- Email: [email protected]
Vascular Plug–Assisted Retrograde Transvenous Obliteration for the Treatment of Gastric Varices and Hepatic Encephalopathy: A Prospective Multicenter Study
CLINICAL STUDY
Vascular Plug–Assisted Retrograde Transvenous Obliteration for the Treatment of Gastric Varices and Hepatic Encephalopathy: A Prospective Multicenter Study Dong Il Gwon, MD, Young Hwan Kim, MD, Gi-Young Ko, MD, Jong Woo Kim, MD, Heung Kyu Ko, MD, Jin Hyoung Kim, MD, Ji Hoon Shin, MD, Hyun-Ki Yoon, MD, and Kyu-Bo Sung, MD
ABSTRACT Purpose: To evaluate technical and clinical outcomes of vascular plug–assisted retrograde transvenous obliteration (PARTO) for the treatment of gastric varices (GVs) and hepatic encephalopathy (HE). Materials and Methods: From March 2012 to June 2014, 73 consecutive patients (47 men, 26 women; mean age, 59 y; range, 28–79 y) who had undergone PARTO were evaluated in a prospective multicenter study. Among 57 patients with GVs, 28 had GVs in danger of rupture, 23 had experienced recent bleeding, and 6 had active variceal bleeding. The 16 patients with HE had been treated unsuccessfully with medical therapies. Results: Placement of the vascular plug and subsequent gelatin sponge embolization were technically successful in all 73 patients. There were no procedure-related complications. Follow-up CT obtained within 1 wk after PARTO showed complete thrombosis of GVs and portosystemic shunts in 72 of 73 patients (98.6%). Sixty patients who underwent follow-up longer than 3 mo showed complete obliteration of GVs and portosystemic shunts. There were no cases of variceal bleeding or HE at the end of follow-up (mean, 544 d). Improvement in Child–Pugh score was observed in 24 patients (40%) at 1-mo follow-up. Worsening of ascites and esophageal varices was observed in 14 (23.3%) and 16 (26.7%) patients at 3-mo follow-up. Conclusions: The present results of PARTO indicate that it can be rapidly performed with high technical success and durable clinical efficacy for the treatment of GVs and HE in the presence of a portosystemic shunt. Therefore, PARTO might be considered a first-line treatment in appropriate patients.
ABBREVIATIONS BRTO = balloon-occluded retrograde transvenous obliteration, EV = esophageal varix, HE = hepatic encephalopathy, GV = gastric varix, PARTO = vascular plug–assisted retrograde transvenous obliteration
Bleeding caused by gastroesophageal varices is a major complication of portal hypertension in patients with From the Department of Radiology and Research Institute of Radiology (D.I.G., G.Y.K., J.W.K., H.K.K., J.H.K., J.H.S., H.K.Y., K.B.S.), University of Ulsan College of Medicine, Asan Medical Center, 86, Asanbyeongwon-gil, Songpa-gu, Seoul 138-736, Korea; and Department of Radiology (Y.H.K.), Keimyung University School of Medicine, Dongsan Medical Center, Daegu, Korea. Received May 15, 2015; final revision received and accepted July 14, 2015. Address correspondence to D.I.G.; E-mail: [email protected] None of the authors have identified a conflict of interest. & SIR, 2015 J Vasc Interv Radiol 2015; XX:]]]–]]] http://dx.doi.org/10.1016/j.jvir.2015.07.011
cirrhosis (1). Although the risk of bleeding from gastric varices (GVs) is relatively low, when rupture of GVs has occurred, the outcome is worse, with a higher mortality rate than seen with esophageal varices (EVs) because there is more severe blood loss and a higher recurrent bleeding rate (1–3). Bleeding of GVs is associated with mortality rates as high as 55%. As most of the gastric fundal varices drain into the left renal vein via a gastrosystemic shunt, the concept of retrograde injection of sclerosing agents into GVs after balloon occlusion of the gastrorenal shunt has been introduced (4). Moreover, gastrosystemic shunts may result in hepatic encephalopathy (HE), a debilitating condition that may be refractory to medical management. The key factor in
2
’
Vascular Plug–Assisted Retrograde Transvenous Obliteration
the pathogenesis of HE is shunting of undetoxified portal blood into the systemic circulation (5,6). Therefore, balloon-occluded retrograde transvenous obliteration (BRTO) has been shown to be a suitable therapeutic option for the control of GVs and HE because a natural interventional target is closure of the GVs and portosystemic shunt (7–14). However, previous studies reported complications associated with the use of liquid and foam sclerosant agents (8,9,13–18). In addition, balloon rupture during the BRTO procedure can result in symptomatic pulmonary edema, treatment failure, and ultimately recurrent variceal bleeding (19,20). Therefore, the absence of a balloon occlusion catheter and sclerosant agents can reduce procedure time, reduce procedure-related morbidity, and simplify procedure-associated logistics. Recently, a modified BRTO procedure was proposed and reported in which a balloon occlusion catheter and sclerosant agents were replaced with a vascular plug/coils and gelatin sponge to minimize some of the complications and logistical issues associated with the balloon catheter (21–23). A study by Gwon et al (21) reported that vascular plug–assisted retrograde transvenous obliteration (PARTO) successfully induced thrombosis of the gastrorenal shunt and GVs with no complications and subsequent obliteration. However, this study (21) has its own limitations (retrospective nature with a small number of study patients). Therefore, the purpose of the present study is to evaluate the technical and clinical outcomes of PARTO for the treatment of GVs and HE in a prospective multicenter study.
Gwon et al
Patient Population A prospective, uncontrolled multicenter study was conducted to evaluate the safety and efficacy of PARTO in 73 consecutive patients (47 men, 26 women; mean age, 59 y; range, 28–79 y) with GVs or HE from March 2012 to Jun 2014. Two academic tertiary referral centers participated in the study. The study was approved by the institutional review board at each participating hospital, and informed consent was obtained from each patient. In each patient, portosystemic shunt was evaluated by contrast-enhanced computed tomography (CT), and GVs were confirmed by endoscopy and CT. Patients were included if they had GVs or HE with a portosystemic shunt. Exclusion criterion included GVs or HE without a portosystemic shunt. Among these 73 patients, 57 had a portosystemic shunt with GVs and 16 had HE. Among 57 patients with GVs, 28 had GVs in danger of rupture, 23 had experienced recent bleeding, and six had active variceal bleeding. GVs showing marked protrusion or red spots and progressive enlargement within a 6-month period on endoscopy were judged to be in danger of rupture. The six patients with ruptured
JVIR
GVs had undergone emergency PARTO as a result of unsuccessful endoscopic hemostasis. Among 16 patients with HE, 11 had only a portosystemic shunt and five had a portosystemic shunt with minimal GVs. The 16 patients who had HE had been treated unsuccessfully with medical therapies such as lactulose and oral branched-chain amino acid supplementation. HE severity was assigned a grade per West Haven criteria: of the 16 patients with HE, four had grade 1, six had grade 2, three had grade 3, and three had grade 4 HE. The baseline demographic and clinical characteristics of the 73 patients are presented in the Table.
Technique Before PARTO, contrast-enhanced CT scans of the abdomen were obtained for all patients, and they confirmed the presence of a portosystemic shunt (Fig 1a). All PARTO procedures were performed with the patient under conscious sedation with intravenous pethidine hydrochloride (Demerol; Keukdong Pharmaceuticals, Seoul, Korea) and local anesthesia with lidocaine (Jeil Pharmaceuticals, Taegu, Korea). The right common femoral vein was punctured initially, after which a 6- or 7-F sheath (Flexor Check-Flo sheath; Cook, Bloomington, Indiana) was inserted. When insertion of a sheath into the portosystemic shunt was not possible through the right common femoral vein because of an acute angle, the Table . Patient Characteristics Characteristic
MATERIALS AND METHODS
’
GVs
HE
37:20 58
10:6 62
38 6
12 2
Alcoholic cirrhosis
8
2
LC/membranous IVC obstruction Other*
2 3
– –
28
11
Sex (M:F) Mean age (y) Underlying liver disease HBV cirrhosis HCV cirrhosis
Concomitant malignancy HCC Child–Pugh classification A
23
–
B C
26 8
13 3
8.2 ⫾ 1.8
7.8 ⫾ 1.4
Indications for GVs Acute bleeding
6
–
Chronic bleeding
23
–
Primary prophylaxis
28
–
MELD score
HE ¼ hepatic encephalopathy; GV ¼ gastric varix; HBV ¼ hepatitis B virus; HCC ¼ hepatocellular carcinoma; HCV ¼ hepatitis C virus; IVC = inferior vena cava; LC ¼ liver cirrhosis; MELD ¼ Model for End-stage Liver Disease. *Biliary cirrhosis (n ¼ 1), splenic vein thrombosis caused by chronic pancreatitis (n ¼ 1), unknown cause (n ¼ 1).
Volume XX
’
Number X
’
Month
’
2015
3
Figure 1. Images from a 55-year-old man with GVs. (a) Contrast-enhanced CT obtained before PARTO shows GVs (asterisk). Note the hypertrophied left gastric vein (arrowhead). (b) After placement of the vascular plug (white arrow) within the narrowest portion of the portosystemic shunt via the left adrenal vein, additional embolization of the gastrorenal shunt, GVs (asterisk), and left gastric vein (arrowhead) was performed by using gelatin sponge particles through the 4-F catheter (black arrow). (c) Contrast-enhanced CT scan obtained 3 months after PARTO shows complete obliteration of the GVs.
right internal jugular vein was then accessed. After negotiation of the left adrenal vein or shunt with a 0.035-inch, 180-cm-long hydrophilic guide wire (Terumo, Tokyo, Japan) and a 4-F angled-tip catheter (Cobra; Terumo), the sheath was then inserted into the shunt. The guide wire and catheter were subsequently removed, and the vascular plug (AMPLATZER Vascular Plug II; AGA Medical, Golden Valley, Minnesota) was deployed. The size of vascular plug was chosen based on the diameter of the narrowest shunt measured on CT, ranging from 8 to 20 mm in diameter or 20% larger than the targeted shunt to prevent migration and gelatin sponge reflux. To obtain easy negotiation of the guide wire and the catheter into the shunt proximal to the vascular plug, the vascular plug was deployed at the most dilated portion of the shunt. The shunt proximal to the vascular plug was then negotiated by using the guide wire and the 4-F angledtip catheter between the vascular plug and the shunt wall. The vascular plug was then pulled down to the narrowest part of the shunt while maintaining the position of the catheter. When contrast medium injection through the catheter showed proper position of the vascular plug at the narrowest portion, an additional embolization was then performed with a mixture of contrast medium and gelatin sponge particles (ranging in size from approximately 1 mm3 to 8 mm3) through the catheter to embolize the shunt, efferent veins, afferent veins, and GVs (Fig 1b). After embolization, retrograde venography through the sheath was performed to confirm occlusion of the shunt. The catheter was subsequently removed, and the delivery cable attached to the end of the vascular plug was then detached.
Follow-up Patients were evaluated by CT during their hospital admission within 1 week after PARTO to evaluate thrombosis of the portosystemic shunt and GVs. Subsequently, clinical and laboratory examinations were
performed 1, 3, 6, and 12 months after the procedures in all patients. Upper-intestinal endoscopy and CT were also performed 3 months after PARTO (Fig 1c). If the shunt and GVs were not completely obliterated at 3month follow-up CT, follow-up CT was performed 6 months after PARTO. Beginning 12 months after PARTO, information regarding patients’ current status or death was obtained by telephone from all patients or their families.
Definitions and Statistical Analysis Primary study endpoints were assessment of technical success, procedure-related complications, and clinical success. Secondary study endpoints were assessment of follow-up clinical results including change of liver function, worsening of EVs, and incidence of ascites. Technical success was defined as successful placement of the vascular plug within the left adrenal vein or the portosystemic shunt and administration of the gelatin sponge particles into the portosystemic shunt and GVs. Complications were classified as major or minor according to the guidelines of the Society of Interventional Radiology Standards of Practice Committee (24). Clinical success was defined as the control of active GV bleeding or complete resolution of all clinical symptoms of HE with complete thrombosis of GVs and portosystemic shunt. Complete resolution of HE was defined as a return to grade 0 HE according to the West Haven criteria as well as normalization of the serum ammonia (ie, NH3) level. Complete obliteration of GVs was defined as the complete eradication of the targeted GVs on follow-up CT or endoscopy 2 months after PARTO or at any time thereafter. Recurrent bleeding from GVs was defined as the presence of hematemesis or melena with endoscopic visualization of the bleeding source originating from the GVs after percutaneous intervention. The paired-sample t test was used to compare preand post-PARTO serum venous ammonia, bilirubin, and
4
’
Vascular Plug–Assisted Retrograde Transvenous Obliteration
albumin levels. Statistical analysis was performed by using SPSS software (version 14.0; SPSS, Chicago, Illinois), with P values less than .05 considered to be statistically significant.
RESULTS Technical Success and Complication The transfemoral approach was used in 66 patients and the transjugular approach was used in seven. Placement of the vascular plug and subsequent embolization by gelatin sponge were technically successful in all 73 patients (Fig 2). A single AMPLATZER Vascular Plug II was used to occlude the left adrenal vein or portosystemic shunt in 70 patients (8 mm [n ¼ 9], 10 mm [n ¼ 7], 12 mm [n ¼ 9], 14 mm [n ¼ 22], 16 mm [n ¼ 11], 18 mm [n ¼ 5], 20 mm [n ¼ 4], and 22 mm in diameter [n ¼ 3]). In the remaining three patients who each had two prominent splenorenal shunts draining into the left adrenal vein, two vascular plugs (one 10 mm and one 8 mm in diameter in one patient, and two 8 mm in diameter in another patient) were used in two patients, and one vascular plug (14 mm) and seven coils were used in one patient. During embolization of portosystemic shunt with/without GVs, 67 patients showed single or multiple collateral veins, and these collateral veins were successfully embolized by gelatin sponge only. However, gelatin sponge embolization was insufficient in two patients because of a prominent left inferior phrenic vein. Additional embolization with microcoils was performed in these two patients. None of the vascular plugs migrated after detachment. The mean procedure time from vascular plug placement to vascular plug detachment was 24 minutes, with a range of 11–124 minutes. There were no procedure-related complications in any patient.
Clinical Success Follow-up CT obtained within 1 week after PARTO showed complete thrombosis of the treated GVs and the
Gwon et al
’
JVIR
gastrorenal shunts in 72 of 73 patients (98.6%). One patient had residual intramural GVs on 1-week followup CT. Additional treatments were not performed, and 12 days after PARTO this patient died of massive hematemesis as a result of recurrent bleeding of GVs. The clinical symptoms of HE completely resolved in all 16 patients with HE, and the mean serum NH3 level decreased significantly (P o .001) from 118.4 mmol/L ⫾ 53.6 before PARTO to 29.8 mmol/L ⫾ 15.9 within 1 week after PARTO (95% confidence interval, 66.3–111.0 mmol/L). Therefore, the clinical success rate was 98.6% (72 of 73 patients).
Follow-up Clinical Results A schematic diagram of clinical outcomes is shown in Figure 3. One patient with GVs underwent liver transplantation 1 month after undergoing PARTO. Two patients were lost to follow-up within 3 months after PARTO. Clinical follow-up until death or the end of the study was available for the remaining 70 patients and lasted 7–1,050 days (mean, 544 d). The cutoff date for data analysis was January 31, 2015. At the end of follow-up, 19 patients had died and 51 remained alive. Ten patients with GVs died within 3 months after PARTO (range, 0.5–3 mo), five from deteriorating hepatic function, three from progression of hepatocellular carcinoma, and two from rupture of a hepatocellular carcinoma. Five patients with HE and four with GVs died more than 3 months after PARTO (range, 7.5–16 months), six from deteriorating hepatic function and three from progression of hepatocellular carcinoma. Laboratory measurements at 1 and 3 months were available for 60 patients. Improvement in Child–Pugh score was observed in 24 of 60 patients (40%) at either follow-up. However, there was no significant change in laboratory values at either follow-up. Follow-up laboratory measurements at 6 and 12 months were available for 51 patients. A 3-month follow-up CT was available for 60 patients and demonstrated complete obliteration of thrombosed
Figure 2. Images from a 68-year-old man with HE caused by a splenorenal shunt. (a) Contrast-enhanced CT shows a hypertrophied splenorenal shunt (asterisks). (b) After placement of a vascular plug (white arrow) within the left adrenal vein, additional embolization of the splenorenal shunt (asterisks) was performed by using gelatin sponge particles through the 4-F catheter (black arrow). (c) Contrastenhanced CT obtained 3 months after PARTO shows complete obliteration of the splenorenal shunt.
Volume XX
’
Number X
’
Month
’
2015
5
Figure 3. Schematic diagram of clinical outcomes.
GVs and portosystemic shunts in 57 patients and marked shrinkage in three patients. A 6-month follow-up CT examination was performed in these three patients and demonstrated complete obliteration of thrombosed GVs and portosystemic shunts. However, ascites newly developed in nine of 60 patients (15%) who did not have ascites before PARTO, and ascites progressed to a larger amount in five of 60 patients (8.3%) who had ascites before PARTO. A 3-month follow-up endoscopy revealed marked shrinkage or disappearance of GVs in all 60 patients. None of these 60 patients experienced bleeding of GVs or HE during the follow-up period. However, 16 patients (26.7%) experienced worsening of EVs. The change in form was from straight and small varices to winding varices in 10 patients, and from no varices to straight and small varices in six patients. Seven of the 10 patients with winding varices had EVs with the appearance of red spots on the esophageal mucosa and were successfully treated endoscopically.
DISCUSSION In the present study, we found a technical success rate (100%) and clinical success rate (98.6%) of PARTO that were in concordance with the result of BRTO studies
with the use of various sclerosing agents (7–14). In those BRTO studies (7–14), clinical success rates ranged from 79.6% to 100% and the relapse rate ranged from 0% to 10%. We also found that there was complete thrombosis of the GVs and portosystemic shunts in the 72 of 73 patients within 1 week after PARTO and complete obliteration of the GVs and portosystemic shunts in all 60 patients who had more than 2 months of follow-up. Among these 60 patients, we observed no cases of variceal bleeding in the 46 patients with GVs and no recurrence of HE in 16 patients. If there was no immediate blood supply into the completely thrombosed GVs and portosystemic shunt from the afferent veins that were embolized by a vascular plug and gelatin sponge, the thrombosed GVs and shunt were completely obliterated. As ruptured GVs are associated with high rates of recurrent bleeding and mortality, quick treatment is essential. Emergency BRTO has been shown effective for rescue in an emergency condition in several reports (25–28). Unlike prophylactic BRTO, it is sometimes difficult to perform emergency BRTO in the setting of active GV bleeding because a sclerosing agent may be unable to control the bleeding as a result of leakage into the gastric lumen (26). Moreover, endoscopic hemostasis or a second BRTO procedure has been required in this situation (25,26,29,30). A study by Kageyama et al (26) reported
6
’
Vascular Plug–Assisted Retrograde Transvenous Obliteration
that patients with poor liver function are more prone to recurrent bleeding from GVs even if they undergo emergency BRTO. In the present study, we successfully performed an emergency PARTO procedure in all six patients with active GV bleeding. Regardless of liver function, there was no recurrent bleeding during the follow-up period as a result of complete thrombosis of the portosystemic shunt and GVs. Therefore, emergency PARTO is also effective for the treatment of active GV bleeding. Although we observed a small amount of leakage of contrast media mixed with gelatin sponge into the gastric lumen, all GVs and portosystemic shunts were successfully filled with gelatin sponge because the particle size we used (1–8 mm3) was large enough to block the leak site. Compared with BRTO, PARTO has several advantages. First, it prevents procedure-related complications and deceases the procedure time because it does not require an indwelling balloon catheter and sclerosing agents. As a result of long balloon indwelling times in BRTO, the indwelling balloon catheter may be a nidus for complications such as an increased risk of balloon rupture, access-site complications, and infection, as well as logistical challenges for hospitals such as patient inconvenience, intensive care unit or higher-level monitoring requirements, and additional staff (7–23). However, the use of a vascular plug, which has been shown to be safe and effective for the treatment of various vascular conditions, provides a permanent embolic effect (21,22,27,28,31,32). This suggests that the complications and logistical issues with an indwelling balloon catheter can successfully be resolved by using a vascular plug. In addition, the use of gelatin sponge instead of sclerosing agents can promote rapid and complete embolization of the portosystemic shunt and GVs without complications (21,22). In the present study, the mean procedure time from vascular plug placement to vascular plug detachment was 24 minutes, with a range of 11–124 minutes. In addition, there were no procedure-related complications in any of our patients. Second, PARTO does not require selective embolization of efferent veins in most cases, as it does not require sclerosing agents. To prevent leakage of the sclerosing agents into the systemic circulation during BRTO, efferent veins should be embolized by using a selectively catheterized microcatheter system. Except in two cases of microcoil embolization of prominent efferent veins, we observed that gelatin sponge particles were sufficient to embolize efferent veins such as left inferior phrenic and paravertebral veins. During BRTO, reducing the indwelling time of the balloon catheter to less than 3 hours can reduce the sclerosing effect, so the frequency of complete thrombosis of GVs with this procedure was reported to be lower, ranging from 55% to 100%, and a repeated procedure was often needed (8,9,11,18). These previous studies have shown that a second or third BRTO procedure
Gwon et al
’
JVIR
may be required to completely obliterate GVs in 10%– 44% of patients. These repeated BRTO procedures are considered burdensome for patients and result in longer hospitalization. Compared with BRTO, a second PARTO procedure may be impossible as a result of the preexisting vascular plug. In the present study, we observed recanalization of GVs in one patient as a result of incomplete filling of GVs with gelatin sponges. Therefore, the distal part of the afferent veins, including left gastric, posterior gastric, and short gastric veins, should be completely filled with gelatin sponge particles to prevent recanalization of the portosystemic shunt and GVs. One key component for BRTO and PARTO is improvement in hepatic function. In previous BRTO reports (10–12,33,34), liver function has been shown to be improved based on an increase in portal hepatic blood flow. However, improvement of hepatic function after BRTO was only temporary (6–12 mo) and seemed to depend on underlying conditions such as the concomitant existence of hepatocellular carcinoma or primary biliary cirrhosis (11). A study by Kumamoto et al (34) provides evidence that BRTO may have a role in preserving liver function. They reported transient improvement in hepatic function that returned to baseline levels by 3 years, and patients without gastrosystemic shunts had stable hepatic function similar to those patients treated with BRTO (34). This suggested that BRTO had a protective longterm role in preserving hepatic function and protecting the liver from portosystemic shunt syndrome. In the present study, improvement in Child–Pugh score was observed in 24 of 60 patients (40%) within 1 month after PARTO. As commonly known, the major drawback of BRTO and PARTO is potential worsening of EVs. The worsening rate of EVs has previously been reported in 30%–68% of patients, with bleeding EVs occurring in 17%–24% of patients who had undergone BRTO, probably because of the increased portal flow and pressures (8,9,11,18). Differences in the rate of EV aggravation after BRTO result from differences in how aggravation of EVs is defined and in the duration of follow-up. In the present study, the worsening rate of EVs was 26.7%, which is within the expected range for this type of procedure. EVs newly developed in six of 60 patients (10%) who did not have EVs before PARTO, and EVs progressed to a larger size in 10 of 60 patients (16.7%) who had EVs before PARTO. Therefore, close monitoring with or without prophylaxis of EVs should be necessary after PARTO. The worsening of ascites and/or hydrothorax after BRTO has been reported to be variable (0–44%) (8,11,13–16,28), but the modality and timing of evaluation between 24 hours and 4 weeks after BRTO varies among facilities, and evaluation has not been standardized. In the present study, on 3-month follow-up CT, ascites had newly developed in nine of 60 patients (15%) who did not have ascites before PARTO,
Volume XX
’
Number X
’
Month
’
2015
and ascites progressed to a larger amount in five of 60 patients (8.3%) who had ascites before PARTO. The present study had a prospective, multicenter design but was really a two-center study. Therefore, randomized comparative trials with large numbers of study centers will be necessary to determine the benefits of PARTO. In conclusion, the present results of PARTO indicate that it can be rapidly performed with high technical success and durable clinical efficacy for the treatment of GVs and HE in the presence of a portosystemic shunt. Therefore, PARTO might be considered a first-line treatment in appropriate patients.
ACKNOWLEDGMENTS The authors thank Ji Young Ham, RN, Eun Joung Kim, RN, and the rest of the nursing staff at the participating centers for their support and assistance in this study; and Youngoh June and Jung Taek Choi for their help with this manuscript.
REFERENCES 1. De Franchis R, Primignani M. Natural history of portal hypertension in patients with cirrhosis. Clin Liver Dis 2001; 5:645–663. 2. Trudeau W, Prindiville T. Endoscopic injection sclerosis in bleeding gastric varices. Gastrointest Endosc 1986; 32:264–268. 3. Sarin SK, Lahoti D, Saxena SP, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology 1992; 16:1343–1349. 4. Watanabe K, Kimura K, Matsutani S, et al. Portal hemodynamics in patients with gastric varices. A study in 230 patients with esophageal and/or gastric varices using portal vein catheterization. Gastroentrology 1988; 95:434–440. 5. Uflacker R, Silva Ade O, d’Albuquerque LA, et al. Chronic portosystemic encephalopathy: embolization of portosystemic shunts. Radiology 1987; 165:721–725. 6. Chandler JG, Fechner RE. Hepatopedal flow restoration in patients intolerant of total portal diversion. Ann Surg 1983; 187:574–583. 7. Kanagawa H, Mima S, Kouyama H, et al. Treatment of gastric fundal varices by balloon-occluded retrograde transvenous obliteration. J Gastroenterol Hepatol 1996; 11:51–58. 8. Chikamori F, Kuniyoshi N, Shibuya S, et al. Eight years of experience with transjugular retrograde obliteration for gastric varices with gastrorenal shunts. Surgery 2001; 129:414–420. 9. Cho SK, Shin SW, Lee IH, et al. Balloon-occluded retrograde transvenous obliteration of gastric varices: outcomes and complications in 49 patients. AJR Am J Roentgenol 2007; 189:W365–W372. 10. Akahane T, Iwasaki T, Kobayashi N, et al. Change in liver function parameters after occlusion of gastrorenal shunts with balloon-occluded retrograde transvenous obliteration. Am J Gastroenterol 1997; 92:1026–1030. 11. Fukuda T, Hirota S, Sugimura K. Long-term results of balloon-occluded retrograde transvenous obliteration for the treatment of gastric varices and hepatic encephalopathy. J Vasc Interv Radiol 2001; 12:327–336. 12. Jang SY, Kim GH, Park SY, et al. Clinical outcomes of balloon-occluded retrograde transvenous obliteration for the treatment of gastric variceal hemorrhage in Korean patients with liver cirrhosis: a retrospective multicenter study. Clin Mol Hepatol 2012; 18:368–374. 13. Choi SY, Won JY, Kim KA, et al. Foam sclerotherapy using polidocanol for balloon-occluded retrograde transvenous obliteration (BRTO). Eur Radiol 2011; 21:122–129.
7
14. Sabri SS, Swee W, Turba UC, et al. Bleeding gastric varices obliteration with balloon-occluded retrograde transvenous obliteration using sodium tetradecyl sulfate foam. J Vasc Interv Radiol 2011; 22:309–316. 15. Bellary SV, Isaacs P. Disseminated intravascular coagulation (DIC) after endoscopic injection sclerotherapy with ethanolamine oleate. Endoscopy 1990; 22:151. 16. Saad WE, Sabri SS. Balloon-occluded retrograde transvenous obliteration (BRTO): technical results and outcomes. Semin Intervent Radiol 2011; 28:333–338. 17. Hashizume M, Kitano S, Yamaga H, et al. Haptoglobin to protect against renal damage from ethanolamine oleate sclerosant. Lancet 1988; 2: 340–341. 18. Hirota S, Matsumoto S, Tomita M, et al. Retrograde transvenous obliteration of gastric varices. Radiology 1999; 211:349–356. 19. Shimoda R, Horiuchi K, Haqiwara S, et al. Short-term complications of retrograde transvenous obliteration of gastric varices in patients with portal hypertension: effects of obliteration of major portosystemic shunts. Abdom Imaging 2005; 30:306–313. 20. Park SJ, Chung JW, Kim HC, et al. The prevalence, risk factors, and clinical outcomes of balloon rupture in balloon-occluded retrograde transvenous obliteration of gastric varices. J Vasc Interv Radiol 2010; 21: 503–507. 21. Gwon DI, Ko GY, Yoon HK, et al. Gastric varices and hepatic encephalopathy: treatment with vascular plug and gelatin sponge-assisted retrograde transvenous obliteration-a primary report. Radiology 2013; 268: 281–287. 22. Park JK, Cho SK, Kee S, et al. Vascular plug-assisted retrograde transvenous obliteration of portosystemic shunts for refractory hepatic encephalopathy: a case report. Case Rep Radiol 2014;391–420. 23. Lee EW, Saab S, Gomes AS, et al. Coil-assisted retrograde transvenous obliteration (CARTO) for the treatment of portal hypertensive variceal bleeding: preliminary results. Clin Transl Gastroenterol 2014; 5:e61. 24. Sacks D, McClenny TE, Cardella JF, et al. Society of interventional radiology clinical practice guidelines. J Vasc Interv Radiol 2003; 14(suppl): S199–S202. 25. Sonomura T, Ono W, Sato M, et al. Emergency balloon-occluded retrograde transvenous obliteration of ruptured gastric varices. World J Gastroenterol 2013; 21:5125–5130. 26. Kageyama K, Nishida N, Yamamoto A, et al. Risk factors for rebleeding and prognostic factors for postoperative survival in patients with balloonoccluded retrograde transvenous obliteration of acute gastric variceal rupture. Cardiovasc Intervent Radiol 2014; 37:1235–1242. 27. Rossi M, Rebonato A, Greco L, et al. A new device for vascular embolization: report on case of two pulmonary arteriovenous fistulas embolization using the amplatzer vascular plug. Cardiovasc Intervent Radiol 2006; 29:902–906. 28. Kessler J, Trerotola SO. Use of the amplatzer vascular plug for embolization of a large retroperitoneal shunt during transjugular intrahepatic portosystemic shunt creation for gastric variceal bleeding. J Vasc Interv Radiol 2006; 17:135–140. 29. Kitamoto M, Imamura M, Kamada K, et al. Balloon-occluded retrograde transvenous obliteration of gastric fundal varices with hemorrhage. AJR Am J Roentgenol 2002; 178:1167–1174. 30. Arai H, Abe T, Shimoda R, et al. Emergency balloon-occluded retrograde transvenous obliteration for gastric varices. J Gastroenterol 2005; 40:964–971. 31. Yoo H, Ko GY, Gwon DI, et al. Preoperative portal vein embolization using an amplatzer vascular plug. Eur Radiol 2009; 19:1054–1061. 32. Kim SK, Chun HJ, Choi BG, et al. Transcatheter venous embolization of a massive hepatic arteriovenous shunt complicating hepatocellular carcinoma using an amplatzer vascular plug. Jpn J Radiol 2011; 29: 156–160. 33. Akahane T, Iwasaki T, Kobayashi N, et al. Change in liver function parameters after occlusion of gastrorenal shunts with balloon-occluded retrograde transvenous obliteration. AJR Am J Roentgenol 1997; 92: 1026–1030. 34. Kumamoto M, Toyonaga A, Inoue H, et al. Long-term results of balloonoccluded retrograde transvenous obliteration for gastric fundal varices: hepatic deterioration links to portosystemic shunt sysndrom. J Gastroenterol Hepatol 2010; 25:1129–1135.
Vascular Plug–Assisted Retrograde Transvenous Obliteration for the Treatment of Gastric Varices and Hepatic Encephalopathy: A Prospective Multicenter Study Dong Il Gwon, MD, Young Hwan Kim, MD, Gi-Young Ko, MD, Jong Woo Kim, MD, Heung Kyu Ko, MD, Jin Hyoung Kim, MD, Ji Hoon Shin, MD, Hyun-Ki Yoon, MD, and Kyu-Bo Sung, MD
ABSTRACT Purpose: To evaluate technical and clinical outcomes of vascular plug–assisted retrograde transvenous obliteration (PARTO) for the treatment of gastric varices (GVs) and hepatic encephalopathy (HE). Materials and Methods: From March 2012 to June 2014, 73 consecutive patients (47 men, 26 women; mean age, 59 y; range, 28–79 y) who had undergone PARTO were evaluated in a prospective multicenter study. Among 57 patients with GVs, 28 had GVs in danger of rupture, 23 had experienced recent bleeding, and 6 had active variceal bleeding. The 16 patients with HE had been treated unsuccessfully with medical therapies. Results: Placement of the vascular plug and subsequent gelatin sponge embolization were technically successful in all 73 patients. There were no procedure-related complications. Follow-up CT obtained within 1 wk after PARTO showed complete thrombosis of GVs and portosystemic shunts in 72 of 73 patients (98.6%). Sixty patients who underwent follow-up longer than 3 mo showed complete obliteration of GVs and portosystemic shunts. There were no cases of variceal bleeding or HE at the end of follow-up (mean, 544 d). Improvement in Child–Pugh score was observed in 24 patients (40%) at 1-mo follow-up. Worsening of ascites and esophageal varices was observed in 14 (23.3%) and 16 (26.7%) patients at 3-mo follow-up. Conclusions: The present results of PARTO indicate that it can be rapidly performed with high technical success and durable clinical efficacy for the treatment of GVs and HE in the presence of a portosystemic shunt. Therefore, PARTO might be considered a first-line treatment in appropriate patients.
ABBREVIATIONS BRTO = balloon-occluded retrograde transvenous obliteration, EV = esophageal varix, HE = hepatic encephalopathy, GV = gastric varix, PARTO = vascular plug–assisted retrograde transvenous obliteration
Bleeding caused by gastroesophageal varices is a major complication of portal hypertension in patients with From the Department of Radiology and Research Institute of Radiology (D.I.G., G.Y.K., J.W.K., H.K.K., J.H.K., J.H.S., H.K.Y., K.B.S.), University of Ulsan College of Medicine, Asan Medical Center, 86, Asanbyeongwon-gil, Songpa-gu, Seoul 138-736, Korea; and Department of Radiology (Y.H.K.), Keimyung University School of Medicine, Dongsan Medical Center, Daegu, Korea. Received May 15, 2015; final revision received and accepted July 14, 2015. Address correspondence to D.I.G.; E-mail: [email protected] None of the authors have identified a conflict of interest. & SIR, 2015 J Vasc Interv Radiol 2015; XX:]]]–]]] http://dx.doi.org/10.1016/j.jvir.2015.07.011
cirrhosis (1). Although the risk of bleeding from gastric varices (GVs) is relatively low, when rupture of GVs has occurred, the outcome is worse, with a higher mortality rate than seen with esophageal varices (EVs) because there is more severe blood loss and a higher recurrent bleeding rate (1–3). Bleeding of GVs is associated with mortality rates as high as 55%. As most of the gastric fundal varices drain into the left renal vein via a gastrosystemic shunt, the concept of retrograde injection of sclerosing agents into GVs after balloon occlusion of the gastrorenal shunt has been introduced (4). Moreover, gastrosystemic shunts may result in hepatic encephalopathy (HE), a debilitating condition that may be refractory to medical management. The key factor in
2
’
Vascular Plug–Assisted Retrograde Transvenous Obliteration
the pathogenesis of HE is shunting of undetoxified portal blood into the systemic circulation (5,6). Therefore, balloon-occluded retrograde transvenous obliteration (BRTO) has been shown to be a suitable therapeutic option for the control of GVs and HE because a natural interventional target is closure of the GVs and portosystemic shunt (7–14). However, previous studies reported complications associated with the use of liquid and foam sclerosant agents (8,9,13–18). In addition, balloon rupture during the BRTO procedure can result in symptomatic pulmonary edema, treatment failure, and ultimately recurrent variceal bleeding (19,20). Therefore, the absence of a balloon occlusion catheter and sclerosant agents can reduce procedure time, reduce procedure-related morbidity, and simplify procedure-associated logistics. Recently, a modified BRTO procedure was proposed and reported in which a balloon occlusion catheter and sclerosant agents were replaced with a vascular plug/coils and gelatin sponge to minimize some of the complications and logistical issues associated with the balloon catheter (21–23). A study by Gwon et al (21) reported that vascular plug–assisted retrograde transvenous obliteration (PARTO) successfully induced thrombosis of the gastrorenal shunt and GVs with no complications and subsequent obliteration. However, this study (21) has its own limitations (retrospective nature with a small number of study patients). Therefore, the purpose of the present study is to evaluate the technical and clinical outcomes of PARTO for the treatment of GVs and HE in a prospective multicenter study.
Gwon et al
Patient Population A prospective, uncontrolled multicenter study was conducted to evaluate the safety and efficacy of PARTO in 73 consecutive patients (47 men, 26 women; mean age, 59 y; range, 28–79 y) with GVs or HE from March 2012 to Jun 2014. Two academic tertiary referral centers participated in the study. The study was approved by the institutional review board at each participating hospital, and informed consent was obtained from each patient. In each patient, portosystemic shunt was evaluated by contrast-enhanced computed tomography (CT), and GVs were confirmed by endoscopy and CT. Patients were included if they had GVs or HE with a portosystemic shunt. Exclusion criterion included GVs or HE without a portosystemic shunt. Among these 73 patients, 57 had a portosystemic shunt with GVs and 16 had HE. Among 57 patients with GVs, 28 had GVs in danger of rupture, 23 had experienced recent bleeding, and six had active variceal bleeding. GVs showing marked protrusion or red spots and progressive enlargement within a 6-month period on endoscopy were judged to be in danger of rupture. The six patients with ruptured
JVIR
GVs had undergone emergency PARTO as a result of unsuccessful endoscopic hemostasis. Among 16 patients with HE, 11 had only a portosystemic shunt and five had a portosystemic shunt with minimal GVs. The 16 patients who had HE had been treated unsuccessfully with medical therapies such as lactulose and oral branched-chain amino acid supplementation. HE severity was assigned a grade per West Haven criteria: of the 16 patients with HE, four had grade 1, six had grade 2, three had grade 3, and three had grade 4 HE. The baseline demographic and clinical characteristics of the 73 patients are presented in the Table.
Technique Before PARTO, contrast-enhanced CT scans of the abdomen were obtained for all patients, and they confirmed the presence of a portosystemic shunt (Fig 1a). All PARTO procedures were performed with the patient under conscious sedation with intravenous pethidine hydrochloride (Demerol; Keukdong Pharmaceuticals, Seoul, Korea) and local anesthesia with lidocaine (Jeil Pharmaceuticals, Taegu, Korea). The right common femoral vein was punctured initially, after which a 6- or 7-F sheath (Flexor Check-Flo sheath; Cook, Bloomington, Indiana) was inserted. When insertion of a sheath into the portosystemic shunt was not possible through the right common femoral vein because of an acute angle, the Table . Patient Characteristics Characteristic
MATERIALS AND METHODS
’
GVs
HE
37:20 58
10:6 62
38 6
12 2
Alcoholic cirrhosis
8
2
LC/membranous IVC obstruction Other*
2 3
– –
28
11
Sex (M:F) Mean age (y) Underlying liver disease HBV cirrhosis HCV cirrhosis
Concomitant malignancy HCC Child–Pugh classification A
23
–
B C
26 8
13 3
8.2 ⫾ 1.8
7.8 ⫾ 1.4
Indications for GVs Acute bleeding
6
–
Chronic bleeding
23
–
Primary prophylaxis
28
–
MELD score
HE ¼ hepatic encephalopathy; GV ¼ gastric varix; HBV ¼ hepatitis B virus; HCC ¼ hepatocellular carcinoma; HCV ¼ hepatitis C virus; IVC = inferior vena cava; LC ¼ liver cirrhosis; MELD ¼ Model for End-stage Liver Disease. *Biliary cirrhosis (n ¼ 1), splenic vein thrombosis caused by chronic pancreatitis (n ¼ 1), unknown cause (n ¼ 1).
Volume XX
’
Number X
’
Month
’
2015
3
Figure 1. Images from a 55-year-old man with GVs. (a) Contrast-enhanced CT obtained before PARTO shows GVs (asterisk). Note the hypertrophied left gastric vein (arrowhead). (b) After placement of the vascular plug (white arrow) within the narrowest portion of the portosystemic shunt via the left adrenal vein, additional embolization of the gastrorenal shunt, GVs (asterisk), and left gastric vein (arrowhead) was performed by using gelatin sponge particles through the 4-F catheter (black arrow). (c) Contrast-enhanced CT scan obtained 3 months after PARTO shows complete obliteration of the GVs.
right internal jugular vein was then accessed. After negotiation of the left adrenal vein or shunt with a 0.035-inch, 180-cm-long hydrophilic guide wire (Terumo, Tokyo, Japan) and a 4-F angled-tip catheter (Cobra; Terumo), the sheath was then inserted into the shunt. The guide wire and catheter were subsequently removed, and the vascular plug (AMPLATZER Vascular Plug II; AGA Medical, Golden Valley, Minnesota) was deployed. The size of vascular plug was chosen based on the diameter of the narrowest shunt measured on CT, ranging from 8 to 20 mm in diameter or 20% larger than the targeted shunt to prevent migration and gelatin sponge reflux. To obtain easy negotiation of the guide wire and the catheter into the shunt proximal to the vascular plug, the vascular plug was deployed at the most dilated portion of the shunt. The shunt proximal to the vascular plug was then negotiated by using the guide wire and the 4-F angledtip catheter between the vascular plug and the shunt wall. The vascular plug was then pulled down to the narrowest part of the shunt while maintaining the position of the catheter. When contrast medium injection through the catheter showed proper position of the vascular plug at the narrowest portion, an additional embolization was then performed with a mixture of contrast medium and gelatin sponge particles (ranging in size from approximately 1 mm3 to 8 mm3) through the catheter to embolize the shunt, efferent veins, afferent veins, and GVs (Fig 1b). After embolization, retrograde venography through the sheath was performed to confirm occlusion of the shunt. The catheter was subsequently removed, and the delivery cable attached to the end of the vascular plug was then detached.
Follow-up Patients were evaluated by CT during their hospital admission within 1 week after PARTO to evaluate thrombosis of the portosystemic shunt and GVs. Subsequently, clinical and laboratory examinations were
performed 1, 3, 6, and 12 months after the procedures in all patients. Upper-intestinal endoscopy and CT were also performed 3 months after PARTO (Fig 1c). If the shunt and GVs were not completely obliterated at 3month follow-up CT, follow-up CT was performed 6 months after PARTO. Beginning 12 months after PARTO, information regarding patients’ current status or death was obtained by telephone from all patients or their families.
Definitions and Statistical Analysis Primary study endpoints were assessment of technical success, procedure-related complications, and clinical success. Secondary study endpoints were assessment of follow-up clinical results including change of liver function, worsening of EVs, and incidence of ascites. Technical success was defined as successful placement of the vascular plug within the left adrenal vein or the portosystemic shunt and administration of the gelatin sponge particles into the portosystemic shunt and GVs. Complications were classified as major or minor according to the guidelines of the Society of Interventional Radiology Standards of Practice Committee (24). Clinical success was defined as the control of active GV bleeding or complete resolution of all clinical symptoms of HE with complete thrombosis of GVs and portosystemic shunt. Complete resolution of HE was defined as a return to grade 0 HE according to the West Haven criteria as well as normalization of the serum ammonia (ie, NH3) level. Complete obliteration of GVs was defined as the complete eradication of the targeted GVs on follow-up CT or endoscopy 2 months after PARTO or at any time thereafter. Recurrent bleeding from GVs was defined as the presence of hematemesis or melena with endoscopic visualization of the bleeding source originating from the GVs after percutaneous intervention. The paired-sample t test was used to compare preand post-PARTO serum venous ammonia, bilirubin, and
4
’
Vascular Plug–Assisted Retrograde Transvenous Obliteration
albumin levels. Statistical analysis was performed by using SPSS software (version 14.0; SPSS, Chicago, Illinois), with P values less than .05 considered to be statistically significant.
RESULTS Technical Success and Complication The transfemoral approach was used in 66 patients and the transjugular approach was used in seven. Placement of the vascular plug and subsequent embolization by gelatin sponge were technically successful in all 73 patients (Fig 2). A single AMPLATZER Vascular Plug II was used to occlude the left adrenal vein or portosystemic shunt in 70 patients (8 mm [n ¼ 9], 10 mm [n ¼ 7], 12 mm [n ¼ 9], 14 mm [n ¼ 22], 16 mm [n ¼ 11], 18 mm [n ¼ 5], 20 mm [n ¼ 4], and 22 mm in diameter [n ¼ 3]). In the remaining three patients who each had two prominent splenorenal shunts draining into the left adrenal vein, two vascular plugs (one 10 mm and one 8 mm in diameter in one patient, and two 8 mm in diameter in another patient) were used in two patients, and one vascular plug (14 mm) and seven coils were used in one patient. During embolization of portosystemic shunt with/without GVs, 67 patients showed single or multiple collateral veins, and these collateral veins were successfully embolized by gelatin sponge only. However, gelatin sponge embolization was insufficient in two patients because of a prominent left inferior phrenic vein. Additional embolization with microcoils was performed in these two patients. None of the vascular plugs migrated after detachment. The mean procedure time from vascular plug placement to vascular plug detachment was 24 minutes, with a range of 11–124 minutes. There were no procedure-related complications in any patient.
Clinical Success Follow-up CT obtained within 1 week after PARTO showed complete thrombosis of the treated GVs and the
Gwon et al
’
JVIR
gastrorenal shunts in 72 of 73 patients (98.6%). One patient had residual intramural GVs on 1-week followup CT. Additional treatments were not performed, and 12 days after PARTO this patient died of massive hematemesis as a result of recurrent bleeding of GVs. The clinical symptoms of HE completely resolved in all 16 patients with HE, and the mean serum NH3 level decreased significantly (P o .001) from 118.4 mmol/L ⫾ 53.6 before PARTO to 29.8 mmol/L ⫾ 15.9 within 1 week after PARTO (95% confidence interval, 66.3–111.0 mmol/L). Therefore, the clinical success rate was 98.6% (72 of 73 patients).
Follow-up Clinical Results A schematic diagram of clinical outcomes is shown in Figure 3. One patient with GVs underwent liver transplantation 1 month after undergoing PARTO. Two patients were lost to follow-up within 3 months after PARTO. Clinical follow-up until death or the end of the study was available for the remaining 70 patients and lasted 7–1,050 days (mean, 544 d). The cutoff date for data analysis was January 31, 2015. At the end of follow-up, 19 patients had died and 51 remained alive. Ten patients with GVs died within 3 months after PARTO (range, 0.5–3 mo), five from deteriorating hepatic function, three from progression of hepatocellular carcinoma, and two from rupture of a hepatocellular carcinoma. Five patients with HE and four with GVs died more than 3 months after PARTO (range, 7.5–16 months), six from deteriorating hepatic function and three from progression of hepatocellular carcinoma. Laboratory measurements at 1 and 3 months were available for 60 patients. Improvement in Child–Pugh score was observed in 24 of 60 patients (40%) at either follow-up. However, there was no significant change in laboratory values at either follow-up. Follow-up laboratory measurements at 6 and 12 months were available for 51 patients. A 3-month follow-up CT was available for 60 patients and demonstrated complete obliteration of thrombosed
Figure 2. Images from a 68-year-old man with HE caused by a splenorenal shunt. (a) Contrast-enhanced CT shows a hypertrophied splenorenal shunt (asterisks). (b) After placement of a vascular plug (white arrow) within the left adrenal vein, additional embolization of the splenorenal shunt (asterisks) was performed by using gelatin sponge particles through the 4-F catheter (black arrow). (c) Contrastenhanced CT obtained 3 months after PARTO shows complete obliteration of the splenorenal shunt.
Volume XX
’
Number X
’
Month
’
2015
5
Figure 3. Schematic diagram of clinical outcomes.
GVs and portosystemic shunts in 57 patients and marked shrinkage in three patients. A 6-month follow-up CT examination was performed in these three patients and demonstrated complete obliteration of thrombosed GVs and portosystemic shunts. However, ascites newly developed in nine of 60 patients (15%) who did not have ascites before PARTO, and ascites progressed to a larger amount in five of 60 patients (8.3%) who had ascites before PARTO. A 3-month follow-up endoscopy revealed marked shrinkage or disappearance of GVs in all 60 patients. None of these 60 patients experienced bleeding of GVs or HE during the follow-up period. However, 16 patients (26.7%) experienced worsening of EVs. The change in form was from straight and small varices to winding varices in 10 patients, and from no varices to straight and small varices in six patients. Seven of the 10 patients with winding varices had EVs with the appearance of red spots on the esophageal mucosa and were successfully treated endoscopically.
DISCUSSION In the present study, we found a technical success rate (100%) and clinical success rate (98.6%) of PARTO that were in concordance with the result of BRTO studies
with the use of various sclerosing agents (7–14). In those BRTO studies (7–14), clinical success rates ranged from 79.6% to 100% and the relapse rate ranged from 0% to 10%. We also found that there was complete thrombosis of the GVs and portosystemic shunts in the 72 of 73 patients within 1 week after PARTO and complete obliteration of the GVs and portosystemic shunts in all 60 patients who had more than 2 months of follow-up. Among these 60 patients, we observed no cases of variceal bleeding in the 46 patients with GVs and no recurrence of HE in 16 patients. If there was no immediate blood supply into the completely thrombosed GVs and portosystemic shunt from the afferent veins that were embolized by a vascular plug and gelatin sponge, the thrombosed GVs and shunt were completely obliterated. As ruptured GVs are associated with high rates of recurrent bleeding and mortality, quick treatment is essential. Emergency BRTO has been shown effective for rescue in an emergency condition in several reports (25–28). Unlike prophylactic BRTO, it is sometimes difficult to perform emergency BRTO in the setting of active GV bleeding because a sclerosing agent may be unable to control the bleeding as a result of leakage into the gastric lumen (26). Moreover, endoscopic hemostasis or a second BRTO procedure has been required in this situation (25,26,29,30). A study by Kageyama et al (26) reported
6
’
Vascular Plug–Assisted Retrograde Transvenous Obliteration
that patients with poor liver function are more prone to recurrent bleeding from GVs even if they undergo emergency BRTO. In the present study, we successfully performed an emergency PARTO procedure in all six patients with active GV bleeding. Regardless of liver function, there was no recurrent bleeding during the follow-up period as a result of complete thrombosis of the portosystemic shunt and GVs. Therefore, emergency PARTO is also effective for the treatment of active GV bleeding. Although we observed a small amount of leakage of contrast media mixed with gelatin sponge into the gastric lumen, all GVs and portosystemic shunts were successfully filled with gelatin sponge because the particle size we used (1–8 mm3) was large enough to block the leak site. Compared with BRTO, PARTO has several advantages. First, it prevents procedure-related complications and deceases the procedure time because it does not require an indwelling balloon catheter and sclerosing agents. As a result of long balloon indwelling times in BRTO, the indwelling balloon catheter may be a nidus for complications such as an increased risk of balloon rupture, access-site complications, and infection, as well as logistical challenges for hospitals such as patient inconvenience, intensive care unit or higher-level monitoring requirements, and additional staff (7–23). However, the use of a vascular plug, which has been shown to be safe and effective for the treatment of various vascular conditions, provides a permanent embolic effect (21,22,27,28,31,32). This suggests that the complications and logistical issues with an indwelling balloon catheter can successfully be resolved by using a vascular plug. In addition, the use of gelatin sponge instead of sclerosing agents can promote rapid and complete embolization of the portosystemic shunt and GVs without complications (21,22). In the present study, the mean procedure time from vascular plug placement to vascular plug detachment was 24 minutes, with a range of 11–124 minutes. In addition, there were no procedure-related complications in any of our patients. Second, PARTO does not require selective embolization of efferent veins in most cases, as it does not require sclerosing agents. To prevent leakage of the sclerosing agents into the systemic circulation during BRTO, efferent veins should be embolized by using a selectively catheterized microcatheter system. Except in two cases of microcoil embolization of prominent efferent veins, we observed that gelatin sponge particles were sufficient to embolize efferent veins such as left inferior phrenic and paravertebral veins. During BRTO, reducing the indwelling time of the balloon catheter to less than 3 hours can reduce the sclerosing effect, so the frequency of complete thrombosis of GVs with this procedure was reported to be lower, ranging from 55% to 100%, and a repeated procedure was often needed (8,9,11,18). These previous studies have shown that a second or third BRTO procedure
Gwon et al
’
JVIR
may be required to completely obliterate GVs in 10%– 44% of patients. These repeated BRTO procedures are considered burdensome for patients and result in longer hospitalization. Compared with BRTO, a second PARTO procedure may be impossible as a result of the preexisting vascular plug. In the present study, we observed recanalization of GVs in one patient as a result of incomplete filling of GVs with gelatin sponges. Therefore, the distal part of the afferent veins, including left gastric, posterior gastric, and short gastric veins, should be completely filled with gelatin sponge particles to prevent recanalization of the portosystemic shunt and GVs. One key component for BRTO and PARTO is improvement in hepatic function. In previous BRTO reports (10–12,33,34), liver function has been shown to be improved based on an increase in portal hepatic blood flow. However, improvement of hepatic function after BRTO was only temporary (6–12 mo) and seemed to depend on underlying conditions such as the concomitant existence of hepatocellular carcinoma or primary biliary cirrhosis (11). A study by Kumamoto et al (34) provides evidence that BRTO may have a role in preserving liver function. They reported transient improvement in hepatic function that returned to baseline levels by 3 years, and patients without gastrosystemic shunts had stable hepatic function similar to those patients treated with BRTO (34). This suggested that BRTO had a protective longterm role in preserving hepatic function and protecting the liver from portosystemic shunt syndrome. In the present study, improvement in Child–Pugh score was observed in 24 of 60 patients (40%) within 1 month after PARTO. As commonly known, the major drawback of BRTO and PARTO is potential worsening of EVs. The worsening rate of EVs has previously been reported in 30%–68% of patients, with bleeding EVs occurring in 17%–24% of patients who had undergone BRTO, probably because of the increased portal flow and pressures (8,9,11,18). Differences in the rate of EV aggravation after BRTO result from differences in how aggravation of EVs is defined and in the duration of follow-up. In the present study, the worsening rate of EVs was 26.7%, which is within the expected range for this type of procedure. EVs newly developed in six of 60 patients (10%) who did not have EVs before PARTO, and EVs progressed to a larger size in 10 of 60 patients (16.7%) who had EVs before PARTO. Therefore, close monitoring with or without prophylaxis of EVs should be necessary after PARTO. The worsening of ascites and/or hydrothorax after BRTO has been reported to be variable (0–44%) (8,11,13–16,28), but the modality and timing of evaluation between 24 hours and 4 weeks after BRTO varies among facilities, and evaluation has not been standardized. In the present study, on 3-month follow-up CT, ascites had newly developed in nine of 60 patients (15%) who did not have ascites before PARTO,
Volume XX
’
Number X
’
Month
’
2015
and ascites progressed to a larger amount in five of 60 patients (8.3%) who had ascites before PARTO. The present study had a prospective, multicenter design but was really a two-center study. Therefore, randomized comparative trials with large numbers of study centers will be necessary to determine the benefits of PARTO. In conclusion, the present results of PARTO indicate that it can be rapidly performed with high technical success and durable clinical efficacy for the treatment of GVs and HE in the presence of a portosystemic shunt. Therefore, PARTO might be considered a first-line treatment in appropriate patients.
ACKNOWLEDGMENTS The authors thank Ji Young Ham, RN, Eun Joung Kim, RN, and the rest of the nursing staff at the participating centers for their support and assistance in this study; and Youngoh June and Jung Taek Choi for their help with this manuscript.
REFERENCES 1. De Franchis R, Primignani M. Natural history of portal hypertension in patients with cirrhosis. Clin Liver Dis 2001; 5:645–663. 2. Trudeau W, Prindiville T. Endoscopic injection sclerosis in bleeding gastric varices. Gastrointest Endosc 1986; 32:264–268. 3. Sarin SK, Lahoti D, Saxena SP, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology 1992; 16:1343–1349. 4. Watanabe K, Kimura K, Matsutani S, et al. Portal hemodynamics in patients with gastric varices. A study in 230 patients with esophageal and/or gastric varices using portal vein catheterization. Gastroentrology 1988; 95:434–440. 5. Uflacker R, Silva Ade O, d’Albuquerque LA, et al. Chronic portosystemic encephalopathy: embolization of portosystemic shunts. Radiology 1987; 165:721–725. 6. Chandler JG, Fechner RE. Hepatopedal flow restoration in patients intolerant of total portal diversion. Ann Surg 1983; 187:574–583. 7. Kanagawa H, Mima S, Kouyama H, et al. Treatment of gastric fundal varices by balloon-occluded retrograde transvenous obliteration. J Gastroenterol Hepatol 1996; 11:51–58. 8. Chikamori F, Kuniyoshi N, Shibuya S, et al. Eight years of experience with transjugular retrograde obliteration for gastric varices with gastrorenal shunts. Surgery 2001; 129:414–420. 9. Cho SK, Shin SW, Lee IH, et al. Balloon-occluded retrograde transvenous obliteration of gastric varices: outcomes and complications in 49 patients. AJR Am J Roentgenol 2007; 189:W365–W372. 10. Akahane T, Iwasaki T, Kobayashi N, et al. Change in liver function parameters after occlusion of gastrorenal shunts with balloon-occluded retrograde transvenous obliteration. Am J Gastroenterol 1997; 92:1026–1030. 11. Fukuda T, Hirota S, Sugimura K. Long-term results of balloon-occluded retrograde transvenous obliteration for the treatment of gastric varices and hepatic encephalopathy. J Vasc Interv Radiol 2001; 12:327–336. 12. Jang SY, Kim GH, Park SY, et al. Clinical outcomes of balloon-occluded retrograde transvenous obliteration for the treatment of gastric variceal hemorrhage in Korean patients with liver cirrhosis: a retrospective multicenter study. Clin Mol Hepatol 2012; 18:368–374. 13. Choi SY, Won JY, Kim KA, et al. Foam sclerotherapy using polidocanol for balloon-occluded retrograde transvenous obliteration (BRTO). Eur Radiol 2011; 21:122–129.
7
14. Sabri SS, Swee W, Turba UC, et al. Bleeding gastric varices obliteration with balloon-occluded retrograde transvenous obliteration using sodium tetradecyl sulfate foam. J Vasc Interv Radiol 2011; 22:309–316. 15. Bellary SV, Isaacs P. Disseminated intravascular coagulation (DIC) after endoscopic injection sclerotherapy with ethanolamine oleate. Endoscopy 1990; 22:151. 16. Saad WE, Sabri SS. Balloon-occluded retrograde transvenous obliteration (BRTO): technical results and outcomes. Semin Intervent Radiol 2011; 28:333–338. 17. Hashizume M, Kitano S, Yamaga H, et al. Haptoglobin to protect against renal damage from ethanolamine oleate sclerosant. Lancet 1988; 2: 340–341. 18. Hirota S, Matsumoto S, Tomita M, et al. Retrograde transvenous obliteration of gastric varices. Radiology 1999; 211:349–356. 19. Shimoda R, Horiuchi K, Haqiwara S, et al. Short-term complications of retrograde transvenous obliteration of gastric varices in patients with portal hypertension: effects of obliteration of major portosystemic shunts. Abdom Imaging 2005; 30:306–313. 20. Park SJ, Chung JW, Kim HC, et al. The prevalence, risk factors, and clinical outcomes of balloon rupture in balloon-occluded retrograde transvenous obliteration of gastric varices. J Vasc Interv Radiol 2010; 21: 503–507. 21. Gwon DI, Ko GY, Yoon HK, et al. Gastric varices and hepatic encephalopathy: treatment with vascular plug and gelatin sponge-assisted retrograde transvenous obliteration-a primary report. Radiology 2013; 268: 281–287. 22. Park JK, Cho SK, Kee S, et al. Vascular plug-assisted retrograde transvenous obliteration of portosystemic shunts for refractory hepatic encephalopathy: a case report. Case Rep Radiol 2014;391–420. 23. Lee EW, Saab S, Gomes AS, et al. Coil-assisted retrograde transvenous obliteration (CARTO) for the treatment of portal hypertensive variceal bleeding: preliminary results. Clin Transl Gastroenterol 2014; 5:e61. 24. Sacks D, McClenny TE, Cardella JF, et al. Society of interventional radiology clinical practice guidelines. J Vasc Interv Radiol 2003; 14(suppl): S199–S202. 25. Sonomura T, Ono W, Sato M, et al. Emergency balloon-occluded retrograde transvenous obliteration of ruptured gastric varices. World J Gastroenterol 2013; 21:5125–5130. 26. Kageyama K, Nishida N, Yamamoto A, et al. Risk factors for rebleeding and prognostic factors for postoperative survival in patients with balloonoccluded retrograde transvenous obliteration of acute gastric variceal rupture. Cardiovasc Intervent Radiol 2014; 37:1235–1242. 27. Rossi M, Rebonato A, Greco L, et al. A new device for vascular embolization: report on case of two pulmonary arteriovenous fistulas embolization using the amplatzer vascular plug. Cardiovasc Intervent Radiol 2006; 29:902–906. 28. Kessler J, Trerotola SO. Use of the amplatzer vascular plug for embolization of a large retroperitoneal shunt during transjugular intrahepatic portosystemic shunt creation for gastric variceal bleeding. J Vasc Interv Radiol 2006; 17:135–140. 29. Kitamoto M, Imamura M, Kamada K, et al. Balloon-occluded retrograde transvenous obliteration of gastric fundal varices with hemorrhage. AJR Am J Roentgenol 2002; 178:1167–1174. 30. Arai H, Abe T, Shimoda R, et al. Emergency balloon-occluded retrograde transvenous obliteration for gastric varices. J Gastroenterol 2005; 40:964–971. 31. Yoo H, Ko GY, Gwon DI, et al. Preoperative portal vein embolization using an amplatzer vascular plug. Eur Radiol 2009; 19:1054–1061. 32. Kim SK, Chun HJ, Choi BG, et al. Transcatheter venous embolization of a massive hepatic arteriovenous shunt complicating hepatocellular carcinoma using an amplatzer vascular plug. Jpn J Radiol 2011; 29: 156–160. 33. Akahane T, Iwasaki T, Kobayashi N, et al. Change in liver function parameters after occlusion of gastrorenal shunts with balloon-occluded retrograde transvenous obliteration. AJR Am J Roentgenol 1997; 92: 1026–1030. 34. Kumamoto M, Toyonaga A, Inoue H, et al. Long-term results of balloonoccluded retrograde transvenous obliteration for gastric fundal varices: hepatic deterioration links to portosystemic shunt sysndrom. J Gastroenterol Hepatol 2010; 25:1129–1135.