May 18, 2026 | At the 2026 International Neuromodulation Society meeting in Lisbon, Portugal earlier this month, a familiar question in pain neuromodulation was revisited directly: do trials and diagnostic blocks improve outcomes, or have they become procedural and reimbursement artifacts embedded in the business model of care?
The question is not new in spinal cord stimulation, where trialing has been standard for decades. But it is increasingly important in peripheral nerve stimulation, where rapid growth, high device margins, variable coverage, and heterogeneous indications are forcing the field to ask whether pre-implant steps are clinically justified, payer-mandated, or commercially convenient.
In SCS, some clinicians and researchers are questioning whether routine trialing should be eliminated because it has limited clinical utility as a prognostic screen. In PNS, the field appears less ready for a universal conclusion because nerve target, device type, indication, mechanism, and patient burden can vary widely.
SCS Trialing: A Standard Pathway Under Pressure
SCS trialing historically gives patients temporary exposure to stimulation before permanent implantation, allows clinicians to assess short-term pain relief, and gives payers a threshold—often 50 percent pain reduction or meaningful functional improvement—and fits comfortably into interventional pain workflow.
However, the INS discussion highlighted a growing challenge to that logic. Ryan D’Souza from Mayo Clinic, in arguing against routine PNS trialing, used SCS trialing as the precedent case. He cited trial-versus-no-trial SCS data suggesting no meaningful difference in pain scores or responder rates between trialed and non-trialed patients. More importantly, he emphasized that SCS trials may have poor specificity. If nearly everyone “passes,” the trial may not effectively identify which patients will fail permanent therapy.
That matters. If the purpose of a trial is to predict durable benefit, but the trial does not improve prediction, its clinical rationale weakens. It may still have value in patient education and expectation-setting, but that is different from saying it improves outcomes.
D’Souza also referenced cost-effectiveness concerns and patient preference data favoring a single-stage implant. The patient logic is straightforward: one procedure rather than two, fewer visits, fewer externalized wires, one recovery period, and less anxiety. For payers and health systems, the question is whether the trial reduces downstream cost enough to justify its own cost and risk.
That has led to discussion within INS circles about whether routine SCS trialing should be reconsidered, particularly given studies such as TRIAL-STIM in which SCS screening trials had 100% sensitivity but only 8% specificity. The TRIAL-STIM publication states that the RCT found no evidence that a screening-trial strategy provided superior patient outcomes versus a no-trial approach. Such a change in trialing policy would be substantial because trialing is built into payer policy, physician workflow, coding, scheduling, and manufacturer sales processes. Eliminating it would require payer acceptance, professional society guidance, and confidence that direct-to-implant pathways do not worsen outcomes.
PNS Trialing: A More Complicated Question
PNS is different. Unlike SCS, where leads are placed routinely within the epidural space, PNS is anatomically and clinically heterogeneous. A PNS system may target many named nerves below the neck, and in some cases above the neck depending on clearance. Targets may include suprascapular, tibial, peroneal, saphenous, sciatic, cluneal, genicular, or other peripheral nerves. The clinical goal, mechanism, and device burden can also differ substantially.
For that reason, INS speakers cautioned against importing the SCS model wholesale. There may be PNS indications where trialing adds little value. If the nerve injury is obvious, pain distribution is anatomically clear, and the patient is well selected, a direct-to-implant pathway may be reasonable. Conversely, trialing may be valuable with unclear diagnosis, multiple possible nerve targets, prior neuromodulation failure, patient hesitancy, compliance concerns, or uncertainty about whether the patient will tolerate device burden.
The pro-trial argument for PNS therefore differs from the traditional SCS argument. It is not simply that the trial predicts long-term analgesic response. It may instead test whether the patient can use the system, tolerate stimulation, manage wearables or charging, accept the sensation, and judge whether functional gains are meaningful. Alaa Abd-Elsayed from University of Wisconsin argument in favor of PNS trialing emphasized predictability, expectation-setting, side-effect identification, device handling, and compliance. In those cases, the trial becomes a structured education tool, not just a pain-score screen.
That distinction matters commercially. A PNS trial may reduce inappropriate permanent implants if it identifies patients who cannot manage the system or whose expectations are unrealistic. It may also increase conversion when patients experience meaningful functional improvement and become more comfortable with therapy. But the evidence base remains immature. The field needs indication-specific studies to determine where trialing improves outcomes and where it adds cost and delay. Until payer policies change, trialing is likely to remain common in PNS, especially where payer policy, diagnostic uncertainty, patient hesitancy, or device-management burden are relevant.
Diagnostic Blocks in PNS: Useful Tool or Payer Gate?
The diagnostic block discussion was clinically nuanced. The central message was that diagnostic blocks should not be viewed as universally predictive of PNS response. A local anesthetic block and electrical neuromodulation are different interventions: a patient may fail a block and still respond to PNS, or respond to a block and fail stimulation.
The practical rule offered at INS was that a diagnostic block should be performed only if it will change decision-making. If a negative block would cause the clinician not to implant, or a positive block would support proceeding, the block may be useful. If the clinician already knows the nerve target and would proceed regardless, the block may add complexity without improving care.
This is particularly relevant in focal PNS. In pain clearly distributed along a surgically injured sural nerve, the target may already be apparent. By contrast, in complex foot pain, where several targets may be plausible, a differential block may help narrow the target. In that setting, the block is a target-selection tool, not a generic payer hurdle.
Payers often treat the issue differently. Some private payers require diagnostic blocks before approving PNS, regardless of whether the physician believes the block is clinically meaningful. That creates a disconnect between clinical reasoning and administrative requirements. The clinician asks: will the result change what I do? The payer asks: has the required step been completed?
Diagnostic Blocks in SCS: A Different Role
Diagnostic blocks in SCS are different and generally less central. SCS does not usually depend on blocking a single named peripheral nerve to identify a stimulation target. Instead, the workup focuses on diagnosis, pain phenotype, failure of conservative management, psychological screening, and appropriateness for neuromodulation. Injections or blocks may be part of the broader treatment history, but they do not serve the same mechanistic role as a peripheral nerve block before PNS.
For SCS, the trial itself has historically functioned as the predictive step. For PNS, blocks may help select the appropriate nerve target, while the stimulation trial may test therapy tolerance, expectations, and device acceptance. A payer policy that simply asks for “a block” or “a trial” may miss the clinical purpose of the step.
Monetary Incentives and Pre-Implant Procedures
INS also raised the uncomfortable issue of financial incentives. In the U.S., trialing has economic consequences for physicians, facilities, manufacturers, and payers. Temporary SCS trials can be financially attractive to procedural practices and are deeply embedded in the SCS pathway. The same is increasingly true in PNS, although the economics differ.
PNS has high gross margins in many commercial models, and manufacturers often discount trial systems, leads, or trial-related components to facilitate adoption and preserve attractive economics on the permanent implant. That creates a predictable commercial incentive: the trial becomes both a clinical step and a sales funnel. A discounted trial can lower the barrier to physician use while increasing conversion.
This does not mean trialing is inappropriate. It means clinical utility must be separated from financial reinforcement. A trial that improves patient selection, reduces explants, improves satisfaction, and supports payer confidence is valuable. A trial that merely creates another reimbursed procedure, delays care, and fails to predict outcome is harder to defend. Diagnostic blocks raise a similar issue. They may be clinically appropriate in selected cases, especially when they identify the correct target. But if blocks are performed primarily because payers require them, or because they fit into a reimbursable procedural sequence, the clinical logic becomes secondary.
Toward a More Rational Pathway
The future pathway is unlikely to be “always trial” or “never trial.” It should be indication-specific, technology-specific, and patient-specific.
For SCS, routine trialing may eventually be reduced or eliminated for well-selected patients if further data confirm limited clinical utility. For PNS, the more likely near-term outcome is stratification. Trialing may remain appropriate for uncertain diagnoses, multiple possible nerve targets, patient hesitancy, prior neuromodulation failure, or systems with meaningful external management requirements. Direct-to-implant may be appropriate where the nerve target is clear, patient selection is strong, and evidence supports durable outcomes. Diagnostic blocks should be reserved for cases where the result will be respected and will guide the treatment plan.
The broader message from INS is that neuromodulation is maturing. Procedural traditions will increasingly be judged against evidence, cost, and patient burden. Trialing and diagnostic blocks may still have important roles, but those roles need to be defended clinically—not simply preserved because they are reimbursed, profitable, or historically familiar.
A preconference workshop at the 2026 INS Congress featured several sessions devoted to commercialization issues. At the workshop, Marc Russo, an Australian pioneer in pain neuromodulation, used his presentation on “the medical entrepreneurship pathway” to address one of the oldest questions in innovation: are entrepreneurs born or made?
Russo’s framework suggests that while individual creativity, persistence, and judgment still matter, medical entrepreneurship can be taught. His evidence was the rise of Biodesign-style programs, particularly Stanford Biodesign, which treat innovation as a repeatable process—identify, invent, implement—rather than as the product of the occasional physician-inventor or academic spinout. For neurotechnology, where elegant engineering often collides with regulatory, reimbursement, and adoption barriers, that distinction is especially important.
Russo, founder and director of Hunter Pain Specialists in Newcastle, Australia, is known in the neuromodulation community as a clinician, investigator, entrepreneur, and society leader. He has treated tens of thousands of patients since establishing Hunter Pain Clinic—now Hunter Pain Specialists—in 1999, and has been involved in a number of world-first spinal cord stimulation procedures and first-in-human device studies.
Russo’s message at INS was not simply that physicians should invent more devices. It was that physicians, engineers, universities, hospitals, investors, and industry partners need better systems for moving clinical observations into commercially viable products.
That ultra-early stage, in this editor’s view, remains one of the hardest phases of medtech development. Universities often generate brilliant ideas but struggle to translate them into commercial successes. Alfred Mann recognized a similar gap when he established institutes designed to move university-originated inventions into commercial ventures. Despite several hundred million dollars of investment, the number of commercial successes was limited. The lesson reinforces Russo’s larger point: translation requires more than invention and capital. It requires structure, operating talent, regulatory planning, reimbursement strategy, and a realistic understanding of adoption.
Russo began by challenging the mythology of the lone physician-inventor. There are, he said, rare examples of people who, through “sheer power of intellect or random luck,” develop an idea and bring it to life through force of personality. But those cases are not a reliable model for repeatable development and commercialization of medical technologies. A more durable approach is to institutionalize the process: identify unmet clinical needs, screen them rigorously, match them to feasible solutions, and move those solutions through regulatory, reimbursement, clinical, and commercial pathways.
For neurotechnology, the message was particularly relevant. Neuromodulation has no shortage of devices—particularly with the recent wave of noninvasive neuromodulation—along with elegant engineering, novel waveforms, and clinical observations. But the field also has a history of products that struggled because the clinical need was poorly framed, reimbursement was considered too late or ignored, the business model was underdeveloped, or the right operating team was assembled after key strategic decisions had already been made.
Russo emphasized that medical entrepreneurship differs from conventional technology entrepreneurship because it must satisfy a broader set of stakeholders. A medical device company must navigate FDA 510(k), de Novo, PMA, or EU MDR pathways; generate evidence; secure reimbursement coding, adequate payment levels, and payer coverage; satisfy patients, physicians, hospitals, payers, and regulators; and deliver a product that clinicians will actually use. Coverage, in particular, is often underestimated in both difficulty and duration. He described the medtech pathway as typically taking three to seven years from concept to first revenue, with longer timelines for biotech.
That estimate is reasonable for many conventional medtech products, but active implantable neurotechnology ventures can take substantially longer. Companies such as Second Sight and SetPoint Medical illustrate how visual restoration and bioelectronic medicine platforms can require 15 to 20 years before meaningful commercialization.
That complexity creates what Russo described as the “valley of death” between proof-of-concept and commercial de-risking. Grant funding is often too limited to complete translational development, while later-stage venture capital waits until clinical, regulatory, and commercial risks have been reduced.
The current federal funding environment may be widening that gap for some early-stage device teams, including those facing uncertainty around NIH Blueprint MedTech award timing after months of delay.
A recurring theme in Russo’s talk was the difference between technology-push and needs-driven innovation. The dominant academic pattern often begins with a technology developed in a lab, followed by a search for a clinical use case and sometimes a premature licensing or spinout effort. By contrast, Biodesign begins with clinical immersion and needs finding. Only after large numbers of needs are identified and screened against clinical, market, intellectual property, regulatory, reimbursement, and stakeholder filters does concept generation begin.
Russo pointed to Stanford Biodesign as the flagship model. Founded in 2001 by Paul Yock and Josh Makower, the program helped make medical technology innovation a teachable process. Russo traced its intellectual lineage back to Thomas Fogarty, whose mentorship of Yock helped inspire the question of how one-on-one inventor mentorship could be scaled into a reproducible institutional model.
Stanford Biodesign’s framework—identify, invent, implement—has become a common language across medtech innovation programs. Fellows immerse themselves in clinical settings, catalog unmet needs, screen concepts, and then develop IP, regulatory, reimbursement, clinical, quality, business, funding, team, and operational strategies. Russo highlighted one discipline in particular: spending three to four months on needs screening before beginning invention. That delay can feel counterintuitive, but it can prevent teams from spending years developing a device for the wrong market, patient population, or reimbursement environment.
According to Russo’s presentation, Stanford Biodesign has trained more than 219 Innovation Fellowship alumni, more than 3,000 Stanford students, 109 faculty, and more than 500 trainees per year. He also cited 57 health-technology companies founded by program trainees and more than 18 million patients reached by Biodesign technologies. Companies associated with the Biodesign network—some directly fellowship-originated and others linked through the broader Biodesign ecosystem—include iRhythm Technologies, NeoTract, Acclarent, Cianna Medical, and EBR Systems.
Russo also examined Case Western Reserve University’s Coulter Translational Research Partnership as a contrasting structure—less boutique fellowship and more embedded institutional engine. At CWRU, the program became a permanent translational research platform supported by a $20 million endowment. Rather than relying primarily on a small fellowship cohort, the university embeds the pathway in master’s programs, biomedical engineering tracks, and translational research funding mechanisms. Russo noted that the program funds six to eight projects per year, typically $50,000 to $200,000 per project, with the expectation that projects move toward licensing or spinout within about three years. He cited more than 22 startups, 26 technologies delivered to patients, and more than $50 million in follow-on investment leveraged from funded projects.
Russo then broadened the frame globally, noting that Stanford-affiliated or Biodesign-inspired programs now exist in Singapore, India, Ireland, Japan, Israel, the U.K., Australia, and continental Europe. (NBR senior contributing editor JoJo Platt helps produce a NeuroDesign program held each year on the CWRU campus.) Each has adapted the model to local clinical, regulatory, funding, and market conditions, but the common DNA is consistent: clinical immersion, multidisciplinary teams, needs-driven development, structured prototyping, integrated business training, and industry mentorship.
For clinicians, Russo’s takeaways were direct. Physicians are well positioned to identify unmet clinical needs, but medical training does not usually cover cap tables, term sheets, freedom-to-operate analysis, FDA submissions, ISO 13485 quality systems, reimbursement coding, manufacturing, supply chain, or sales channels. Structured fellowships and translational programs can compress years of business and regulatory learning while surrounding clinicians with disciplines they typically lack.
He also cautioned physician-founders against assuming they should remain chief executive officers as companies scale. In many successful medtech startups, the physician becomes chief medical officer, scientific founder, board member, or clinical advisor rather than long-term CEO.
For investors, Russo suggested that Biodesign-style training can serve as a useful diligence signal, but fellowship pedigree should not be confused with operating capability. Founder-market fit, execution discipline, capital efficiency, and team-building ability still matter. For engineers and founders, the message was even simpler: do not build first and search for a clinician later. Find a clinical co-founder before building anything. Spend months on the need before spending years on the device. Master at least one regulatory pathway early. And treat reimbursement not as a post-market detail, but as a feature of the product itself. As Russo put it, reimbursement is a feature, not an afterthought; an implantable device without a credible reimbursement pathway is unlikely to be commercially viable.
Russo’s presentation carried an implicit warning for neurotechnology. The next generation of successful neuromodulation companies may be distinguished less by invention alone than by disciplined translation: identifying a meaningful need, matching the product to a realistic regulatory pathway, generating evidence that persuades clinicians, hospitals, and payers, and designing reimbursement into the venture from the beginning.
Russo closed with the proposition that Yock’s original hypothesis—that medical technology innovation is a teachable discipline—is no longer really in question. For neurotechnology, as devices become more complex, competition increases, and evidence standards rise, the likely winners will be companies that treat innovation not as a single inspired act, but as a structured process that begins with the patient and ends only when a clinically useful product reaches the market.
