Publication and the PI: Edouard Aboian, MD
Yale Department of Surgery’s Edouard Aboian, MD, evaluated abdominal aortic aneurysm surveillance in postoperative settings. The findings were presented earlier this year at the Vascular and Endovascular Surgery Society’s 48th Annual Meeting. We asked Dr. Aboian, assistant professor of surgery, about the reidentification of at-risk aortic aneurysms.
Can you tell us about your background and what inspired you to pursue this particular research topic?
One paper in the past that struck me reported around 1,400 deaths per year after abdominal aortic aneurysms repair procedure. One of the potential gaps in our understanding of abdominal aortic aneurysms is a subset of patients that have changes in sac volume measurements that we cannot pick up with current methods of monitoring.
So, that pushed me to find more sensitive ways of evaluating abdominal aortic aneurysms repair. We also explored how to measure it more efficiently and reduce interobserver variability, meaning variability among physicians that care for those patients.
Can you give us a brief overview of your recent article in Annals of Vascular Surgery?
We compared the standard for postoperative abdominal aortic aneurysm surveillance, maximum aortic diameter, to volumetric analysis of abdominal aortic aneurysm. Essentially, we believe that aortic volume analysis is more sensitive because it reflects not only diameter changes but also three-dimensional changes elongation along the different axis that happens to the abdominal aneurisms. That was the hypothesis that we tested.
The literature on volumetric aneurysm analysis is ambiguous. There are reports that say its more sensitive, and there are others that say there's no significant difference. We analyzed pre-and-postoperative computed tomography angiography (CTA) scans and then evaluated that over one year to see if there was a significant difference in diameter change and compared it with the change of volume.
The change of volume was more sensitive compared to diameter change, especially after endovascular abdominal aortic aneurysm repair. On average, approximately 14 percent of patients did not have diameter change where they did have a volume change. Thus, the volumetric changes can tell us whether patients are at higher risk for potential ruptures, hence need to be monitored more closely.
We didn't have enough data to validate our findings in long term. However, what we saw in another randomized trial that after eight years aortic related mortality increased in a group of patients that had endovascular abdominal aortic aneurysm repair versus open repair. Our volumetric aneurysm analysis confers that you can detect that and potentially intervene early on.
If you see early volume changes, you may need to follow-up with these patients twice a year instead of once a year. Or instead of an ultrasound you may use CTA scans in that group of patients. On the other hand, you may need to follow-up less frequently with patients that don’t have volumetric changes.
During the course of your research did you encounter any setbacks or unexpected results?
We saw a decrease in the aortic volume and diameter in patients who used metformin to treat type 2 diabetes. I believe that these patients have sac shrinkage or a decrease in size. But it hasn't been consistently proven. In our study, we saw a volume decrease that was statistically significant associated with metformin use. Which was surprising because we didn't look to find that specifically.
There are ongoing studies of metformin use. Our finding could potentially corroborate with that study, but we don't have the results. But it was exciting that what we found is being actively investigated.
Your team recently presented an abstract titled, “Artificial Intelligence-Based Morpho-Volumetric Analysis of Pre- and Post-Evar Infrarenal Abdominal Aortic Aneurysms Characterized on Computed Tomography Angiography” at the Yale Department of Surgery Research Day.
Volumetric evaluation has less interobserver variability compared to maximum diameter because diameter can be measured in different planes, oftentimes not orthogonal to the centerline of flow, and that may introduce some false findings in aortic aneurysm behavior. To rectify the aneurysm diameter findings additional steps maybe needed to compare two studies.
When you're comparing apples to apples, volume has the lesser variability, and as such provides more precise provider independent monitoring of abdominal aortic aneurysm behavior. On the other hand, it’s part of our work to make volumetric measurements more efficient. We trained an artificial intelligence-based deep learning platform that has an incredible accuracy of 99.7 percent, and that is agnostic to images acquisition and sites of acquisition. It improves efficiency by approximately 58 percent.
When you're comparing apples to apples, volume has the lesser variability, and as such provides more precise provider independent monitoring of abdominal aortic aneurysm behavior.
Edouard Aboian, MD
What advice would you give to young surgeons interested in pursuing similar research endeavors?
Don't get disappointed. And don't stop. Research is a complex process. Sometimes it gives you the results that you expected and sometimes the results are negative. You shouldn't feel discouraged. The second thing that I advise is consistency. Once you have a specific question in mind, you have to stay consistent until you have exhausted all available options to find the answer to that research question. Otherwise, you cannot succeed in research.
Lastly, can you discuss any ongoing or future research projects that build upon the findings of your recent study?
We are developing predictive clinical tools for a more sensitive analysis of aneurysm post intervention. That's number one, number two is behavior of non-ruptured non-repaired aneurysms based on the extensive database that we have at Yale. And then identifying medications or other risk factors that come not just with imaging but with a patient's clinical history and prior interventions that potentially can put them at risk for progression and rupture.