Pediatr Radiol (2015) 45:422–429 DOI 10.1007/s00247-014-3181-z
ORIGINAL ARTICLE
Safety and efficacy of expanded polytetrafluoroethylene-covered transjugular intrahepatic portosystemic shunts in children with acute or recurring upper gastrointestinal bleeding Luis J. Zurera & Juan J. Espejo & Sagrario Lombardo & Juan J. Gilbert & Miguel Canis & Concepción Ruiz
Received: 26 December 2013 / Revised: 28 July 2014 / Accepted: 28 August 2014 / Published online: 28 November 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background The transjugular intrahepatic portosystemic shunt (TIPS) is effective for treating complications of portal hypertension in cirrhotic adults but the experience in children is limited. Objective To retrospectively review the safety and efficacy of expanded polytetrafluoroethylene (ePTFE)-covered TIPS in children with acute or recurrent gastrointestinal bleeding. Materials and methods We reviewed the medical records of children who received implants of 10-mm-diameter PTFEcovered endoprostheses for acute or recurring upper gastrointestinal bleeding caused by medically or endoscopically uncontrollable varices. The recurrence of upper gastrointestinal bleeding, associated complications and permeability were assessed with Doppler sonography sequentially or up to transplantation. Results In all children (n=12; mean age 9 years; mean weight 30 kg) a single endoprosthesis was implanted with no associated mortality. The mean initial transhepatic gradient was 15 mmHg (range 3–21 mmHg), dropping to 7 mmHg (range 1–12 mmHg) after TIPS. Immediate complications were mild encephalopathy (n=1) and acute occlusion of the TIPS (n=1). Stenosis of the
L. J. Zurera : J. J. Espejo : S. Lombardo (*) : M. Canis Interventional Radiology Unit, Hospital Universitario Reina Sofía, Menendez Pidal s/n, 14004 Córdoba, Spain e-mail: [email protected] J. J. Gilbert Paediatric Gastroenterology Unit, Hospital Universitario Reina Sofía, Córdoba, Spain C. Ruiz Anaesthesiology Department, Hospital Universitario Reina Sofía, Córdoba, Spain
TIPS was observed in two children, at 9 months and 54 months follow-up, and thrombosis was observed in two children, at 7 months and 12 months follow-up. All four stenoses/ occlusions were resolved with coaxial endoprostheses. Conclusion The safety profile and efficacy of expanded polytetrafluoroethylene (ePTFE)-covered TIPS were satisfactory in this small series of children with acute or recurrent gastrointestinal bleeding. Keywords Intrahepatic portosystemic shunt . Portal vein thrombosis . Liver transplantation . Cavernomatosis . Venography . Fluoroscopy . Child
Introduction Creation of a transjugular intrahepatic portosystemic shunt (TIPS) is an effective procedure for treating complications of portal hypertension in adults with cirrhosis (mainly upper gastrointestinal bleeding and refractory ascites) when medical treatment and conventional endoscopic therapy have failed [1, 2]. However, experience with TIPS in the paediatric population is limited and the largest series of cases in the literature is 12 children [3]. Although the implant technique and the indications are similar in adults and children, children have a series of special considerations [4]. First, their liver and veins are smaller than those of adults. Second, the aetiology of cirrhosis in children is different from that in adults (extrahepatic biliary atresias, cystic fibrosis, cystic bile malformation), so children have major periportal fibrosis. It has also been shown that the permeability of TIPS in adults is considerably greater when using polytetrafluoroethylene (PTFE)-covered endoprostheses than when using uncovered ones [5]. However, most of the
Pediatr Radiol (2015) 45:422–429
published case series in children use the latter, with a high rate of recurring stenosis [6], although two of the largest and most recent series of TIPS in children use covered stents, with a good midterm patency rate [7, 8]. Our objectives are to present our initial experience with 12 children with portal hypertension of different causes treated with ePTFE-covered TIPS and to analyse their safety and efficacy retrospectively, and to present some technical considerations regarding their implantation.
Materials and methods Ethical approval was waived for this retrospective observational study, in accordance with our hospital guidelines. We reviewed the medical records of children who between May 2009 and May 2012 had TIPS implanted for the treatment of upper gastrointestinal bleeding secondary to portal hypertension with rebleeding despite secondary prophylaxis. We defined the failure to prevent rebleeding as a recurrent melena or hematemesis resulting in hospital admission, blood transfusion and Three gram drop in haemoglobin [9]. All procedures were performed with general anaesthetic and informed consent for TIPS was obtained from parents. The implant technique and equipment used were the same as for adults (Ring TIPSS-200 set; Cook Inc., Bjaeverskov, Denmark) [10, 11]. The right internal jugular vein was punctured and US guidance was used to control a 10-French introducer. Access was gained through a hepatic vein (right or middle). The needle used for the intrahepatic puncture and access to the portal system was the 16-gauge Colapinto included in the set. Indirect portography or wedged suprahepatic venography was used for guidance. A direct portography using a conventional 5-F calibrated pigtail catheter was performed in all children to select the right length for the endoprosthesis and view the portal anatomy, and to perform a hepatic haemodynamic study to establish the transhepatic gradient between the portal vein and the right atrium/inferior cava vein before and after the TIPS implantation. Intraoperative sonography was required to access the periportal veins in three children with portal cavernomatosis. The intrahepatic tract between the suprahepatic vein and the portal branch was dilated with a conventional 8-mm-diameter balloon catheter. There was no embolisation of gastric varices in any of the patients. The maximum dose of iodinated contrast agent used throughout the procedure was 3 ml/kg of weight. The average diameter of the main portal vein was 13 mm (range 10–17 mm) and the average of the collateral vein chosen in cases of cavernomatosis was 8 mm (range 7– 10 mm). The mean diameter of suprahepatic vein was 14 mm (range 9–25 mm). All of these measurements were performed using a calibrated catheter in direct and indirect portography under fluoroscopy.
423
In all cases a 10-mm-diameter ePTFE-covered nitinol endoprosthesis (Viatorr R; W.L. Gore and Associates, Flagstaff, AZ) was used, dilated initially to 8 mm and to 10 mm if the transhepatic gradient was greater than 12 mmHg; only in the youngest boy (a 2-year-old) was it dilated to 6 mm. The length of the ePTFE graft stent was variable, determined by tract measurement from the portal vein entry point to the hepatic vein inferior cava junction, leaving the proximal end in the suprahepatic vein and avoiding its protrusion into the right atrium at all times (to prevent technical complications during future transplantation). The lengths used for implanted ePTFE-covered nitinol endoprostheses were 10 mm (n=7), 9 mm (n=3) and 8 mm (n=2). The purpose of the procedure was to obtain a porto-systemic gradient below 12 mmHg. In children with chronic portal thrombosis and cavernomatosis, we chose a collateral vein of major caliber and the wire was advanced through this vein to the extrahepatic portal vein remnant or superior mesenteric or splenic veins. No antibiotics were administered. Infant aspirin was administered (3 mg/kg/day) after the TIPS procedure to the three children with portal cavernomatosis. All the children were followed clinically and with Doppler US within 4–6 days after the TIPS implantation and every 3 months up to the time of this study or up to transplantation; endoscopy was also performed at 3–6 months and at 1 year. Permeability of the TIPS for the first week after implantation was considered technical success and the control of upper gastrointestinal bleeding as a clinical success.
Results The sample consisted of 9 boys and 3 girls with an average age of 9 years (range 2–16 years) and a mean weight of 30 kg (range 11–60 kg). Portal hypertension was caused by cirrhosis in 10 children: there were 3 cases of extrahepatic biliary atresia (treated in infancy using Kasai procedure), two of cystic fibrosis, 1 of congenital hepatic fibrosis, 1 of complex Caroli disease cystic bile malformation, 1 of ductopenia, 1 of cirrhosis of unknown origin after biopsy and 1 of liver transplant in a newborn because of tyrosinaemia with chronic portal thrombosis and cavernomatosis. In the other two children liver disease was non-cirrhotic with thrombosis and portal cavernomatosis (one of them presented 6 years after a living-donor liver transplantation following biliary atresia and the other had umbilical catheterisation in the neonatal period). All children had hypersplenism with associated thrombocytopaenia and endoscopic evidence of gastroesophageal varices causing repeated episodes of upper gastrointestinal bleeding that were not stopped with standard medical treatment or endoscopic variceal ligation (n=2) or sclerotherapy (n=1) (Table 1). Seven of the children showed upper
424
Pediatr Radiol (2015) 45:422–429
Table 1 Demographic characteristics of 12 children treated with transjugular intrahepatic portosystemic shunt (TIPS), and results of the procedure Case Age
Weight (kg) Underlying cause
TIPS indication Gradient
Complications
Follow-up
Mild encephalopathy
36 m Permeable US 7 m Permeable TX 58 m Permeable US 9 m Permeable TX 10 m Permeable TX 30 m Permeable US 25 m Permeable US
Pre- Post1 2 3 4 5 6 7
12 y 7 m 11 y 2 m 16 y 12 y 9y 6 y 10 m 13 y 5 m
27 29 60 35 30 20 55
8 9 10 11 12
5y8m 10 y 10 m 4y 10 y 2 m 2y
14 30 22 30 11
Cystic fibrosis Cystic fibrosis Congenital hepatic fibrosis Biliary atresia Caroli disease Liver TX Cavernomatosis Portal cavernomatosis due to umbilical catheter Biliary atresia Cirrhosis of unknown origin Biliary atresia Liver TX Cavernomatosis Ductopenia
Recurring UGB Acute UGB Acute UGB Acute UGB Acute UGB Acute UGB Acute UGB
12 15 21 11 21 18 15
1 12 11 5 11 8 8
Acute UGB Recurring UGB Recurring UGB Recurring UGB Recurring UGB
20 3 20 12 18
10 3 8 8 6
TIPS stenosis at 54 m/UGB
TIPS thrombosis at 24 h
TIPS stenosis at 9 m/UGB
17 m Permeable TX 24 m Permeable US TIPS thrombosis at 12 m 24 m Permeable US TIPS thrombosis at 7 m/UGB 23 m Permeable US 1 m Permeable TX
mmonths, TX liver transplantation, UGB upper gastrointestinal bleeding, US Doppler ultrasonography, y years
gastrointestinal bleeding at the time of inserting the TIPS and required transfusion. None of the children showed signs of liver encephalopathy or ascites before the TIPS implant. The portal axis was studied by Doppler US in all of the children, being patent in nine, and chronically occluded in three children. The technical success was 100%, with a single ePTFEcovered endoprosthesis implanted in a single session in each child with permeability of the TIPS in the first week, with no
associated mortality (Fig. 1). A girl receiving transplantation with portal thrombosis and cavernomatosis sustained an acute TIPS thrombosis 24 h post implantation that appeared to be severe liver failure. This was recannulated using an 8-mmdiameter coaxial endoprosthesis (Dynamic; Biotronik, Black, Sweden), which remained patent as assessed with US during 30 months of follow-up. The mean initial transhepatic gradient was 15 mmHg (range 3–21 mmHg), dropping to 7 mmHg (range 1–
Fig. 1 A 10-year and 9-month-old 30-kg boy (case 9 in Table 1) with spontaneous splenorenal shunt from endoscopic evidence of varices and recurring upper gastrointestinal bleeding. a Indirect portography with contrast agent injection in the superior mesenteric artery shows portal permeability with opacification of the gastric varices (star) and a spontaneous
splenorenal shunt (arrow) caused by portal hypertension. The transhepatic gradient was 3 mmHg. b Direct portography shows a 10-mm diameter TIPS implant (final transhepatic gradient: 3 mmHg) without any opacification of gastric varices, and the splenorenal shunt after the TIPS implantation. This child did not experience post-TIPS encephalopathy
Pediatr Radiol (2015) 45:422–429 Table 2 Radiation exposure in children
Procedure time: includes both fluoroscopy and anaesthesia time
425
Case
Age
Procedure time (min)
Fluoroscopy time (s)
Kerma area product total (Gy·cm2)
1 2 3 4 5
12 y 7 m 11 y 2 m 16 y 12 y 9y
180 240 210 150 180
1,374 2,100 2,946 1,416 1,740
562,6 1,356, 1 1,571, 2 880, 9 849, 5
6 7 8 9 10 11 12
6 y 10 m 13 y 5 m 5y8m 10 y 10 m 4y 10 y 2 m 2y
180 190 240 150 210 230 196
2,136 2,154 2,142 900 3,600 2,118 1,740
706, 5 1,662, 3 1,383, 2 411, 1 1,672, 6 1,534, 5 534, 4
12 mmHg) after the TIPS implant (Table 1). It was decided to implant the TIPS in one child with a transhepatic gradient of 3 mmHg and a spontaneous splenorenal shunt because of endoscopic evidence of varices and recurring upper gastrointestinal bleeding (Fig. 1). The procedure involved a mean Kerma-area-product of 1,093.7 Gy·cm2 (range 411.1–1,672 Gy⋅cm2) with a mean fluoroscopy time of 2,030 s (range 900–2,946 s). The procedure involved a mean full time (including fluoroscopy and anaesthesia time) of 196 min (range 150–240 min) (Table 2). Immediate complications observed in one child were a post-TIPS transhepatic gradient of 1 mmHg and a slight ammonaemia, as well as symptoms of mild hepatic encephalopathy 24 h after the TIPS implantation, which resolved after a week with medical treatment using lactulose enemas and antibiotics. Late complications observed were recurrence of the upper gastrointestinal bleeding from enlarging varices in three children. In two of them, 9 months and 54 months after the TIPS implant, respectively, the bleeding was caused by an intrastent stenosis, which was recannulated with the implantation of two uncovered coaxial endoprostheses (Wallstent 10 × 94; Boston Scientific, Galway, Ireland, and Omnilink 9 × 59; Abbott, Santa Clara, CA). In the third child, who had liver transplantation with portal cavernomatosis, thrombosis of the TIPS was identified at 7 months post-implant (Fig. 2). This was recannulated by implanting another uncovered coaxial endoprosthesis (Protege EverFlex; eV3, Plymouth, MN). Permeability was maintained in all three children during the follow-up. The first one was transplanted 8 months later and the third was on the transplant waiting list at the time of this report. The second child is actually out of transplantation waiting list. In a fourth child, Doppler US detected thrombosis of the TIPS after 12 months. This thrombosis was recannulated using a US-guided transhepatic route (it wasn’t possible to recannalise it by the usual endovascular route) for
implantation of an uncovered coaxial stent (Protege EverFlex 7 × 120; eV3, Plymouth MN) (Fig. 3). Initial clinical success was 75%, with only three children sustaining episodes of upper gastrointestinal bleeding during the clinical follow-up period or before transplantation. Five children underwent liver transplantation without technical complications at 1, 7, 9, 10 and 17 months after TIPS implantations, respectively, and macroscopic analysis demonstrated liver permeability (Table 1). Four of the 12 TIPS required recanalization, so primary permeability of our PTFE-covered TIPS was 67%, and secondary permeability was 100% during a mean follow-up of 22 months (range 1–58 months). All the children had thrombocytopaenia of 1,000– 110,000/mm3 (normal range: 150,000–500,000/mm3) before the TIPS implantation but a improvement after implantation was only observed in one (from 1,000/mm3 to 85,000/mm3), while none of the children got worse. Also, in five children there was a slight increase in post-TIPS ammonaemia (125– 265 μg/dL, normal range: 30–123 μg/dL), which normalised one or two weeks after the procedure.
Discussion Before TIPS implantation, hepatic functional insufficiency and clinically overt hepatic encephalopathy should be excluded. TIPS have been used to treat most of the complications of portal hypertension, the main indication in our series being acute or recurrent variceal bleeding that was medically or endoscopically uncontrollable. Absolute contraindications for the TIPS procedure include severe coagulopathy; however portal thrombosis and cavernomatosis are not contraindications. In our study the use of ePTFE-covered TIPS controlled the upper gastrointestinal bleeding in 75% of children, with only
426
Pediatr Radiol (2015) 45:422–429
Fig. 2 A 10-year and 2-month-old boy (case 11 in Table 1) weighing 30 kg with a liver transplant. a Indirect portography with contrast agent injection into the superior mesenteric artery shows chronic portal vein thrombosis (arrow) with cavernomatosis (star), which recannulates the intrahepatic portal branches. b Portogram demonstrates the first step in creating the transjugular intrahepatic portosystemic shunt (TIPS) with the catheter in the start of the portal thrombosis (arrow). c Access is achieved through a thin periportal collateral vein using a calibrated catheter to the superior mesenteric vein (arrows). The periportal cavernomatosis (star) and small gastroesophageal varices are opacified. Transhepatic gradient is
12 mmHg. d Shown is a Viatorr (W.L. Gore & Associates, Flagstaff, Arizona) TIPS implant with coaxial prolongation Wallstent (Boston Scientific, Galway, Ireland) towards the superior mesenteric vein. The gradient is 8 mmHg, ceasing opacification of the cavernomatosis and gastroesophageal varices. e At 7 months post implant, a TIPS thrombosis was suspected by Doppler US after a considerable enlargement of varices was demonstrated by endoscopy. Here the thrombosis is confirmed by portography with opacification of the cavernomatosis (arrowhead) and varices (star). f End result after coaxial recannulation of the TIPS with two uncovered stents. Two months later, the TIPS remained patent
three episodes of rebleeding (25%) after revision during follow-up. Our results are similar to those published for this type of covered endoprosthesis [7]. However, with uncovered endoprostheses, rebleeds have been detected in up to 67% of children [6] (Table 3). The technical procedure for TIPS implantation in our series included blind puncture of the portal vein under fluoroscopy through a hepatic vein, based on our experience with TIPS in adults, which created fewer Colapinto needle passages than using US guidance. We only used US guidance in cases of cavernomatosis, in which we had to choose the collateral vein
with greater caliber. This is contrary to some other authors who during the portal vein puncture performed continuous US monitoring [8]. Our technical success was also 100%; although we had an acute TIPS thrombosis at 24 h, it was resolved in the first 72 h by recannulation. Although others have reported sporadic cases of the impossibility of implanting TIPS and higher mortality and morbidity associated with hemoperitoneum [6, 12], we found it to be a safe technique. One of our patients sustained transitory encephalopathy, which has been described in up to 11% of cases, always associated with end gradients 12 mmHg; we insist on this strategy because not all patients are candidates for liver transplantation, hence
Pediatr Radiol (2015) 45:422–429
in a minority of growing children a time may come when the 8-mm endoprosthesis is under-sized for portal vein growth, making recurrence of portal hypertension theoretically possible. Thus we always over-dilate the endoprosthesis to 10 mm in case this occurs. Only in the youngest boy of our series (a 2year-old weighing 11 kg) we dilated to a maximum of 6 mm and obtained a good gradient of 6 mmHg. The minimal experience with TIPS implants in children with liver grafts [8, 18] and chronic portal thrombosis represents an additional technical difficulty with TIPS. In children with cavernomatosis, who have very small-calibre collateral veins, there is higher theoretical risk of thrombosis and rupture of the vein. In our series there were two children in whom implant was achieved with technical difficulty but without complications (Fig. 2), and recannulation of the occluded portal segment was achieved. However, in both children there was thrombosis of the TIPS (acute in one of those) during follow-up and we were able to recannulate it by implanting coaxial endoprostheses. The occurrence of portal vein thrombosis, however, does not affect the outcome nor lessen the value of the treatment [19]. TIPS can be considered an efficient long-term alternative to surgery in children with a good overall prognosis when surgical shunting is not feasible, when there are medical complications that increase the risk of surgical shunting, and in small children as a bridge to liver transplantation [9]. Thrombocytopaenia does not normally improve with TIPS implantation [6, 20], even though this clearly occurred in one of our patients. A transitory and almost always asymptomatic increase in post-implant ammonaemia [7] is common, as occurred in five of our patients, with only symptoms being mild encephalopathy in one. TIPS implantation is an interventional radiology procedure known to have a higher radiation dose than other radiologic procedures. In our series the mean Kerma-area-product, a measure of the total X-ray energy absorbed, was 1,093.7 Gy⋅cm2. Sometimes technical difficulty leads to prolonged procedures and radiation doses that are higher than expected [21]. It is essential to manage radiation dose during paediatric fluoroscopy and to optimise dose [22]. The use of US to monitor TIPS in every child could minimise the fluoroscopy time and consequently radiation exposure. The major limitations of our study are the small number of patients and the fact that it was a retrospective single-centre study. There are also no long-term follow-up results.
Conclusion This small case series suggests that ePTFE-covered endoprostheses are safe in children and their permeability is good, but reintervention is frequently necessary in children with portal cavernomatosis.
Pediatr Radiol (2015) 45:422–429 Acknowledgements The authors thank J.A. Miñano for supporting the estimation of fluoroscopic radiation doses. Conflicts of interest None
References 1. Garcia-Pagan JC, Caca K, Bureau C et al (2010) Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med 362: 2370–2379 2. Rossle M, Ochs A, Gulberg V et al (2000) A comparison of paracentesis and transjugular intrahepatic portosystemic shunting in patients with ascites. N Engl J Med 342:1701–1707 3. Hackworth CA, Leef JA, Rosenblum JD et al (1998) Transjugular intrahepatic portosystemic shunt creation in children: initial clinical experience. Radiology 206:109–114 4. Lorenz J (2008) Placement of transjugular intrahepatic portosystemic shunts in children. Technol Cardiovasc Intervent Radiol 11:235–240 5. Bureau C, Garcia-Pagan JC, Layrargues GP et al (2007) Patency of stents covered with polytetrafluorethylene in patients treated by transjugular intrahepatic portosystemic shunts: long-term results of a randomised multicentre study. Liver Int 27:742–747 6. Huppert PE, Goffette P, Astfal W et al (2002) Transjugular intrahepatic portosystemsic shunt in children with biliary atresia. Cardiovasc Intervent Radiol 25:484–493 7. Vo NJ, Shivariam GM, Andrews RT et al (2012) Midterm follow-up of transjugular intrahepatic portosystemic shunts using polytetrafluoroethylene endografts in children. J Vasc Interv Radiol 23:919–924 8. Di Giorgio A, Agazzi R, Alberti D et al (2012) Feasibility and efficacy of transjugular intrahepatic porto-systemic shunt (TIPS) in children. J Pediatr Gastroenterol Nutr 54:594–600 9. Shneider BL, Bosch J, de Franchis R et al (2012) Portal hypertension in children: expert pediatric opinion on the report of the Baveno v Consensus workshop on methodology of diagnosis and therapy in portal hypertension. Pediatr Transplant 16:426–437
429 10. Eesa M, Clark T (2011) Transjugular intrahepatic portosystemic shunt: state of the art. Semin Roentgenol 46:125–132 11. Kalva SP, Salazar GM, Walker G (2009) Transjugular intrahepatic portosystemic shunt of acute variceal hemorrhage. Technol Cardiovasc Intervent Radiol 12:92–101 12. Pozler O, Krajina A, Vanicek H et al (2003) Transjugular intrahepatic portosystemic shunt in five children with cystic fibrosis: long term results. Hepatogastroenterology 50:1111–1114 13. Heyman M, LaBerge J, Sombely K et al (1997) Transjugular intrahepatic portosystemic shunt in children. J Pediatr 131:914–919 14. Mileti E, Rosenthal P (2011) Management of portal hypertension in children. Curr Gastroenterol Rep 13:10–16 15. Chlapoutaki CE, Franchi-Abella S, Habes D et al (2009) Custommade covered transjugular intrahepatic portosytemic shunt (TIPS) in an infant with trisomy 22 and biliary atresia. Pediatr Radiol 39:739– 742 16. Maleux G, Perez-Gutierrez NA, Evrard S et al (2010) Covered stents are better than uncovered stents for transjugular intrahepatic portosystemic shunts in cirrhotic patients with refractory ascites: a retrospective cohort study. Acta Gastroenterol Belg 73:336–341 17. Mermuys K, Maleux G, Heye S et al (2008) Use of the Viatorr expanded polytetrafluoroethylene-covered stent-graft for transjugular intrahepatic portosystemic shunt creation in children: initial clinical experience. Cardiovasc Intervent Radiol 31:S192–S196 18. Van Ha TG, Funaki BS, Ehrhardt J et al (2005) Transjugular intrahepatic portosystemic shunt placement in liver transplant recipients: experiences with pediatric and adult patients. AJR Am J Roentgenol 184:920–925 19. Rössle M (2013) TIPS: 25 years later. J Hepatol 59:1081–1093 20. Schweizer P, Brambs HJ, Schweizer M et al (1995) TIPS: a new therapy for esophageal variceal bleeding caused by EHBA. Eur J Pediatr Surg 5:211–215 21. Miller DL, Balter S, Schueler BA et al (2010) Clinical radiation management for fluoroscopically guided interventional procedures. Radiology 257:321–332 22. Hernanz-Schulman M, Goske MJ, Bercha IH et al (2011) Pause and pulse: ten steps that help manage radiation dose during pediatric fluoroscopy. AJR Am J Roentgenol 197:475–481
ORIGINAL ARTICLE
Safety and efficacy of expanded polytetrafluoroethylene-covered transjugular intrahepatic portosystemic shunts in children with acute or recurring upper gastrointestinal bleeding Luis J. Zurera & Juan J. Espejo & Sagrario Lombardo & Juan J. Gilbert & Miguel Canis & Concepción Ruiz
Received: 26 December 2013 / Revised: 28 July 2014 / Accepted: 28 August 2014 / Published online: 28 November 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background The transjugular intrahepatic portosystemic shunt (TIPS) is effective for treating complications of portal hypertension in cirrhotic adults but the experience in children is limited. Objective To retrospectively review the safety and efficacy of expanded polytetrafluoroethylene (ePTFE)-covered TIPS in children with acute or recurrent gastrointestinal bleeding. Materials and methods We reviewed the medical records of children who received implants of 10-mm-diameter PTFEcovered endoprostheses for acute or recurring upper gastrointestinal bleeding caused by medically or endoscopically uncontrollable varices. The recurrence of upper gastrointestinal bleeding, associated complications and permeability were assessed with Doppler sonography sequentially or up to transplantation. Results In all children (n=12; mean age 9 years; mean weight 30 kg) a single endoprosthesis was implanted with no associated mortality. The mean initial transhepatic gradient was 15 mmHg (range 3–21 mmHg), dropping to 7 mmHg (range 1–12 mmHg) after TIPS. Immediate complications were mild encephalopathy (n=1) and acute occlusion of the TIPS (n=1). Stenosis of the
L. J. Zurera : J. J. Espejo : S. Lombardo (*) : M. Canis Interventional Radiology Unit, Hospital Universitario Reina Sofía, Menendez Pidal s/n, 14004 Córdoba, Spain e-mail: [email protected] J. J. Gilbert Paediatric Gastroenterology Unit, Hospital Universitario Reina Sofía, Córdoba, Spain C. Ruiz Anaesthesiology Department, Hospital Universitario Reina Sofía, Córdoba, Spain
TIPS was observed in two children, at 9 months and 54 months follow-up, and thrombosis was observed in two children, at 7 months and 12 months follow-up. All four stenoses/ occlusions were resolved with coaxial endoprostheses. Conclusion The safety profile and efficacy of expanded polytetrafluoroethylene (ePTFE)-covered TIPS were satisfactory in this small series of children with acute or recurrent gastrointestinal bleeding. Keywords Intrahepatic portosystemic shunt . Portal vein thrombosis . Liver transplantation . Cavernomatosis . Venography . Fluoroscopy . Child
Introduction Creation of a transjugular intrahepatic portosystemic shunt (TIPS) is an effective procedure for treating complications of portal hypertension in adults with cirrhosis (mainly upper gastrointestinal bleeding and refractory ascites) when medical treatment and conventional endoscopic therapy have failed [1, 2]. However, experience with TIPS in the paediatric population is limited and the largest series of cases in the literature is 12 children [3]. Although the implant technique and the indications are similar in adults and children, children have a series of special considerations [4]. First, their liver and veins are smaller than those of adults. Second, the aetiology of cirrhosis in children is different from that in adults (extrahepatic biliary atresias, cystic fibrosis, cystic bile malformation), so children have major periportal fibrosis. It has also been shown that the permeability of TIPS in adults is considerably greater when using polytetrafluoroethylene (PTFE)-covered endoprostheses than when using uncovered ones [5]. However, most of the
Pediatr Radiol (2015) 45:422–429
published case series in children use the latter, with a high rate of recurring stenosis [6], although two of the largest and most recent series of TIPS in children use covered stents, with a good midterm patency rate [7, 8]. Our objectives are to present our initial experience with 12 children with portal hypertension of different causes treated with ePTFE-covered TIPS and to analyse their safety and efficacy retrospectively, and to present some technical considerations regarding their implantation.
Materials and methods Ethical approval was waived for this retrospective observational study, in accordance with our hospital guidelines. We reviewed the medical records of children who between May 2009 and May 2012 had TIPS implanted for the treatment of upper gastrointestinal bleeding secondary to portal hypertension with rebleeding despite secondary prophylaxis. We defined the failure to prevent rebleeding as a recurrent melena or hematemesis resulting in hospital admission, blood transfusion and Three gram drop in haemoglobin [9]. All procedures were performed with general anaesthetic and informed consent for TIPS was obtained from parents. The implant technique and equipment used were the same as for adults (Ring TIPSS-200 set; Cook Inc., Bjaeverskov, Denmark) [10, 11]. The right internal jugular vein was punctured and US guidance was used to control a 10-French introducer. Access was gained through a hepatic vein (right or middle). The needle used for the intrahepatic puncture and access to the portal system was the 16-gauge Colapinto included in the set. Indirect portography or wedged suprahepatic venography was used for guidance. A direct portography using a conventional 5-F calibrated pigtail catheter was performed in all children to select the right length for the endoprosthesis and view the portal anatomy, and to perform a hepatic haemodynamic study to establish the transhepatic gradient between the portal vein and the right atrium/inferior cava vein before and after the TIPS implantation. Intraoperative sonography was required to access the periportal veins in three children with portal cavernomatosis. The intrahepatic tract between the suprahepatic vein and the portal branch was dilated with a conventional 8-mm-diameter balloon catheter. There was no embolisation of gastric varices in any of the patients. The maximum dose of iodinated contrast agent used throughout the procedure was 3 ml/kg of weight. The average diameter of the main portal vein was 13 mm (range 10–17 mm) and the average of the collateral vein chosen in cases of cavernomatosis was 8 mm (range 7– 10 mm). The mean diameter of suprahepatic vein was 14 mm (range 9–25 mm). All of these measurements were performed using a calibrated catheter in direct and indirect portography under fluoroscopy.
423
In all cases a 10-mm-diameter ePTFE-covered nitinol endoprosthesis (Viatorr R; W.L. Gore and Associates, Flagstaff, AZ) was used, dilated initially to 8 mm and to 10 mm if the transhepatic gradient was greater than 12 mmHg; only in the youngest boy (a 2-year-old) was it dilated to 6 mm. The length of the ePTFE graft stent was variable, determined by tract measurement from the portal vein entry point to the hepatic vein inferior cava junction, leaving the proximal end in the suprahepatic vein and avoiding its protrusion into the right atrium at all times (to prevent technical complications during future transplantation). The lengths used for implanted ePTFE-covered nitinol endoprostheses were 10 mm (n=7), 9 mm (n=3) and 8 mm (n=2). The purpose of the procedure was to obtain a porto-systemic gradient below 12 mmHg. In children with chronic portal thrombosis and cavernomatosis, we chose a collateral vein of major caliber and the wire was advanced through this vein to the extrahepatic portal vein remnant or superior mesenteric or splenic veins. No antibiotics were administered. Infant aspirin was administered (3 mg/kg/day) after the TIPS procedure to the three children with portal cavernomatosis. All the children were followed clinically and with Doppler US within 4–6 days after the TIPS implantation and every 3 months up to the time of this study or up to transplantation; endoscopy was also performed at 3–6 months and at 1 year. Permeability of the TIPS for the first week after implantation was considered technical success and the control of upper gastrointestinal bleeding as a clinical success.
Results The sample consisted of 9 boys and 3 girls with an average age of 9 years (range 2–16 years) and a mean weight of 30 kg (range 11–60 kg). Portal hypertension was caused by cirrhosis in 10 children: there were 3 cases of extrahepatic biliary atresia (treated in infancy using Kasai procedure), two of cystic fibrosis, 1 of congenital hepatic fibrosis, 1 of complex Caroli disease cystic bile malformation, 1 of ductopenia, 1 of cirrhosis of unknown origin after biopsy and 1 of liver transplant in a newborn because of tyrosinaemia with chronic portal thrombosis and cavernomatosis. In the other two children liver disease was non-cirrhotic with thrombosis and portal cavernomatosis (one of them presented 6 years after a living-donor liver transplantation following biliary atresia and the other had umbilical catheterisation in the neonatal period). All children had hypersplenism with associated thrombocytopaenia and endoscopic evidence of gastroesophageal varices causing repeated episodes of upper gastrointestinal bleeding that were not stopped with standard medical treatment or endoscopic variceal ligation (n=2) or sclerotherapy (n=1) (Table 1). Seven of the children showed upper
424
Pediatr Radiol (2015) 45:422–429
Table 1 Demographic characteristics of 12 children treated with transjugular intrahepatic portosystemic shunt (TIPS), and results of the procedure Case Age
Weight (kg) Underlying cause
TIPS indication Gradient
Complications
Follow-up
Mild encephalopathy
36 m Permeable US 7 m Permeable TX 58 m Permeable US 9 m Permeable TX 10 m Permeable TX 30 m Permeable US 25 m Permeable US
Pre- Post1 2 3 4 5 6 7
12 y 7 m 11 y 2 m 16 y 12 y 9y 6 y 10 m 13 y 5 m
27 29 60 35 30 20 55
8 9 10 11 12
5y8m 10 y 10 m 4y 10 y 2 m 2y
14 30 22 30 11
Cystic fibrosis Cystic fibrosis Congenital hepatic fibrosis Biliary atresia Caroli disease Liver TX Cavernomatosis Portal cavernomatosis due to umbilical catheter Biliary atresia Cirrhosis of unknown origin Biliary atresia Liver TX Cavernomatosis Ductopenia
Recurring UGB Acute UGB Acute UGB Acute UGB Acute UGB Acute UGB Acute UGB
12 15 21 11 21 18 15
1 12 11 5 11 8 8
Acute UGB Recurring UGB Recurring UGB Recurring UGB Recurring UGB
20 3 20 12 18
10 3 8 8 6
TIPS stenosis at 54 m/UGB
TIPS thrombosis at 24 h
TIPS stenosis at 9 m/UGB
17 m Permeable TX 24 m Permeable US TIPS thrombosis at 12 m 24 m Permeable US TIPS thrombosis at 7 m/UGB 23 m Permeable US 1 m Permeable TX
mmonths, TX liver transplantation, UGB upper gastrointestinal bleeding, US Doppler ultrasonography, y years
gastrointestinal bleeding at the time of inserting the TIPS and required transfusion. None of the children showed signs of liver encephalopathy or ascites before the TIPS implant. The portal axis was studied by Doppler US in all of the children, being patent in nine, and chronically occluded in three children. The technical success was 100%, with a single ePTFEcovered endoprosthesis implanted in a single session in each child with permeability of the TIPS in the first week, with no
associated mortality (Fig. 1). A girl receiving transplantation with portal thrombosis and cavernomatosis sustained an acute TIPS thrombosis 24 h post implantation that appeared to be severe liver failure. This was recannulated using an 8-mmdiameter coaxial endoprosthesis (Dynamic; Biotronik, Black, Sweden), which remained patent as assessed with US during 30 months of follow-up. The mean initial transhepatic gradient was 15 mmHg (range 3–21 mmHg), dropping to 7 mmHg (range 1–
Fig. 1 A 10-year and 9-month-old 30-kg boy (case 9 in Table 1) with spontaneous splenorenal shunt from endoscopic evidence of varices and recurring upper gastrointestinal bleeding. a Indirect portography with contrast agent injection in the superior mesenteric artery shows portal permeability with opacification of the gastric varices (star) and a spontaneous
splenorenal shunt (arrow) caused by portal hypertension. The transhepatic gradient was 3 mmHg. b Direct portography shows a 10-mm diameter TIPS implant (final transhepatic gradient: 3 mmHg) without any opacification of gastric varices, and the splenorenal shunt after the TIPS implantation. This child did not experience post-TIPS encephalopathy
Pediatr Radiol (2015) 45:422–429 Table 2 Radiation exposure in children
Procedure time: includes both fluoroscopy and anaesthesia time
425
Case
Age
Procedure time (min)
Fluoroscopy time (s)
Kerma area product total (Gy·cm2)
1 2 3 4 5
12 y 7 m 11 y 2 m 16 y 12 y 9y
180 240 210 150 180
1,374 2,100 2,946 1,416 1,740
562,6 1,356, 1 1,571, 2 880, 9 849, 5
6 7 8 9 10 11 12
6 y 10 m 13 y 5 m 5y8m 10 y 10 m 4y 10 y 2 m 2y
180 190 240 150 210 230 196
2,136 2,154 2,142 900 3,600 2,118 1,740
706, 5 1,662, 3 1,383, 2 411, 1 1,672, 6 1,534, 5 534, 4
12 mmHg) after the TIPS implant (Table 1). It was decided to implant the TIPS in one child with a transhepatic gradient of 3 mmHg and a spontaneous splenorenal shunt because of endoscopic evidence of varices and recurring upper gastrointestinal bleeding (Fig. 1). The procedure involved a mean Kerma-area-product of 1,093.7 Gy·cm2 (range 411.1–1,672 Gy⋅cm2) with a mean fluoroscopy time of 2,030 s (range 900–2,946 s). The procedure involved a mean full time (including fluoroscopy and anaesthesia time) of 196 min (range 150–240 min) (Table 2). Immediate complications observed in one child were a post-TIPS transhepatic gradient of 1 mmHg and a slight ammonaemia, as well as symptoms of mild hepatic encephalopathy 24 h after the TIPS implantation, which resolved after a week with medical treatment using lactulose enemas and antibiotics. Late complications observed were recurrence of the upper gastrointestinal bleeding from enlarging varices in three children. In two of them, 9 months and 54 months after the TIPS implant, respectively, the bleeding was caused by an intrastent stenosis, which was recannulated with the implantation of two uncovered coaxial endoprostheses (Wallstent 10 × 94; Boston Scientific, Galway, Ireland, and Omnilink 9 × 59; Abbott, Santa Clara, CA). In the third child, who had liver transplantation with portal cavernomatosis, thrombosis of the TIPS was identified at 7 months post-implant (Fig. 2). This was recannulated by implanting another uncovered coaxial endoprosthesis (Protege EverFlex; eV3, Plymouth, MN). Permeability was maintained in all three children during the follow-up. The first one was transplanted 8 months later and the third was on the transplant waiting list at the time of this report. The second child is actually out of transplantation waiting list. In a fourth child, Doppler US detected thrombosis of the TIPS after 12 months. This thrombosis was recannulated using a US-guided transhepatic route (it wasn’t possible to recannalise it by the usual endovascular route) for
implantation of an uncovered coaxial stent (Protege EverFlex 7 × 120; eV3, Plymouth MN) (Fig. 3). Initial clinical success was 75%, with only three children sustaining episodes of upper gastrointestinal bleeding during the clinical follow-up period or before transplantation. Five children underwent liver transplantation without technical complications at 1, 7, 9, 10 and 17 months after TIPS implantations, respectively, and macroscopic analysis demonstrated liver permeability (Table 1). Four of the 12 TIPS required recanalization, so primary permeability of our PTFE-covered TIPS was 67%, and secondary permeability was 100% during a mean follow-up of 22 months (range 1–58 months). All the children had thrombocytopaenia of 1,000– 110,000/mm3 (normal range: 150,000–500,000/mm3) before the TIPS implantation but a improvement after implantation was only observed in one (from 1,000/mm3 to 85,000/mm3), while none of the children got worse. Also, in five children there was a slight increase in post-TIPS ammonaemia (125– 265 μg/dL, normal range: 30–123 μg/dL), which normalised one or two weeks after the procedure.
Discussion Before TIPS implantation, hepatic functional insufficiency and clinically overt hepatic encephalopathy should be excluded. TIPS have been used to treat most of the complications of portal hypertension, the main indication in our series being acute or recurrent variceal bleeding that was medically or endoscopically uncontrollable. Absolute contraindications for the TIPS procedure include severe coagulopathy; however portal thrombosis and cavernomatosis are not contraindications. In our study the use of ePTFE-covered TIPS controlled the upper gastrointestinal bleeding in 75% of children, with only
426
Pediatr Radiol (2015) 45:422–429
Fig. 2 A 10-year and 2-month-old boy (case 11 in Table 1) weighing 30 kg with a liver transplant. a Indirect portography with contrast agent injection into the superior mesenteric artery shows chronic portal vein thrombosis (arrow) with cavernomatosis (star), which recannulates the intrahepatic portal branches. b Portogram demonstrates the first step in creating the transjugular intrahepatic portosystemic shunt (TIPS) with the catheter in the start of the portal thrombosis (arrow). c Access is achieved through a thin periportal collateral vein using a calibrated catheter to the superior mesenteric vein (arrows). The periportal cavernomatosis (star) and small gastroesophageal varices are opacified. Transhepatic gradient is
12 mmHg. d Shown is a Viatorr (W.L. Gore & Associates, Flagstaff, Arizona) TIPS implant with coaxial prolongation Wallstent (Boston Scientific, Galway, Ireland) towards the superior mesenteric vein. The gradient is 8 mmHg, ceasing opacification of the cavernomatosis and gastroesophageal varices. e At 7 months post implant, a TIPS thrombosis was suspected by Doppler US after a considerable enlargement of varices was demonstrated by endoscopy. Here the thrombosis is confirmed by portography with opacification of the cavernomatosis (arrowhead) and varices (star). f End result after coaxial recannulation of the TIPS with two uncovered stents. Two months later, the TIPS remained patent
three episodes of rebleeding (25%) after revision during follow-up. Our results are similar to those published for this type of covered endoprosthesis [7]. However, with uncovered endoprostheses, rebleeds have been detected in up to 67% of children [6] (Table 3). The technical procedure for TIPS implantation in our series included blind puncture of the portal vein under fluoroscopy through a hepatic vein, based on our experience with TIPS in adults, which created fewer Colapinto needle passages than using US guidance. We only used US guidance in cases of cavernomatosis, in which we had to choose the collateral vein
with greater caliber. This is contrary to some other authors who during the portal vein puncture performed continuous US monitoring [8]. Our technical success was also 100%; although we had an acute TIPS thrombosis at 24 h, it was resolved in the first 72 h by recannulation. Although others have reported sporadic cases of the impossibility of implanting TIPS and higher mortality and morbidity associated with hemoperitoneum [6, 12], we found it to be a safe technique. One of our patients sustained transitory encephalopathy, which has been described in up to 11% of cases, always associated with end gradients 12 mmHg; we insist on this strategy because not all patients are candidates for liver transplantation, hence
Pediatr Radiol (2015) 45:422–429
in a minority of growing children a time may come when the 8-mm endoprosthesis is under-sized for portal vein growth, making recurrence of portal hypertension theoretically possible. Thus we always over-dilate the endoprosthesis to 10 mm in case this occurs. Only in the youngest boy of our series (a 2year-old weighing 11 kg) we dilated to a maximum of 6 mm and obtained a good gradient of 6 mmHg. The minimal experience with TIPS implants in children with liver grafts [8, 18] and chronic portal thrombosis represents an additional technical difficulty with TIPS. In children with cavernomatosis, who have very small-calibre collateral veins, there is higher theoretical risk of thrombosis and rupture of the vein. In our series there were two children in whom implant was achieved with technical difficulty but without complications (Fig. 2), and recannulation of the occluded portal segment was achieved. However, in both children there was thrombosis of the TIPS (acute in one of those) during follow-up and we were able to recannulate it by implanting coaxial endoprostheses. The occurrence of portal vein thrombosis, however, does not affect the outcome nor lessen the value of the treatment [19]. TIPS can be considered an efficient long-term alternative to surgery in children with a good overall prognosis when surgical shunting is not feasible, when there are medical complications that increase the risk of surgical shunting, and in small children as a bridge to liver transplantation [9]. Thrombocytopaenia does not normally improve with TIPS implantation [6, 20], even though this clearly occurred in one of our patients. A transitory and almost always asymptomatic increase in post-implant ammonaemia [7] is common, as occurred in five of our patients, with only symptoms being mild encephalopathy in one. TIPS implantation is an interventional radiology procedure known to have a higher radiation dose than other radiologic procedures. In our series the mean Kerma-area-product, a measure of the total X-ray energy absorbed, was 1,093.7 Gy⋅cm2. Sometimes technical difficulty leads to prolonged procedures and radiation doses that are higher than expected [21]. It is essential to manage radiation dose during paediatric fluoroscopy and to optimise dose [22]. The use of US to monitor TIPS in every child could minimise the fluoroscopy time and consequently radiation exposure. The major limitations of our study are the small number of patients and the fact that it was a retrospective single-centre study. There are also no long-term follow-up results.
Conclusion This small case series suggests that ePTFE-covered endoprostheses are safe in children and their permeability is good, but reintervention is frequently necessary in children with portal cavernomatosis.
Pediatr Radiol (2015) 45:422–429 Acknowledgements The authors thank J.A. Miñano for supporting the estimation of fluoroscopic radiation doses. Conflicts of interest None
References 1. Garcia-Pagan JC, Caca K, Bureau C et al (2010) Early use of TIPS in patients with cirrhosis and variceal bleeding. N Engl J Med 362: 2370–2379 2. Rossle M, Ochs A, Gulberg V et al (2000) A comparison of paracentesis and transjugular intrahepatic portosystemic shunting in patients with ascites. N Engl J Med 342:1701–1707 3. Hackworth CA, Leef JA, Rosenblum JD et al (1998) Transjugular intrahepatic portosystemic shunt creation in children: initial clinical experience. Radiology 206:109–114 4. Lorenz J (2008) Placement of transjugular intrahepatic portosystemic shunts in children. Technol Cardiovasc Intervent Radiol 11:235–240 5. Bureau C, Garcia-Pagan JC, Layrargues GP et al (2007) Patency of stents covered with polytetrafluorethylene in patients treated by transjugular intrahepatic portosystemic shunts: long-term results of a randomised multicentre study. Liver Int 27:742–747 6. Huppert PE, Goffette P, Astfal W et al (2002) Transjugular intrahepatic portosystemsic shunt in children with biliary atresia. Cardiovasc Intervent Radiol 25:484–493 7. Vo NJ, Shivariam GM, Andrews RT et al (2012) Midterm follow-up of transjugular intrahepatic portosystemic shunts using polytetrafluoroethylene endografts in children. J Vasc Interv Radiol 23:919–924 8. Di Giorgio A, Agazzi R, Alberti D et al (2012) Feasibility and efficacy of transjugular intrahepatic porto-systemic shunt (TIPS) in children. J Pediatr Gastroenterol Nutr 54:594–600 9. Shneider BL, Bosch J, de Franchis R et al (2012) Portal hypertension in children: expert pediatric opinion on the report of the Baveno v Consensus workshop on methodology of diagnosis and therapy in portal hypertension. Pediatr Transplant 16:426–437
429 10. Eesa M, Clark T (2011) Transjugular intrahepatic portosystemic shunt: state of the art. Semin Roentgenol 46:125–132 11. Kalva SP, Salazar GM, Walker G (2009) Transjugular intrahepatic portosystemic shunt of acute variceal hemorrhage. Technol Cardiovasc Intervent Radiol 12:92–101 12. Pozler O, Krajina A, Vanicek H et al (2003) Transjugular intrahepatic portosystemic shunt in five children with cystic fibrosis: long term results. Hepatogastroenterology 50:1111–1114 13. Heyman M, LaBerge J, Sombely K et al (1997) Transjugular intrahepatic portosystemic shunt in children. J Pediatr 131:914–919 14. Mileti E, Rosenthal P (2011) Management of portal hypertension in children. Curr Gastroenterol Rep 13:10–16 15. Chlapoutaki CE, Franchi-Abella S, Habes D et al (2009) Custommade covered transjugular intrahepatic portosytemic shunt (TIPS) in an infant with trisomy 22 and biliary atresia. Pediatr Radiol 39:739– 742 16. Maleux G, Perez-Gutierrez NA, Evrard S et al (2010) Covered stents are better than uncovered stents for transjugular intrahepatic portosystemic shunts in cirrhotic patients with refractory ascites: a retrospective cohort study. Acta Gastroenterol Belg 73:336–341 17. Mermuys K, Maleux G, Heye S et al (2008) Use of the Viatorr expanded polytetrafluoroethylene-covered stent-graft for transjugular intrahepatic portosystemic shunt creation in children: initial clinical experience. Cardiovasc Intervent Radiol 31:S192–S196 18. Van Ha TG, Funaki BS, Ehrhardt J et al (2005) Transjugular intrahepatic portosystemic shunt placement in liver transplant recipients: experiences with pediatric and adult patients. AJR Am J Roentgenol 184:920–925 19. Rössle M (2013) TIPS: 25 years later. J Hepatol 59:1081–1093 20. Schweizer P, Brambs HJ, Schweizer M et al (1995) TIPS: a new therapy for esophageal variceal bleeding caused by EHBA. Eur J Pediatr Surg 5:211–215 21. Miller DL, Balter S, Schueler BA et al (2010) Clinical radiation management for fluoroscopically guided interventional procedures. Radiology 257:321–332 22. Hernanz-Schulman M, Goske MJ, Bercha IH et al (2011) Pause and pulse: ten steps that help manage radiation dose during pediatric fluoroscopy. AJR Am J Roentgenol 197:475–481
