Robotic surgery has established itself as one of the most documented medical advances of the decade. Fewer complications, shortened hospital stays, reduced scarring: clinical evidence accumulates. Yet a recent French national study published in Cancers reveals a significant oversight. Across a significant number of analyzed hospitals, a considerable volume of robotic procedures per institution is required to guarantee an acceptable complication rate — namely four to seven times more than the 25 interventions commonly accepted as the mastery threshold. The majority of French hospitals that have purchased a surgical robot do not use it enough to achieve the promised clinical benefits.

The question is no longer technological. The robot operates well when a trained surgeon uses it frequently in a well-organized institution. It is institutional: how do health systems create conditions for this potential to spread beyond centers of excellence, without wasting public resources or concentrating benefits on a handful of well-positioned patients?

The Essentials

  • A French study (Pages et al., Cancers) establishes that the real mastery threshold sits at a level significantly higher than the 25 procedures retained in prior literature, with variability depending on specialty.
  • NHS England has set a target of 90% of eligible robotic surgeries by 2035 as part of its first national guidance published in 2025; NICE conditions 11 platforms on producing evidence within 3 years.
  • In France, there is no dedicated pricing for robotic surgery: institutions absorb the additional costs without tariff compensation, which mechanically concentrates activity in university hospitals.
  • The robotic learning curve is individually shortened through simulation and in-room mentoring, but institutional validation protocols remain absent in most countries.

The 25-Procedure Threshold Was a Comfortable Illusion

For years, surgical literature fixed at 25 the number of procedures needed to master a robotic intervention. This was reasonable as an individual estimate for a trained surgeon in a favorable context. It was insufficient as an institutional criterion for evaluating whether a hospital should acquire a robot and how much it should use it.

The Pages et al. study changes this landscape. By exploiting medical-administrative data covering a broad set of French hospitalizations, the authors reconstructed the history of a significant number of hospitals practicing robotic surgery in urology and gynecology. Their method: measuring not the learning curve of an isolated surgeon, but that of the institution as an entity. The result is unequivocal. The complication rate only stabilizes at an acceptable level beyond a substantial annual volume of procedures depending on specialty, with significant variability based on team composition, complexity of cases treated, and program age.

This figure has an immediate implication. If a medium-sized hospital performs 40 robotic prostatectomies per year, it is permanently operating on the declining portion of the institutional learning curve, where complications are more frequent and stays longer. The patient receives robotic surgery — technically modern — without benefiting from the mastery that justified the investment.

When the Robot Costs More Than It Brings

Robotic surgery is not free. A latest-generation da Vinci system represents approximately 2 million euros for purchase, to which are added between 150,000 and 200,000 euros of annual maintenance, not counting single-use consumables that its manufacturer, Intuitive Surgical, bills per procedure. In 2024, Intuitive Surgical achieved approximately 8.35 billion dollars in revenue, according to its annual results published in early 2025 — a 17% growth over one year, driven largely by these recurring consumables. The manufacturer’s economic model relies on volume: the more the hospital operates, the more consumables it purchases.

This model is not inherently illegitimate. But it enters direct tension with French hospital logic, where activity-based tarification does not distinguish between robotic prostatectomy and conventional laparoscopic prostatectomy. The hospital investing in a robot absorbs the additional cost alone — and recovers it primarily through volume and attractiveness to surgeons and patients. This logic mechanically favors university hospitals and high-volume private facilities, to the detriment of community hospitals that have neither sufficient patient flow nor amortization capacity.

The High Health Authority published a nuanced evaluation of robotic surgery in 2023, recognizing clinical benefits in several indications while highlighting the absence of adapted pricing and the difficulty in evaluating cost-effectiveness in the French hospital context. The situation has not evolved since. There still does not exist in France a nomenclature specific to robotic surgery, meaning the decision to invest remains essentially a financial gamble that each institution assumes alone.

What NHS and NICE Understood That Others Are Slow to Implement

The United Kingdom chose a different trajectory. In May-June 2025, NHS England published its first national guidance on robotic surgery, setting a target of 90% of eligible surgeries performed robotically by 2035, backed by significant public investment to equip structured regional centers. The logic is explicit: concentrate robots in hospitals capable of achieving required volumes, train teams in dedicated centers, then evaluate results.

Simultaneously, the National Institute for Health and Care Excellence adopted a conditionality approach that breaks with usual practice. Rather than authorizing or prohibiting robotic platforms following static evaluation, NICE engaged 11 manufacturers in a conditional access mechanism over 3 years: systems are authorized to operate on condition that hospitals using them feed real-time data registries. At the deadline, if evidence does not confirm clinical benefits under actual British use conditions, authorization may be suspended.

This is a form of evidence-based governance that few health systems have implemented as clearly. It creates two virtuous effects. First, it forces the generation of real-world data, where most countries settle for retrospective studies or meta-analyses on selected cohorts. Second, it aligns the manufacturer’s interest — continuing to sell — with the health system’s interest — proving it works under ordinary conditions, not only in the world’s best teams.

The Learning Curve Shortens, Governance Remains to Be Built

On the technical front, the situation is not static. Several research teams have shown that robotic surgical simulation significantly reduces the time needed to achieve mastery. A meta-analysis published in Surgical Endoscopy in 2023 estimated that surgeons trained on simulators reached competency criteria in 30 to 40% fewer real procedures than counterparts trained solely in the operating room. Structured mentoring programs, where an expert surgeon guides in real time from a remote console, are beginning to deploy in certain European centers, notably in the Netherlands and Germany.

Emerging players are also beginning to challenge Intuitive Surgical’s dominance. CMR Surgical, a British company founded in 2014, commercializes the Versius system in about ten European countries with a modular architecture designed to reduce installation constraints and fixed costs. Avatera Medical, a German company, obtained CE marking for its system in 2022. Hugo RAS, developed by Medtronic, is actively deploying in Europe and Asia. Growing market competition could drive down acquisition and maintenance costs over the next five years — but it does not solve the problem of minimum required volume.

Because individual and institutional learning curves are two distinct problems. A surgeon can technically master a robot in 60 well-supervised procedures. But if their hospital performs only 30 per year, it will never reach the institutional threshold identified by Pages et al. The solution is not individual. It passes through reorganization of patient flows, hierarchization of institutions authorized to practice certain robotic interventions, and a system of continuous training allowing surgeons to maintain competency even in moderate-volume institutions.

This type of territorial governance of robotic surgery does not formally exist in any Continental European country. Exercise authorizations are national or regional; they do not condition the right to use a robot on achievement of a minimum volume. This is a gap that the French study contributes to documenting, and that public health policies will have to fill.

Intra-operative Data, an Untapped Wealth

There is an advantage that robotic surgery possesses uniquely and that conventional surgery does not have: each intervention generates data. Speed of gestures, forces applied, trajectory errors, duration of sequences — all is recorded by the console. This data theoretically allows objective measurement of a surgeon’s progression, early detection of deviations, and construction of quality benchmarks comparable across institutions.

In practice, this data remains largely captured by manufacturers and poorly accessible to hospitals or regulators. Intuitive Surgical possesses a considerable intra-operative database, drawn from over 12 million procedures performed worldwide since 1999, according to figures published by the company. This database feeds its own research and development programs, but it is not systematically shared with user institutions or public health agencies. The logic is that of proprietary assets — where public health logic would want a common good.

A few initiatives attempt to change this. The European Association of Urology launched a European robotic registry that aggregates post-operative outcome data across several countries; it remains voluntary and under-resourced. In France, medical-administrative data provide valuable information — precisely what the Pages et al. study exploited — but they do not capture intra-operative data. The junction between anonymized clinical data and real-time surgical performance data remains to be built.

This is an issue analogous to other domains where data is produced by private actors but generates public value that no one has yet successfully captured effectively. The Draghi report on European competitiveness had indeed identified this type of sectoral data fragmentation as one of the structural brakes on European industrial innovation. Robotic surgery is a further clinical illustration.

What Health Systems Can Do Now

Robotic surgery does not need to be slowed. It needs to be organized. Three levers are available and several countries have begun activating at least one.

The first is pricing. Recognizing the true additional cost of robotic surgery in national nomenclatures would make visible what hospitals currently finance in opacity, and could condition reimbursement on achieving volume and quality criteria. France is clearly behind on this point. Germany and the Netherlands have advanced in this direction with specific packages for certain robotic indications, without completely solving the problem but at least creating financial transparency.

The second lever is volume certification. Several surgical specialties have adopted minimum activity thresholds for authorization of certain complex interventions — cardiac surgery, liver transplantation, pancreatic surgery. The same logic applies to robotic surgery in indications where clinical benefit is proven but volume-dependent. This means accepting that some hospitals will not practice certain procedures, and organizing referral pathways accordingly. It is politically difficult; it is clinically justified.

The third lever is shared data. NICE’s conditional access mechanism is a serious path: requiring manufacturers to feed common registries as a condition of market authorization, rather than leaving this data in proprietary silos. The European Union could play a role here, under the medical device regulation, by imposing post-commercialization data-sharing requirements for class III surgical technologies.

The issue, fundamentally, is transforming robotic surgery from a race to equipment into a coherent public health policy. The countries succeeding first will not necessarily be those with the most robots. They will be those that have managed to decide where to install them, how to train teams, and how to measure that it works. Institutions doing this well deserve to be named — and imitated.

Sources

  1. Pages et al., Cancers — French national study on institutional learning curve of robotic surgery (PMSI database): https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11674775/
  2. High Health Authority — Evaluation of robotic surgery, 2023: https://www.has-sante.fr
  3. NHS England — Official announcement robotic surgery June 2025: https://www.england.nhs.uk/2025/06/millions-to-benefit-from-nhs-robot-drive/
  4. Intuitive Surgical — 2024 Annual Results, Globe Newswire: https://www.globenewswire.com/news-release/2025/01/15/3010092/7637/en/Intuitive-Announces-Preliminary-Fourth-Quarter-and-Full-Year-2024-Results.html
  5. NICE — Conditional approval 11 robotic systems, April 2025: https://www.nice.org.uk/news/articles/cutting-edge-robotic-surgery-gets-green-light-as-11-systems-are-recommended
  6. Surgical Endoscopy — Meta-analysis on robotic simulation and learning curve reduction, 2023
  7. SNITEM — Robot-Assisted Surgery File, Spring 2025: https://www.snitem.fr/wp-content/uploads/2025/04/SI-237-Dossier-Robotique.pdf
  8. Annals of Surgery 2021 — Robotic learning curve (25 cases): https://journals.lww.com/annalsofsurgery/abstract/2021/05000/comprehensive_learning_curve_of_robotic_surgery_.18.aspx
  9. ACS Bulletin 2026 — Cost da Vinci 5: https://www.facs.org/for-medical-professionals/news-publications/news-and-articles/bulletin/2026/february-2026-volume-111-issue-2/cost-of-robotic-surgery-remains-complex-equation/
  10. ScienceDirect — T2A and robotic surgery France: https://www.sciencedirect.com/science/article/pii/S0040595724001975