Efficacy and Safety of Venous Sac Embolization With or Without Feeding Artery Embolization Versus Feeding Artery Embolization Alone in the Management of Pulmonary Arteriovenous Malformations: A Systematic Review and Meta-analysis

Introduction

Pulmonary arteriovenous malformations (PAVMs) are characterized by a direct communication between pulmonary artery and vein bypassing the normal intervening capillary bed. ¹ It results in right-to-left shunt leading to hypoxemia, causing cyanosis, and paradoxical embolization, causing events such as transient ischemic attacks, stroke, and brain abscess. ² PAVMs can be single or multiple and simple or complex. ² Approximately 70% are associated with hereditary hemorrhagic telangiectasia (HHT). ³ PAVMs may be idiopathic, post-traumatic, post-infective, or secondary to hepatopulmonary syndrome.

Transcatheter embolotherapy is the standard of treatment for PAVMs, and traditionally, feeding artery embolization (FAE) has been done to treat the PAVMs. ⁴ Another technique employing the embolization of the venous sac alone or venous sac embolization (VSE) assisted with FAE has also been reported. Hayashi et al ⁵ first demonstrated the efficacy and safety of VSE in comparison with FAE. Subsequently, additional 2 studies comparing VSE with FAE versus FAE alone have been published.^6,7 To the best of our knowledge, the safety and efficacy of VSE (with or without FAE) versus FAE alone in the management of PAVMs have not been systematically evaluated. Hence, the present systematic review and meta-analysis sought to evaluate the safety and efficacy of VSE vis-à-vis FAE in patients with PAVMs.

Materials and Methods

The current systematic review and meta-analysis was reported according to the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines. ⁸ The study protocol was registered on the international prospective register of systematic reviews (CRD42021267375). As per our institutional policy, ethical approval is not required for performing systematic reviews and meta-analysis.

Assessment of Study Quality

Two authors (A.T. and M.V.) independently rated the methodological quality (risk of bias) of the included studies based on the National Institutes of Health (NIH) Quality Assessment tool for observational cohort studies. ⁹ All 3 studies were rated fair (Table 1). ¹⁰

OPEN IN VIEWER

Table 1.

Quality Rating of the Included Studies Using the NIH Quality Assessment Tool for Observational Cohort Studies.

Study	Questions
	1	2	3	4	5	6	7	8	9	10	11	12	13	14
Hayashi et al 5	Yes	Yes	Yes	No	No	Yes	Yes	NA	Yes	NA	Yes	NR	Yes	NR
Roberts et al 6	Yes	Yes	Yes	Yes	No	Yes	Yes	NA	Yes	NA	Yes	NR	Yes	NR
Nagai et al 7	Yes	Yes	Yes	No	No	Yes	Yes	NA	Yes	NA	Yes	NR	Yes	NR
Study					Overall rating
					Reviewer 1					Reviewer 2
Hayashi et al 5					Fair					Fair
Roberts et al 6					Fair					Fair
Nagai et al 7					Fair					Fair

Abbreviations: NA, not applicable; NIH, National Institutes of Health; NR, not reported.

Results

Literature Search and Study Selection

The study selection process is depicted in Figure 1. The systematic search yielded 37 results (MEDLINE: 10, Embase: 16, and Scopus: 11). After removal of duplicates and screening of title and abstracts, full-text review was done for 7 studies. Four articles that studied the efficacy and safety of VSE without comparison with FAE were excluded.^11–14 Finally, 3 studies were included for the final analysis.^5–7

OPEN IN VIEWER

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart. Literature search and study selection process.

Study Characteristics

The 3 included studies had a total of 169 patients (119 females and 50 males). The total number of PAVMs was 298—156 in FAE group and 142 in VSE (with or without FAE) group. Hereditary hemorrhagic telangiectasia was present in 119 patients. The number of simple PAVMs was more in FAE group (77%; 120/156) as compared to VSE group (61%; 87/142). The study and patient characteristics of the 3 studies are summarized in Table 2.

OPEN IN VIEWER

Table 2.

Study and Patient Characteristics of Included Studies.

	Study 1		Study 2		Study 3
Author	Hayashi et al 5		Roberts et al 6		Nagai et al 7
Year of publication	2012		2021		2021
Type of study	Retrospective		Retrospective		Retrospective
	FAE	VSE	FAE	FAE+VSE	FAE	FAE+VSE
No. of patients	14	11	53	63	16	12
Female	12	8	33	45	13	8
Male	2	3	20	18	3	4
Age of patients (mean or median)	44.1	42.8	42.6	42.7	52.5	56
No. of PAVMs	22	15	114	105	20	22
Simple PAVMs	11 a	10 a	89	55	20	22
Complex PAVMs	3 a	1 a	25	50	0	0
HHT	3	1	50	54	5	6
Location (upper/middle/lower)	22 (7/2/13)	15 (0/6/9)	114 (20/24/70)	105 (22/24/59)	20 (6/2/12)	22 (7/1/14)
Pre-embolization: venous sac diameter (mm)	21±14	15±7	8.4±5.3	9.2±4.3	7.1	6.7
Pre-embolization: feeding artery diameter (mm)	5±2	4±1	2.7±0.9	3.6±0.7	2.7	3.2
Pre-embolization: draining vein (diameter: mm)	8±5	5±2	NA	NA	2.9	3.9
No. of PAVMs with decrease in size of draining vein or sac ≥70%	15	15	93	99	4	8
No. of coils used/PAVM	4±4	8±4	3.9±2	7.1±5.8	NR	NR
Follow-up (in months)	58±54	42±42	20.6	16.2	12	12
Reperfusion or recurrence	11	0	NR	NR	2	0
Minor complication (pleurisy)	2	2	16	16	0	0
Major complication	0	0	1	0	0	0

Abbreviations: FAE, feeding artery embolization; HHT, hereditary hemorrhagic telangiectasia; NA, not applicable; NR, not reported; PAVM, pulmonary arteriovenous malformation; VSE, venous sac embolization.

a

Number of patients reported instead of number of PAVMs.

Outcomes

Only coils were used as embolic agents, and the number of coils used in VSE was more as compared to FAE. The clinical success was higher with VSE (with or without FAE) as compared to FAE alone (OR=3.54, 95% CI=1.66–7.56; Figure 2). The minor complications were similar among both the groups while major complication was seen in 1 patient in the FAE group.

OPEN IN VIEWER

Figure 2.

Random-effects forest plot of ≥70% decrease in the size of draining vein/sac. The summary effect was expressed as odds ratio. Weights are from the random-effects model: continuity correction applied to studies with zero cells. CI, confidence interval; DL, DerSimonian-Laird.

Discussion

The present meta-analysis has consolidated and summarized the evidence that VSE with or without FAE is safe and a more effective approach than FAE alone. Venous sac embolization (with or without FAE) has outperformed FAE alone despite the fact that there were a greater number of complex PAVMs in the VSE group. Transcatheter embolization is the standard of care for treating PAVMs. ⁴ The benefit of embolotherapy includes mitigation of dyspnea and prevention of paradoxical embolization. Conventionally, the endovascular technique involves embolization of the distal end of the feeding artery, beyond the branches supplying the normal lung parenchyma and within 1 cm of the PAVM nidus. There is no consensus regarding the optimal embolic agent or the optimal site of embolization in relation to the nidus. ¹⁵ However, there are few concerns regarding FAE as it does not completely eliminate the risk of reperfusion. Since the nidus remains patent, it serves as a potential sink for adjacent collaterals without significant reduction in the risk of paradoxical embolization. ⁴ To encounter these issues, another technique for treatment of PAVMs was devised wherein embolization of the venous sac is performed with or without embolization of feeding artery which theoretically reduces the chance of recanalization of PAVM as the sac itself is occluded. In addition, VSE allows greater amount of embolic agent to be placed within the nidus compared to FAE where the amount of embolic agent that can be used is limited by the presence of a proximal non-target branch. ¹⁵

In the present meta-analysis, ≥70% reduction in draining vein/sac diameter was taken as a surrogate marker for completeness of occlusion according to the previously described computed tomography (CT) criteria by Lee et al and also adopted by 2 of the included studies.^5,6,16 This does not necessarily implicate cessation of flow with a possible residual risk of paradoxical-embolism–related events. However, using this threshold, there were no reports of events as stroke, transient ischemic attack, or brain abscess in 298 embolized PAVMs. Although the 70% metric seems to be imperfect, it is a clinically relevant one. ¹⁵ In the included studies, computed tomography angiography (CTA) was used for follow-up. The detection of reperfusion is challenging with CTA due to possible obscuration of sac resulting from metallic artifacts from the coil. There is evidence in the literature to suggest that catheter angiography or time-resolved magnetic resonance angiography (MRA) may be better imaging modalities for follow-up. ¹⁷ However, CTA is a readily available modality, fast and relatively cheap as compared to MRA which is less time and cost-efficient.

Although VSE appears to be more beneficial, it is associated with the following challenges. Various possible complications include coil dislodgement and in situ thrombosis with subsequent systemic embolization. Furthermore, the sac is prone to rupture during the packing of the sac with coils. ¹⁸ However, no evidence of coil migration or PAVM rupture was observed in any patient in the present meta-analysis. It is quite possible that the above-mentioned complications were prevalent in past decades when stiffer coils and detachable balloons were used. With the advent of microcatheters and super soft coils, the complication rates are negligible. ¹⁵ In the present meta-analysis, only minor complications were observed in either group, the most common being pleuritic chest pain. Due to greater number of coils being used in VSE, it adds to the cost of the procedure along with prolonging the procedure time and increasing the radiation dose. ¹⁹ However, this increased cost should be viewed in the light of expenses involved for a repeat embolization procedure in cases where the venous sac is not embolized. ¹⁵

The current systematic review and meta-analysis reveals 3 fair-quality single-center retrospective studies, reflecting a sustained benefit in the VSE group. The various limitations of this meta-analysis are as follows. First, the primary outcome measure of reduction in outflow vein diameter for “completeness” of occlusion is not ideal. ¹⁹ Reduction of the diameter of the draining vein is an indirect sign and is not a well-established outcome of success of embolization. The presence or absence of residual or recurrent flow should be the true gold standard that can only be evaluated on catheter angiography or time-resolved MRA. Second, in the included studies secondary to small sample size and absence of randomization, the operator preference would introduce a risk of selection bias. Hence, a prospective, multicentric, randomized controlled trial with an adequately powered sample size comparing the 2 techniques will provide more definitive evidence and conclusively resolve the equipoise. ¹⁹ In addition, better imaging surrogates of efficacy are warranted for developing an evidence-based approach to management. In all the 3 studies included, FAE was performed by coils which are known to have high reperfusion rates of 30% to 50%. Vascular plugs, which are shown to have high success rates with FAE alone, were not used in any of the included studies. Also, majority of the patients in the included studies were in the fifth to sixth decades of their lives despite the fact that PAVMs are frequently present since childhood. The response of PAVMs to coil embolization may not be identical to those observed in adults as it has been suggested that reperfusion rates post-embolotherapy may be higher in children.^20,21 Radiation exposure is another factor to be considered in children, and as a result, the use of CT/catheter angiographies should be kept to a minimum. ²¹

In conclusion, the present systematic review and meta-analysis has consolidated the evidence that VSE (with or without FAE) is more efficacious and superior than FAE alone in the management of PAVMs without any associated risk of sac rupture or coil migration. Until high-quality data from a prospective, multicentric, randomized controlled trial become available, the existing evidence may be taken into consideration while formulating the course of management in patients with PAVMs on an individual case-to-case basis.