DOI: https://doi.org/10.21203/rs.3.rs-2523576/v1
Background: Transcatheter arterial embolization (TAE) is a standard treatment for acute lower gastrointestinal bleeding (LGIB) in situations where endoscopic approaches are impossible or ineffective. Various embolic materials, such as metallic coils and N-butyl cyanoacrylate, are used. This study aimed to evaluate the clinical outcomes of an imipenem/cilastatin (IPM/CS) mixture as an embolic agent in TAE for acute LGIB.
Results: Twelve patients (mean age, 67 years) with LGIB treated with TAE using IPM/CS were retrospectively evaluated between February 2014 and September 2022. All patients showed evidence of extravasation on computed tomography and 50% (6/12) also showed evidence on angiography. The technical success rate for TAE in this study was 100%, including in patients who showed active extravasation on angiography. The clinical success rate was 83.3% (10/12), with two patients experiencing re-bleeding within 24 h after the procedure. No ischemic complications were observed and no bleeding episodes or other complications were reported during the follow-up period.
Conclusions: This study revealed that using IPM/CS as an embolic agent in TAE for acute LGIB may be safe and effective, even in cases of active bleeding.
Transcatheter arterial embolization (TAE) is a standard treatment for lower gastrointestinal bleeding (LGIB) in situations where endoscopic approaches are impossible or ineffective. Various embolic materials, such as metallic coils and N-butyl cyanoacrylate (NBCA), are used (Nagata et al. 2019; Oakland et al. 2019; Lee et al. 2022; Chevallier et al. 2021; Funaki et al. 2001; Kinoshita et al. 2021). However, in some cases, superselective embolization using coils may be impossible or unfeasible. In these situations, the use of NBCA may be considered, but it is associated with the risk of ischemic complications in non-superselective situations (Ikoma et al. 2010; Kodani et al. 2016). In 2013, the efficacy of imipenem/cilastatin (IPM/CS) as an embolization agent for treating acute LGIB was reported (Woodhams et al. 2013).
This study aimed to determine the effectiveness of a mixture of IPM/CS as an alternative embolization agent in TAE for acute LGIB in situations where superselective embolization using microcoils is impossible or unfeasible. In addition, this study also aimed to evaluate the clinical outcomes of using the IPM/CS mixture as an embolic agent in TAE for acute LGIB.
This single-center retrospective study evaluated the outcomes of TAE using IPM/CS as an embolic agent for acute LGIB. A total of 47 patients underwent angiography to evaluate acute LGIB, including 12 patients (nine males and three females; age range, 37–93 years; mean age, 67 years) treated with TAE using IPM/CS as an embolic agent. All patients were treated at the National Hospital Organization Disaster Medical Center in Tokyo, Japan, and provided written informed consent for the procedure, which included information regarding the potential risks and benefits.
TAE is typically performed when there is active extravasation in the lower gastrointestinal tract on a dynamic contrast-enhanced computed tomography (CT) scan and when catheter access is possible. Angiography was performed by a team of radiologists, including at least one diagnostic specialist who is a member of the Radiological Society and a specialist who is a member of the Japan Society of Interventional Radiology. These specialists have extensive training and experience in performing and interpreting angiography examinations and are authorized to be involved in planning and performing interventional procedures, including TAE.
All procedures were performed using an angiographic system (Allura Xper FD 20/10 or Azurion 7 M20, Philips Healthcare, Best, The Netherlands) under local anesthesia administered through the femoral artery using a 4- or 5-French (Fr) sheath (Medikit Super Sheath, Medikit, Tokyo, Japan; Radiforcus Introducer IIH, Terumo, Tokyo, Japan). A 4-Fr angiographic catheter shepherd hook-type catheter (SHK or SHA, Medikit, Tokyo, Japan) or 5-Fr guiding catheter Cobra (C1, Medikit, Tokyo, Japan) was injected into the superior mesenteric artery (SMA) or inferior mesenteric artery (IMA), and diagnostic angiography was performed to identify the bleeding site. Even if the bleeding site could not be identified using SMA or IMA arteriography, arteriography using the SMA or IMA branches (right colic, middle colic, ileocolic, and sigmoid arteries) or selective angiography using the vasa recta or marginal artery was performed using 1.9- to 2.9-Fr microcatheters (BISHOP HF, PIOLAX, Kanagawa, Japan; LEONIS Mova HF, Sumitomo Bakelite, Tokyo, Japan; Carnelian® MARVEL Non-Taper or Carnelian® MARVEL S, Tokai Medical Products, Aichi, Japan).
Our team had a standard approach for managing cases of acute LGIB using TAE. Our priority was to identify the bleeding vessel using contrast-enhanced CT and selective angiography, using the vasa recta or marginal artery; and, if possible, stopping the bleeding using superselective embolization with microcoils. If this was impossible or unfeasible, the team would use IPM/CS as an alternative embolic material. Based on this management technique, the final decision regarding which embolization agent to use and whether to perform the embolization procedure was at the operators’ discretion, based on their clinical judgment.
After the embolization procedure was completed, angiography was performed to confirm the technical success of the procedure and check for any other possible sources of bleeding.
The use of IPM/CS as an embolic agent is not typically covered by insurance in Japan; therefore, we obtained approval from the institutional review board before using this treatment.
We prepared an intravenous injection of a mixture of IPM/CS (0.5 g) and nonionic contrast medium (5 mL). The mixture was drawn into a syringe and gently pumped approximately 10 times to ensure it was mixed properly. The mixture was mainly injected through a high-flow microcatheter or a microcatheter placed in a marginal artery near the bleeding site until the blood flow in the bleeding site and the surrounding marginal artery became stagnant. However, in some cases, it was necessary to inject the mixture tightly to achieve this effect.
Technical success was defined as the success of the angiographic procedure based on the disappearance of extravasation or blood flow in the target artery. Clinical success was defined as the cessation of bleeding without needing additional hemostatic treatment within 7 days. Early recurrent bleeding was defined as bleeding from the treated area within 3 days after TAE. Complications from embolization procedures can be divided into two categories: "non-specific" and "specific". Non-specific complications are those that are due to the angiographic procedure itself, such as bleeding, infection, or allergic reactions to the materials used. Specific complications refer to those that are specific to the area being embolized, such as intestinal ischemia in the case of intestinal embolization.
Patients’ backgrounds are shown in Table 1. This study included 12 patients, of whom six were taking antithrombotic medications. The etiology of bleeding was colonic diverticulosis in eight patients and bleeding in the small bowel due to inflammatory, ulcerative, cancerous, or unknown causes in the remaining four patients. All but one patient had evidence of extravasation on the CT scan. This patient had micro-extravasation on a trans-arterial contrast-enhanced CT with a catheter in the SMA.
| n (%) | ||||
|---|---|---|---|---|
| Sex | ||||
| Male | 9 (75%) | |||
| Female | 3 (25%) | |||
| Age (range: 37–93 years) | ||||
| > 60 | 7 (58%) | |||
| ≦ 60 | 5 (42%) | |||
| Antithrombotic medications | ||||
| No | 6 (50%) | |||
| Yes | 6 (50%) | |||
| Etiology of bleeding | ||||
| colonic diverticulosis | 8 (68%) | |||
| inflammatory | 1 (8%) | |||
| ulcerative | 1 (8%) | |||
| cancerous | 1 (8%) | |||
| unknown | 1 (8%) | |||
| Site of bleeding | ||||
| small bowel | 4 (34%) | |||
| Right-sided colon | 5 (42%) | |||
| Left-sided colon | 1 (8%) | |||
| Transverse colon | 1 (8%) | |||
| Rectosigmoid | 1 (8%) | |||
The summary of TAE outcomes is shown in Table 2. Extravasation appeared on angiography in 50% of the patients (6/12), most of which occurred in the SMA, which supplies blood to the small intestine and right-sided colon. In all patients, the microcatheter was located at the marginal artery of the bleeding site during embolization. IPM/CS was used as the embolization agent for six patients due to the disappearance of extravasation on angiography and for the remaining six patients due to difficulty catheterizing the bleeding vasa recta.
| n(%) | ||||
|---|---|---|---|---|
| Angiographic extravasation | ||||
| No | 6 (50%) | |||
| Yes | 6 (50%) | |||
| Bleeding territory | ||||
| SMA | 10 (83%) | |||
| IMA | 2 (17%) | |||
| The reason why the IPM/CS was chosen | ||||
| disappearance of extravasation on angiography | 6 (50%) | |||
| difficulty in catheterizing the bleeding vasa recta | 6 (50%) | |||
The technical success rate for TAE in this study was 100% (12/12), which included patients in whom active extravasation was identified on angiography (Fig. 1, 2). The clinical success rate was 83.3% (10/12), with two patients experiencing re-bleeding within 24 h after the procedure. The physical examination and laboratory data showed no complications associated with the TAE technique or ischemia. In addition, no bleeding episodes or other complications were observed during the follow-up period.
IPM/CS is an antibiotic and was used as an embolic agent for chemoembolization in a 1999 animal study (Aihara 1999; Aihara 1999). Previous studies have reported that IPM/CS may have unique characteristics that make it a useful embolic agent, including the production of particles predominantly < 60 µm in size and the relatively short embolic effect of < 48 h (Aihara 1999; Yamada et al. 2021). In 2013, its efficacy as an embolization agent for treating acute LGIB was reported (Woodhams et al. 2013). Since 2014, IPM/CS has been used as an alternative embolic agent by our team in cases where superselective embolization using microcoils was difficult to achieve.
In the present study, we initially attempted to use the IPM/CS mixture for TAE to treat LGBI by injecting the mixture until extravasation or blood flow of the bleeding vasa recta ceased. This is because TAE using conventional embolization materials has been reported to potentially cause ischemic damage if embolization is performed on > 3 vasa recta. However, this approach resulted in re-bleeding within 24 h in the first two patients. Therefore, we changed our approach and injected the mixture until the blood flow of the bleeding vasa recta and the surrounding vasa recta became stagnant. This new approach successfully prevented re-bleeding in the patients, including the five patients with massive extravasation on angiography. No ischemic complications were reported.
A histological study of the effects of IPM/CS as an embolic agent in animals reported that almost all microvessels < 60 µm in diameter were embolized immediately after the injection of the IPM/CS mixture and that the mixture was found in some relatively large arterial main trunks. However, after 5 min, no embolic material was found in the main arterial trunks, only in the microvessels (Aihara 1999). This suggests that IPM/CS is effective as a micro-embolizing agent but may not be as effective in larger vessels.
Another study on the safety of proximal embolization using an IPM/CS mixture in humans was conducted to treat tendinopathy and enthesopathy (Okuno et al. 2013; Fujiwara et al. 2021). The study revealed that even when IPM/CS was injected from the proximal point of the branches feeding the target site, no complications associated with IPM/CS embolization occurred. This finding supports the safety of IPM/CS as an embolic agent. In our study, no ischemic complications were observed, and no bleeding episodes or other complications were reported during the follow-up period.
However, IPM/CS has some disadvantages as an embolization agent, including a lack of insurance coverage in Japan and a transient embolic effect that may lead to re-bleeding. IPM/CS may be effective in occluding small blood vessels; however, it has a transient effect, indicating that it is only effective for a limited period (< 48 h). This may increase the risk of re-bleeding because the blood vessel may re-open after IPM/CS has been removed. If superselective catheterization can be achieved, embolization using microcoils or other conventional embolic materials may be preferred because they have a longer-lasting effect. However, if superselective embolization is impossible, IPM/CS is a viable option for stopping the bleeding in cases of acute LGBI.
This study had some limitations, including its retrospective design and small sample size. In addition, only IPM/CS was used as the embolic agent in this study; therefore, comparing the results with those obtained using other embolic materials is impossible. The possibility that colonoscopy after TAE was not routinely performed might have resulted in some ischemic complications being undetected if they did not cause clinical symptoms. These limitations may affect the reliability and generalizability of this study's findings.
This study revealed that using IPM/CS as an embolic material in TAE for acute LGIB may be safe and effective, even in cases of active bleeding.
CT: computed tomography
Fr: French
IMA: inferior mesenteric artery
IPM/CS: imipenem/cilastatin
LGIB: lower gastrointestinal bleeding
NBCA: N-butyl cyanoacrylate
SMA: superior mesenteric artery
TAE: transcatheter arterial embolization
Ethics approval and consent to participate
All patients provided written informed consent for the procedure, which included information regarding the potential risks and benefits. This study was approved by the institutional review board.
Consent for publication
For this type of study, consent for publication is not required.
Availability of data and material
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests
The authors declare that they have no interests.
Funding
The authors received no financial support for the preparation of this article for publication.
Authors' contributions
All authors engaged in treatment and contributed to the accumulation of cases. FS and YI contributed to the composition of the dissertation and the content of the discussion. All authors have read and approved the final manuscript.
Acknowledgements
We would like to thank Editage (www.editage.com) for English language editing.
Authors' information
Department of radiology, National Hospital Organization Disaster Medical Center, 3256, Midoricho, Tachikawa, Tokyo, 190-0014, Japan
Sakiko Hiraki, Fumie Sato, Masaya Osugi, Yoshiya Watanabe, Yoshiaki Ichinose