The 4K-3D exoscope

Almost every operating-room technology is justified by what it does to the tumour. The exoscope is the exception: its central argument is what it does to the surgeon.

The surgical microscope imposes a posture. The eyepieces are where they are, and the surgeon's head has to go to them and stay there for hours, neck flexed and shoulders raised. The exoscope takes the optics away from the surgeon's eye and places them over the field, with the image projected on a large high-definition screen with stereoscopic depth. The surgeon operates looking forward, standing or seated, head in a neutral position. And the whole theatre sees what they see.

The interesting question is not whether that is a good thing. It is what was measured, what was demonstrated and what was taken for granted.

The problem it addresses is real

It is worth starting there, because it is usually treated as a comfort detail and it is not. An international survey of 409 neurosurgeons found that 87.9 per cent had suffered at least one work-related musculoskeletal disorder¹. Of those, nearly three in four reported pain; 27.4 per cent began experiencing it during residency; 8.4 per cent reduced their surgical volume and about 8 per cent missed work because of it. Only 3.4 per cent used a chair with lumbar support during their most frequent operation.

A systematic review of the specialty's ergonomics literature completes the picture with figures from spine surgeons: 62 per cent with low back pain, 59 per cent with neck pain, and 23 per cent of those who developed a lumbar disc herniation ended up in surgery themselves². In endoscopic surgery, up to 80 per cent report musculoskeletal problems.

This is an occupational health problem of a magnitude that in any other industry would have produced regulation. That the solution comes in the form of a device costing hundreds of thousands of dollars rather than a redesign of the workstation says something about how medical innovation works, but it takes nothing away from the solution.

What actually improves

The ergonomic evidence in favour exists and is reasonably solid, though not unanimous.

The study closest to a randomised trial in neurosurgery compared operating through eyepieces or on a monitor in twenty randomised supratentorial tumours³. During resection the surgeon kept the trunk upright 80 per cent of the time on the monitor against 56 per cent with eyepieces, a statistically significant difference. A study with wearable sensors on the nape and upper back of ten neurosurgeons found that in spine surgery, where all used the exoscope standing, time in neutral posture was significantly greater⁴. That work carries a limitation its authors flag: the device is confounded with the type of procedure, so better posture cannot be causally attributed to the exoscope.

And there is a counterpoint worth not hiding. The most rigorous comparative trial of exoscope versus microscope is not in neurosurgery but in ear surgery: 62 patients, 31 per arm⁵. There the exoscope was significantly inferior in visibility of detail, depth perception and illumination, and — against all expectation — musculoskeletal discomfort in the upper back was greater with the exoscope, not less. One study, a different specialty, a different device. But it is the only one with a randomised design, and it does not say what the rest of the literature says.

What does not change

Here the answer is clean and should be stated without ornament: there is no randomised trial in neurosurgery comparing exoscope and microscope with patient outcomes. None.

The most recent meta-analysis in spine surgery gathered seven studies and 785 patients⁶. Operating time: no difference. Complications: no difference. Length of stay: no difference. Postoperative axial and radicular pain: no difference. The only variable reaching statistical significance was blood loss, with a difference of 5.9 millilitres favouring the exoscope, which is an almost textbook example of a result that is statistically significant and clinically irrelevant.

One methodological detail betrays the state of the field better than any statement: the authors assessed the seven studies with the Newcastle-Ottawa scale, the instrument designed for observational studies. Had there been randomised trials they would have used a different one. The quality tool reveals the absence.

In that same meta-analysis's qualitative assessment, across 36 domain judgements, the exoscope came out superior in ten, equivalent in sixteen and inferior in ten. Consistently better in ergonomics and teaching; the most cited drawback, the mandatory 3D glasses.

The price of learning

The transition is not free, and the figures vary in a way that says more about teams than about devices.

A series of 39 procedures involving ten different surgeons reported that 69.2 per cent of operations had to be converted to the microscope⁷. The proportion fell from 90 per cent in the first half of the study to 52.6 in the second, a significant improvement. The stated reason was almost always the same: impaired hand-eye coordination and insufficient depth perception. One surgeon out of the ten performed 41 per cent of the cases, half of them without conversion.

At the other extreme, a series of 243 spine operations with the exoscope recorded no conversions at all⁸. Between 0 and 69.2 per cent there are two orders of magnitude, and that spread does not measure a property of the device: it measures the team's learning curve and the centre's policy.

The optical limitations are well documented and consistent. Depth perception is the main complaint, and illumination in narrow, deep corridors is worse than the microscope's: in anterior cervical fusion and long approaches, image quality was rated inferior⁹. An early robotic exoscope was described by its own evaluators as slower and riskier in the first cases, precisely because of the lack of stereopsis¹⁰. It is worth noting that this device offered no three-dimensional image: stereopsis, not magnification, is the variable separating these machines from one another. "The exoscope" does not exist as a homogeneous category.

One figure sets the scale of the effort. A randomised crossover study with 17 novice surgeons and seven experts found that the exoscope's learning curve is non-inferior to the microscope's in a laboratory task¹¹. But the stated preferences split revealingly: three in four preferred the microscope for visualisation, and nearly two in three preferred the exoscope for ergonomics. And a study that followed two already-trained neurosurgeons with no prior exoscope experience for a year measured mean dissection time falling from 34 to 26 minutes: a 24 per cent improvement over twelve months¹².

The teaching argument

It is the most cited and the least supported. That the whole theatre — scrub nurse, resident, anaesthetist — sees exactly what the surgeon sees is a correct physical description of the device and an obvious advantage over a machine with two eyepieces.

What is not demonstrated is that this translates into better learning. A narrative review devoted to the topic gathers the available evidence and it is thin: a training study with twenty neurosurgeons that found no difference in duration or in the amount of haematoma evacuated; a suturing study with eight participants in which six reported greater ease; a learning-curve study with no published scores¹³. The authors of the largest spine series acknowledge it verbatim: the teaching value of the exoscope has not been objectively measured.

There is reasonable indirect evidence — theatre nursing staff rate the visualisation of the field as superior to the microscope's — and there is a strong pedagogical intuition. What there is not, is a measured outcome.

What to take away

The exoscope is an interesting case because it inverts the usual order. Almost every operating-room technology is sold on the promise of a better clinical result that is later not demonstrated. This one is sold on an ergonomic promise that is demonstrated, and drags behind it the suspicion of not having proved a clinical benefit it never, strictly speaking, promised.

A series of 52 aneurysm clippings by a single surgeon found equivalent outcomes with both devices¹⁴. That is what one should expect, and it is the right result: if the exoscope is a safe alternative and the surgeon works better, clinical equivalence is a pass, not a fail.

The problem is not the absence of clinical evidence. It is the habit of demanding it only from technologies that lack it, and never from those already installed.