a randomized trial of MILD vs “sham” MILD, with the sham procedure involving epidurography, contrast injection, and port placement, but no bone or tissue removal. Until that time when such studies are performed, contributions such as those of Wang et al. and Durkin et al. should be encouraged by the pain community.


  1. Top of page
  2. References