Combined DSA- and US-guided management of acute bleeding: effectiveness of percutaneous glue embolization in six cases

Abstract

Aims

This study aims to report on safety and effectiveness of glue embolization of acute hemorrhages performed by US-guided percutaneous direct puncture, in patients where the standard endovascular approach is technically unfavourable.

Methods

In this single center retrospective analysis, patients affected by traumatic or non traumatic acute hemorrhages were treated with glue embolization technically performed by US-guided direct puncture. Patients suffered from active bleeding detected at contrast-enhanced Computer Tomography and confirmed at Digital Subtracted Arteriography, with concomitant hemoglobin drop and blood pressure reduction. Six patients were reviewed; hemorrhages occurred in liver parenchima (1), gallbladder (1) and lower limbs (4). Bleedings etiologies were post-traumatic (5) and inflammatory (1); four had extraluminal blushes while 2 were pseudoaneurysms. In five cases (4 limbs and 1 hepatic bleedings) the direct-puncture approach was adopted because of technically unfavourable vascular anatomy for superselective embolization; in one case (cystic artery pseudoaneurysm), an endovascular embolization would entail a high risk of gallbladder ischemia because of its terminal arterial supply. Technical success was defined disappearance of bleeding signs at last arteriography; clinical success was considered stabilization and/or improvement of hemoglobin values and arterial pressure without additional interventions.

Results

Both technical and clinical successes were obtained in 100% of the cases without major complications.

No needle occlusion occurred during glue injection.

Conclusions

In this study percutaneous US-guided embolization by direct puncture was a safe and effective approach to manage acute bleedings; it could be considered as an alternative in patients with unfavourable vascular anatomy for the standard catheter-directed endovascular embolization.

Introduction

Endovascular arterial embolization is nowadays considered a mainstay in the non-operative management of active bleedings, both traumatic and non traumatic, included in multiple guidelines and standards of practice documents [1, 2].

The technological developments in the last decades have allowed to perform superselective procedures with high precision and clinical success. However, sometimes challenging vascular anatomies, especially in the elderly population, can make the endovascular approach technically demanding and time consuming with consequent increase in X-ray exposure and costs; furthermore, in visceral districts with terminal arterial supply, endovascular embolization may induce organ necrosis.

In these cases, if the hemorrhagic lesion is appreciable at ultrasound (US), a percutaneous US-guided direct puncture embolization could be considered as an alternative; indeed, some case series in literature have already described the direct puncture embolization, under US-guidance or computed tomography (CT)-guidance, in elective scenarios [3–9] with encouraging results.

This study aims now to report on safety and effectiveness of glue embolization of acute hemorrhages performed by combined digital subtracted arteriography (DSA)- and US-guided percutaneous direct puncture, in those cases where endovascular approach is technically unfavourable.

Materials and methods

Sample

This is a single center retrospective analysis of patients affected by traumatic or non traumatic acute hemorrhages and treated with glue embolization technically performed by US-guided direct puncture.

The institutional ethical committee approved this study; all patients signed a written informed consent before the procedure.

Patients were referred to the Interventional Radiology department because of active bleeding detected at contrast-enhanced computer tomography (CT), concomitant hemoglobin drop and blood pressure reduction.

Six patients (Table 1), 2 women and 4 men, mean age: 79.3 (range: 71–87 years) were reviewed; one was affected by liver parenchima active bleeding, one by intramural gallbladder hemorrhage and 4 by lower limbs hemorrhages. Bleedings etiologies were post-traumatic (5) and inflammatory (1); four had extraluminal blushes, while 2 were pseudoaneurysms.

Table 1.

Sample data

Pt	Sex	Age	Lesion location	Lesion etiology	Lesion type	Antiplatelet therapy
1	F	87	Leg	Trauma	Blush	No
2	M	71	Gallbladder	Inflammatory	PA	No
3	M	80	Knee	Trauma	Blush	Yes
4	M	73	Leg	Trauma	PA	Yes
5	F	86	Liver	Trauma	Blush	No
6	M	79	Thigh	Trauma	Blush	No

Pt patient, F female, M male, PA pseudoaneurysm

Two patients assumed antiplatelet therapy.

Technique

All procedures were performed by an interventional radiologist with more than 7 years of experience.

A preprocedural contrast-enhanced CT scan was performed in all cases.

The whole interventional procedure was accomplished in the angio-suite. First, an endovascular approach through a 5Fr femoral arterial access was conducted; the target district was catheterized and selective DSA was obtained; signs of active bleeding (pseudoaneurysm or contrast extravasation) were detected. After uneffective attempts to reach superselectively and/or safely the target artery, the standard catheter directed endovascular embolization was considered technically unfeasible and time consuming because of vascular anatomical features (Fig. 1) or risky in terms of post-embolization necrosis (Fig. 2); so, the endovascular procedure was suspended leaving the diagnostic catheter onsite, switching to US-guided direct puncture approach.

Fig. 1.

80 years old male, assuming antiplatelet therapy, with hematoma of the right knee associated with active bleeding. A Axial CT scan showing contrast agent extravasation (white circle). B DSA of the femoral artery confirms arterial blush (white circle); the reversed acute origin of the external sovragenicular artery is appreciable (black thick arrow), unfavourable for superselective microcatheter selection. C Preparation of the mixture of glue and Lipiodol by three-way stopcock and glucosate into the syringes, ready to be injected. D US-guided (black asterisk) puncture of the lesion with a 20 G 10 cm needle (black arrow), using the free-hand technique. E US scan showing knee hematoma (white asterisk) and the echogenic needle tip positioned into an anechoic area (black arrow). F Blood drop outsourcing from needle hub (white arrow) confirming proper needle position (black arrow). G Fluoroscopic scan showing glue injection (white dotted circle) through the needle (black arrow). H Final DSA run detecting arterial blush resolution, glue cast being appreciable (white dotted circle)

Fig. 2.

71 years old male with acute cholecystitis, causing intramural pseudoaneurysm associated with gallbladder hemobilia. A US scan showing gallbladder lumen fulfilled with blood (white asterisk) and an anechoic round-shaped pseudoaneurysm (white circle). B Color-Doppler scan confirming arterial flow into the lesion (white circle). C Arterial phase axial CT scan confirming intramural gallbladder hematoma (white asterisk) and pseudoaneurysm (white circle). D US-guided puncture of the pseudoaneurysm, the echogenic tip of the 21 G 15 cm needle being evident (black arrow). E After percutaneous cholecystostomy drainage positioning (white arrow), under fluoroscopy glue was injected into the pseudoaneurysm (white dotted circle) through the needle (black arrow). F Final hepatic DSA detecting pseudoaneurysm resolution, glue cast being appreciable (white dotted circle); white arrow indicating cholecystostomy drainage

At US, in case of extraluminal blush the hemorragic lesion appeared as an anechoic focus with pulsatile arterial sign at color-Doppler, surrounded by disomogeneous hematoma (Fig. 1); in case of pseudoaneurysm, a round-shaped anechoic lesion with arterial pattern at color-Doppler US was appreciable (Fig. 2).

A convex probe (3.5 MHz) was adopted in case of liver procedures while a linear probe (7.5 MHz) in case of inferior limb district.

Once the target lesion was recognized at US, in sterile conditions, local anesthesia (10 ml Mepivacaine) was performed; then, a 20G or a 21G needle, different lenghts (10 cm or 15 cm) according to target depth, was selected to puncture the bleeding foci. Needle tip was scrubbed to improve its US visibility [10], while needle lumen was flushed with glucosate 5% to avoid that lumen stucked. Under US-guidance the needle tip was positioned into the target with a free-hand technique. Fluoroscopic check confirmed the proper matching between needle tip position and angiographic site of bleeding; furthermore, blood drop outsourced from needle hub after removing the needle stylet.

Glue embolization was performed by injecting a mixture of N-butyl-cianoacrylate (NBCA, Glubran2-GEMS^®, Italy) mixed with Lipiodol (Guerbet^®, France) (ratio 1:2), under combined fluoroscopic- and US-guidance; a homogeneous cast of glue was obtained by applying a continuous and gentle pression on the syringue piston. The glue injection speed was faster compared to the rate usually adopted in the endovascular setting, to avoid needle lumen occlusion.

Arteriographic checks from the diagnostic catheter, left previously onsite, were performed to detect bleeding resolution; if bleeding signs persisted at DSA, needle glue injection was repeated. Once bleeding source was solved, needle was withdrawn under continuous glue injection to seal also the needle percutaneous tract.

Finally, a bandaid and a compressive elastic dressing were positioned on the needle skin puncture and on the femoral arterial access respectively.

In five cases (4 limbs and 1 hepatic bleedings) the direct-puncture approach was adopted because of technically unfavourable vascular anatomy for superselective embolization: the target vessel presented a stenotic origin facing with an acute angle opposite to catheter direction (Fig. 1).

In one case (cystic artery pseudoaneurysm), an endovascular embolization would entail a high risk of gallbladder ischemia because of its terminal arterial supply, therefore the sole pseudoaneurysm occlusion was performed by direct puncture (Fig. 2).

Outcomes definition

Technical success was considered as disappearance of bleeding signs at final DSA control.

Clinical success was considered as stabilization and/or improvement of the hemoglobin values and arterial pressure values without additional interventions during the hospital recovery.

Intra- and post-procedural complications were evaluated according to the CIRSE classification system for complications [11].

Results

Technical success was achieved in 100% of the cases; no residual bleeding signs were evident at the last DSA control and no additional embolization maneuvers were necessary after glue injection. Furthermore, a post-procedural CT scan performed at 24–48 h confirmed technical success and excluded other sources of bleeding.

Clinical success was obtained in 100% of the cases; blood pressure parameters improved after the procedure while hemoglobin values rised in the follow-up monitoring during the hospital recovery.

No needle occlusion occurred during glue injection.

No major complications were recorded; one minor complication (grade I) was registered: in one case, a small amount of glue migrated into an inferior limb vein while withdrawing the needle, without clinical sequelae.

Discussion

In this sample, glue embolization by combined DSA- and US-guided percutaneous direct puncture was a safe and effective approach to treat active bleeding in patients where the standard endovascular strategy was technically unfavourable. Technical and clinical successes were achieved in all six patients without major complications.

The percutaneous approach was adopted in all cases as a secondary strategy; therefore it was never planned in advance.

The needle caliper (20 G or 21 G) allowed proper handling and US visibility; at the same time, it reduced the risk of tissues damage.

Glue was choosen as embolic agent, being a cost-effective and permanent agent, especially for patients under anticoagulation therapy. The dilution ratio with lipiodol was low (1:2) because the needle tip was already into the target and no distalization was required; this allowed to reduce the rate of non-target embolization, occurring only in one case without sequelae.

This technique has been already described in literature, especially in neuroradiological procedures with published case series regarding patients affected by hypervascularized tumors of head and neck districts: Pedicelli et al. [3] reported successful direct puncture preoperative embolization with SQUID 12^® in 11 patients; similarly, Pérez-Garcìa et al. [4] treated 6 patients with carotid paraganglioma, while Elhammady et al. [5] used Onyx^® in 18 patients.

In peripheral interventions, literature data report mainly about case reports. In elective setting, direct puncture embolization has been proposed for type IIIb arterio-venous malformations [6], for type-II endoleaks [7, 8] and iliac aneurysm unaccessible via endovascular approach [9] under CT guidance. US-guided direct puncture embolization has been proposed also in stomal variceal hemorrhages by Thouveny et al. [12] in 7 patients and by Dolan et al. [13] in one patient using glue, while Ryan et al. [14] used coils and gelfoam in one patient. Different case reports have described successfull US-guided direct puncture embolization of pseudoaneurysms located in the liver with coils [15], coils and thrombin [16] or glue [17]; in the gastroduodenal artery with thrombin [18]; in a superior mesenteric artery branch with glue [19]; in the inferior gluteal artery with coils [20]; in a radiocephalic arteriovenous dialytic fistula pseudoaneurysm with wide neck managed with thrombin injection after endovascular balloon inflation [21].

Contrast-enhanced ultrasound (CEUS) could have a potential relevant role in this scenario. Lv et al. [22] have reported their experience in percutaneous hemostasis of blunt hepatic and splenic trauma with interesting results: they injected successfully cyanoacrylate under sole CEUS-guidance without concomitant DSA acquisitions. Furthermore, CEUS has been proposed to detect hemorrhages in suspected bleeding patients, especially immediately after abdominal parenchimal ablation or biopsy, that could be subsequently managed with CEUS-guided percutaneous embolization [23] and also in splenic lesions [24]. Therefore, the role of ultrasound in interventional radiology procedures could expand well beyond the already known techniques of image guidance for biopsy/ablation [25] and peripheral pseudoaneurysm compression [26].

This is now the first series of patients affected by peripheral hemorrhagic lesions treated with percutaneous US-guided direct puncture glue embolization. This approach should be considered as a secondary alternative in case of unfavourable vascular anatomies for standard endovascular embolization, as seldomly occurs in elederly population; in these scenarios, it could reduce procedural time and save devices related costs.

The main technical drawback is that it can be applied only in districts accessible with US; furthermore operators should be skilled both in US-guided interventions and glue handling.

This study presents limitations: first, the number of analyzed patients is low; then, a direct comparison with the endovascular approach in terms of clinical outcome, procedural time and costs has not been performed.

Conclusions

In conclusion, in this study percutaneous combined DSA- and US-guided embolization by direct puncture was a safe and effective approach to manage acute bleedings; it could be considered as an alternative in patients with unfavourable vascular anatomy for the standard catheter-directed endovascular embolization. Further experiences with larger samples are needed to confirm the reported findings.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Research involving human participants

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual patients included in the study.