Parametric Analysis of Intraprocedural DSA in GAE

Quantifying Change in Perfusion after Genicular Artery Embolization with Parametric Analysis of Intraprocedural Digital Subtraction Angiograms

Clinical question

After Genicular Artery Embolization, will perfusion be decreased in hyperemic target vessels but preserved in parent vessels?

Take away point

Objective parametric measures showed patent parent vessel with unchanged flow after embolization of distal hyperemic target vessels, validating the pruning technique in genicular artery embolization and demonstrating utility in quantification of intraoperative perfusion.

Reference

Quantifying Change in Perfusion after Genicular Artery Embolization with Parametric Analysis of Intraprocedural Digital Subtraction Angiograms. Badar W M.D. et al. Journal of Vascular and Interventional Radiology, Volume 34, Issue 12, 2189-2195.

Click here for abstract

Study design

Single center, single arm, retrospective study with 12 patients

Funding Source

No reported funding.

Setting

Academic hospital, University of Chicago, Illinois, United States

Figures

Figure 1:(a) Pre–genicular artery embolization selective digital subtraction angiography run of the right superior medial genicular artery, with region of interest (ROI) 1 (red) corresponding to the parent vessel (PV) and ROI 2 (green) corresponding to the hyperemic target vessel (TV). (b) Post-GAE selective digital subtraction angiography run of the same vessel, with ROI 3 (red) corresponding to the PV and ROI 4 (green) corresponding to the “pruned” TV.



Figure 2: (a) Time–intensity curves for the parent vessel before genicular artery embolization (GAE) (red) and after GAE (blue). (b) Time–intensity curves for the target vessel before GAE (red) and after GAE (blue). The deep spikes in the post curves represented artifact from patient movement at the end of the digital subtraction angiography acquisition when the regions of interest were shifted momentarily to empty or subtracted areas where there was no contrast density.

Summary

Geniculate Artery Embolization has shown promising results for the treatment of medically refractory knee osteoarthritis. Instead of attempting near-complete stasis, genicular artery embolization aims to achieve “pruning” where distal hyperemic vessels are embolized, but parent vessel flow is preserved. Parametric analysis of digital subtraction angiography is an established post processing method to quantitatively evaluate tissue perfusion changes and determine angiographic endpoints. But it has never been applied in the context of genicular artery embolization, specifically to validate the pruning method and to correlate with clinical success.

The authors performed this study in a single academic center with 12 patients who were diagnosed with mild to moderate osteoarthritis, age > 40, and pain refractory to 6 months of conservative management that included physical therapy and/or non-steroidal anti-inflammatory medication.

The genicular artery embolization procedure was performed through ipsilateral femoral access with a DSA run to identify the genicular arteries of interest, evaluate hyperemic vasculature, and ensure the resolution of hyperemia after administration of the embolic agent. genicular artery embolization was performed with 200-µm microspheres with a 1:3 dilution in contrast medium, in 0.1ml aliquoted to near stasis. Each patient received a call 1-day post-procedure and follow up visits at 1 week and 1, 3, 6, and 12 months.

For each vessel treated, regions of interest were marked within the lumen of the parent vessels and hyperemic vessels on digital subtraction angiography images. Perfusion parameters were computed for each region of interest before and after the embolization. These included area under the curve, peak intensity, and time to arrival, a measure of total blood volume, of highest blood concentration, and of blood velocity, respectively. 36 total vessels were identified with 2 vessels excluded due to motion artifacts.

Paired student t-tests were used to compare the area under the curve, peak intensity, and time to arrival. In the hyperemic vessels, there were significantly lower area under the curve and peak intensity, as well as significantly longer time to arrival on the post-embolization images compared to the pre-embolization images. There was no significant difference in area under the curve, peak intensity, or time to arrival in the parent vessels. These results provided quantitative support for the “pruning method” for genicular artery embolization. Skin changes were the most common adverse events and all instances resolved within 3 weeks. Each incidence occurred with an embolic volume greater than 3.0 ml. A Pearson correlation coefficient test showed no correlation between the volume of embolic agent administered and the parent vessel diameter. Unpaired student T-tests were used to show that higher volumes of embolic administered were associated with increased non-target embolization but no further reduction in hyperemia. Hence, overtreating vessels provides no additional benefit but increases the risk of non-target embolization. Baseline clinical symptoms were assessed with the Western-Ontario and McMaster Universities Arthritis index, showing significant reductions in total and pain scores at 1 and 3 months postoperatively.

Commentary

The authors in this study provided quantitative evidence of the pruning method often described in genicular artery embolization. These findings have implications in minimizing non-target embolization while ensuring treatment efficacy. The results also support the utility of intraprocedural quantitative perfusion parametric imaging to provide objective measures of embolization endpoints. While these results are encouraging, the study acknowledges that future studies with a larger sample size, knee immobilization techniques to reduce motion artifact, and use of different types of embolic agents would be helpful. Future research in other areas of interventional radiology may learn from this study and experiment the use of intraprocedural parametric imaging for objective procedural endpoint measures.

Post Author

Anthony M. Camargo, BA
MD candidate, Class of 2025
University of Massachusetts Chan Medical School
@anthonymcamargo