There are many myths and stories about how 3D printing is used by physicians and surgeons in caring for our patients. For this article I will be using veterinary examples, both because I do my research in clinical 3D printing in veterinary orthopedic surgery and this of course avoids medical privacy laws. In this article I will describe a typical workflow and how you can get involved in clinical 3D printing if that is something that interests you professionally. The image above is a dog's nose on a CT (those delicate structures are called turbinates)
I am a practicing attending physician and Assistant Professor, at a Harvard Medical School affiliated hospital, (for European types, you would refer to me as a “consultant” which is not what we mean in the US by that term, we use the term attending) who is board certified in Hospital Medicine (inpatient medicine) and Clinical Informatics (the fusion of clinical care and computer science/information theory). I teach a course in CAD/CAM and medical device design to surgeons. My earlier career was in the industrial control industry at a joint venture with Allen-Bradley (now Rockwell Automation) which made high-level factory control systems that sat on top of the controllers such as PLCs, Robots, etc (as such as car factories, etc). It was natural that I would gravitate to 3D printing and fuse that with medicine. I do most of my research and am faculty at the Cummings School of Veterinary Medicine at Tufts University, where I am an Assistant Professor as well in the Surgical Innovation Center.
It is important to understand, as this often causes confusion, I am not doing research on animals in the way the horror stories of animals being exposed to toxins, etc are, these are patients of the hospital (a large academic veterinary hospital) who have an orthopedic or other injury/disease who need our care, and we are massively advancing their care where honestly we do more sophisticated 3D printing usage on the small and large animals, than we do on the semi-sentient bipedal primates (that is humans for non-native English speakers) I work on in my other practice. All of the research on the animals is governed by IACUC (the regulatory body that covers research on animals) but this allows us to fix injured animals that would never otherwise be treatable. The philosophy of Tufts University between the Veterinary and Medical schools is called “One Health” where one should inform the other, so here animal care is improving human care as I will describe later.
there are 3 primary kinds of projects I engage in at both institutions:
We will now cover each of these:
This is how I got started, what I jokingly refer now to “stupid pieces of plastic” type projects, where surgeons and other specialists would ask me to design a “thing to hold the thing”. And while these sound like silly little 3D printing projects (similar to the brackets and holders on printables) they turn out to be a huge time-saver for a lot of medical care. The very first project I got asked to make was a cup (yes a cup). The project was for one of my colleagues who is a thoracic surgeon and they use an electro-cautery knife, knowns as a Bovie knife (a knife that uses RF to cauterize while it cuts) that is much longer than the typical surgeon as the normal model is for skin and easily reached tissue, rather than deep inside the chest. So the tip is about 4x as long, and the nylon cup that comes with the device (sterile) which clips on the surgical table to hold the device while not in use, was too short so when he threw the device into the normal cup it would bounce out and fall on the floor which caused a 5-10 minute delay while the got another one from the closet and connected and tested it. So could I design a cup that could be sterilized and would hold the long tip safely. That was a pretty easy first project, which I printed in Taulman3D Nylon 910 which came with a FDA Certificate of Origin which helped to get the project approved by the hospital. We demonstrated it was safe to autoclave (125C) and with the help of our microbiology lab could prove it was sterile. The steam didn't seem to affect the strength in any way (and this should only ever need to hold 10-15g of force). Once that project was completed, numerous surgeons came out of the woodwork to find me and have me make <plastic thing here> which was the in I needed to start getting involved in actual clinical printing with them.
If you get involved this way here are things to keep in mind:
This is the most fun as you combine both 3D printing, CAD and creativity. This is the ultimate cosplay in many ways as your models are functional while also possibly having to look realistic (more on that in a second). Simulation is often highly desired at teaching hospitals, and will immerse you in the world where you can learn in a much lower stakes environment (killing a patient is a huge no-no, killing a medical student, well new ones arrive every year [that is a joke]), but seriously, the worst thing with a bad simulation design is the students don't like it, or the device doesn't get used much. Simulation centers at hospital are also great testing grounds for designs of all sorts of projects and a great way to try out a device with a surgeon in a realistic setting without consequence.
Example project: this project was to produce simulated small bowel to teach surgical residents how to repair the bowel (suturing the ends together, known as anastomosing) and the 3D printing aspect of this was to produce the spin-mold that this silicone/powermesh bowel segment
Your friendly author showing the inside of the colon after demolding from the PLA spin mold
Yes, it looks obscene, but that is a 3D model of the “cecum” the connection between the small and large intestines (made from a CT scan of a patient). To make it function like actual colon tissue, we put power mesh (Dacron/spandex blend, microweave) to simulate the fascia in the tissue. and then had several surgeons test suturing into it in the laparoscopic simulator and on the table by hand.
As an outsider you might offer an interesting alternative view, but always respect learning of the past 4000 years, most of the lessons in medicine are written in blood as they say. So one example was one of our cardiac surgeons approached me “could you 3D print a realistic heart model” “Sure, why?” (the why was mostly because since anatomic models of the heart are available in the bookstore so not sure why he wanted another one). He noted that a particular surgery (mitral valve repair) was probably the most dangerous surgery done in any specialty and it was so hazardous they don't allow trainees to even learn it (um??) and even new attending surgeons are paired with an experienced surgeon to supervise. And if we could produce a simulator it would enable frequent practice and training before performing it on a live patient. So my first task was of course googling it, and there was one on the market costing around $1500 (cheap for surgical simulators) but the issue was it was $300/reload as the part is damaged during the procedure as it is is in real life.
So we quickly set about making the simulated anatomically correct heart, it was harder than you'd think as it had to be both anatomically correct and functionally correct. We had a high-school student intern in the lab, and he was helping model it and asked the unobvious question of why did it have to look like a heart, since presumably anyone training for cardiac surgery knew what a heart was. This prompted an exercise to watch a surgeon perform the procedure seeing what the tricky part of the surgery was and realizing the outside of the heart was meaningless but several other aspects were the skills to master letting us make an easily printable structure (we ended up making most of it from laser cut plywood with a cast silicone part that was the disposable part which had embedded power mesh valve tissue. The casting was in a PLA mold on the MK3S.
Testing the Mitral Valve Repair simulator
This is the easiest area to get involved in, as there are no regulations from the FDA for simulation devices, and educational devices are excluded from most of the research rules.
This is the highest risk activity to get involved in and is highly regulated. So why do we need 3D printing in clinical care. I frequently hear stories of “My surgeon didn't know how to fix my X until I printed a model” If that was true your surgeon, probably shouldn't be working as a surgeon! We are all trained to use flat imaging (x-rays, CT, ultrasound) and with our knowledge of anatomy to visualize everything in 3D. So who uses 3D printing the most? Well dentists are some of the largest users of 3D printers, period. But if we broaden dentists to people who work with bones (teeth are bones). Orthopedic surgeons use it heavily (which is what my research is around).
The printing is not because the surgeon doesn't know the anatomy, (I promise we do!) as a colleague once said, “I know what the tumor looks from a 140 character tweet!”
The parts we print for the OR is actual surgical guides for driving the drill exactly in the plane we need on a bone (imagine getting the bit exactly normal to a curved surface or the plate will rock slightly impairing bone healing). So we use a specialized CAD software from Materialize, called Mimics, which understands bones and plates derived from CT scans and will generate a conformal drill and saw guide that we print in FDA approved surgical guide resin (in fact FormLabs did the clinical trials with our lab) and additionally the software can develop custom conformal plates as well that can be machined. The print is immediately sent to central sterilization where it is autoclaved and sent to the OR. Additionally we print the bones in stiff resin, not because the surgeon doesn't understand what the bone looks like, but before performing the operation, they can dry fit the plate and guide on the bone to dry out the angle with the drill and see how the guide locks onto the bone to see if the incision is large enough. Once the surgery is complete the guide is kept for records (it has the patient's ID number in the resin).
However, before you casually start printing things for the OR, this required an IND (Investigational Novel Device) certificate and approval from the FDA, IRB and NIH, and the hospital required frequent microbiologic testing to assure sterility. Additionally most polymers do not tolerate autoclaving without speaking with the materials scientist at the manufacturer. PLA does a particularly poor job as well all know when wet. Think about starting in one of the first 2 areas and the next area.
Now there are items to help surgeons that aren't actually in contact with the patient, and here is a perfect example using Prusa polymers. One challenge you have when doing orthopedic surgery is we use a fluoroscope (x-ray video camera) which lets us see bones in real-time. However it is hard to know if a fixture (such as a drill guide) is exactly perpendicular to the camera. We first DMLS printed this gunsight like fixture (a drill guide slips into the inner ring) in 316L stainless steel, which turns out to be far less opaque on x-ray than we thought. So when Prusa came out with the Prusament PETG with Tungsten in it (specifically for radiation blocking) we reprinted it (the darker of the two images):
The top Prusament tungsten-PETG fixture has tiny blades that let the x-rays through if they aren't perfectly square to the camera (about 0.1 degrees is enough to shine through) making it instantly apparent that the drill guide is not in line as needed (the guide would be on the patient)
Left is the Prusament PETG-Tungsten, Right is 316L Stainless via dMLS that took the photo above. The slip of paper under one side is enough of a tilt to make the blade gaps visible on screen.
This was a great project to work on as the initial testing did not require clinical approval (other than getting permission to use the fluoroscope) and if you are under 18, OSHA does not permit under 18-year-olds to work with ionizing radiation. (note the initial testing actually just used a light table as light and x-rays perform pretty similarly here).
A question I frequently get from makers is how can I get involved. Well if you are affiliated with an academic institution that is easiest particularly if you are at an institution that is affiliated with a medical school. Then faculty of both can meet in a faculty setting and hopefully find overlapping needs and skills. It is important to set expectations of it will be a long while until someone is letting you print things used in patients (if ever) and even as a doctor I did a lot of years of making little clips, fittings and things for other doctors until we did our first real phase 1 clinical trial device. Even then you need to take this seriously as real harm can happen if things go awry. However, it is exciting and there are lots of opportunities to help patients. The first thing you need to do is research the area of medicine you want to work with, and learn as much as you can (but also understand doing some heavy reading is not the same as our decade of training!) and then see if you have a friend who is a surgeon and begin tossing around idea.
The most fruitful place I found to get ideas is attend the surgical M&M (Morbidity and Mortality conference) which is a weekly or so meeting in a department that goes over adverse outcomes (some avoidable some not) that have something to learn from. Sometimes they will involve not having a thing (hey, you know how to make things!) or a thing broke (reinforce the thing). Often the problem seems simplistic (the thing fell on the floor like the cup I talked about above), but understand that cup was resulting in an entire thoracic surgery team standing around for 10 minutes staring at the walls with a patient under anesthesia, while someone got the new device hooked up. That's literally thousands of dollars of wasted time over a 50 cent cup! When you hear one of the M&M stories ask a surgeon why that thing was not present (was it because nobody brought the thing to the OR or that thing doesn't exist). If you have the ability to make the thing, see if you can get an introduction.
Comments
█
█
█
█
█
█
█
█
█
█