By Nicolas Borenstein, DVM, PhD
By the time your novel, or generational medical device arrives at a preclinical facility, the commitment is already made. The protocol has been developed, the in vivo model and subjects and test articles are prepared, the team is assembled, the suite is booked. You may well have crossed a continent, and even an ocean, to be there. What happens next depends on a capability some sponsors fail to evaluate: what the team can do when something goes wrong mid-study.
Innovation all but guarantees an unanticipated problem to solve. If a device has never been implanted in a living cardiovascular system, no model development, selection, preparation or protocol can anticipate everything it will do once it gets there. That is not a failure in planning. It is the nature of the work.
The question worth asking of every preclinical partner is not just whether their protocols are sound including the experience or rationale behind the protocol, or whether they’re intimately familiar with the model. It is what happens in the room when the surprise arrives. A team that solves the problem in-flight sends you home with data. A team that cannot sends you home with an invoice, a program that has regressed or stalled, and nothing for your investors, the FDA or notified body. Same device, same day(s), similar spend, opposite outcome.
The ethical arithmetic points in the same direction. Everything committed to that study, the animals included, deserves to produce something of value. Declining to proceed does not undo that commitment. It wastes it.
Solving problems mid-study relies on proficiencies in place before anyone enters the OR: the procedure and the sequence of steps it demands; model selection (and occasionally novel model development) and the comparative anatomy behind it; imaging, so the team sees what is happening while it happens; anesthesia, so someone can adapt as needed when pressure drops and keep the animal viable; and fast, useful analysis and action when something fails.
Anatomy sits underneath all of it. When people ask me what matters most in this work, my answer is anatomy, then anatomy, then anatomy. Model selection is where that knowledge does its earliest work. By now we have seen enough devices in enough anatomies to say in advance that a design will abrade, or perforate, or never fit, and the same understanding is what lets a team respond when it happens anyway. The words for the rest of it are adaptable, nimble, and capable. In practice they mean the credentials and the willingness to change approach in the middle of a procedure.
Ventricular fibrillation. A functional cardiac arrest is not a rare event when new cardiac technology meets a live heart. In some preclinical laboratories, the response is to record the loss and bring in the next animal. Ours is to fight for it. I tell the younger members of this team the same thing every year: never capitulate. If it takes thirty minutes of cardiac massage, you do thirty minutes. The anesthetist calls it out and everyone comes, including colleagues from other operating rooms. Very often, I would estimate 75 percent of the time, we revive and keep the animal viable through the protocol. Instead of a study or test subject that ended ten minutes in, we complete the full protocol.
Context matters to be sure. If the animal fibrillates because the device made life impossible, we aren’t magicians. But if ventricular fibrillation comes from excessive manipulation or momentary low blood pressure for example, we usually are quite successful in maintaining the viability of the subject.
Hemorrhage. The classic cardiac version starts with a nose cone that is not in line with the shaft, and ends with a substantial tear in the jugular, the inferior vena cava, or an iliac artery. Another begins with a delivery system far too large for the anatomy. We place the test article and complete the delivery, then cannot withdraw the delivery system without tearing the vessel. Now someone has to save that artery, or at minimum ligate it and recover the animal. That takes surgical skill, not interventional skill alone. We came into interventional work from surgery rather than practicing only as interventionalists, and that is where it shows.
A device that is not ready for a live system. The delivery system is almost a specialty within the specialty. It has to be blood tight, with no step between the nose cone and the shaft. Teams get there, but it takes time, so I ask early whether the sponsor’s delivery system holds pressure. The answer is usually yes, because on the bench it did, where the fluid was water, and a small leak was hard to see. Inside a large animal at 130 mmHg systolic, that same leak announces itself, on every pass, and more so if the wire sits slightly off center. You can lose a liter of blood in two minutes, and you have to solve it right then.
The blunter version is a specification that does not hold. A sponsor tells us the sheath is 33 French. It arrives at 40 French and will not pass the vessel. One option is to stop the study. The other is to change the route: open the sternum and deliver transapically, if the delivery can be inverted from antegrade to retrograde.. Moving between percutaneous, transvascular with a cutdown, transapical, transatrial, and transseptal approaches is what turns an impossible session into a usable one.
Being good at this cuts both ways, and I will say so plainly. When we make a crude delivery system work regardless of its deficiencies, the sponsor does not feel the pressure to improve it as quickly as they should. The other side is that they get usable data from the very beginning, from a prototype that still needs work. I would rather hand a team early evidence and then tell them honestly what must change.
Sometimes the problem cannot be solved in flight, and then the priority becomes speed of understanding. We open the organ and reconstruct the sequence: it began here, abraded there, perforated at this point under tachycardia, held for a while, then a cleft formed.
Then we suggest changes. The three words I repeat most often about device design are padding, padding, and padding. Devices arrive full of metal, with abrasive surfaces and angles that sit against a papillary muscle or an atrial wall. Protect the device and protect the anatomy. A sponsor should leave with either success including highly informative, usable data or something genuinely learned. The only real failure is having done all of it for nothing.
1. What do you do when an animal fibrillates during an implant? One partially acceptable answer is that defibrillation and cardiac massage would damage the device and skew the results. If that is truly the reason, I can understand it. More often it means that they might not be completely comfortable with how to address it with a reasonable expectation of successful intervention to salvage the animal to successful protocol completion.
2. If a delivery system tears a vessel on withdrawal, who in the room can repair it? A surgical question, not an interventional one. Ask who holds those credentials, and further, is there a blood bank?
3. If our test article (device) arrives outside its intended specification, what are our alternatives that day? You are testing willingness and capability to change approach, not only to follow the protocol as written.
One Caveat: Adaptability Belongs to Preclinical Feasibility and Proof of Concept Pilots, Not to GLP
Everything above belongs to feasibility, R&D, and proof of concept preclinical studies, frequently called pilots. When we move to a GLP-compliant study, the design is frozen, and it should be. There you want rigor, repeatability, and every question settled well ahead of time, not adaptation in the moment. Knowing which phase you are in, and holding the line in the phase that demands it, is its own proficiency.
But a pilot [or non-GLP] study is where a device meets a living system for the first time. For those moments when the problem-solving attitude and acumen of the preclinical lab will make or break the results of the day, it is good to have a solid partnership with a highly experienced preclinical lab, especially in your device category or type, technology and therapeutic area of interest.
About the author: Nicolas Borenstein, DVM, PhD, is Co-President of Preclinical Contract Research Services at Veranex. He has led cardiovascular preclinical research at the Paris facility for more than 25 years and is a widely published author and peer-reviewed journal contributor in surgical and transcatheter preclinical science.
Regulatory bodies require safety and performance evidence before human trials can begin, but poorly designed preclinical studies don't just create delays - they waste the lives of animals and the hopes of patients waiting for better options. Veranex Preclinical Services approaches this responsibility with the intensity it deserves.
We deliver GLP and non-GLP studies across 11 therapeutic areas, with particular depth in cardiovascular technologies including TAVR and structural heart, where we set the industry standard. Our teams don't just execute protocols; they understand that translational science isn't copy-and-paste. Models have limitations, regulators know this, and our job is to design studies that generate the specific evidence your submission needs while respecting both the animals in our care and the patients waiting for your innovation.
Upstream, we can work with product development teams to ensure device readiness. Our integrated pathology services deliver insights on model optimization that improves interpretation and analysis without handoff delays, and our Regulatory Affairs teams use these findings to shape pre-sub meetings, regulatory submissions and trial designs. Downstream, our Clinical Research teams design first-in-human and pivotal trials informed by preclinical findings, while Commercial Strategy & Market Access ensures the evidence we generate aligns with payer requirements and reimbursement realities from the earliest stages. With over 100 regulatory submissions returned with no questions asked from preclinical data, we deliver evidence that clears the path to human trials and builds toward revenue.
If you're evaluating preclinical partners especially for cardiovascular or structural heart devices, or if you're navigating the path from prototype to GLP studies, contact Veranex Preclinical Services to learn how our teams can accelerate your regulatory pathway.