Safety and Efficacy Profile of Off-Label Use of the Pipeline Embolization Device: A Systematic Review and Meta-Analysis

Abstract

Objective:

The off-label utilization of the Pipeline Embolization Device (PED) is a common practice in numerous medical centers globally. Therefore, we conducted a systematic review and meta-analysis to evaluate the overall outcomes of this off-label usage of PEDs.

Methods:

PubMed, Web of Science, Ovid Medline, Ovid Embase, and Scopus were searched up to February 2023 using the Nested Knowledge platform to identify studies assessing the off-label use of PEDs. Any use of PED outside of the FDA-approved indication granted in 2018 is considered off-label use. Overall angiographic occlusion rates, ischemic and hemorrhagic complications, mortality, retreatment rates, and favorable clinic outcomes were included. Statistical analyses were performed using R 4.2.1 software to compare the overall outcome rates of anterior cerebral artery(ACA) vs. middle cerebral artery(MCA) and anterior vs posterior circulation subgroups.

Results:

We included 26 studies involving a total of 1,408 patients. The overall rate of complete occlusion was 80.3% (95% CI= 76.0–84.1). Subgroup analysis demonstrated a statistically significant difference in the rate of complete occlusion between anterior circulation (78.9%) and posterior circulation (69.2%) (p value=0.02). The rate of good clinical outcomes was 92.8% (95% CI= 88.8–95.4). The mortality rate was 1.4% (95% CI= 0.5–2.7). The overall rate of ischemic complications was 9.5% (95% CI= 7.7–11.6), with a comparable difference between anterior circulation (7.7%) and posterior circulation (12.8%) (p value=0.07). There was no statistically significant difference in MCA vs ACA subgroups in all parameters.

Conclusions:

Off-label use of PEDs can be a safe and effective treatment option for intracranial aneurysms. However, there is a need for more prospective, high-quality, non-industry-funded registry studies and randomized trials to test the efficacy and safety of off-label usage of PEDs and to expand its indications.

INTRODUCTION

It has been more than ten years since the initial Food and Drug Administration (FDA) approved the first flow diverters (FDs) for treating brain aneurysms¹. The approval of these devices was made possible due to the safety and efficacy demonstrated in both the single-arm Pipeline Embolization Device(PED) for the intracranial treatment of aneurysm trial² and the Pipeline for Uncoilable or Failed aneurysms (PUFs) trial^3,4. The FDA initially approved the indications for the flow diverters based on the inclusion criteria of the PED for the intracranial treatment of aneurysms trial (PITA) and extrapolated from its results². In 2015, the second-generation PED, known as PED Flex (Medtronic, Minneapolis, Minnesota, USA), also received FDA approval for the same indications. Subsequently, in 2018, the FDA expanded the approved indications for the PED Flex to include aneurysms with a size of ≤10 mm, those located up to the ICA terminus, and fusiform aneurysms⁵.

Over time, as neuro-interventionists have gained more experience, there has been a gradual increase in both the frequency of usage and the locations where such devices are employed. Apart from treating ICA aneurysms, the use of FDs in other locations, such as a distal anterior cerebral or middle cerebral artery, ruptured, and posterior circulation, are considered “off-label” uses. Retrospective studies are available in the literature assessing the PED’s safety and efficacy profile in these cases.

In current literature, the use of PED is employed for different aneurysm localizations, age groups, and aneurysm morphologies. The safety and efficacy of off-label use have been published in retrospective studies. There are currently no prospective studies in the literature investigating the safety and efficacy of off-label Pipeline use. The objective of this systematic review and meta-analysis is to identify and evaluate studies that have used the Pipeline Embolization Device (PED) for off-label purposes and to objectively analyze the safety and efficacy profiles of such use in comparison to clinical trials of on-label PED use.

METHODS

Search strategy

On February 18, 2023, a systematic literature review of English-language literature was conducted in accordance with the PRISMA guidelines for systematic reviews⁶. The review was carried out using the Nested Knowledge Autolit software⁷, following the drafted protocol, and covered the period from inception to the present day. Four major databases - PubMed, Embase, Web of Science, and Scopus - were utilized with different combinations of possible keywords and/or MeSH terms, including “off-label”, “on-label”, “Pipeline”, “PED”, “middle cerebral artery”, “MCA”, “anterior cerebral artery”, “ACA”, “posterior circulation”, “on-label”, and “intracranial aneurysms”. Additionally, an extensive manual search was conducted through the references of the included articles to retrieve any potentially missed papers.

In 2018, the indications for PED were expanded, and the primary approved or “on-label” uses of PED include endovascular treatment of small and medium wide-necked (neck width >= 4 mm or dome-to-neck ratio < 2) saccular or fusiform intracranial aneurysms. These aneurysms arise from a parent vessel with a diameter between 2.0 mm and 5.0 mm, and they are found in the Internal Carotid Artery up to the terminus for adults aged 22 years or older⁸ (Figure 1). Studies were included if they were randomized or had an observational prospective or retrospective study design and reported outcomes following “off-label” or “not on-label” use of PED in patients with intracranial aneurysms. Single-arm studies with off-label usage or comparative studies with off-label vs on-label usage of PEDs were included. Exclusion criteria included pediatric cases to reduce heterogeneity, case reports, case series with less than 10, review articles, conference abstracts, animal studies, editorials and non-peer-reviewed publications.

Figure 1:

The evolution of the Pipeline Embolization Device over time

The title and abstract screening process was carried out by two authors (AO, SA) according to pre-defined criteria. After the initial screening, any studies that were retained underwent full-text screening. The senior author (RK) was consulted during both stages to resolve any conflicts in the decisions.

Outcomes and Definitions

The study extracted the following data from each study: study design, country of origin, patient eligibility, patient sex and age, presenting neurologic status, total number of aneurysms, previous treatments for aneurysms, aneurysm locations and morphology, number of devices used, angiographic outcomes, clinical outcomes, complications, retreatment rates, and mortality rates and aneurysm characteristics. The meta-analysis focused on angiographic outcomes, clinical outcomes, complications, retreatment rates, and mortality rates.

Long-term results were assessed by evaluating angiographic outcomes at least 6 months after treatment. Immediate angiographic data was not collected. A favorable outcome was defined as an mRS score of 0–2, and clinical outcomes were evaluated at least 3 months after treatment. In-stent stenosis was defined as long-term angiographic severe stenosis (>%50 stenosis) related to the off-label use of PEDs. Early and late complications, whether ischemic or hemorrhagic, were not differentiated in at least 6 months follow-up time. Only permanent ischemic complications were considered and transient ischemic events or silent ischemic occurrences in magnetic resonance imaging are not included the meta-analysis

Quality Assessment

For quality assessment for our observational non-randomized included studies, we used the Risk Of Bias In Non-Randomized Studies – of Interventions (ROBINS-I) tool to assess the risk of bias. Based on seven domains, studies with a “Low,” “Moderate,” “Serious,” or “Critical” risk of bias were examined and appraised (confounding bias, selection bias, measurement classification of interventions bias, deviation from intended intervention bias, missing data bias, measurement of outcomes bias, and selection of the reported result bias. Quality assessments were completed separately by two authors (YS, AO). Discussions between the two authors were successful in resolving any disagreements of opinion.

Statistical Analysis

The data were analyzed using R software version 4.2.1 and the ‘meta’ package. We computed rate ratios (RRs) and their corresponding 95% confidence intervals (CIs) using a random-effects model. Subgroup analyses were performed, comparing anterior vs posterior circulation, anterior cerebral artery (ACA) vs middle cerebral artery (MCA), and adult (>22 years) vs pediatric patients. We assessed heterogeneity using Q statistics and the I² test, where I² >50% or a p value <0.05 were considered significant. Due to the small number of included studies (<10 per analysis), we perform Egger’s regression test for publication bias or meta-regression.

RESULTS

Search results and risk of bias assessment

Our initial search returned 940 results. After removing duplicates (243), a total of 686 studies were evaluated for screening. As a result of this initial screen, 652 entries were excluded. The remaining 45 articles were read in full text, and 19 more studies were excluded. A systematic review was prepared with the remaining 26 studies. Our risk of bias assessment identified 19 studies as being low risk of bias and 7 studies as being moderate risk of bias (Table 1).

Table 1.

Summary of Aneurysm Characteristics

Authors	Aneurysm Location (n)	Aneurysm morphology (n)	Aneurysm diameter (mean, SD)	Parent Artery Diameter (mean, SD)	No of Devices Used / Adjunctive Technique	Deployed Device Diameter	Number of patients followed-up clinically (%)	Number of patients/aneurysms followed-up radiologically (%)	Clinical or Radiologic Follow-up Time	Risk Of Bias
Authors	MCA (6), ACA (7), PCA (7), Acom (2), Vertebral (8), Basilar (6)	N/A	N/A	N/A	Single: 28, Dual: 7, Quad device: 1	N/A	36/36(100)	19/36(52.7)	5.9 (range, 3–8) months	Low
Al Kasab et. al. 2020 40	Acom (8), A1/A2(1), A2(2)	Saccular (9), fusiform (1), Blister (1)	5mm	Proximal: 2.0 Distal: 1.8 (0.4) Average: 2.4 (0.4)	Single:11	2.5–3mm	11/11(100)	11/11(100)	6.5 (range; 4–9) months	Moderate
Amuluru et. al. 2021 41	MCA (11), ACA (3), PCA (6), PICA (3)	Fusiform (9), Saccular (12), Dissecting (2)	12.5 mm (6.3)	N/A	Adjunctive Coiling: 6	N/A	23/23(100)	At least 2 Angiographic views	3/6/12/18/60 months	Low
Atallah et. al. 2019 42	VA (11), BA (10), PICA (1), PCA (1)	N/A	7 mm	N/A	Single: 22, Dual: 3, Adjunctive coilins for 7	2.5–4.5mm	16/18(88.8)	18/23(78.2)	13 months	Low
Baker et. al. 2019 24	ACA:51 (A1:2, A1-A2:38, A2-A39, Distal:2), MCA:14(M1:3, M2:4, Bifurcation:6, Distal: 1), P1:2	N/A	4.1 mm (3.0)	Proximal and distal vessel diameters were 2.10 and 1.96 mm	PED (Single: 64, Dual: 2; Adjunctive coiling: 5; Post deployment balloon angioplasty: 4	2.5mm(only)	57(100)	47/67(70.1)	19(10) months	Low
Bender et. al. 2019 43	A1(2), A1/A2(3), A2(1), Pericallosal (2), M1(5), MCA Bifurcation (5), M2(2), M2-M3(1), P1(1), P1/P2(1), P2/P3(4), PICA (2)	Fusiform (7), Saccular (21), Dissecting (1)	5.3 mm (3.9)	2.1 mm (0.37)	PED (Single: 12, Dual: 1); p64: 8; FRED Jr: 5; Surpass (Single: 2, Dual: 1)	2.5–3.25 PED	29/29(100)	22/29(75.8)	19.4 months	Low
Bhogal et. al. 2018 44	M1/M2(13), M1(2)	Fusiform (1), Saccular (14)	7 mm (0.8)	N/A	Single:13, Dual:1	N/A	15/15(100)	15/15(100)	6–48 months	Moderate
Briganti et. al. 2016 35	ICA (76), MCA (17), ACA (19), VA (20), PICA (5), BA (18), PCA (5)	Fusiform (35), Saccular (105), Other (20)	7.1 mm (range; 4.2–12.0)	N/A	N/A	N/A	105/140(75)	131(81.8)	13.7 (range; 6.7–28.1) months	Low
Cler et. al. 2022 18	Acom (13), A1(1), A2(4), Pericallosal (2)	Fusiform (5), Saccular (15)	7.3 mm (4.6)	1.8 mm (range, 1 to 2.6)	N/A	N/A	20/20(100)	16/20(80)	10 months	Low
Dabus et. al. 2017 45	15(Acom), A1/A2(11), A2/A3(1)	N/A	4mm (range; 1.5–15)	N/A	N/A	N/A	N/A	20/28(71.4)	7.5 (range; 3–20) months	Moderate
Dakay et. al. 2021 46	BA (38), PCA (14), PICA (13), SCA (1), VA (39) VBJ (11)	Fusiform/blister (78), Saccular (136)	N/A	N/A	PED (Single:84, Multiple:33)	N/A	112/117((95.7)	97/117(82.9)	12 months	Moderate
Dmytriw et. al. 2022 47	ICA (98), MCA (M1: 14, M2: 3, M3: 1), ACA (A1: 4, A2: 2, A3: 1, Acom: 1), PCA (P1: 4, P2: 4, P3: 1), PICA (8), VA (14), Basilar trunk (11), Basilar tip (5), Bifurcation aneurysm (13), Daughter sac (14),	Fusiform/blister (36), Saccular (135)	N/A	3.5 (range: 2.74–4.0)	N/A	N/A	165/165(100)	157/171(91.8)	13.4 (range: 12.0–27.9) months	Low
Enriquez-Marulanda et. al. 2022 8	VA (77), VBJ (8), BA (34), SCA (3), PICA (12), PCA (15)	Fusiform (42), Saccular (53), Dissecting/blister (54)	9 mm (range; 6–14)	N/A	Single: 125, Dual: 16, Triple: 4, Quad: 1; Adjunctive coiling for 29	N/A	145/146(99.3)	131/149(87.9)	12 (range; 6–25.6) months	Low
Griessenauer et. al. 2020 19	MCA (M1: 17, Bifurcation: 8)	Fusiform (6), Saccular (19)	9.4 mm (range 3.3–20)	2.66 mm (0.1)	Single:23, Dual:2	… Diameters ranging from 2.5–5.0mm and lengths ranging from 10–35 mm	24/25(96)	22/25(88)	33 months (range 1–93 months)	Low
Lauzier et. al. 2022 39	Ophthalmic (12,), Pcom (12,), supra-clinoid (11), superior hypophyseal (8,), cavernous (6), para-clinoid (6), Dural ring (6), M1 (1)	Saccular (46), lobulated (8), Tandem (3), Blister (3), Fusiform (2)	7.6 mm (range; 1.4–25.0)	N/A	Single:62, Adjunctive Coiling:2	N/A	48/48(100)	52 (83.9)	12.5 months (383 days)	Moderate
Liang et. al. 2019 48	MCA (M1: 12, M1/M2 junction: 4, M2: 4), ACA (A1: 1, A1/A2 junction: 1, A2: 2, A3: 2), Acom (2)	Fusiform (15), Saccular (8), Dissecting (5)	12.3 mm (8.1)	N/A	PED (Single:15, double: 11, Triple: 1, 5 devices: 1); Adjunctive coilina: 6	2.5–3.25 for 34 PED, 3.5–4.25 for 10 PED, 4.5 for 1 PED	28/28(100)	27(96.4)	7.7 (range 1–25) months	Moderate
Lin et. al. 2016 49	MCA (38), ACA (7), PCA (5), Acom (3)	Fusiform (40), Saccular (13)	12.3 mm (5.7)	Proximal: 2.5 (0.5) Distal: 2.3 (0.4) Average: 2.3 (0.4)	Single:43, Multiple:10	N/A	53/53(100)	51/53(96.2)	12 (range; 4–24) months	Low
Ma et. al. 2022 50	P1 (3), BA (4), Acom (1), ACA (2), ICA (2)	Fusiform (9), Saccular (2), Pseudo (1)	18mm (range; 2–38)	<3mm	Single: 8, Dual: 3, Triple: 1	<3mm PED	12/12(100)	12/12(100)	6–42 months	Low
Martin et al. 2015 51	MCA (M2: 28, M3: 1), ACA (A2: 8, A3: 16), PCA (P2: 9, P3: 3)	Fusiform (19), Saccular (38), Dissecting (8)	7.8 mm	2.4 mm	Single: 57, Dual: 5, Triple: 1; Adjunctive coiling for 6; Other stents for 2	N/A	60/60(100)	53/65(81.5)	11.7 (1.9) months	Low
Primiani et. al. 2019 52	P1 (3), P1/P2 (3), P2 (2), P2/P3(2)	Fusiform (4), Saccular (6)	10.2 mm (2.4)	1.9 mm (0.29)	Single:8, Dual: 2	2.5mm: 4, 2.75mm: 4, 3mm:1, 4mm:2, 4,25mm:1	9/9(100)	9/10(90)	2–33 months	Low
Wallace et. al. 2017 53	MCA Bifurcation (49)	N/A	7.0 mm (3.9)	N/A	Single:48, Dual: 1; Adjunctive coiling:17, Adjunctive Stent-coil: 3	N/A	46/46(100)	41/49(89.1)	6 months	Moderate
Wang et. al. 2022 36	MCA Bifurcation (21), Distal MCA (4)	N/A	N/A	N/A	Single:23, Dual:2	N/A	21/21(100)	25/25(100)	6/12/18 months (Up to angiographic result)	Moderate
Yavuz et. al. 2014 37	MCA (M1: 6, M2: 2, M3: 6), ACA (A2: 3), PCA (P1: 3, P2: 2), PICA (P3: 1)	N/A	12.2 mm (7.1)	2.0 mm (0.6)	Single: 18, Dual:3; Adjunctive Coiling:3	N/A	22/22(100)	23/23(100)	10.9 months (11.4)	Moderate
Yu et. al. 2021 54	Acom (1), BA (1), ICA (53), MCA (3), PCA (5), Pcom (14), PICA (3), VBJ (5), VA (7), Other (17)	Saccular (59), Fusiform (19)	8.4 mm (7.4)	N/A	N/A	N/A	96/109(88)	96/109(88)	35.6 weeks (22.4)	Low
Zammar et. al. 2018 20	M1 (5), MCA Bifurcation (3), Distal M2 (2)	Saccular (3), Fusiform (7)	15 mm	N/A	Mean PED device number is 1.60	N/A	10/10(100)	10/10(100)	7.5 (range; 7–12) months	Low
Zanaty et. al. 2014 38	ACA (72), PCA (35)	Fusiform (42), Saccular (51), Blister (14)	6.6 mm (6.2)	N/A	N/A	N/A	103/107 (96.2)	113/121 (93.3)	6.5 months	Low

ACA: Anterior Cerebral artery, MCA: Middle Cerebral Artery, PCA: Posterior Cerebral Artery, Acom: Anterior communicating artery, PICA: Posterior inferior cerebellar artery, BA: Basilar Artery, ICA: Internal Carotid Artery, VBJ: Vertebrobasilar Junction

In the systematic review, there were only 2 comparative studies included. Four studies focused on posterior circulation only, while 10 studies were related to anterior circulation aneurysms only. Three studies specifically investigated ACA aneurysms, and 5 studies examined the use of the Pipeline in MCA aneurysms. The studies were conducted between the years 2013 and 2022. Only 1 study was designed prospectively, and the remaining 25 studies had a retrospective design. Fourteen studies were multicenter, and 12 studies were single center studies. (Table 2).

Table 2.

Summary of Study Characteristics

Articles	Authors	Risk Of Bias	Year	Country	Study Design	Patient Number	Age (mean, SD, range)	Female Gender (N, %)	Presenting symptoms	Number of total aneurysms	Number of previously treated aneurysms
Safety and Efficacy of the Pipeline Embolization Device Use in the Outside Circle of Willis Located Intracranial Aneurysms: A Single-Center Experience.	Al Kasab et. al. 2020 40	Moderate	2020	USA	RS (SC)	36	58.2 (14.6)	19 (52.8)	SAH (7)	36	2 (Coiling: 1, PED: 1)
Flow diversion treatment of anterior communicating artery region aneurysms.	Amuluru et. al. 2021 41	Moderate	2021	USA	RS(SC)	10	N/A	N/A	N/A	11	N/A
Pipeline for Distal Cerebral Circulation Aneurysms.	Atallah et. al. 2019 42	Low	2019	USA	RS (SC)	23	60.4 (12.1)	6 (26.1)	Previous Treatment (4), Previous SAH (2), SAH (1)	23	3 (Coiling)
Pipeline Embolization in Patients with Posterior Circulation Subarachnoid Hemorrhages: Is Takotsubo Cardiomyopathy a Limiting Factor?	Baker et. al. 2019 24	Low	2020	USA, Canada, Austria,	RS(MC)	23	55(range; 24–75)	14 (60.9)	N/A	23	6(Clipping:2), coiling:3, Clipping+coiling:1)
Tiny Pipes: 67 Cases of Flow Diversion for Aneurysms in Distal Vessels Measuring Less Than 2.0 mm.	Bender et. al. 2019 43	Low	2019	USA	PS (SC)	57	55.8 (10.8) (range; 33–78)	40 (60)	SAH (1), Previously Treated (30), Previously Ruptured (25)	67	30 (Clipping: 7, Coiling: 21, FD: 2)
The Use of Flow Diversion in Vessels ≤2.5 mm in Diameter-A Single-Center Experience.	Bhogal et. al. 2018 44	Moderate	2018	Germany, Argentina	RS(MC)	29	56.2(15.9) (range;21–83)	22 (76)	N/A	29	4 (Coiling)
Flow diverter device for the treatment of small middle cerebral artery aneurysms.	Briganti et. al. 2016 35	Low	2014	Italy	RS(MC)	14	58.5(8.8) (range;39–71)	10 (71.4)	N/A	15	4 (Coiling)
Comparative study of on-label versus off-label treatment of intracranial aneurysms with the Pipeline embolization device.	Cler et. al. 2022 18	Low	2022	USA	RS (SC)	140/330	57.5 (range; 48.0–67.3)	103 (73.6)	N/A	160	31
Treatment of complex anterior cerebral artery aneurysms with Pipeline flow diversion: mid-term results	Dabus et. al. 2017 45	Low	2017	USA	RS(MC)	20	N/A	13 (65)	N/A	20	6 (Coiling: 4, Clipping: 2)
Flow diversion in anterior cerebral artery aneurysms.	Dakay et. al. 2021 46	Low	2021	USA	RS(SC)	27	57(range; 38–85)	11(40.7)	13(SAH)	28	13 (Coiling)
The Pipeline Embolization Device: a decade of lessons learned in the treatment of posterior circulation aneurysms in a multicenter cohort.	Dmytriw et. al. 2022 47	Low	2022	USA	RS(MC)	117	60 (range; 53–69)	59(50.4)	18(SAH)	117	16 (Both:1, Endovascular:13, Clipping:2)
Safety and Efficacy of the Off-Label Use of Pipeline Embolization Device Based on the 2018 Food and Drag Administration-Approved Indications for Intracranial Aneurysms: A Single-Center Retrospective Cohort Study.	Enriquez-Marulanda et. al. 2022 8	Low	2022	USA	RS (SC)	165/391	N/A	N/A	N/A	171	71 (Microsurgical clipping: 13, Coiling: 30, Coiling + clipping: 2, PED: 22, Stent-assisted coiling: 3, Standard stent: 1)
Experience With the Pipeline Embolization Device for Posterior Circulations Aneurysms: A Multicenter Cohort Study.	Griessenauer et. al. 2020 19	Low	2020	Austria	RS(MC)	146	56 (range; 46–62)	71 (48.6)	24(SAH), 52(Incidental), 14(Headache). 15(Dizziness), Brainstem Compression (13), CN palsy (12), Stroke(9), Other(7)	149	10 (Endovascular)
Pipeline embolization of proximal middle cerebral artery aneurysms: A multicenter cohort study	Lauzier et. al. 2022 39	Low	2022	USA	RS(MC)	25	53.7 (range 16–72)	15 (60)	1 (Seizures), 1 (SAH)	25	10 (Coiling: 7, Clipping: 1, FD (with PED) 2)
Off-Label Application of Pipeline Embolization Device for Intracranial Aneurysms.	Liang et. al. 2019 48	Moderate	2019	USA	RS (SC)	48	54.3(range; 21–75)	44 (91.7)	Headache (47), Vision changes 9), Dizziness/vertigo (7), Cranial nerve III deficits (5), Tinnitus (5), Nausea/vomiting (3), Dysphasia (1), Asymptomatic (10)	62	20 (Coiled: 12, Stent only: 2, Stent-assisted Coiling: 2, Stent-asissted coil + clipped: 2, Clipped: 1, Coiled + clipped: 1)
Treatment of Distal Anterior Circulation Aneurysms with the Pipeline Embolization Device: A US Multicenter Experience.	Lin et. al. 2016 49	Low	2016	USA	RS (MC)	28	51.7 (16.7) (range 13–71)	18 (64.3)	Headaches (10), TIA or neurological deficit (9), SAH (2)	28	11 (Coiling or stent-assisted coiling: 6, clipping: 4, clipping and
Pipeline for the treatment of distal cerebral circulation aneurysms: A multicenter study focusing on periprocedural Complications.	Ma et. al. 2022 50	Low	2022	China	RS (MC)	53	53 (16–69)	29 (54.7)	Headaches and Dizziness (24), Incidental (29)	53	9 (Clipping: 4, coiling: 4, stent-assisted coiling:1,
Small Pipes: Preliminary Experience with 3-mm or Smaller Pipeline Flow-Diverting Stents for Aneurysm Repair prior to Regulatory Approval	Martin et al. 2015 51	Moderate	2015	Canada	RS(MC)	12	48(11.2) (range 22–62)	8(66.6)	4 (SAH), 3 (Headache), 3 (Previous Treatment), 2 (Stroke-TIA)	12	5(coiling or stent-coil)
A2, M2, P2 aneurysms and beyond: results of treatment with pipeline embolization device in 65 patients.	Primiani et. al. 2019 52	Moderate	2019	USA	RS (MC)	65	54.7 (15.2)	42 (64.6)	9(SAH), 51(Incidental), 5(Unruptured Symptomatic)	65	N/A
Endovascular treatment of posterior cerebral artery aneurysms with the pipeline embolization device	Wallace et. al. 2017 53	Low	2018	USA	RS(MC)	9	56(range; 18–74)	6 (66.7)	5 (Incidental), 1 (3.CN Palsy), 1 (Aneurysmal Mass Effect), 1 (Stroke), 1 (Previous Treatment)	10	1 (Stent-assisted coiling)
Pipeline embolization of complex, wide-necked middle cerebral artery bifurcation aneurysms: A single-center experience.	Wang et. al. 2022 36	Low	2022	China	RS(SC)	46	51(14.2)	34(73.9)	N/A	49	N/A
Endovascular treatment of middle cerebral artery aneurysms with flow modification with the use of the pipeline embolization device	Yavuz et. al. 2014 37	Low	2013	Turkey	RS(SC)	21	56(4.2) (range; 34–74)	12(57.1)	21 (Headache)	25	1 (Balloon-Assisted Coiling)
Flow Diversion for Intracranial Aneurysms Beyond the Circle of Willis.	Yu et. al. 2021 54	Moderate	2021	China	RS (MC)	22	44.5 (12.7) (range; 16–66)	16 (72.7)	17 (Incidental), 4 (Previous Treatment), 1 (SAH)	23	4 (Coiling)
Outcomes After Off-Label Use of the Pipeline Embolization Device for Intracranial Aneurysms: A Multicenter Cohort Study.	Zammar et. al. 2018 20	Moderate	2018	USA	RS (MC)	109	54.4 (12.3)	91 (83.5)	N/A	109	1 (Stent)
Flow diversion for complex middle cerebral artery aneurysms.	Zanaty et. al. 2014 38	Low	2014	USA	RS(SC)	10	47.5	4(40)	N/A	10	N/A
The off-label uses of pipeline embolization device for complex cerebral aneurysms: Mid-term follow-up in a single center.	Zhang et. al. 2022 21	Moderate	2022	China	RS (SC)	107	54 (range; 16–81)	63 (58.9)	N/A	121	7 (Coiling)

RS: Retrospective Studies, PS: Prospective Studies, MC: Multicenter, SC: Single Center, N/A: Not applicable, FD: Flow diverter, SAH: Subarachnoid Hemorrhage, PED: Pipeline Embolization Device

Complete Occlusion

The analysis of 26 studies and a total of 1269 patients showed an overall rate of aneurysm’s complete occlusion at 80.3% (95% CI= 76.0–84.1). However, significant heterogeneity was observed in the studies (I²= 58%; p value < 0.05. In terms of the location of the occlusion, a statistically significant difference was observed between anterior circulation (78.9%; 95% CI= 73.3–83.7) and posterior circulation (69.2%; 95% CI= 62.7–74.9) (p value=0.017). However, no statistically significant difference was found in the rate of complete occlusion between ACA (80.0%; 95% CI= 61.5–91.0) and MCA aneurysms (76.3%; 95% CI= 67.6–83.2) (p value=0.67) (Figure 2).

Figure 2:

A: Forest plot illustrating overall complete occlusion rates for the off-Label use of the Pipeline Device. B: Comparison between the Anterior Cerebral Artery and Middle Cerebral Artery regarding complete occlusion. C: Comparison of Posterior versus Anterior Circulation for the Off-Label usage of the Pipeline Embolization Device.

Clinical Outcomes

This study assessed data from 21 studies with 1,055 patients to evaluate the rate of positive clinical outcomes. The rate was determined to be 92.8% (95% CI= 88.8–95.4). Regarding location, a significant difference was found between anterior circulation (95.9%; 95% CI= 87.6–98.7) and posterior circulation (84.0%; 95% CI= 79.4–87.8) (p value=0.02). Nevertheless, no significant differences were observed in the rate of favorable outcomes between ACA (90.0%; 95% CI= 73.1–96.7) and MCA aneurysms (97.3%; 95% CI= 81.7–99.6) (p value=0.26) (Figure 3)

Figure 3:

A: Forest plot depicting rates of good clinical outcomes (mRS 0–2) for the off-Label use of the Pipeline Device. B: Comparison between the Anterior Cerebral Artery and Middle Cerebral Artery for good clinical outcomes. C: Comparison of Posterior versus Anterior Circulation for the off-Label usage of the Pipeline Embolization Device in relation to good clinical outcomes.

Mortality Rates

This study examined data from 24 studies with 1,293 patients to establish the rate of mortality. The rate was determined to be 1.4% (95% CI= 0.5–2.7. Regarding location, a significant difference was found between anterior circulation (0.06%; 95% CI= 0–1.48) and posterior circulation (5.8%; 95% CI= 0.95–13.39) (p value=0.02)(Online Supplementary Figure 1). There were no significant differences were observed in the rate of mortality between ACA (2.7%; 95% CI= 0–10.0) and MCA aneurysms (0%; 95% CI= 0–1.5) (p value=0.09).

Ischemic Complications

This study analyzed data from 27 studies involving 1,337 patients to determine the rate of ischemic complications. The rate was found to be 9.5% (95% CI= 7.7–11.6) (Figure 4). In terms of location, no statistically significant difference was observed between anterior circulation (7.7%; 95% CI= 4.8–12.0) and posterior circulation (12.8%; 95% CI= 9.4–17.1) (p value=0.07). However, there was no statistically significant difference in the rate of ischemic complications between ACA (4.1%; 95% CI= 0–19.2) and MCA (8.9%; 95% CI= 3.9–15.3) (p value=0.51)

Figure 4:

A: Forest plot demonstrates rates of ischemic complications for the off-Label use of the Pipeline Device. B: Comparison between the Anterior Cerebral Artery and Middle Cerebral Artery for ischemic complications. C: Comparison of Posterior versus Anterior Circulation for the off-Label usage of the Pipeline Embolization Device in relation to ischemic complications.

Hemorrhagic Complications

This study analyzed data from 25 studies involving 1,231 patients to determine the rate of hemorrhagic complications. The rate was found to be 2.6% (95% CI= 1.6–3.8). A subgroup analysis revealed no statistically significant difference was observed between anterior circulation (0.9%; 95% CI= 0–3.1) and posterior circulation (3.8%; 95% CI= 1.6–6.7) (p value=0.11). However, there was no statistically significant difference in the rate of hemorrhagic complications between ACA (2.6%; 95% CI= 0–9.6) and MCA (0.7%; 95% CI= 0–3.3) (p value=0.26).

Retreatment Rates

This study assessed data from 16 studies with 838 patients to evaluate the rate of retreatments. The rate was determined to be 3.4% (95% CI= 1.4–5.9)(Online Supplementary Figure 2). A subgroup analysis showed significant differences in the rate of retreatment between anterior circulation (0.5%; 95% CI= 0–4.2) and posterior circulation (7.2%; 95% CI= 4.1–11.0) (p value=0.01).

In Stent-Stenosis

18 studies with 716 patients to establish the incidence of in-stent stenosis. The incidence was determined to be 2.8% (95% CI= 1.1–5.1. A subgroup analysis revealed no significant differences of in-stent stenosis between ACA (4.6%; 95% CI= 0–21.0) and MCA aneurysms (3.2%; 95% CI= 0.08–9.1) (p value = 0.76) (Online Supplementary Figure 3).

DISCUSSION

This current systematic review of published literature regarding the off-label use of PED for aneurysm treatment offers several important findings. First, off-label use is associated with a relatively high rate of complete occlusion and good clinical outcomes overall. Secondly, off-label usage of PEDs resulted in lower complete occlusion rates in the posterior circulation than in the anterior circulation, but the complication rates were found to be higher in posterior circulation. Third, hemorrhagic and ischemic complications were relatively lower in pediatric patient group. Most studies on off-label usage of PED for aneurysms are retrospective clinical series, and there are currently no randomized controlled trials to support their effectiveness. However, some evidence suggests that off-label PEDs may be a safe and effective alternative for certain cases when other techniques are not feasible. Further research is necessary to establish the safety and effectiveness of off-label PED for aneurysms, through well-designed studies, including randomized controlled trials.

Our current study is the first meta-analysis to report clinical and radiological outcomes associated with off-label usage of PED, providing valuable insights into the safety and effectiveness of this technique. The results of this meta-analysis highlight the need for further research to optimize patient selection criteria and determine the best technique and device selection for each specific indication. The off-label PED concept started to be used for indications other than those defined by the FDA in 2011 and those indications expanded in 2018. However, the use of flow diverters in the posterior circulation remains off-label, and concerns persist regarding their application in treating aneurysms in this area⁹. While there have been descriptions of using flow-diverter devices for specific types of aneurysms in the posterior circulation, the outcomes remain unclear due to complex deployment techniques, leading to higher mortality and morbidity rates^10,11. Despite recent technological advances indicating the potential safe and effective use of flow diversion in treating certain ruptured aneurysms in the posterior circulation and distal anterior circulation aneurysms¹², debates persist concerning its application for aneurysms in the posterior circulation. This is primarily due to significantly higher rates of ischemic stroke and perforator infarction compared to anterior circulation aneurysms¹³. Furthermore, flow diverters have shown promise in the anterior circulation, reports suggest incomplete occlusion could be a limitation when dealing with certain types of aneurysms in the posterior circulation¹⁴. Another limitation for off-label use of PED is posterior circulation aneurysms are often associated with a higher incidence of morbidity and mortality compared to their anterior circulation counterparts, primarily due to higher rates of aneurysm rupture and neurovascular compression caused by large dolichoectatic aneurysms¹⁵. Hopefully, studies reported that flow diversion might be a possible treatment in carefully selected patients with high-risk atypical posterior circulation aneurysms, which have a poor natural history and no optimal treatment strategy¹⁶. In conclusion, while there is growing evidence supporting the feasibility and efficacy of flow diversion for posterior circulation and distal cerebral aneurysms, the off-label status and associated concerns continue to pose challenges to its widespread use.

Studies in the literature have investigated off-label use of PED for previously treated aneurysms, fusiform and dissecting aneurysms, small aneurysms, distal cerebral aneurysms, ruptures and pseudoaneurysms, and carotid-cavernous fistulas¹⁷. To best of our knowledge, no meta-analysis has been conducted on the use of off-label PEDs. However, there have been several large multicenter cohort studies and systematic reviews on this topic, which have reported complete occlusion rates ranging from 78% to 82%^18–21. In our study, we found a similar rate of 80.7%. The overall clinical outcome rates reported in these studies ranged from 90% to 95%, which was also comparable to our findings of 91.7%. The literature examining the treatment of posterior circulation aneurysms over the years reveals observed mortality rates (0–20%) and long-term complete occlusion rates (52–57%) in cases treated with flow diverters^16,22. In the latest meta-analysis conducted in 2021 with disaggregated patients group, it was reported that patients treated with flow diverters for posterior circulation aneurysms had a recorded major complication rate of 22% and an angiographic occlusion rate of 67%²³. These findings, while still relatively high compared to the anterior circulation, were comparable to our study. The rates of good clinical outcomes for posterior circulation aneurysms were statistically significantly lower than those for anterior circulation aneurysms (95.6% vs. 84.0%). Moreover, posterior circulation group had higher rates of retreatment and hemorrhagic complications. The high retreatment rates could have led to the reapplication of alternative endovascular treatments and an increase in complication rates^24–26. However, it’s important to note that this study only covers the period from 2011 to 2023 and recent studies have shown that new generation PEDs have lower complication rates for aneurysms in posterior circulation^19,26.

Notably, the effectiveness of PED treatment for both anterior and posterior circulation aneurysms have improved significantly over the years. Especially in cases of bleeding blister and blister-like distal anterior cerebral artery aneurysms where there is no alternative treatment and due to the low chance of successful coiling, studies using flow diversion have achieved favorable clinical outcomes^27,28. This shows that off-label usage, particularly in these types of aneurysms, is more frequent. Initial clinical trials for PED showed higher complication rates for posterior circulation aneurysms comparing to anterior. The first clinical trial for the Pipeline Embolization PED, called PUFS, only included internal carotid artery aneurysms³. However, the subsequent PITA study included distal anterior cerebral aneurysms and posterior circulation aneurysms. In this study, after PED deployment in the single treated M1 aneurysm, the patient experienced grey matter infarction which is one of the 2 complications of 31 patient²⁹. In the PREMIER trial, which consisted of two separate follow-up studies at one year and three years, retreatment was performed in two out of four V4 aneurysms after 3 years follow-up³⁰.

Neointimal hyperplasia, post flow-diverter deployment, stands as a crucial factor in assessing stent safety and efficacy, potentially contributing to in-stent stenosis. Caroff et al. delved into this by exploring its correlation with cardiovascular risk factors and stent design³¹. They observed a 9.8% in-stent stenosis occurring around 6 months post-PED treatment, notably higher than our studies with 2.8%. This disparity could potentially stem from differences in categorizing stent stenosis levels (mild <25%, moderate 25–50%, severe >50%) and our exclusion of stent stenosis above 25% in this study. his difference might arise from variations in the classification of stent stenosis severity (mild <25%, moderate 25–50%, severe >50%) and our deliberate exclusion of stent stenosis above 25% in this specific study. Our main goal in not including stent stenosis above 50% in our research was to focus on instances where, based on previous studies, stent stenosis generally remained benign and did not lead to any complications if it’s not severe^32,33. However, clinically significant stent stenosis was predominantly observed in cases where the stenosis exceeded 50%.

In recent years, endovascular treatment has been increasingly used for distal anterior circulation aneurysms and MCA aneurysms. These aneurysms present challenges similar to flow diversion in the posterior circulation due to regional anatomy and arterial branches, including small lenticulostriate arteries³⁴. The literature reports ischemic complication rates ranging from 0% to 8% for MCA aneurysms^35–38. In our study, this rate was found to be 10%, but there was no statistically significant difference compared to ACA aneurysms. The complete occlusion rates for MCA aneurysms ranged from 59% to 80%, which was similar to the literature, with our study finding a rate of 76.3%, comparable to that of ACA aneurysms^37–39.The only statistically significant result found was that the mortality rate for MCA aneurysms was lower than that for ACA aneurysms.

There are several limitations to our study. Firstly, all studies except one study on off-label usage were retrospective cohort studies. Secondly, subgroup analysis could not be performed due to the heterogeneity of the studies, including the posterior circulation, anterior distal circulation, different age groups, ruptured/unruptured aneurysms, and different aneurysm morphologies and sizes. Thirdly, there are ongoing debates about the concept of off-label usage, especially with the increase in PED usage in the posterior circulation. To address this issue, off-label indications were defined as indications other than the FDA premarket approval indications. Finally, the lack of direct comparisons between off-label and on-label PED usage in the literature limits our study to only focus on off-label PED outcomes.

CONCLUSION

In conclusion, our meta-analysis suggests that the off-label use of PEDs can be a safe and effective treatment option for intracranial aneurysms. However, the scarcity of direct comparison studies between on-label and off-label usage makes it challenging to draw definitive conclusions. Despite the limitations of the studies included in the meta-analysis, our findings support the notion that off-label use of PEDs may have good angiographic and clinical outcomes. Nevertheless, there is a need for more prospective, high-quality, non-industry funded registry studies and randomized trials to test the efficacy and safety of off-label usage of PEDs and to expand its indications.

Supplementary Material

1

Online Supplementary Figure 1: Forest plot of Mortality rates for Posterior vs Anterior Circulation

2

Online Supplementary Figure 2: Forest plot demonstrates retreatment rates for the off-Label use of the Pipeline Device and subgroup analysis for anterior vs posterior circulation

3

Online Supplementary Figure 3: Forest plot demonstrates stent stenosis percentage for the off-Label use of the Pipeline Device

Highlights.

• In the realm of the flow diverter era, case reports and retrospective case for off-label usage of pipeline series have been published. However, for a considerable time, the scope of application for these flow diverters has not been expanded. This study demonstrates that the effectiveness and safety of off-label use of flow diverters are comparable to on-label uses.

• The use of the Pipeline device in posterior circulation has a statistically lower complete occlusion rate compared to its use in anterior circulation. However, in terms of ischemic complications, there was no statistically significant difference found between anterior and posterior circulations.

• There was no statistical difference observed in subgroup analyses concerning off-label Pipeline usage between the anterior cerebral artery(ACA) and middle cerebral artery(MCA).