Coil Embolization for Unruptured Intracranial Aneurysms at the Dawn of Stent Era: Results of the Japanese Registry of Neuroendovascular Therapy (JR-NET) 3.

Endosaccular coiling is recognized as a feasible method for treating unruptured intracranial aneurysms (UIAs). We retrospectively reviewed cases of UIAs treated by coiling in the Japanese Registry of Neuroendovascular Therapy (JR-NET) 3, a nationwide survey of NET between 2010 and 2014, the beginning period of intracranial stents in Japan. Data were extracted for 6844 UIAs (6619 procedures) from 40,169 registered records of all NETs in the JR-NET 3 databases. The features of the aneurysms and procedures, immediate radiographic findings, procedure-related complications, and clinical outcomes at 30 days after the procedures were assessed. Of 6844 UIAs, 81.8% were located in the anterior circulation. The mean patient age was 61.3 years (72.4% females). Compared with the preceding JR-NET 1 and 2, there were significant increases (P <0.05) in the rates of the following in JR-NET 3: wide-necked and small UIAs measuring <10 mm (from 56.4% to 58.8%), adjunctive techniques (54.8% to 71.8%), and stent usage (1.1% to 22.1%). Both pre- (85.6% to 96.7%) and post-procedural (84.0% to 94.6%) antiplatelet therapy were more frequently administered in JR-NET 3. Although procedure-related complication rates did not differ between the two groups, ischemic complication rates increased from 4.6% to 5.9%, leading to an increase in the 30-day morbidity (modified Rankin Scale >2) from 2.1% to 2.8%. In conclusion, introduction of neck-bridge stent was associated with an increase in cases of wide-necked aneurysms. However, the ischemic complication rate increased despite the greater use of periprocedural antiplatelet therapy.

Introduction

The management of asymptomatic, unruptured intracranial aneurysms (UIAs) has been controversial. Several reports regarding the natural history of UIAs showed that factors such as size, location, can be the predictors of their rupture including those among Japanese people who are believed to be at higher risk.[1–3)]

Nowadays, endosaccular coiling of intracranial aneurysms is prevailing worldwide with its less invasiveness, lower incidence of complications compared with clipping[4,5)] and long-term durability.[6,7)] However, the complication rate and the possibility of retreatment due to recanalization have each been reported at 5–10%.[6,8–10)] Therefore, the assessment of current status of coiling is important for the management of patients with UIAs to compare the benefits and risks of this modality with surgical clipping or conservative management.

The Japanese Registry of Neuroendovascular Therapy (JR-NET) study group endorsed by Japanese Society for Neuroendovascular Therapy (JSNET) have conducted retrospective studies (JR-NET 1&2) to clarify the general status of neuroendovascular therapy delivered by JSNET-certified physicians, and to standardize endovascular procedures and to assist with education planning for Japanese neurointerventionists based on outcomes. This report details the results of JR-NET 3, in which clinical and procedural data were collected retrospectively from January 2010 through December 2014 for all endovascular procedures to further assess and guarantee the quality of NET in Japan. The primary end point was the 30-day clinical outcome [modified Rankin Scale (mRS)] and secondary end points comprised technical success, adverse events arising within 30 days.[11)]

Here, we collected a considerable amount of clinical data on NET for UIAs through the JR-NET 3 investigations and evaluated the outcomes of NET, especially endosaccular coiling for patients with asymptomatic UIAs in Japan.

Materials and Methods

JR-NET 3 protocols

The JR-NET 3 was conducted between 2010 and 2014 as the third nationwide survey of neuroendovascular treatment in Japan, targeting all endovascular procedures during the study period by JSNET board-certified physicians, who were instructed to register their endovascular procedures in the JR-NET 3 database. Patient-identifying information was anonymized, retrospectively registered via website (https://jr-net.tri-kobe.net/jr-net, and collected at the Translational Research Informatics Center (http://www.tri-kobe.org/). The Institutional Review Board (IRB) at the participating centers approved the use of the patient data for research (local IRB approval number: M25-024-2). A total of 40 169 endovascular procedures were registered for the JR-NET 3.[11)] For the current report, we extracted data for all cases of elective endosaccular coiling of UIAs. Cases with incomplete data were excluded.

Primary and secondary endpoints

The primary endpoint was activities of daily living, as indicated by the mRS scores. The secondary endpoints were the technical success of procedures and major adverse events within 30 days of the procedures. Adverse events were classified as minor and major when mRS scores deteriorated by 1 and ≥2 points, respectively.

Statistical methods

Variables are presented as mean ± standard deviation (SD), count, or percentage, as appropriate. The χ2-test was used to compare the categorical variables. The Wilcoxon signed-rank test was used to compare the ordinal variables while Student’s t-test was used to compare continuous variables. P-values <0.05 were considered statistically significant. Highly significant data with P-values <0.001 are indicated in the text. We used the JMP 10 software (SAS Institute Inc., Cary, NC, USA) to perform all the analyses.

Results

Patients characteristics and multiplicity of the procedure

In total, 6844 cases were selected for analysis, among which 6619 procedures were performed. Of these, 1811 (27.4%) were for men and 4808 (72.6%) were for women. The flowchart for data selection is shown in Fig. 1. The mean patient age was 61.3 ± 11.6 years (range 17–89 years), and 98.9% were independent (mRS 0–2) at the time of the procedure. The numbers of treated UIAs were one (96.8%) of 6404, two of 205, and three of 10 procedures.

Fig. 1.

Figure shows the flowchart of data extraction from the JR-NET 3. The pre-procedural status included sex, age, date of treatment, pre-procedural modified Rankin Scale (mRS), antithrombotic therapy, and aneurysm characteristics. The intra-procedural status included the techniques and antithrombotic therapy. The post-procedural status included the radiographic outcome, procedure-related adverse events and mRS 30 days after the procedures. JR-NET: Japanese Registry of Neuroendovascular Therapy, UIA: unruptured intracranial aneurysm.

Aneurysm characteristics

The features of the aneurysms are summarized in Table 1. Most UIAs were located in the anterior circulation (n = 5599; 81.8%), while the remainder were located in the posterior circulation (n = 1245; 18.2%). Specifically, 33.8% were in the paraclinoid region, followed by the internal carotid artery–posterior communicating artery (14.7%), anterior communicating artery (13.9%), bifurcation of the basilar artery (9.0%), and middle cerebral artery (5.2%). As for the maximal diameters, 2318 (33.9%) were 5–6 mm and 1436 (21.0%) were 7–9 mm. Regarding the minimal diameters, 2125 (31.0%) were 3–4 mm and 124 (1.8%) were <3 mm. Notably, 3548 (58.8%) UIAs with diameters <10 mm had wide necks (neck diameter >4 mm or dome-to-neck ratio <1.5).[12)]

Table 1

Characteristics of unruptured intracranial aneurysms treated by endosaccular coiling, based on data in JR-NET 3

	n (%)
Location
Anterior circulation	5599 (81.8)
ICA-cav	295 (4.3)
Paraclinoid	2316 (33.8)
ICA-PCom	1006 (14.7)
ICA-AChA	233 (3.4)
ICA-bif	191 (2.8)
MCA	354 (5.2)
A1	76 (1.1)
ACoA	952 (13.9)
DACA	130 (1.9)
Other (AC)	46 (0.7)
Posterior circulation	1245 (18.2)
VA	154 (2.3)
BA-trunk	64 (0.9)
BA-SCA	208 (3.0)
BA-bif	617 (9.0)
PCA	56 (0.8)
VA-PICA	131 (1.9)
Other (PC)	15 (0.2)
Size	(r, mm)*
<3	124 (1.8)
3 to <5	2125 (31.0)
5 to <7	2318 (33.9)
7 to <10	1436 (21.0)
10 to <25	827 (12.1)
≥25	14 (0.2)
Appearance (% of UIA <10 mm)	(n = 6003)
Narrow neck†	2475 (41.2)
Wide neck‡	3528 (58.8)

r: maximal diameter,

Neck: ≤4 mm and dome-to-neck (D/N) ratio ≥1.5,

Neck: >4 mm or D/N ratio <1.5. ACoA: anterior communicating artery, A1: anterior cerebral artery proximal to anterior communicating artery, BA: basilar artery, BA-bif: basilar bifurcation, BA-SCA: junction of basilar artery and superior cerebellar artery, BA-trunk: trunk of basilar artery, DACA: anterior cerebral artery distal to anterior communicating artery, IC-AChA: anterior choroidal artery, ICA-bif: bifurcation of internal carotid artery, ICA-cav: cavernous segment of ICA, ICA-Pcom: posterior communicating artery, JR-NET 3: Japanese Registry of Neuroendovascular Therapy 3, MCA: middle cerebral artery, Other (AC): other locations in anterior circulation, Other (PC): other locations in posterior circulation, PCA: posterior cerebral artery, UIA: unruptured intracranial aneurysm, VA: vertebral artery.

Endovascular modalities

The techniques used for endosaccular coiling are shown in Table 2. Simple technique by which a single catheter and coils were used were performed in 1933 cases (28.2%), while adjunctive techniques were used in the remaining 4911 (71.8%) cases. Among them, balloon assist, double catheter, stent-assisted, stent monotherapy techniques were used in 2806 (41.0%), 439 (6.4%), 1253 (18.3%), 23 (0.3%), respectively. A combination of techniques was used in 391 cases (5.7%). Specifically, neck bridge stents were used in 1512 cases (22.1%).

Table 2

The techniques and periprocedural antithrombotic therapy used according to data in JR-NET 3

	n (%)
Technique (pa, 6844 cases)
Simple	1933 (28.2)
Adjunctive	4911 (71.8)
DCT	439 (6.4)
BAT	2806 (41.0)
SAC	1253 (18.3)
DCT + BAT	129 (1.9)
DCT + SAC	84 (1.2)
BAT + SAC	146 (2.1)
DCT + BAT + SAC	15 (0.2)
Stenting only	23 (0.3)
Other	17 (0.3)
Antithrombotic therapy (pp, 6619 cases)
Anticoagulation
Intraprocedural systemic heparinization	6457 (97.6)
Continuous anticoagulation	3995 (60.4)
heparin	949 (14.3)
argatroban	2730 (41.2)
heparin + argatroban	107 (1.6)
Antiplatelet therapy
Preprocedural	6398 (96.7)
SAPT	1647 (24.9)
DAPT	4405 (66.6)
TAPT or more	346 (5.2)
Postprocedural	6262 (94.6)
SAPT	1738 (26.3)
DAPT	3872 (58.5)
TAPT or more	652 (9.9)

With mRS deterioration. BAT: balloon assist technique, DAPT: dual antiplatelet therapy, DCT: doble catheter technique, EVT: endovascular therapy, mRS: modified Rankin scale, pa: per aneurysm, pp: per procedure, RA: residual aneurysm, RN: residual neck, SAC: stent-assisted coiling, SAPT: single antiplatelet therapy, TAPT: triple antiplatelet therapy, uPAO: unpredicted parent artery occlusion.

Periprocedural antithrombotic management

Table 2 also shows the antithrombotic therapy. Systemic heparinization was used in 6458 (97.6%) procedures. Continuous anticoagulation therapy was applied in 3786 procedures (55.3%). Argatroban, a direct thrombin inhibitor, was preferentially used (41.2%).

Notably, pre- and post-procedural antiplatelet therapy (APT) was executed in 6398 (96.6%) and 6263 (94.6%), respectively.

Mode of the APT in the pre-procedural period were single in 1647 (24.9%), dual in 4405 (66.5%), and triple or more in 346 (5.2%); the corresponding figures in the 30-day post-procedural period were 1739 (26.3%), 3872 (58.5%), and 652 (9.8%), respectively.

Feasibility and immediate angiographic outcome

Failure of endosaccular coiling was noted in for 78 (1.1%) aneurysms (Table 3) at rates of 5.4%, 1.2%, 0.8%, 0.7%, 1.4%, and 0% for aneurysms with diameters of <3, 3–4, 5–7, 7–9, 10–25, and >25 mm, respectively (Table 4).

Table 3

Outcomes of coiling for unruptured intracranial aneurysms

	n	(%)
Feasibility (pa, 6844 cases)
Success	6766	98.9
Failure	78	1.1
Anatomic outcome (per successfully treated UIA, 6766 cases)
Complete occlusion	3147	46.5
Residual neck	2324	34.3
Residual aneurysm	1288	19.0
Unpredicted parent artery occlusion	7	0.1
Adverse events (pp, 6619 cases)
Procedure-related complications	664 (154)*	10.0 (2.3)*
Hemorrhagic	143 (46)*	2.2 (0.7)*
Intraprocedural aneurysmal rupture	86 (20)*	1.3 (0.3)*
Aneurysmal rupture in post-treatment period	5 (3)*	0.1 (0.05)*
Ischemic	389 (103)*	5.9 (1.6)*
Puncture site	89	1.3
Other	67	1.0
mRS 30 days after EVT (pp, 6619 cases)
0	5988	90.5
1	337	5.1
2	153	2.3
3	58	0.9
4	45	0.7
5	23	0.3
6	15	0.2
Clinical outcome (pp = 6619)
30-day morbidity	186	2.8
30-day mortality	15	0.2

With mRS deterioration. EVT: endovascular therapy, mRS: modified Rankin scale, pa: per aneurysm, pp: per procedure.

Table 4

Failure rate, anatomic outcome, and complication rate by the aneurysm location and maximal radius (6404 cases)

	Location of the aneurysms (n, %)		P-value	Maximal radius (n, %)						P-value	Overall (%)
	Anterior Circ.	Posterior Circ.		<3	3 to <5	5 to <7	7 to <10	10 to <25	≥25
Number	5216	1188		92	1943	2174	1379	802	14		6404
Failure rate	59 (1.1)	8 (0.7)	0.20	5 (5.4)	24 (1.2)	18 (0.8)	9 (0.7)	11 (1.4)	0	0.008	67 (1.0)
CO	2452 (47.0)	498 (41.9)	0.002	54 (58.7)	1030 (53.0)	1018 (46.8)	573 (41.6)	273 (34.0)	2 (14.3)	<.001	2950 (46.1)
RA	977 (18.7)	232 (19.5)	0.54	10 (10.9)	291 (15.0)	399 (18.4)	286 (20.7)	215 (26.8)	8 (57.1)	<.001	1209 (18.9)
Procedure-related complications	465 (8.9)	163 (13.7)	<.001	10 (10.9)	151 (7.8)	182 (8.4)	155 (11.2)	129 (16.1)	1 (7.1)	<.001	628 (9.8)
Hemorrhagic complications	101 (2.9)	31 (2.6)	0.142	5 (5.4)	45 (2.3)	42 (1.9)	21 (1.5)	19 (2.4)	0 (0)	0.146	132 (2.1)
Ischemic complications	264 (5.1)	105 (8.8)	<.001	4 (4.3)	75 (3.9)	105 (4.8)	98 (7.1)	86 (10.7)	1 (7.1)	<.001	369 (5.8)

circ.: circulation, CO: complete occlusion, RA: residual aneurysm.

The failure rate decreased significantly with increasing aneurysmal size. There was no significant difference in failure rate between aneurysms in the anterior (1.1%) and posterior (0.7%) circulations. The immediate radiographic outcomes of 6766 successfully treated aneurysms demonstrated that 3147 (46.5%) were completely occluded, 2324 (34.3%) had residual necks, 1288 (19.0%) had residual aneurysmal domes, and 7 (0.1%) resulted in unpredicted parent artery occlusion (Table 3).

Adverse events related to endosaccular coiling

Procedure-related adverse events occurred in 664 patients (10.0%), of whom 23.2% (n = 154; 2.3% of total) had increases in their 30-day mRS scores (Table 3). Intracranial hemorrhage and ischemia were noted in 143 (2.2%) and 389 (5.9%), respectively; there were 89 intra-procedural aneurysmal ruptures, or 1.3% per procedure. The anterior communicating artery was the most frequent site of these ruptures (30 cases or 3.0% of aneurysms in this region). Delayed aneurysmal rupture occurred in five cases within 30 days (Table 3).

Ischemic complication rates occurred at a lower rate in smaller aneurysms (3–7 mm) and at a higher rate in larger aneurysms (7–25 mm); however, there was no statically significant correlation between hemorrhagic complications and aneurysmal size (Table 4).

Concerning the location, aneurysms in the posterior circulation (8.8%) were more prone to ischemic complications than those in the anterior circulation (5.1%), which resulted in a higher incidence of procedure-related complications (13.7% versus 8.9%, respectively; Table 4).

30-Day clinical outcomes

The 30-day morbidity and mortality rates were 2.8% and 0.2% among 6404 procedures, respectively (Table 3). Of the 15 deaths within 30 days of treatment, 12 were associated with hemorrhagic complications (11 aneurysmal ruptures and one vessel rupture). A total of 6478 patients (97.9%) remained independent (mRS score 0–2).

Impact of stent-assisted coiling in JR-NET 3

Table 5 shows the comparison between stented and non-stented cases in this study.

Table 5

Comparison of stented and non-stented coiling procedure based on data in JR-NET 3 (pa, 6404 cases)

	Stents used (n = 1462)	Stents not used (n = 4942)	P-value
Age	61.5 ± 11.6	61.3 ± 11.6	0.500
Anterior circulation*	1058 (72.4)	4158 (84.1)	<0.001
Posterior circulation*	404 (27.6)	784 (15.9)	<0.001
Maximum radius >7 mm	906 (62.0)	1289 (26.1)	<0.001
Unfavorable anatomy†	1373 (93.9)	2772 (56.1)	<0.001
DAPT or more	1368 (93.6)	3211 (65.0)	<0.001
Complete obliteration	581 (39.7)	2369 (47.9)	<0.001
Residual aneurysm	391 (26.7)	818 (16.6)	<0.001
Procedure-related complications	204 (14.0)	424 (8.6)	<0.001
Hemorrhagic complications	31 (2.1)	101 (2.0)	0.834
Ischemic complications	130 (8.9)	239 (4.8)	<0.001
30-day morbidity	88 (6.0)	94 (1.9)	<0.001
30-day mortality	4 (0.3)	9 (0.2)	0.510

Significantly different between anterior circulation and posterior circulation (P <0.001),

Aneurysms with wide neck and/or maximal diameter >10 mm. DAPT: dual antiplatelet therapy, pa: per aneurysm.

Neck bridge stents were used in 1462 cases, but were used more frequently in the posterior circulation (404 of 1188 cases; 34.0%) than in the anterior circulation (1058 of 5116 cases; 20.3%) (P <0.001). Also, the proportion of aneurysms >7 mm and aneurysms with unfavorable anatomy (i.e., either wide-necked or >10 mm) in terms of initial occlusion status and recanalization for non-stented coiling[13)] were larger in the stented cases than those in non-stented cases (62% versus 26.1%; P <0.001, 93.9% versus 56.0%; P <0.001, respectively).

The rate of complete obliteration was lower in the stented cases (39.7%) than in the non-stented cases (47.9%) (P <0.001). Compared with patients who treated without stents, those who treated with stents had higher rates of procedure-related (14.0% versus 8.6%, P <0.001) and ischemic (8.9% versus 4.8%, P <0.001) complications. Although the overall rates of hemorrhagic complications were similar between the two groups (2.1% versus 2.0%, P = 0.834), there was a significant difference in 30-day morbidity between the stented (6.0%) and non-stented (1.9%) cases (P <0.001).

Comparison between JR-NET 3 and JR-NET 1&2

Table 6 shows the comparison of data from JR-NET 3 with that from JR-NET 1&2,[12)] which covered the data of UIAs treated by endosaccular coiling between 2005 and 2009.

Table 6

Comparison of the data from JR-NET 1&2 and JR-NET 3

	JR-NET 1, 2 (2005–2009)	JR-NET 3 (2010–2014)	P-value
UIAs (n)	4767	6844
Procedures (n)	4573	6619
Age (y, mean ± SD, range)	60.6 ± 11.1 (6–93)	61.3 ± 11.6 (17–89)	0.001
Female (n, %)	3311 (72.4)	4808 (72.6)	0.796
Anterior circulation	3814 (80.0)	5614 (82.0)	0.006
Size	(n = 4767)	(n = 6844)
<3	119 (2.5)	124 (1.8)	0.012
3 to <5	1569 (32.9)	2125 (31.0)	0.035
5 to <10	2476 (51.9)	3754 (54.9)	0.002
≥10	603 (12.6)	841 (12.3)	0.568
Appearance (% of UIA <10 mm)	n = 4164	n = 6003
Narrow neck	1816 (43.6)	2475 (41.2)	0.017
Wide neck	2348 (56.4)	3528 (58.8)
Technique (pa)	n = 4767	n = 6844
Adjunctive	2612 (54.8)	4911 (71.8)	<0.001
Use of stent	51 (1.1)	1512 (22.1)	<0.001
Antithrombotic regimen (pp)	n = 4573	n = 6619
PRE antiplatelet therapy	3914 (85.6)	6398 (96.7)	<0.001
INTRA systemic heparinization	4488 (98.1)	6457 (97.6)	0.043
CONT anticoagulation	3108 (68.0)	3995 (60.4)	<0.001
POST antiplatelet therapy	3841 (84.0)	6262 (94.6)	<0.001
Feasibility (pa)	n = 4767	n = 6844
Technical success	4665 (97.9)	6766 (98.9)	<0.001
Anatomic outcome (st)	n = 4665	n = 6766
CO	2690 (57.7)	3147 (46.5)	<0.001
RN	1490 (31.9)	2324 (34.3)	0.007
RA	468 (10.0)	1288 (19.0)	<0.001
uPAO	17 (0.4)	7 (0.1)	0.003
Adverse events (pp)	n = 4573	n = 6619
Procedure-related complications	417 (9.1)	664 (10.0)	0.111
Hemorrhagic complications	90 (2.0)	143 (2.2)	0.500
Ischemic complications	210 (4.6)	389 (5.9)	0.003
mRS 30 days after EVT (pp)	n = 4573	n = 6619
0 to 2	4481 (98.0)	6478 (97.9)	0.686
3 to 6	92 (2.0)	141 (2.1)
30-day morbidity (pp)	97 (2.1)	186 (2.8)	0.024
30-day mortality (pp)	14 (0.3)	13 (0.2)	0.247

CO: complete occlusion, CONT: continuous, EVT: endovascular therapy, INTRA: intraprocedural, JR-NET: Japanese Registry of Neuroendovascular Therapy, mRS: modified Rankin scale, pa: per aneurysm, POST: postprocedural, pp: per procedure, PRE: preprocedural, RA: residual aneurysm, RN: residual neck, UIA: unruptured intracranial aneurysms, uPAO: unpredicted parent artery occlusion.

The number of patients with UIAs in JR-NET 3 (n = 6844) was higher than in JR-NET 1&2 (n = 4767). The mean age was also older (P <0.001), and the ratio of anterior circulation aneurysms was higher (P = 0.006), in JR-NET 3. The rate of involvement of anterior communicating arteries increased from 12.3% in JR-NET 1&2 to 13.9% in JR-NET 3. As to the size and appearance of the treated aneurysms, there was a decrease in aneurysms smaller than 5 mm from JR-NET 1&2 to JR-NET 3 (35.4–32.8%, P = 0.004) while the rate of wide-necked, small aneurysms (<10 mm) were higher in JR-NET 3 than that in JR-NET 1&2 (58.8% versus 56.4%, P = 0.017). Angiographically, the rate of complete obliteration was lower in JR-NET 3 (46.5%) compared with JR-NET 1&2 (57.7%) (P <0.001).

The use of adjunctive technique was more frequent in JR-NET 3 than in the previous two cohorts (71.8% versus 54.8%, P <0.001), and the use of stent markedly increased from JR-NET 1&2 to JR-NET 3 (1.1–22.1%, P <0.001).

Regarding antithrombotic therapy, the rates of both pre- and post-procedural antiplatelet therapy increased from JR-NET 1&2 (85.6% and 84.0%, respectively) to JR-NET 3 (96.7% and 94.6%, respectively), whereas the rate of continuous heparinization decreased (68.0% in JR-NET 1&2 to 60.4% in JR-NET 3).

As to the adverse events and 30-day clinical outcomes, ischemic complication rates were higher in JR-NET 3 (5.9%) than in JR-NET 1&2 (4.6%) (P <0.001), but there was no significant difference in the procedure-related complication rate (10.0% and 9.1%, respectively). The 30-day morbidity rate was also higher in JR-NET 3 (2.8%) than in JR-NET 1&2 (2.1%) (P = 0.024), but there was no significant difference in the independence rates (mRS 0–2) and 30-day mortality rates between JR-NET 3 and JR-NET 1&2.

Discussion

We have reported the current status of endosaccular coiling for UIAs in Japan between 2010 and 2014. The most noteworthy event in this study period was that reimbursement was introduced for two neck bridge stents—the Enterprise VRD (Cerenovus, Johnson and Johnson, New Brunswick, NJ, USA) in 2010 and the Neuroform EZ (Stryker Neurovascular, Fremont, CA, USA) in 2012—for treating intracranial aneurysms. Flow diverters were used in some cases, but these were not reimbursed during the study period. Thus, we can consider the main theme of this study as being to verify the feasibility and safety of endovascular treatment for UIAs after the two neck bridge stents were introduced.

The increase in the rate of treated aneurysms in the anterior circulation was considered to be multifactorial, with stents potentially of benefit for aneurysms of both the anterior and posterior circulations.[14)] Other factors, such as the increased number of neuroendovascular physicians and the increased awareness of coil embolization in Japan, probably contributed to this result. The increases in the number of wide-necked aneurysms and in the use of adjunctive techniques likely resulted from the introduction of stents, which were used in 22.1% of all procedures in the JR-NET 3 period.

The size of the treated aneurysms significantly changed, with the proportion of aneurysms measuring <3 and 3–5 mm decreasing from JR-NET 1&2 to JR-NET 3 (2.5–1.8% and 32.9–31.0%, respectively). This followed the publication of the UCAS Japan in 2012,[2)] which reported on the natural history of UIAs in Japan. That study reported that the annual rupture rate of UIAs was 0.95%, with aneurysms measuring >7 mm having the greatest risk of rupture. At the same time, the Japanese Guidelines for the Management of Stroke recommended continuing to treat UIAs measuring 5–7 mm.[15)] Thus, fewer UIAs measuring <5 mm and more UIAs measuring 5–7 mm were treated.

Concerning the antithrombotic therapy, periprocedural APT became more frequent in JR-NET 3 period. Several reports about the efficacy of antiplatelet therapy[16–18)] including its use with stents,[19,20)] likely accelerated this increased usage. However, it should be noted that the use of heparinization decreased both intra- and post-procedurally. Although the reduction in post-procedural anticoagulation can be explained by the lack of verification in the literature, no clear explanation can be found to account for the reduction in intra-procedural anticoagulation.

The technical success rate increased marginally from 97.9% in JR-NET 1&2 to 98.9% in JR-NET 3. This may be explained by greater technical skill among Japanese physicians, even though the use of stents will have been associated with a learning curve.[21)]

That said, the percentage of complete occlusion decreased from 57.7% to 46.5%, with several plausible causes. First, physicians likely avoided the risk of thromboembolic complications by meticulous coiling because progressive thrombosis could be expected, even if complete occlusion was not achieved immediately,[22)] and because retreatment is not associated with excessive risk.[23)] Second, stent-assisted coiling tends to show progressive thrombosis of the aneurysms at follow-up.[24,25)] The fact that stented cases had a significantly lower incidence of complete occlusion compared with non-stented cases probably reflected such reports.

Procedure-related complications were noted in 10.0% in JR-NET 3, representing a non-significant increase from 9.1% in JR-NET 1&2. These rates are consistent with previously reported data, although comparisons between studies are difficult because of the different criteria use for complications.[9,10,26)] Unfortunately, the prevalence of ischemic complications increased to 5.9% in JR-NET 3 from a level of 4.6% in JR-NET 1&2. When stents were used, this rate was as high as 8.9% and the 30-day morbidity rate was 6.0% despite dual or triple antiplatelet therapy being given in 93.6% of the cases that received stent-assisted coiling. Coupled with this, the rates of ischemic events associated with non-stented cases were comparable between JR-NET 3 (4.8%) and JR-NET 1&2 (4.6%). We concluded that stent introduction led to an expanded indication of endosaccular coiling, especially for wide-necked aneurysms, but that this was at the expense of an increased incidence of ischemia. By contrast, the rate of hemorrhagic complications, including intra-procedural, was comparable to that in previous studies.[12,27)]

Although there were no statistically significant differences in patient independence between JR-NET 3 and the preceding JR-NET 1&2 studies, the 30-day morbidity rate was significantly higher in the former (2.8% versus 2.1%, respectively). Given that the ischemic complication rate contributed heavily to this, we propose using a tailored antiplatelet therapy regimen based on pre-procedural platelet function,[28,29)] data of which we could not obtain in this study.

Limitations

There are several limitations in this study. First, we relied on retrospectively collected registry data that covered only about 35% of all cases receiving endovascular therapy in Japan. Second, our results could be biased because the radiographic findings, clinical outcomes, and procedure-related complications were only assessed by the treating physicians. Decisions on therapeutic indications might also have introduced inclusion bias. Third, clinical outcomes were recorded at the 30-day follow-up, and there was a lack of information on long-term outcomes after endosaccular coiling, after which recurrence rates are more important than after surgical clipping. In the future, a more complete and prospective registry would help to assess the therapeutic standards in the endovascular treatment of UIAs. The impact of using new devices, such as flow diverters and intrasaccular flow disruptors, should also be assessed.[30,31)]

Conclusion

The recent status of endosaccular coiling for UIAs in Japan is reported. The introduction of neck bridge stents has led to an increase in the number of cases of wide-necked aneurysms being treated and has been associated with a higher technical success rate compared with previous cohorts. However, the rate of ischemic complications has increased despite the increased use of periprocedural APT.