- Email: [email protected]
Stent Placement for the Treatment of Portal Vein Stenosis or Occlusion in Pediatric Liver Transplant Recipients
Stent Placement for the Treatment of Portal Vein Stenosis or Occlusion in Pediatric Liver Transplant Recipients Gi-Young Ko, MD, PhD, Kyu-Bo Sung, MD, PhD, SungGyu Lee, MD, PhD, Hyun-Ki Yoon, MD, PhD, Kyung Rae Kim, MD, Kyung Mo Kim, MD, PhD, and Young-Joo Lee, MD, PhD
PURPOSE: To evaluate the efficacy of stent placement for the treatment of portal vein (PV) stenosis or occlusion in pediatric liver transplant recipients. MATERIALS AND METHODS: Written informed consent was obtained from a legal guardian, and our institutional review board approved this study. Percutaneous (n ⴝ 10) or intraoperative (n ⴝ 2) stent placement was attempted in 12 pediatric recipients (age range, 6 –102 months) via the percutaneous transhepatic or inferior mesenteric vein route. Stents 6 –10 mm in diameter were placed. Technical and clinical success, complications, and patency of the PV were retrospectively analyzed. RESULTS: Technical success was achieved in 10 of 12 patients (83%) and clinical success was achieved in eight patients (67%). Eight of the 10 patients in whom technical success was achieved (80%) remained healthy with a patent PV during the 10 –58-month clinical follow-up period. One patient with technical success died of acute rejection without recurrent PV complications and another died of acute rejection after stent replacement as a result of an hourglass deformity of a deployed stent with partial thrombosis. No major procedural complications occurred. CONCLUSIONS: Based on this study in a relatively small number of patients, PV stent placement seems to be a safe and effective method for the treatment of posttransplantation PV stenosis or occlusion in pediatric patients. J Vasc Interv Radiol 2007; 18:1215–1221 Abbreviations:
LT ⫽ liver transplantation, PV ⫽ portal vein
LIVER transplantation (LT) has become a valuable treatment modality for pediatric patients with liver failure. However, portal vein (PV) stenosis is
From the Department of Radiology and Research Institute of Radiology (G.Y.K., K.B.S., H.K.Y., K.R.K.), Division of Hepatobiliary Surgery and Liver Transplantation, Department of Surgery (S.L., Y.J.L.), and Department of Pediatrics (K.M.K.), Asan Medical Center, University of Ulsan College of Medicine, 388-1 Poongnap-2-dong, Songpa-ku, Seoul 138-040, Republic of Korea. Received February 22, 2007; final revision received June 20, 2007; accepted June 21, 2007. Address correspondence to G.Y.K.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2007 DOI: 10.1016/j.jvir.2007.06.029
still one of the significant causes of graft failure or posttransplantation morbidity in pediatric patients who undergo left lateral segment transplantation. This complication has traditionally been managed by surgery; however, in recent years, percutaneous transhepatic balloon angioplasty has been found to be a safe and effective alternative for the treatment of PV stenosis (1–7). Conversely, stents have usually been used to treat elastic or recurrent stenosis after balloon angioplasty (2– 4,8), and there are only a few reports of stent placement for elastic or recurrent PV stenosis in pediatric patients after LT (3,4). The purpose of this study is to evaluate the efficacy of stent placement for the treatment of PV stenosis or occlusion in pediatric patients after LT .
MATERIALS AND METHODS Patient Population Between December 1994 and June 2006, 126 pediatric patients (age ⬍15 years) underwent LT with living (n ⫽ 118) or cadaver (n ⫽ 8) donors at our institution. Except for two patients who underwent balloon angioplasty only, 11 patients (8.7%) underwent percutaneous or intraoperative stent placement to treat PV stenosis or occlusion. One patient who had undergone LT at another hospital also underwent percutaneous PV stent placement at our institute. The study included four male patients and eight female patients ranging in age from 6 to 102 months (mean ⫾ SD, 51 months ⫾ 36). All these pa-
1215
1216
•
October 2007
PV Stent Placement in Pediatric Liver Transplant Recipients
JVIR
Demographic Data of All Patients
Pt. No.
Age (months)
Interval after Transplantation
Severity of stenosis (%)
1 2 3 4 5 6 7 8 9
12 25 26 50 53 63 65 98 98
1.2 months 3.8 months 14.2 months 36.4 months 42.0 months 52.8 months 1.0 months 76.0 months 81.7 months
90 100 90 100 100 100 80 100 100
10
10
43.8 months
100
Intraoperative treatment
Symptoms
Procedure
Ascites, effusion LFT, ascites SPM SPM, melena SPM SPM, pancytopenia LFT, ascites Melena, gastric varices Hematochezia, esophageal varices SPM, melena, pancytopenia
PTA/stent PTA/stent PTA/stent Not completed PTA/stent PTA/stent Stent PTA/stent Not completed PTA/stent
Indications
11
6
1 day
70
12
10
2 days
100
Periportal hematoma, LFT, tension PV thrombosis, LFT, kinking
Hematoma evacuation, stent Surgical thrombectomy, stent
Note.—LFT ⫽ liver function tests; NA ⫽ not available; PTA ⫽ percutaneous transluminal angioplasty; SPM ⫽ splenomegaly. * Five- and 3-month follow-up volumetric CT scans of these patients revealed shrinkage of the splenic volume by approximately 66% and 56%, respectively, of splenic volume before stent placement. † The patient was lost to follow-up approximately 12 months after stent placement. ‡ PV was patent, but the placed stent showed an hourglass deformity.
tients had received a living donor left lateral segment graft for underlying biliary atresia. Reconstruction of the PV during transplantation was performed by end-to-end anastomosis of the recipient and donor PV in 11 patients. In one patient, the type of PV reconstruction was unknown. The demographic data for all patients are shown in the Table. Stent Placement We obtained institutional review board approval at our hospital to conduct a retrospective review of the patients’ medical and imaging records. Written informed consent for the treatment of PV stenosis or occlusion was obtained from a legal guardian. All procedures were performed under general anesthesia. Ten patients underwent stent placement through a percutaneous transhepatic route and two patients underwent this procedure intraoperatively. Diagnosis of PV stenosis or occlusion was based on a combination of Doppler ultrasonogra-
phy (US) and computed tomography (CT) in nine patients and CT alone in three patients. On US, stenosis or occlusion was diagnosed if Doppler US showed no flow in the PV or an acceleration of the flow rate in the poststenotic PV of more than three times that in the prestenotic PV with a narrowing of the PV diameter of less than 2.5 mm. On CT, PV occlusion was diagnosed if CT demonstrated discontinuity of the PV. In two patients, PV stenosis or occlusion was confirmed by direct portography through the inferior mesenteric vein performed after surgical periportal bleeding control and thrombectomy, respectively. The average interval between LT and stent placement was 35 months ⫾ 30 (range, 1– 82 months) in the 10 patients who underwent percutaneous transhepatic stent placement. The remaining two patients underwent the procedure on day 1 and day 2 after LT, respectively. Percutaneous transhepatic puncture of the intrahepatic PV was performed with a 21-gauge Chiba needle (Cook, Bloomington, Ind) under US
and fluoroscopic guidance. The needle was exchanged for a 4-F coaxial dilator and a 6- or 7-F sheath over a 0.018-inch guide wire (Cook) or a 0.035-inch angled hydrophilic guide wire (Terumo, Tokyo, Japan). Then, a direct main portal venogram was obtained and the pressure gradient across the stenosis or occlusion was measured. For intraoperative procedures, the inferior mesenteric vein was punctured with an 18-gauge Angiocath (Becton Dickinson Korea, Seoul, Korea) and then the Angiocath was exchanged for a 7-F sheath over a 0.035-inch angled hydrophilic guide wire under fluoroscopic guidance with a portable C-arm (OEC 9800; General Electric, Salt Lake City, Utah). A direct main portal venogram was then obtained. A bolus of heparin (50 U/kg) was then administered directly into the PV if there was no coagulopathy. The 0.035-inch guide wire and a 5-F Cobra catheter (Cook) or Kumpe catheter (Cook) were used to traverse stenosis or occlusion of the PV. Three patients who underwent this proce-
Volume 18
Number 10
Ko et al
•
1217
Pressure gradient (mm Hg) Stent size (mm)
Before stenting
After stenting
Technical success
Clinical success
Clinical follow-up duration (months)
Image follow-up duration (months)
PV status
10 ⫻ 43 8 ⫻ 50 10 ⫻ 60 NA 10 ⫻ 50 10 ⫻ 70 10 ⫻ 70 10 ⫻ 50 10 ⫻ 60
16 5 14 NA 11 10 8 12 NA
2 0 2 NA 0 0 0 2 NA
Yes Yes Yes No Yes Yes Yes Yes No
Yes No Yes No Yes* Yes Yes Yes No
53.1 2.5 49.3 NA 58.1 11.7 51.6 9.7 NA
50.9 0.7 1.5 NA 49.4 3.0 46.3 9.7 NA
Patent Patent Patent NA Patent Patent† Patent Patent NA
NA
15
3
Yes
Yes*
56.5
53.6
Patent
6 ⫻ 40
NA
NA
Yes
Yes
22.8
21.4
Patent
10 ⫻ 40
NA
NA
No
No
dure within 1 month after LT (patients 7, 11, and 12) underwent primary stent placement without balloon angioplasty because of the risk of anastomotic rupture during balloon angioplasty. The remaining nine patients underwent balloon angioplasty initially. Balloon inflation was continued until there was a loss of waist deformity of the balloon catheter. Each dilation lasted 30 – 60 seconds, with two to three inflations for each procedure. If there was elastic recoil of more than 50% of normal extrahepatic PV or a residual pressure gradient of more than 5 mm Hg after balloon angioplasty, stent placement was performed. A balloon catheter and a selfexpandable stent (Wallstent; Boston Scientific, Galway, Ireland; or Zilver stent; William Cook Europe, Bjaeverskov, Denmark) with the same diameter or a 1–2 mm larger diameter than the nonstenotic extrahepatic PV was used. Stents 6 –10 mm in diameter and 4 –7 cm in length were used to cover a stenotic or occluded segment with minimal angulation between the PV
⬍1
and the proximal and distal edges of the deployed stent. After the procedure, a portal venogram was obtained and the pressure gradient was measured. A percutaneous transhepatic tract was embolized with several coils (Cook) and the inferior mesenteric vein was ligated with 9 – 0 or 10 – 0 nylon. Anticoagulation was not routinely performed after stent implantation. Data Analysis The following parameters were documented retrospectively: pre- and postprocedural pressure gradients across the stenosis or occlusion, technical and clinical success, complications, and patency of PV inflow. Technical success was defined as successful stent placement in the intended location of the PV with subsequent improvement of PV inflow and pressure gradient threshold less than 5 mm Hg. Clinical success was defined as subsequent normalization of liver function and disappearance of clinical signs
⬍1
Patent‡
and symptoms relating to portal hypertension. Daily physical examination, US, and/or splenic volume analysis with use of a CT analysis system (PetaVision for Diagnosis; Asan Medical Center, Seoul, Korea) were performed to evaluate the change of splenic size and ascites. Improvement of splenomegaly was defined by shrinkage of the splenic size by at least a two-finger width on physical examinations by the same physician in charge in each patient. Major complications were defined as those (i) necessitating an increased level of care or an additional surgical or interventional manipulation and/or (ii) resulting in adverse sequelae or death. All other complications were defined as minor complications. Patency of the PV was evaluated by means of Doppler US and/or CT. Doppler US was routinely performed the day after stent placement, weekly until the patient was discharged from our hospital, and then 1, 6, and 12 months after discharge. However, CT was performed at random intervals.
1218
•
PV Stent Placement in Pediatric Liver Transplant Recipients
October 2007
JVIR
Figure 1. Axial (a) and coronal (b) multiplanar reconstruction CT images obtained 43 months after LT in patient 10 show the occluded extrahepatic PV (arrow) with multiple collateral veins (arrowheads). L, grafted liver; P, intrahepatic PV; Sp, spleen. (c) Direct portogram from a percutaneous transhepatic approach reveals the occluded extrahepatic PV (arrow) with the collateral veins (arrowheads). (d) Direct portogram after balloon angioplasty and stent placement shows normalized PV inflow. The collateral veins have disappeared. (e) Oblique coronal reformatted CT image obtained 45 months after stent placement shows a patent PV (arrows indicate the stent in the PV).
Statistical Analysis The paired Student t test was used to assess the differences between preand postprocedural pressure gradients. Analysis was conducted with SPSS software (version 12.0; SPSS, Chicago, Ill), with P values lower than .05 considered to be significant.
RESULTS Outcomes of stent placement are shown in the Table. Technical success was achieved in 10 of 12 patients (83.3%; Figs 1, 2). Stent placement failed in two patients because of failed negotiation of the occluded PV. One patient underwent mesocaval shunt surgery and then died of acute rejection after repeat LT necessitated by chronic rejection 18 months after failure of percutaneous stent placement. The other patient is still alive with con-
servative management, although she has had intermittent hematochezia for more than 3 years. The mean pressure gradients in eight of the 10 patients before and after stent implantation were 11.4 mm Hg ⫾ 3.7 (range, 5–16 mm Hg) and 1.1 mm Hg ⫾ 1.2 (range, 0 –3 mm Hg), respectively (P ⬍ .001). In the remaining two patients who underwent the procedure intraoperatively, the pressure gradient was not measured because direct portal venography demonstrated occlusion in one patient and severe stenosis of more than 70% versus the nonstenotic extrahepatic PV in the other patient. Clinical success was achieved in eight of 12 patients (67%). These eight patients were discharged 2–90 days (mean, 14 d; median, 5 d) after stent placement. In four patients with splenomegaly, daily physical examinations revealed improved spleno-
megaly in all patients within 4 days after stent placement. Three- and 5-month follow-up volumetric CT scans of two patients (patients 5 and 10) revealed shrinkage of the splenic volume by approximately 56% and 66%, respectively, compared with the splenic volume before stent placement. Other clinical signs and symptoms related to portal hypertension, including ascites, pancytopenia, and melena, also ameliorated gradually after stent placement. One of the eight patients in whom clinical success was achieved was lost to follow-up as a result of emigration 12 months after stent placement, without recurrent symptoms. In the remaining seven patients, there were no recurrent symptoms during the 10 –58month clinical follow-up period (mean, 43 months ⫾ 19). The last follow-up images obtained 1.5–53.6 months after stent placement (mean, 33 months ⫾
Volume 18
Number 10
Ko et al
•
1219
Figure 2. Oblique axial (a) and coronal (b) multiplanar reconstructed CT images obtained on postoperative day 1 in patient 11 show diffuse stenosis (arrows) of the extrahepatic PV and a large-dimension hematoma (arrowheads) in the portocaval space. P, donor PV. (c) Direct portogram after hematoma evacuation still shows diffuse stenosis (arrows) of the extrahepatic PV. (d) Direct portogram obtained after primary stent placement shows much-improved stenosis (arrows indicate the PV stent). (e) Doppler US image obtained 15 months after stent placement shows patent PV inflow through the stent (arrows).
22) also revealed a patent PV in these seven patients. The remaining two patients in whom clinical success was not achieved died of pathologically proven acute rejection: One patient (patient 2) had a patent PV at the last follow-up CT examination, but the other patient (patient 12) had to undergo stent replacement. In the latter patient, venography after stent placement showed an hourglass deformity of a deployed stent that was approximately 40% of the normal extrahepatic PV. However, balloon angioplasty was not performed after stent placement based on the expectation of self-expansion of the deployed stent and a risk of anastomotic disruption during balloon angioplasty. However, this patient’s liver function began to deteriorate from the sixth day after stent placement, and follow-up US and indirect portography through the superior mesenteric artery at 9 days revealed poor PV inflow with persistent hourglass stent deformity. Repeat intraoperative portal venography also demonstrated
partial thrombosis of the PV. Intraoperative balloon-expandable stent replacement was then performed on day 10 after primary stent placement; however, this patient died of acute rejection the day after stent replacement. No other major procedural complications occurred. As for minor complications, four patients had transient fever for 1–3 days after the procedure.
DISCUSSION Although recent improvements in surgical techniques have contributed to decreased vascular anastomotic complications after LT, PV complications are still a significant cause of postoperative graft failure and morbidity. The incidence of PV stenosis or occlusion in pediatric patients has been reported to be as high as 33% and is generally much greater than the incidence in adult recipients (9). In addition, patients who undergo living-donor LT have a higher risk of PV complications than those who
undergo cadaveric-donor LT. In one series, the greatest incidence of PV stenosis was in cases of living-donor LT, with a 27% incidence, followed by reducedsize cadaveric grafts at 1% and wholesized cadaveric grafts at 1% (4). Recently, percutaneous transhepatic balloon angioplasty has become widely accepted as a safe and effective procedure for the treatment of PV stenosis after LT (2,3,5,7). However, stent placement has usually been used restrictively to treat recurrent or elastic stenosis after balloon angioplasty because the procedure has several potential complications, such as PV thrombosis, instent or stent edge restenosis, and difficult reanastomosis if repeat LT becomes necessary (2– 4,7,8). In addition, there is a risk of functional stent stenosis in the PV of a growing child. In our patients who were treated with percutaneous stent placement, we therefore initially performed balloon angioplasty in all patients except one who underwent this procedure within 1 month after LT.
1220
•
PV Stent Placement in Pediatric Liver Transplant Recipients
However, because balloon angioplasty alone was not sufficient to treat PV stenosis or occlusion in most of our study patients, we had to perform stent placement after balloon angioplasty. The rate of stent placement after balloon angioplasty in our study was higher those in previous studies (1–3). This discrepancy might have been related to the high frequency of PV occlusion in our study: five of eight study patients who underwent percutaneous transhepatic stent placement had PV occlusion. Conversely, we preferred to perform primary stent placement rather than balloon angioplasty in three patients because of the possibility of anastomotic rupture of a relatively fresh anastomosis. Although, to our knowledge, there has been no consensus regarding the optimal interval for balloon angioplasty after vascular anastomosis, we believe there may not be sufficient healing of a vascular anastomosis within 1 month after LT. In addition, we considered that the possible etiologies of PV stenosis and occlusion in two patients who underwent stent placement intraoperatively were tension on the PV anastomosis and kinking of the PV, respectively. Therefore, we chose primary stent placement because we thought balloon angioplasty alone might not be effective to treat a condition related to these etiologies. After primary stent placement, two of the three patients subsequently showed a patent PV on CT images at 46.3 months and 21.4 months, respectively. Gomez-Gutierrez et al (10) reported a 100% patency rate in seven adults with intraoperative stent placement during cadaveric LT during a follow-up of 10 –54 months. We therefore assume that intraoperative primary stent placement may be valuable for the treatment of PV stenosis or occlusion. However, the remaining patient had to undergo stent replacement as a result of a persistent hourglass deformity of the stent and partial thrombosis. In fact, PV venography after primary stent placement in this patient showed an approximate 40% hourglass deformity of the stent (compared with nonstenotic extrahepatic PV), which was likely a result of underlying atrophic PV in the recipient. We did not perform addi-
tional balloon angioplasty with the expectation of self-expansion of the stent. Thereafter, the deformity did not improve and we had to perform stent replacement. We therefore assume that careful balloon angioplasty in a case of significant hourglass deformity of a deployed stent may be valuable to preserve the patent PV. Although one patient had partial PV thrombosis, other patients showed a continually patent PV during a mean imaging follow-up period of 33 months ⫾ 22 even though they were not given anticoagulants after stent placement. Several investigators have also reported a 100% PV patency rate after stent placement with various follow-up periods (3,8,10 –12). These patency rates are superior to previously reported recurrence rates of balloon angioplasty alone (27%– 50%) (3,5–7). However, there is still a risk of functional stenosis of a stent in a growing vessel in pediatric patients. To treat such stenoses, some investigators have placed balloon-expandable stents in pediatric patients, as these stents can subsequently be dilated to achieve a larger diameter (2,13,14). However, most of our study patients had relatively sufficient PV diameters to allow deployment of a self-expandable stent with a diameter as large as those used in adults. Previous reports usually described placement of stents 9 –12 mm in diameter in adults (8,11,12,15) and 8 –9 mm in diameter in pediatric patients (3,4,13). In addition, Funaki et al (3) reported a 100% PV patency rate after 8-mm-diameter stent placement in 12 pediatric patients during a mean follow-up of 47 months. Eight of our study patients received a 10-mm-diameter stent, one received an 8-mm-diameter stent, and one received a 6-mm-diameter stent. Thereafter, functional stenosis did not occur in any of these patients during the follow-up period. Therefore, although our follow-up period was limited, we assume that functional stenosis in pediatric patients may be uncommon when the stent placed is 8 mm or greater in diameter. In addition, we also believe that balloon-expandable stents may be helpful to treat subsequent functional stenosis if the PV diameter is too small to allow placement of a stent at least 8 mm in diameter, although one of our patients who received a 6-mm-diameter stent has not shown any subsequent functional stenosis.
October 2007
JVIR
In summary, although we had a relatively small number of study patients, PV stent placement seems to be a safe and effective method for the treatment of posttransplantation PV stenosis or occlusion in pediatric recipients. The intermediate-term PV patency after stent placement is also excellent. However, additional experience and long-term follow-up studies are needed to verify the usefulness of this technique in pediatric patients.
Acknowledgment: The authors thank Bonnie Hami, MA, for editorial assistance in preparing this manuscript.
References 1. Olcott EW, Ring EJ, Roberts JP, Ascher NL, Lake JR, Gordon RL. Percutaneous transhepatic portal vein angioplasty and stent placement after liver transplantation: early experience. J Vasc Interv Radiol 1990; 1:17–22. 2. Zajko AB, Sheng R, Bron K, Reyes J, Nour B, Tzakis A. Percutaneous transluminal angioplasty of venous anastomotic stenoses complicating liver transplantation: intermediate-term results. J Vasc Interv Radiol 1994; 5:121–126. 3. Funaki B, Rosenblum JD, Leef JA, et al. Percutaneous treatment of portal venous stenoses in children and adolescents with segmental hepatic transplantation: long-term results. Radiology 2000; 215:147–151. 4. Buell JF, Funaki B, Cronin DC, et al. Long-term venous complications after full-size and segmental pediatric liver transplantation. Ann Surg 2002; 236: 658 – 666. 5. Park KB, Choo SW, Do YS, Shin SW, Cho SG, Choo IW. Percutaneous angioplasty of portal vein stenosis that complicates liver transplantation: the mid-term therapeutic results. Korean J Radiol 2005; 6:161–166. 6. Ueda M, Egawa H, Ogawa K, et al. Portal vein complications in the longterm course after pediatric living donor liver transplantation. Transplant Proc 2005; 37:1138 –1140. 7. Shibata T, Itoh K, Kubo T, et al. Percutaneous transhepatic balloon dilation of portal venous stenosis in patients with living donor liver transplantation. Radiology 2005; 235:1078 – 1083. 8. Shan H, Xiao XS, Huang MS, Ouyang Q, Jiang ZB. Portal venous stent placement for treatment of portal hypertension caused by benign main portal vein stenosis. World J Gastroenterol 2005; 7: 11:3315–3318.
Volume 18
Number 10
9. Millis JM, Seaman DS, Piper JB, et al. Portal vein thrombosis and stenosis in pediatric liver transplantation. Transplantation 1996; 27:62:748 –754. 10. Gomez-Gutierrez M, Quintela J, Marini M, et al. Portal vein thrombosis in patients undergoing orthotopic liver transplantation: intraoperative endovascular radiological procedures. Transplant Proc 2005; 37:3906 –3908. 11. Godoy MA, Camunez F, Echenagusia A, et al. Percutaneous treatment of
Ko et al
benign portal vein stenosis after liver transplantation. J Vasc Interv Radiol 1996; 7:273–276. 12. Wang JF, Zhai RY, Wei BJ, et al. Percutaneous intravascular stents for treatment of portal venous stenosis after liver transplantation: midterm results. Transplant Proc 2006; 38:1461–1462. 13. Carnevale FC, Borges MV, Moreira AM, Cerri GG, Maksoud JG. Endovascular treatment of acute portal vein thrombosis after liver transplantation in a child.
•
1221
Cardiovasc Intervent Radiol 2006; 29: 457– 461. 14. Cwikiel W, Solvig J, Schroder H. Stent recanalization of chronic portal vein occlusion in a child. Cardiovasc Intervent Radiol 2000; 23:309 –311. 15. Cherukuri R, Haskal ZJ, Naji A, Shaked A. Percutaneous thrombolysis and stent placement for the treatment of portal vein thrombosis after liver transplantation: long-term follow-up. Transplantation 1998; 27;65:1124 –1126.
PURPOSE: To evaluate the efficacy of stent placement for the treatment of portal vein (PV) stenosis or occlusion in pediatric liver transplant recipients. MATERIALS AND METHODS: Written informed consent was obtained from a legal guardian, and our institutional review board approved this study. Percutaneous (n ⴝ 10) or intraoperative (n ⴝ 2) stent placement was attempted in 12 pediatric recipients (age range, 6 –102 months) via the percutaneous transhepatic or inferior mesenteric vein route. Stents 6 –10 mm in diameter were placed. Technical and clinical success, complications, and patency of the PV were retrospectively analyzed. RESULTS: Technical success was achieved in 10 of 12 patients (83%) and clinical success was achieved in eight patients (67%). Eight of the 10 patients in whom technical success was achieved (80%) remained healthy with a patent PV during the 10 –58-month clinical follow-up period. One patient with technical success died of acute rejection without recurrent PV complications and another died of acute rejection after stent replacement as a result of an hourglass deformity of a deployed stent with partial thrombosis. No major procedural complications occurred. CONCLUSIONS: Based on this study in a relatively small number of patients, PV stent placement seems to be a safe and effective method for the treatment of posttransplantation PV stenosis or occlusion in pediatric patients. J Vasc Interv Radiol 2007; 18:1215–1221 Abbreviations:
LT ⫽ liver transplantation, PV ⫽ portal vein
LIVER transplantation (LT) has become a valuable treatment modality for pediatric patients with liver failure. However, portal vein (PV) stenosis is
From the Department of Radiology and Research Institute of Radiology (G.Y.K., K.B.S., H.K.Y., K.R.K.), Division of Hepatobiliary Surgery and Liver Transplantation, Department of Surgery (S.L., Y.J.L.), and Department of Pediatrics (K.M.K.), Asan Medical Center, University of Ulsan College of Medicine, 388-1 Poongnap-2-dong, Songpa-ku, Seoul 138-040, Republic of Korea. Received February 22, 2007; final revision received June 20, 2007; accepted June 21, 2007. Address correspondence to G.Y.K.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2007 DOI: 10.1016/j.jvir.2007.06.029
still one of the significant causes of graft failure or posttransplantation morbidity in pediatric patients who undergo left lateral segment transplantation. This complication has traditionally been managed by surgery; however, in recent years, percutaneous transhepatic balloon angioplasty has been found to be a safe and effective alternative for the treatment of PV stenosis (1–7). Conversely, stents have usually been used to treat elastic or recurrent stenosis after balloon angioplasty (2– 4,8), and there are only a few reports of stent placement for elastic or recurrent PV stenosis in pediatric patients after LT (3,4). The purpose of this study is to evaluate the efficacy of stent placement for the treatment of PV stenosis or occlusion in pediatric patients after LT .
MATERIALS AND METHODS Patient Population Between December 1994 and June 2006, 126 pediatric patients (age ⬍15 years) underwent LT with living (n ⫽ 118) or cadaver (n ⫽ 8) donors at our institution. Except for two patients who underwent balloon angioplasty only, 11 patients (8.7%) underwent percutaneous or intraoperative stent placement to treat PV stenosis or occlusion. One patient who had undergone LT at another hospital also underwent percutaneous PV stent placement at our institute. The study included four male patients and eight female patients ranging in age from 6 to 102 months (mean ⫾ SD, 51 months ⫾ 36). All these pa-
1215
1216
•
October 2007
PV Stent Placement in Pediatric Liver Transplant Recipients
JVIR
Demographic Data of All Patients
Pt. No.
Age (months)
Interval after Transplantation
Severity of stenosis (%)
1 2 3 4 5 6 7 8 9
12 25 26 50 53 63 65 98 98
1.2 months 3.8 months 14.2 months 36.4 months 42.0 months 52.8 months 1.0 months 76.0 months 81.7 months
90 100 90 100 100 100 80 100 100
10
10
43.8 months
100
Intraoperative treatment
Symptoms
Procedure
Ascites, effusion LFT, ascites SPM SPM, melena SPM SPM, pancytopenia LFT, ascites Melena, gastric varices Hematochezia, esophageal varices SPM, melena, pancytopenia
PTA/stent PTA/stent PTA/stent Not completed PTA/stent PTA/stent Stent PTA/stent Not completed PTA/stent
Indications
11
6
1 day
70
12
10
2 days
100
Periportal hematoma, LFT, tension PV thrombosis, LFT, kinking
Hematoma evacuation, stent Surgical thrombectomy, stent
Note.—LFT ⫽ liver function tests; NA ⫽ not available; PTA ⫽ percutaneous transluminal angioplasty; SPM ⫽ splenomegaly. * Five- and 3-month follow-up volumetric CT scans of these patients revealed shrinkage of the splenic volume by approximately 66% and 56%, respectively, of splenic volume before stent placement. † The patient was lost to follow-up approximately 12 months after stent placement. ‡ PV was patent, but the placed stent showed an hourglass deformity.
tients had received a living donor left lateral segment graft for underlying biliary atresia. Reconstruction of the PV during transplantation was performed by end-to-end anastomosis of the recipient and donor PV in 11 patients. In one patient, the type of PV reconstruction was unknown. The demographic data for all patients are shown in the Table. Stent Placement We obtained institutional review board approval at our hospital to conduct a retrospective review of the patients’ medical and imaging records. Written informed consent for the treatment of PV stenosis or occlusion was obtained from a legal guardian. All procedures were performed under general anesthesia. Ten patients underwent stent placement through a percutaneous transhepatic route and two patients underwent this procedure intraoperatively. Diagnosis of PV stenosis or occlusion was based on a combination of Doppler ultrasonogra-
phy (US) and computed tomography (CT) in nine patients and CT alone in three patients. On US, stenosis or occlusion was diagnosed if Doppler US showed no flow in the PV or an acceleration of the flow rate in the poststenotic PV of more than three times that in the prestenotic PV with a narrowing of the PV diameter of less than 2.5 mm. On CT, PV occlusion was diagnosed if CT demonstrated discontinuity of the PV. In two patients, PV stenosis or occlusion was confirmed by direct portography through the inferior mesenteric vein performed after surgical periportal bleeding control and thrombectomy, respectively. The average interval between LT and stent placement was 35 months ⫾ 30 (range, 1– 82 months) in the 10 patients who underwent percutaneous transhepatic stent placement. The remaining two patients underwent the procedure on day 1 and day 2 after LT, respectively. Percutaneous transhepatic puncture of the intrahepatic PV was performed with a 21-gauge Chiba needle (Cook, Bloomington, Ind) under US
and fluoroscopic guidance. The needle was exchanged for a 4-F coaxial dilator and a 6- or 7-F sheath over a 0.018-inch guide wire (Cook) or a 0.035-inch angled hydrophilic guide wire (Terumo, Tokyo, Japan). Then, a direct main portal venogram was obtained and the pressure gradient across the stenosis or occlusion was measured. For intraoperative procedures, the inferior mesenteric vein was punctured with an 18-gauge Angiocath (Becton Dickinson Korea, Seoul, Korea) and then the Angiocath was exchanged for a 7-F sheath over a 0.035-inch angled hydrophilic guide wire under fluoroscopic guidance with a portable C-arm (OEC 9800; General Electric, Salt Lake City, Utah). A direct main portal venogram was then obtained. A bolus of heparin (50 U/kg) was then administered directly into the PV if there was no coagulopathy. The 0.035-inch guide wire and a 5-F Cobra catheter (Cook) or Kumpe catheter (Cook) were used to traverse stenosis or occlusion of the PV. Three patients who underwent this proce-
Volume 18
Number 10
Ko et al
•
1217
Pressure gradient (mm Hg) Stent size (mm)
Before stenting
After stenting
Technical success
Clinical success
Clinical follow-up duration (months)
Image follow-up duration (months)
PV status
10 ⫻ 43 8 ⫻ 50 10 ⫻ 60 NA 10 ⫻ 50 10 ⫻ 70 10 ⫻ 70 10 ⫻ 50 10 ⫻ 60
16 5 14 NA 11 10 8 12 NA
2 0 2 NA 0 0 0 2 NA
Yes Yes Yes No Yes Yes Yes Yes No
Yes No Yes No Yes* Yes Yes Yes No
53.1 2.5 49.3 NA 58.1 11.7 51.6 9.7 NA
50.9 0.7 1.5 NA 49.4 3.0 46.3 9.7 NA
Patent Patent Patent NA Patent Patent† Patent Patent NA
NA
15
3
Yes
Yes*
56.5
53.6
Patent
6 ⫻ 40
NA
NA
Yes
Yes
22.8
21.4
Patent
10 ⫻ 40
NA
NA
No
No
dure within 1 month after LT (patients 7, 11, and 12) underwent primary stent placement without balloon angioplasty because of the risk of anastomotic rupture during balloon angioplasty. The remaining nine patients underwent balloon angioplasty initially. Balloon inflation was continued until there was a loss of waist deformity of the balloon catheter. Each dilation lasted 30 – 60 seconds, with two to three inflations for each procedure. If there was elastic recoil of more than 50% of normal extrahepatic PV or a residual pressure gradient of more than 5 mm Hg after balloon angioplasty, stent placement was performed. A balloon catheter and a selfexpandable stent (Wallstent; Boston Scientific, Galway, Ireland; or Zilver stent; William Cook Europe, Bjaeverskov, Denmark) with the same diameter or a 1–2 mm larger diameter than the nonstenotic extrahepatic PV was used. Stents 6 –10 mm in diameter and 4 –7 cm in length were used to cover a stenotic or occluded segment with minimal angulation between the PV
⬍1
and the proximal and distal edges of the deployed stent. After the procedure, a portal venogram was obtained and the pressure gradient was measured. A percutaneous transhepatic tract was embolized with several coils (Cook) and the inferior mesenteric vein was ligated with 9 – 0 or 10 – 0 nylon. Anticoagulation was not routinely performed after stent implantation. Data Analysis The following parameters were documented retrospectively: pre- and postprocedural pressure gradients across the stenosis or occlusion, technical and clinical success, complications, and patency of PV inflow. Technical success was defined as successful stent placement in the intended location of the PV with subsequent improvement of PV inflow and pressure gradient threshold less than 5 mm Hg. Clinical success was defined as subsequent normalization of liver function and disappearance of clinical signs
⬍1
Patent‡
and symptoms relating to portal hypertension. Daily physical examination, US, and/or splenic volume analysis with use of a CT analysis system (PetaVision for Diagnosis; Asan Medical Center, Seoul, Korea) were performed to evaluate the change of splenic size and ascites. Improvement of splenomegaly was defined by shrinkage of the splenic size by at least a two-finger width on physical examinations by the same physician in charge in each patient. Major complications were defined as those (i) necessitating an increased level of care or an additional surgical or interventional manipulation and/or (ii) resulting in adverse sequelae or death. All other complications were defined as minor complications. Patency of the PV was evaluated by means of Doppler US and/or CT. Doppler US was routinely performed the day after stent placement, weekly until the patient was discharged from our hospital, and then 1, 6, and 12 months after discharge. However, CT was performed at random intervals.
1218
•
PV Stent Placement in Pediatric Liver Transplant Recipients
October 2007
JVIR
Figure 1. Axial (a) and coronal (b) multiplanar reconstruction CT images obtained 43 months after LT in patient 10 show the occluded extrahepatic PV (arrow) with multiple collateral veins (arrowheads). L, grafted liver; P, intrahepatic PV; Sp, spleen. (c) Direct portogram from a percutaneous transhepatic approach reveals the occluded extrahepatic PV (arrow) with the collateral veins (arrowheads). (d) Direct portogram after balloon angioplasty and stent placement shows normalized PV inflow. The collateral veins have disappeared. (e) Oblique coronal reformatted CT image obtained 45 months after stent placement shows a patent PV (arrows indicate the stent in the PV).
Statistical Analysis The paired Student t test was used to assess the differences between preand postprocedural pressure gradients. Analysis was conducted with SPSS software (version 12.0; SPSS, Chicago, Ill), with P values lower than .05 considered to be significant.
RESULTS Outcomes of stent placement are shown in the Table. Technical success was achieved in 10 of 12 patients (83.3%; Figs 1, 2). Stent placement failed in two patients because of failed negotiation of the occluded PV. One patient underwent mesocaval shunt surgery and then died of acute rejection after repeat LT necessitated by chronic rejection 18 months after failure of percutaneous stent placement. The other patient is still alive with con-
servative management, although she has had intermittent hematochezia for more than 3 years. The mean pressure gradients in eight of the 10 patients before and after stent implantation were 11.4 mm Hg ⫾ 3.7 (range, 5–16 mm Hg) and 1.1 mm Hg ⫾ 1.2 (range, 0 –3 mm Hg), respectively (P ⬍ .001). In the remaining two patients who underwent the procedure intraoperatively, the pressure gradient was not measured because direct portal venography demonstrated occlusion in one patient and severe stenosis of more than 70% versus the nonstenotic extrahepatic PV in the other patient. Clinical success was achieved in eight of 12 patients (67%). These eight patients were discharged 2–90 days (mean, 14 d; median, 5 d) after stent placement. In four patients with splenomegaly, daily physical examinations revealed improved spleno-
megaly in all patients within 4 days after stent placement. Three- and 5-month follow-up volumetric CT scans of two patients (patients 5 and 10) revealed shrinkage of the splenic volume by approximately 56% and 66%, respectively, compared with the splenic volume before stent placement. Other clinical signs and symptoms related to portal hypertension, including ascites, pancytopenia, and melena, also ameliorated gradually after stent placement. One of the eight patients in whom clinical success was achieved was lost to follow-up as a result of emigration 12 months after stent placement, without recurrent symptoms. In the remaining seven patients, there were no recurrent symptoms during the 10 –58month clinical follow-up period (mean, 43 months ⫾ 19). The last follow-up images obtained 1.5–53.6 months after stent placement (mean, 33 months ⫾
Volume 18
Number 10
Ko et al
•
1219
Figure 2. Oblique axial (a) and coronal (b) multiplanar reconstructed CT images obtained on postoperative day 1 in patient 11 show diffuse stenosis (arrows) of the extrahepatic PV and a large-dimension hematoma (arrowheads) in the portocaval space. P, donor PV. (c) Direct portogram after hematoma evacuation still shows diffuse stenosis (arrows) of the extrahepatic PV. (d) Direct portogram obtained after primary stent placement shows much-improved stenosis (arrows indicate the PV stent). (e) Doppler US image obtained 15 months after stent placement shows patent PV inflow through the stent (arrows).
22) also revealed a patent PV in these seven patients. The remaining two patients in whom clinical success was not achieved died of pathologically proven acute rejection: One patient (patient 2) had a patent PV at the last follow-up CT examination, but the other patient (patient 12) had to undergo stent replacement. In the latter patient, venography after stent placement showed an hourglass deformity of a deployed stent that was approximately 40% of the normal extrahepatic PV. However, balloon angioplasty was not performed after stent placement based on the expectation of self-expansion of the deployed stent and a risk of anastomotic disruption during balloon angioplasty. However, this patient’s liver function began to deteriorate from the sixth day after stent placement, and follow-up US and indirect portography through the superior mesenteric artery at 9 days revealed poor PV inflow with persistent hourglass stent deformity. Repeat intraoperative portal venography also demonstrated
partial thrombosis of the PV. Intraoperative balloon-expandable stent replacement was then performed on day 10 after primary stent placement; however, this patient died of acute rejection the day after stent replacement. No other major procedural complications occurred. As for minor complications, four patients had transient fever for 1–3 days after the procedure.
DISCUSSION Although recent improvements in surgical techniques have contributed to decreased vascular anastomotic complications after LT, PV complications are still a significant cause of postoperative graft failure and morbidity. The incidence of PV stenosis or occlusion in pediatric patients has been reported to be as high as 33% and is generally much greater than the incidence in adult recipients (9). In addition, patients who undergo living-donor LT have a higher risk of PV complications than those who
undergo cadaveric-donor LT. In one series, the greatest incidence of PV stenosis was in cases of living-donor LT, with a 27% incidence, followed by reducedsize cadaveric grafts at 1% and wholesized cadaveric grafts at 1% (4). Recently, percutaneous transhepatic balloon angioplasty has become widely accepted as a safe and effective procedure for the treatment of PV stenosis after LT (2,3,5,7). However, stent placement has usually been used restrictively to treat recurrent or elastic stenosis after balloon angioplasty because the procedure has several potential complications, such as PV thrombosis, instent or stent edge restenosis, and difficult reanastomosis if repeat LT becomes necessary (2– 4,7,8). In addition, there is a risk of functional stent stenosis in the PV of a growing child. In our patients who were treated with percutaneous stent placement, we therefore initially performed balloon angioplasty in all patients except one who underwent this procedure within 1 month after LT.
1220
•
PV Stent Placement in Pediatric Liver Transplant Recipients
However, because balloon angioplasty alone was not sufficient to treat PV stenosis or occlusion in most of our study patients, we had to perform stent placement after balloon angioplasty. The rate of stent placement after balloon angioplasty in our study was higher those in previous studies (1–3). This discrepancy might have been related to the high frequency of PV occlusion in our study: five of eight study patients who underwent percutaneous transhepatic stent placement had PV occlusion. Conversely, we preferred to perform primary stent placement rather than balloon angioplasty in three patients because of the possibility of anastomotic rupture of a relatively fresh anastomosis. Although, to our knowledge, there has been no consensus regarding the optimal interval for balloon angioplasty after vascular anastomosis, we believe there may not be sufficient healing of a vascular anastomosis within 1 month after LT. In addition, we considered that the possible etiologies of PV stenosis and occlusion in two patients who underwent stent placement intraoperatively were tension on the PV anastomosis and kinking of the PV, respectively. Therefore, we chose primary stent placement because we thought balloon angioplasty alone might not be effective to treat a condition related to these etiologies. After primary stent placement, two of the three patients subsequently showed a patent PV on CT images at 46.3 months and 21.4 months, respectively. Gomez-Gutierrez et al (10) reported a 100% patency rate in seven adults with intraoperative stent placement during cadaveric LT during a follow-up of 10 –54 months. We therefore assume that intraoperative primary stent placement may be valuable for the treatment of PV stenosis or occlusion. However, the remaining patient had to undergo stent replacement as a result of a persistent hourglass deformity of the stent and partial thrombosis. In fact, PV venography after primary stent placement in this patient showed an approximate 40% hourglass deformity of the stent (compared with nonstenotic extrahepatic PV), which was likely a result of underlying atrophic PV in the recipient. We did not perform addi-
tional balloon angioplasty with the expectation of self-expansion of the stent. Thereafter, the deformity did not improve and we had to perform stent replacement. We therefore assume that careful balloon angioplasty in a case of significant hourglass deformity of a deployed stent may be valuable to preserve the patent PV. Although one patient had partial PV thrombosis, other patients showed a continually patent PV during a mean imaging follow-up period of 33 months ⫾ 22 even though they were not given anticoagulants after stent placement. Several investigators have also reported a 100% PV patency rate after stent placement with various follow-up periods (3,8,10 –12). These patency rates are superior to previously reported recurrence rates of balloon angioplasty alone (27%– 50%) (3,5–7). However, there is still a risk of functional stenosis of a stent in a growing vessel in pediatric patients. To treat such stenoses, some investigators have placed balloon-expandable stents in pediatric patients, as these stents can subsequently be dilated to achieve a larger diameter (2,13,14). However, most of our study patients had relatively sufficient PV diameters to allow deployment of a self-expandable stent with a diameter as large as those used in adults. Previous reports usually described placement of stents 9 –12 mm in diameter in adults (8,11,12,15) and 8 –9 mm in diameter in pediatric patients (3,4,13). In addition, Funaki et al (3) reported a 100% PV patency rate after 8-mm-diameter stent placement in 12 pediatric patients during a mean follow-up of 47 months. Eight of our study patients received a 10-mm-diameter stent, one received an 8-mm-diameter stent, and one received a 6-mm-diameter stent. Thereafter, functional stenosis did not occur in any of these patients during the follow-up period. Therefore, although our follow-up period was limited, we assume that functional stenosis in pediatric patients may be uncommon when the stent placed is 8 mm or greater in diameter. In addition, we also believe that balloon-expandable stents may be helpful to treat subsequent functional stenosis if the PV diameter is too small to allow placement of a stent at least 8 mm in diameter, although one of our patients who received a 6-mm-diameter stent has not shown any subsequent functional stenosis.
October 2007
JVIR
In summary, although we had a relatively small number of study patients, PV stent placement seems to be a safe and effective method for the treatment of posttransplantation PV stenosis or occlusion in pediatric recipients. The intermediate-term PV patency after stent placement is also excellent. However, additional experience and long-term follow-up studies are needed to verify the usefulness of this technique in pediatric patients.
Acknowledgment: The authors thank Bonnie Hami, MA, for editorial assistance in preparing this manuscript.
References 1. Olcott EW, Ring EJ, Roberts JP, Ascher NL, Lake JR, Gordon RL. Percutaneous transhepatic portal vein angioplasty and stent placement after liver transplantation: early experience. J Vasc Interv Radiol 1990; 1:17–22. 2. Zajko AB, Sheng R, Bron K, Reyes J, Nour B, Tzakis A. Percutaneous transluminal angioplasty of venous anastomotic stenoses complicating liver transplantation: intermediate-term results. J Vasc Interv Radiol 1994; 5:121–126. 3. Funaki B, Rosenblum JD, Leef JA, et al. Percutaneous treatment of portal venous stenoses in children and adolescents with segmental hepatic transplantation: long-term results. Radiology 2000; 215:147–151. 4. Buell JF, Funaki B, Cronin DC, et al. Long-term venous complications after full-size and segmental pediatric liver transplantation. Ann Surg 2002; 236: 658 – 666. 5. Park KB, Choo SW, Do YS, Shin SW, Cho SG, Choo IW. Percutaneous angioplasty of portal vein stenosis that complicates liver transplantation: the mid-term therapeutic results. Korean J Radiol 2005; 6:161–166. 6. Ueda M, Egawa H, Ogawa K, et al. Portal vein complications in the longterm course after pediatric living donor liver transplantation. Transplant Proc 2005; 37:1138 –1140. 7. Shibata T, Itoh K, Kubo T, et al. Percutaneous transhepatic balloon dilation of portal venous stenosis in patients with living donor liver transplantation. Radiology 2005; 235:1078 – 1083. 8. Shan H, Xiao XS, Huang MS, Ouyang Q, Jiang ZB. Portal venous stent placement for treatment of portal hypertension caused by benign main portal vein stenosis. World J Gastroenterol 2005; 7: 11:3315–3318.
Volume 18
Number 10
9. Millis JM, Seaman DS, Piper JB, et al. Portal vein thrombosis and stenosis in pediatric liver transplantation. Transplantation 1996; 27:62:748 –754. 10. Gomez-Gutierrez M, Quintela J, Marini M, et al. Portal vein thrombosis in patients undergoing orthotopic liver transplantation: intraoperative endovascular radiological procedures. Transplant Proc 2005; 37:3906 –3908. 11. Godoy MA, Camunez F, Echenagusia A, et al. Percutaneous treatment of
Ko et al
benign portal vein stenosis after liver transplantation. J Vasc Interv Radiol 1996; 7:273–276. 12. Wang JF, Zhai RY, Wei BJ, et al. Percutaneous intravascular stents for treatment of portal venous stenosis after liver transplantation: midterm results. Transplant Proc 2006; 38:1461–1462. 13. Carnevale FC, Borges MV, Moreira AM, Cerri GG, Maksoud JG. Endovascular treatment of acute portal vein thrombosis after liver transplantation in a child.
•
1221
Cardiovasc Intervent Radiol 2006; 29: 457– 461. 14. Cwikiel W, Solvig J, Schroder H. Stent recanalization of chronic portal vein occlusion in a child. Cardiovasc Intervent Radiol 2000; 23:309 –311. 15. Cherukuri R, Haskal ZJ, Naji A, Shaked A. Percutaneous thrombolysis and stent placement for the treatment of portal vein thrombosis after liver transplantation: long-term follow-up. Transplantation 1998; 27;65:1124 –1126.