Why I Hate Spine Surgery Robots in 2021, 2022 and 2023

Having been a spine surgeon for over 20 years I’ve seen it all. Robots are the handy dandy new shiny thing in spine surgery. I hate them. Why? Read on.

What is a Robot?

From Wikipedia: “A robot is a machine—especially one programmable by a computer— capable of carrying out a complex series of actions automatically. Robots can be guided by an external control device or the control may be embedded within. Robots may be constructed on the lines of human form, but most robots are machines designed to perform a task with no regard to their aesthetics.”

Robots do things. They make dangerous or repetitive maneuvers for humans easier by executing those maneuvers.

Tesla Uses Robots

Watch the video below. The cars are assembled by robots, repetitively doing heavy, dangerous, synchronized actions again and again. These are true robots:

Do Robots Have Artificial Intelligence?

Not yet. Most have computers programmed by humans and work within a specific frame of reference and cannot make independent decisions. They are not sentient, not conscious and able to feel and perceive things.

In classic science fiction, Isaac Asimov created the Three Laws of Robotics. In 2021 these are still  more science fiction than reality as autonomous decision making is not seen in most mechanical structures we call robots.

For a fascinating read, look at this interview by Stephen Hawking on the dangers of artificial intelligence:

A more detailed analysis is here, ironically on Reddit!

Even Tesla struggles with its Autopilot feature or what they call Fully Autonomous Driving. It’s a misnomer. I love Teslas and own one but my Tesla won’t drive itself yet from home to work and would most likely violate the first law or robotics if it did today.

The video below is what a Tesla can allegedly do on Autopilot. I wish it were true in 2021 but it’s fantasy for my Tesla. Mine can’t:

What is a Surgical Robot?

The Da Vinci robot in general surgery, cardiac surgery and urology has been around a while. It has gained acceptance. The surgeon places ports inside the patient and then sits at a workstation, which can be remote. The robot arms then mimic the surgeon’s hand movements in the patient. The ‘robot’ does not act autonomously but mimics surgeon movements although it may provide some smoothing of those movements.

Watch this video:


What We Try and Do in Spine Surgery

One of the most challenging things a spine surgeon does is put screws into patients. The screws are anchors in bone. They must often be placed with mm accuracy missing nerves, blood vessels and organs and must be secure. In the lumbar spine we typically place screws in pedicles. Imagine a desk turned upside down. The vertebral body would be the flat table. The legs would be pedicles. As a surgeon we train for years in placing screws straight down the table legs into the flat part of the table.

This is the anatomy of pedicles in the lower back (lumbar spine):

(Case courtesy of Dr Matt Skalski, https://radiopaedia.org)

As trainees we learned to use anatomical landmarks to guide screw placement. The miss rate was 20%. With the use of x-rays (real time is called fluoroscopy) the miss rate was probably 5%.

If you are bored watch this video which shows the nuances of putting in screws using traditional methods:

In the past 5-10 years surgeons have been able to use a spine GPS system in the operating room  ( called an O- arm) which in some cases gives the surgeon a 3 dimensional CT scan. The patient has a CT scan in the operating rooms. Those images are matched on the screen to the patient and the surgeon works on the patient he can see where he is on the screen of the CT scanner. We call this spinal navigation or CT navigation or simply the “O-Arm”. The miss rate on using navigation is probably less than 1%.

Watch the video below to see how surgeons use navigation like GPS to put in screws:


This is true navigated spine surgery– the surgeon being guided but in control and being able to feel and control the course of screws and implants as he works. If he doesn’t like things he can change them.

The O-Arm was a game changer.

This is a side view and a front view of a patient into whom I placed screws. The CT scanner guided them but I placed them free hand, not with a robot. There are 6 screws but on the side view it can look like 3 as they overlap:


Here is a case I did, before and after. The after shows putting screws in that thread the needle to miss screws already in place. This was not a robot. This was freehand with CT navigation:

What is a Spine Robot?

Imagine getting into what you are told is a robotic car and being asked the destination and the car just drives you there. That’s a robotic car.

Now imagine that you are asked the destination but rather than taking you there it just shows you the way and you actually have to do the driving. On top of that once you decide the destination there is no changing the route. That’s a spine robot in 2021.

A spine robot in 2021 is not a robot in the sense it does anything.

A spine surgery robot in 2021 is a GPS system with a guidance arm.

It’s a GPS system with a fixed guidance arm. Once you choose the destination, the arm will look the trajectory in and you use that trajectory and hope that what you chose on the computer matches the patient. It usually does but if the patient moves, the computer will not know and the wrong destination can be reached. There is no feedback. It’s rare but can easily happen. The robot will give a channel to work down. You can put drills and taps down that channel and at the end of the case hope the channel you made matches that on the screen. It usually does.

The main 3 robots all look the same:

Mazor Robot:

Pretty well known. I have used this one.

Globus Robot:

Again, in the top 3. I trained with this one.

Rosa Robot:

This I know of but never used.

Note that all are basically a computer connected to a mechanized arm. The patient is scanned. The scan is fed into a computer. The surgeon picks the path he wants. The arm shows the path. That’s it. The surgeon then drills down the hole and puts screws in. The robot doesn’t do anything. it shows the path.

If you watch this video you will know what I mean:

Bingo. Pretending to be one thing but being another!

Spine Robots don’t do anything. They are GPS systems on Fixed Guidance Arms.

Look at the definition from dictionary.com and see that it applies to spine robots. They use a form of GPS to show the routes in the patients body:

I would call them  GPS Fixed Arm Guidance Systems but GASY is  not sexy.

What is the Value Proposition?

A value proposition is a simple statement that summarizes why a customer would choose your product or service.

I break things down in terms of new spine technologies into adding value in one of 3 ways without sacrificing things:

  1. Cheaper
  2. Better
  3. Faster

Cheaper is cheaper for everyone. American healthcare is expensive. Capital expenditures have opportunity costs. Operating time is thousands per hour and prolonged surgeries cost more money. Disposables add to this expense. I don’t see the perceived financial benefits of a robot over CT navigation. If you want to deep dive on how crazy US healthcare costs are and how little we get for our money, read this.

Better means that it’s more robust and fails less often and may be safe. Our own hospital has had patients injured by using a robot. CT Navigation (where the surgeon uses his hands, tactile feel and adjustments) is for most surgeons who navigate, safer.

Faster is faster for the surgeon. Less time under anaesthesia means a quicker recovery for the patient. I wish robots were faster than what we do but that’s not the case.

The value proposition of spinal robots centers around the perception that the general public has that a robot ‘does things’, is automated and safe and free from ‘human error’. It’s a myth for spine robots.

Who Uses a Robot?

  1. Surgeon’s who are not comfortable with traditional screw placement techniques or don’t use navigation. Makes a poor surgeon safer.
  2. Those looking for a marketing gimmick.
  3. Early adopters. We have a saying- don’t be the first one on the bus and don’t be the last on the bus.

Should Patients Look For Surgeons Using Robots?

About as much as the should look for surgeons who use lasers, surgeons who tout ultrasonic drilling over regular drills, awake surgery, augmented reality or other marketing gimmicks. The irony is the best surgeons I have seen use anatomical landmarks or navigation. With navigation putting in screws is not the main part of the case. Robots in spine surgery made something easy harder.I won’t go into technical nuances but those surgeons comfortable with just navigation, are often the fastest and best. See my blog on this.

Why the Tail Should Not Wag the Dog

This is food for thought. Why is the medical device industry pushing robots in spine surgery? Well, each robot works best with a particular company’s widgets, usually screws. If a company places a robot in a hospital it only works with that companies implants- you sell more widgets. This is like giving the car away for free but charging for mileage, maintenance, consumption etc. The big companies do their math and know how to make money. The ultimate goal is selling more widgets. More widgets sold is more profits which satisfies the fiduciary duty of most spinal implant companies who are trying to give their shareholders the greatest returns. The big companies have a duty to sell stuff and spine robots may help them do that.

Want a Good Deal on an Palm Pilot?

The Palm Pilot was ahead of it’s time but limited. Few apps. Limited utility. Today’s spine robots are Palm Pilots with million dollar price tags. In 5 years they may still be Palm devices. Might be a good idea for a new model? I can sell you a Newton or iPhone 1 if you don’t mind Gen 1 technology.

Consumer Beware the Snake Oil

I’ve written on this before. Consumers need to be beware of gimmicks. Lasers, robots, augmented reality. See my blog article here.

The Future

True robotic spine surgery will come. Patients will be able to have implants placed safely as outpatients prior to surgery. Surgeons will do the planning. Robots will do the drilling. They will place screws and decompress neural structures safely. Surgeons will oversee it. It will come but it’s science fiction in 2021.

So… …Why Do I Hate Spine Robots in 2021?

  1. They are not robots. They are GPS systems with fixed arms (GASY)

  2. Competent surgeons don’t need to use a robot to be efficient and safe

  3. I’m not a Luddite but I don’t like to be a beta tester for new technologies on my patients

  4. The public should not be fooled that ‘robotic’ spine surgery is superior to what we do now, especially with navigation access.

  5. They are not robots (I know I said that, it’s important).

  6. They cost a million bucks each.

  7. The tail should not wag the dog. Unbiased surgeons not megacorporations or industry megaphones need to guide healthcare.

  8. Healthcare is expensive enough. Technology should make things cheaper, faster or better; not longer, more involved and more expensive

  9. I don’t want to buy an Palm Pilot (and not for a million bucks)

  10. CT navigation with an O-arm already does it all

Give it 5-10 years and it will improve. I want to be sipping coffee in the surgeon’s lounge whilst a real robot puts in screws, does the decompression and calls me to inspect when it’s done. Now that’s a robot!

My Disclosure

I have no relationships with industry and speak my mind. These are personal opinions, not my workplace or employer. Note that this opinion applies to 2021. I have consulted over the last 20 years with most of the major players but in the quest for remaining neutral don’t consult with the spinal medical device industry anymore. I don’t lecture for industry as most of these educational opportunities became infomercials. You can look at what I have gotten from industry here and if you see someone criticize what I say, have a look at their profile at the same website and look at their disclosures. I teach residents/fellows/surgeons through a multinational not-for-profit called the AO Spine. That entity focuses on non-industry driven knowledge and principles without the secondary gain of selling more widgets.

I welcome any comments or criticism. I just ask that anyone who comments disclose any financial ties to the spine industry. If you are consultant on spinal robots or work with a company say so. I don’t have any of those conflicts. Thank you for reading!

8 responses to “Why I Hate Spine Surgery Robots in 2021, 2022 and 2023”

  1.  Avatar

    Great review Lali

  2. dentonrdavenport Avatar

    Great article on the use of robots in surgery. On the anesthesia side of the table, I was not a fan of robots being used in the OR, regardless of the case. With extremely rare exception were the result better or faster than could have been accomplished by traditional laparoscopic approaches. Often, the docking of the robot and set up added 30 minutes to a case. Not to mention the extremes of head tilt on the OR table needed to facilitate the surgeon. Once the robot is docked, the OR table positioning is fixed unless you undock to robot. Your take is fantastic, its basically a guiding arm, that arm isn’t needed but as you say, it makes a bad surgeon safer, not a great marketing tool because what surgeon is going to claim they are bad.

    1. Lali Sekhon Avatar
      Lali Sekhon

      Hi Denton! Thanks for taking the time to read it and your honest take. Could not agree more! Here is an interesting article I saw today that says the same!

      1. dentonrdavenport Avatar

        I’m not sure I told you but I do own a good amount of stock in the company that owns the Da Vinci system and it has tripled since I bought. Its the current cool toy that surgeons want to play with.

  3. Mark Goodwin Avatar
    Mark Goodwin

    Hi Dr. Sekhon,

    Loved your review. I am a retired surgical device representative and spent years selling spine implants; I loved working with surgeons! I recall being aghast the first time I observed a spinal fusion in 1999! Since then, I have been searching for a non-ionizing mode of imaging to facilitate real time navigation and interrogation of bone with a hand-held device with which the operator can efficiently and accurately locate the “opening” (pedicle os) to the vertebral pedicle BEFORE cortical bone breach followed by creation of a purely intra-cancellous channel—ideally into a second layer of cortical bone for so-called bi-cortical purchase, e.g., cortical bone of the vertebral body. No device or technology yet exists to achieve this—obviously a big ask. I have identified a real time, non-ionizing, mode of imaging capable of penetrating cortical and cancellous bone prior to its disruption essential for such a device.

    I am well versed on the pros and cons of image guidance technology and wholly agree with your review—I sold the very first FDA cleared image guidance system (Viewing Wand™) as an alternative to rigid stereotaxis for localizing brain tumors, etc.

    I like to use google maps GPS technology to help explain the difference between “real time” image guidance versus GPS navigation: Google Maps is a marvel at plotting and guiding a path for an operator to his/her target destination—even including a reasonable assessment of traffic conditions along the way. What it cannot do is provide real time instantaneous road conditions. So, if a large sink hole opens minutes before the driver reaches that spot, he/she may be doomed to drive into it. The google path (trajectory) is accurate but limited by what is “historical” data—even if by only a few minutes.

    I welcome your thoughts, comments, and/or input.

    Mark Goodwin

    1. Lali Sekhon Avatar
      Lali Sekhon

      Hi Mark! Thank you for taking the time to read the blog and your thoughtful reply. I think you nailed it all in your comments. Image guidance really is the bomb- it’s not real time and you need to watch for those sinkholes! Take care. Lali Sekhon

  4. Nicholas Theodore Avatar
    Nicholas Theodore

    Hey Lali! Nice review but you are missing a major piece of the puzzle. As we automate spine surgery, first with navigation, then with robotics for screw placement and soon for osteotomies and bone cutting we are increasing the accuracy and safety of the procedure. My team at Johns Hopkins and many others have published extensively on this topic. Robotic assisted spine surgery is more accurate than freehand and image-guided surgery. You are also overlooking another major fact which is that we do not have as many hours to train our residents as we used to. Teaching the safe and proper use of enabling technologies is the future and my millennial trainees are embracing and insisting on using these technologies. The adoption rate of spinal robotics has been an enormous success story. When you see the work being done at NYU, Hopkins, Duke and other academic medical centers, as well as countless community surgeons you soon realize that despite your “hate” of the technology, it is here to stay. The Luddites shunned technology as well and look what happened to them!

    Certainly as the inventor and founder of Excelsius Surgical and Globus’ ExcelsiusGPS system, I may be a bit biased, but after seeing the field explode with over 100,000 robotic cases being done worldwide the trends in adoption are such that as in countless other cases we will not regress. Continue doing what you do by there is no need to be a hater!


    Nicholas Theodore, MD, FAANS, FACS
    Professor of Neurosurgery, Orthopedics and Biomedical Engineering
    Johns Hopkins University

    1. Lali Sekhon Avatar
      Lali Sekhon

      Hi Nick: Thanks for taking the time to read it and reply! Totally agree; the future is what’s missing. Current price point as a drill guide doesn’t translate to a value proposition for community docs but as you said, for teaching it may make trainee surgeons safer (as will NAV). Once it can do the do the decompression and ideally insert the hardware we are somewhere! The Palm Pilot was a great start but most of us dreamed of the Star Trek tricorder. In 2023 major capital purchases have a significant opportunity cost and I dream of a small robot that inserts screws, decompresses, allows for NAV and is under 100K in today’s dollars. It’s a great start but a long way to go.
      BTW I don’t really hate robots- i just want more! Lali

Leave a Comment or Reply Below!


Enjoy this blog? Please spread the word :)

%d bloggers like this: