Early angiographic changes of intra-aneurysmal flow after flow-diverter stent treatment are not predictive of therapeutic success

Abstract

Background and purpose

Flow-diverter stents (FDS) are new devices for the endovascular treatment of intracranial aneurysms (IAs) promoting progressive aneurysmal thrombosis. To date, the delay of aneurysmal exclusion remains unclear. We evaluated the correlation between angiographic changes in the first 24 hours and 12-month occlusion in aneurysms treated with FDS.

Methods

We retrospectively analyzed the intra-aneurysmal flow by evaluating the in-flow and out-flow delays on preoperative, immediate postoperative, 24-hour and 12-month follow-up angiography. Dichotomy of in-flow and out-flow within the aneurysm was considered as the time of contrast filling and time of contrast washing relatively to the parent artery. The delay times were compared and correlated with the therapeutic success of FDS at 12 months of follow-up.

Results

Out of 14 treated IAs, in 13 consecutive patients, n = 10 (71%) aneurysms showed complete occlusion at 12 months. Between immediate postoperative and 24-hour control, 10 aneurysms (71%) demonstrated in-flow modification, with eight increasing, two decreasing and four having no change. There were no statistical differences in therapeutic success in relation to the different flow-related profiles of intra-aneurysmal flux.

Out-flow modifications were found in 11 aneurysms (79%) between immediate postoperative and 24-hour control, with five increasing, six decreasing and three having no change. Similar to the in-flow changes, there were no statistical differences in therapeutic success relative to the flow-related profiles.

Conclusions

Early angiographic changes after FDS placement are very frequent, but are not correlated with the 12-month technical success of flow-diversion techniques.

Introduction

Flow-diverter stents (FDS) are new and important tools for the treatment of intracranial aneurysms (IAs).^1,2 The mechanism of intra-aneurysmal thrombosis is complex and still not completely explained.^3,4 Some studies suggest that disruption of flow at the aneurysm neck provokes thrombosis into the aneurysm sac while preserving physiological flow in the parent vessel and adjacent branches.^4,5 In vitro and computational studies have shown that thrombosis within the aneurysm after placement of FDS seems to be a complex combination of hemodynamic, physiological and biological mechanisms.^6–8 These mechanisms are supposed to change along the time scale; however, the delay of aneurysmal thrombosis in not well known.^9,10

In the present study, we evaluate whether early intra-aneurysmal flow (IAF) modifications following FDS placement are related to complete aneurysm occlusion and can predict technical success of flow diversion strategies.

Material and methods

Patients

This was a retrospective analysis based on a prospective register (DIVERSION)¹¹ of all consecutive patients treated by FDS and followed up from December 2012 to December 2014 in our institution. Patients were enrolled both for treatment and retreatment procedure of re-canalized aneurysms. Patients who underwent a 12-month clinical and angiographic follow-up were included. A single operator performed the procedure (PC). Antiplatelet protocol was administrated three days preoperatively, with a loading dose of dual-anti-aggregation therapy clopidogrel (75 mg) and aspirin (160 mg), followed by three-month clopidogrel (75 mg) and 12-month aspirin 160 mg. An antiplatelet therapy sensibility test wasn’t performed in the routine. The study was approved by the local ethics committee.

Assessment of angiographic variables

All angiograms were performed on a flat detector monoplane angiographic system (ALLURA 400©, Philips, USA), using the same image rate protocol (two images/second), which permits flow delay evaluation without maximizing patient exposure to X-rays. Intra-arterial injections were performed with an automatized injector system, with a fixed flow-rate protocol of 4 cc/sec and volume 8 cc of the same contrast media (Iodixanol, Visipaque© 270 mg, GE Healthcare, UK). For all patients, angiogram in the antero-posterior, lateral and working projection of the treated aneurysm was performed preoperatively, immediate postoperatively (after FDS placement), 24 hours postoperatively and at 12 months. The images were reviewed and analyzed blindly and consensually by two experienced neuroradiologists (PEL and BG).

IAF was analyzed based on the concept that contrast media circulation within the aneurysm can depict the intra-aneurysmal blood flow, as previously reported.^7,9 IAF analysis was indirectly assessed using the digital subtraction angiography (DSA) images by studying the contrast media flow characteristics within the aneurysm and its parent artery. Thus, contrast media behavior was dichotomized in two time phases: in-flow delay and out-flow delay inside the aneurysm sac. These times were quantitatively assessed and quoted in seconds.

In-flow delay was defined as the time of contrast full opacification of the aneurysmal sac after a full opacification of the adjoined segment of the parent artery. Likewise, out-flow delay was defined as the time of contrast total disappearance (washing) in the aneurysmal sac after initiation of contrast washing from the adjoined segment of the parent artery. Those delays were standardized in the angiographic reviewing protocol as presented in Figure 1. This scale was determined by counting each image on the angiographic run with a difference in contrast enhancement between the aneurysm and direct parent vessel. In both time phases, delays were assessed per 0.5 second because all angiographies were performed at two images per second, allowing accurate and constant time measurements for the delay assessment.

Figure 1.

Assessment of early intra-aneurysmal flow modification at 24-hour angiographic control. (a) In-flow assessment by counting the images from full parent vessel opacification to full aneurysmal opacification. (b) Out-flow assessment by counting the images from the beginning of parent vessel contrast washing to complete aneurysmal contrast washing.

Angiographic changes were assessed comparing angiographic aspect before the procedure, immediately after FDS placement and at 24-hour angiographic control. Early angiographic changes were defined as modification of IAF between immediate postoperative and 24-hour control. On both in-flow and out-flow phases, variations of delay times between postoperative and 24-hour control were classified as increase, decrease or stable.

Therapeutic success of FDS placement was assessed at 12-month control angiograms, and considered as success for a complete (grade = 4) aneurysmal occlusion according to the angiographic Kamran occlusion scale.¹² All other grades of occlusion were considered as incomplete aneurysmal occlusion.

Endpoint and statistical analysis

The objective of this study was to analyze if early angiographic changes of IAF can predict therapeutic success. Therefore, we assessed subgroups of patients dichotomized by stability, and increase or decrease of both aneurysmal in-flow and out-flow delay times. The occurrence of therapeutic success on all the subgroups was analyzed using the Fisher’s exact test for countable data to determine if the early angiographic changes have a predictive value, and which trend (increase, decrease or stability) is more significant to foretell it at 24-hour angiographic control. All statistical analysis was conducted on SPPS© (SPPS software, USA).

Results

During the examination period, 13 consecutive patients harboring 14 aneurysms treated with FDS were included in the study.

Five of the aneurysms (36%) were located in the supraclinoid segment of the internal carotid artery, five were located in the carotid ophthalmic segment (36%), three were carotid-cavernous segments (21%) and one was a vertebra-basilar location (7%). The mean transverse diameter was 8.84 mm (minimum 1.5, maximum 20), and the average neck size was 7.58 mm (minimum 1.2, maximum 20). No aneurysm was treated during the acute phase of subarachnoid hemorrhage.

Almost all patients treated by FDS in our center during this period were included since only one patient did not complete the 12-month angiographic follow-up. Fourteen FDS embolization devices were used (PIPELINE© (Covidien, USA) n = 12 and SILK© (Balt, France) n = 2). At 12-month angiographic control, 10 aneurysms presented with complete occlusion (Kamran grade = 4) and four aneurysms were partially occluded (Kamran grade = 0–3).¹²

In-flow modifications results are shown in Table 1. Ten aneurysms (71%) presented in-flow modification at 24 hours. In these aneurysms, the rate of 12-month complete occlusion was 80% (n = 8), and was not significantly different from aneurysms that did not present early angiographic change of in-flow, 75% (n = 3). Subgroup analysis shows that neither increase nor decrease of the in-flow delay time correlated with a significant modification of the 12-month complete occlusion rate and thus the technical success of the procedure (p = 0.7538).

Table 1.

Twenty-four-hour angiographic in-flow modification correlated to 12-month complete occlusion of intracranial aneurysms (IAs).

Sub-groups of aneurysm	Postoperative to 24-hours in-flow modification	Rate of 12-month complete occlusion on IAs with 24-hour in-flow modification
Aneurysms with in-flow modifications on 24 hours/total	10/14	80% (8/10)
Increased in-flow time	8/14	75% (6/8)
Decreased in-flow time	2/14	50% (1/2)
Stable in-flow time	4/15	75% (3/4)

The same methodology was applied on out-flow modification as shown in Table 2. Eleven aneurysms (79%) presented out-flow modification at day 1. On these aneurysms, the rate of 12-month complete occlusion was 73% (n = 8), and was not statistically different from the aneurysms that did not present early angiographic change in out-flow, 100% (n = 3). As demonstrated for in-flow modification, none of the subgroup were significantly associated with a 12-month complete occlusion (p = 0.6043).

Table 2.

Twenty-four-hour angiographic out-flow modification correlated to 12-month complete occlusion of intracranial aneurysms (IAs).

Sub-groups of aneurysm	Postoperative to 24-hour out-flow modification	Rate of 12-month complete occlusion on IAs with 24 hours out-flow modification
Aneurysms with out-flow modifications at 24 hours/total	11/14	73% (8/11)
Increased out-flow time	5/14	80% (4/5)
Decreased out-flow time	6/14	67% (4/6)
Stable out-flow time	3/14	100% (3/3)

Discussion

In this small series, we describe a new angiographic approach for understanding the hemodynamics inside the aneurysm following flow-diverter deployment. This approach advises the delay of aneurysm thrombosis following FDS placement. The present study is the first, to our knowledge, to report early angiographic analysis at 24 hours postoperatively and its correlation to 12-month complete occlusion.^1,5,13 Because control angiogram at 24 hours is performed under local anesthesia and with the patient on baseline antiplatelet therapy without any additive heparin therapy, it is interesting to consider it as a representative state of the physiological hemodynamic conditions, which leads to aneurysmal thrombosis after FDS placement. Under these conditions we report that early angiographic changes in IAF after FDS placement are very common. In fact, 24-hour modification of in-flow and out-flow were observed in 71% and 79% of aneurysms, respectively. Similar findings were previously experimentally observed in animal studies.^7,8,14 Arising from these observations, a supposed mechanism of aneurysm thrombosis after FDS placement was suggested, as increasing blood stagnation inside the aneurysm sac promotes the thrombosis. Some authors suggested that early blood stagnation in the aneurysm is predictive for therapeutic success,^1,12 although immediate angiographic exclusion of the IAs at the end of the therapeutic procedure is observed in only a minority of patients (8%–21%).^2,13 Of note, in our study those modifications consist not only of increasing blood stagnation but also a decrease in the intra-aneurysmal flux. We report that 14% of treated aneurysms present an in-flow time decrease and 42% out-flow time decrease. In these situations the blood stagnation within the IA is supposed to decrease. Thus, in a significant proportion, FDS seems to provoke the opposite effect than expected. One hypothesis is that the phenomenon of intra-aneurysmal thrombosis after the placement of an FDS is influenced not only by flow stagnation within the aneurysm sac.^7,15 Regardless of the characteristics of the IAF, these early modifications are not associated with complete aneurysm occlusion.

To our knowledge, there are no perioperative data that can predict the complete occlusion and the success of flow diversion treatment. The rates of complete occlusion are often variable, ranging from 55% to 95%.^1,13 In the meta-analysis by Brinjikji et al., including 1451 patients with 1654 aneurysms, the rate of complete aneurysm occlusion was 76% at six-month follow-up irrespective of aneurysm size, a similar rate compared to the 71.4% of our study.¹ We found that even though the early angiographic changes are frequent and variable, they do not correlate with therapeutic success. In both in-flow and out-flow analysis, none of the three subgroups of aneurysm considered in the study (stability, increase and decrease) were found to be statistically associated with 12-month complete occlusion. In light of these results, early angiographic change should not be highly considered as a determinant of therapeutic success and IAs thrombosis.

The limits of this study include the methodology of assessment of the early angiographic changes. Indeed, in-flow and out-flow delays were indirectly assessed by studying the contrast media flow characteristics within the aneurysm. This method was previously described¹⁶ but with a more precise “aneurysmal washout curve” technique on computer-assisted post-processing of DSA at the rate of 10 frames per seconds. In our angiographic protocol the too-low frame rate (two frames/second) wasn’t compatible with such a more precise model.¹⁶ But the main limitation of our study is the small sample size. In light of these results and using the same methodology, we manage to verify this hypothesis in all the aneurysms included in the DIVERSION registry.

Conclusion

Early IAF modifications are frequent and variable after flow-diverter placement. These phenomena are not correlated with complete aneurysm occlusion at 12-month angiographic follow-up, suggesting that the variability of thrombosis process delay and its lack of predictability are possibly because it is a more acute and less progressive phenomenon. More studies are required to further substantiate our understanding of the mechanism and timing of aneurysm occlusion after FDS deployment.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.