- Email: [email protected]
Percutaneous Transcatheter Hepatic Artery Embolization for Liver Cysts in Autosomal Dominant Polycystic Kidney Disease
ORIGINAL INVESTIGATIONS Pathogenesis and Treatment of Kidney Disease
Percutaneous Transcatheter Hepatic Artery Embolization for Liver Cysts in Autosomal Dominant Polycystic Kidney Disease Ryoji Takei, MD,1 Yoshifumi Ubara, MD,2 Junichi Hoshino, MD,2 Yasushi Higa, MD,2 Tatsuya Suwabe, MD,2 Yoko Sogawa, MD,2 Kazufumi Nomura, MD,2 Shohei Nakanishi, MD,2 Naoki Sawa, MD,2 Hideyuki Katori, MD,2 Fumi Takemoto, MD,2 Shigeko Hara, MD,2 and Kenmei Takaichi, MD2 Background: We have achieved renal contraction therapy in patients with autosomal dominant polycystic kidney disease (ADPKD) by means of renal transcatheter arterial embolization (TAE) using intravascular coils, decreasing renal size and improving quality of life in almost all patients. We presently perform hepatic TAE in patients with intractable symptomatic polycystic liver. Study Design: Uncontrolled trial. Setting & Participants: 30 patients with ADPKD referred for arteriography to an academic medical center. 22 patients had kidney failure treated by means of dialysis. Intervention: We embolized arteries supplying hepatic segments replaced by cysts that were associated with well-developed hepatic arteries, but obstructed intrahepatic portal veins. Outcomes & Measurements: Various volumes before and after TAE were compared by using computed tomography and National Institutes of Health Image software in 30 patients with follow-up computed tomography 18 to 37 months after therapy. Results: Total liver volume and total intrahepatic cyst volume decreased from 7,882 ⫾ 2,916 and 6,677 ⫾ 2,978 to 6,041 ⫾ 2,282 and 4,625 ⫾ 2,299 cm3, respectively (P ⬍ 0.0001 for both). Fractions of remaining (FR) total liver volume and FR of intrahepatic cyst volume were 78.8% ⫾ 17.6% and 70.4% ⫾ 20.9%, respectively. Hepatic parenchyma increased from 1,205 ⫾ 250 to 1,406 ⫾ 277 cm3 (P ⫽ 0.0004). In 29 of 30 patients, both total liver volume and intrahepatic cyst volume decreased; in 1 patient, total liver volume increased from 5,755 to 7,069 cm3, whereas cysts enlarged from 4,500 to 5,531 cm3. No serious complications were experienced. In 24 patients, the post-TAE course was favorable. TAE failed to benefit 6 patients because of unrelated hepatic infection, peritonitis, hepatic failure, acute leukemia, or pelvic fracture. Limitations: Absence of a control group. Conclusions: TAE may be an option for patients with ADPKD with symptomatic polycystic liver who are not candidates for surgical treatment. Am J Kidney Dis 49:744-752. © 2007 by the National Kidney Foundation, Inc. INDEX WORDS: Autosomal dominant polycystic kidney disease (ADPKD); transcatheter arterial embolization (TAE); intravascular treatment; percutaneous transcatheter hepatic artery embolization; polycystic liver disease.
K
Editorial, p. 725
From the 1Department of Radiology and 2Nephrology Center, Toranomon Hospital Kajigaya, Kanagawa, Japan. Received October 22, 2006; accepted in revised form March 2, 2007. Originally published online as doi:10.1053/j.ajkd.2007.03.018 on May 8, 2007. Support: None. Potential conflicts of interest: None. Address correspondence to Yoshifumi Ubara, MD, Nephrology Center, Toranomon Hospital Kajigaya, 1-3-1, Kajigaya, Takatu-ku, Kawasaki, Kanagawa, 213-0015, Japan. E-mail: [email protected] © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/4906-0005$32.00/0 doi:10.1053/j.ajkd.2007.03.018 744
idneys of patients with autosomal dominant polycystic kidney disease (ADPKD) typically are supplied by well-developed arteries. Accordingly, we performed renal cyst contraction therapy by means of renal transcatheter arterial embolization (TAE) in patients with renal enlargement, including those with cyst hemorrhage, beginning in 1996.1,2 Renal TAE was effective in itself for patients with organomegaly that was predominantly renal. Although renal TAE was effective in patients with similar enlargement of both kidneys and liver, additional treatment was required for hepatomegaly. Renal TAE did not benefit patients with primarily hepatic enlargement who similarly required treatment for hepatomegaly. Several conventional surgical treatments have been performed as therapy for patients with
American Journal of Kidney Diseases, Vol 49, No 6 (June), 2007: pp 744-752
Intravascular Treatment of Polycystic Liver
symptomatic polycystic liver, including percutaneous cyst aspiration and sclerosis, laparoscopic fenestration, open surgical cyst fenestration and partial hepatectomy, and hepatic transplantation.3-7 Because renal TAE was successful in treating enlarged kidneys, many symptomatic patients came to our institution seeking other intravascular treatment. In June 2001, we obtained approval from our hospital ethics committee for evaluation of a new treatment for symptomatic enlarged polycystic liver; hepatic TAE. After fully informed consent was obtained, TAE was performed for patients with ADPKD with symptomatic polycystic liver for the first time in the world, beginning in 2001. After success in 2 patients reported in the American Journal of Kidney Diseases,8,9 hepatic TAE was performed in 30 patients with polycystic liver; the last patient included in the present report underwent TAE in October 2004. METHODS Patient Characteristics Considering that very few cadaveric donors for hepatic transplantation are available in Japan and open surgical cyst fenestration and partial hepatectomy were not clinically acceptable because of concomitant renal dysfunction, hepatic TAE was performed in 30 patients with ADPKD who strongly preferred to undergo that procedure to alleviate compression symptoms related to liver enlargement, even after multiple conventional surgical treatments were proposed as alternatives. The latter included hepatic transplantation and open surgical cyst fenestration and partial hepatectomy. Percutaneous cyst aspiration followed by local sclerosing agent injection was performed for cyst mass reduction at other institutions in 6 patients, but this proved ineffective. Patients gave consent after being fully informed about the new procedure, including its frequent complications of fever and pain. From June 2001 to October 2004, hepatic TAE with microcoil was performed in 5 men and 25 women with ages ranging from 41 to 73 years (mean age, 58 ⫾ 7 [SD] years). Of 30 patients, 22 were treated with dialysis, showing serum creatinine levels (measured before hemodialysis) ranging from 8.2 to 13.6 mg/dL (725 to 1,202 mol/L; mean, 10.9 ⫾ 2.2 mg/dL [964 ⫾ 194 mol/L]); 19 patients had undergone renal TAE because of kidney enlargement according to a previously described method.1,2 Two patients had increased serum creatinine levels of 3.0 and 1.3 mg/dL (265.2 and 114.9 mol/L). Six patients had serum creatinine levels less than 1.0 mg/dL (⬍88.4 mol/L). All patients were referred for hepatic TAE from various other institutions in Japan, including dialysis clinics. Massive ascites and severe spleno-
745 megaly most likely caused by portal hypertension with hepatocellular dysfunction were observed in 9 patients. On the basis of a questionnaire survey, we determined the prevalence of several symptoms in patients before they underwent TAE. Patients reported such mass lesion symptoms as abdominal distention and discomfort (86.7%), poor fit of slacks or skirt (86.7%), inability to see the feet and cut toenails (33.3%), and inability to pick up objects from the floor because of loss of balance (36.7%); such upper digestive tract symptoms as heartburn, nausea, vomiting, loss of appetite, and early satiety (70.0%); such intestinal symptoms as constipation (56.7%); such respiratory symptoms as shortness of breath and loud snoring (56.7%); lumbar spine symptoms, including “restless legs” (33.3%) and low-back pain (53.3%); such nutritional problems as thinness of legs and chest (70%); and such disturbances possibly related to anxiety as insomnia (50.0%).
Interventional Procedure for Hepatic TAE The diagnosis of ADPKD was established by means of computed tomography, magnetic resonance imaging, and ultrasonography. We performed hepatic angiography using Seldinger’s technique. After local anesthesia was achieved, the femoral artery was cannulated using a 5-Fr long-sheath catheter (Medikit Super Sheath; Yushima, Tokyo, Japan). Selective angiography of the celiac artery and superior mesenteric artery was performed using a 4-Fr Cobra catheter (Terumo; Shibuya, Tokyo, Japan) inserted through the longsheath catheter. First, portal venography was performed as the late phase of superior mesenteric artery angiography. Second, hepatic arteriograms were obtained by means of celiac arteriography. A 2.6-Fr microcatheter (Excelsior; Boston Scientific, Boston, MA) was inserted into the smaller peripheral branches of the hepatic artery through an intraCobra catheter led by a guidewire (Transend EX; Boston Scientific). After the guidewire catheter was removed, a platinum microcoil (Tornado; Cook Group Co, Bloomington, IN and C-Stopper Coil, Solution Corp, Yokohama, Japan) was introduced using a pusher (Trupush; Johnson & Johnson, Miami Lakes, FL). Coils were 0.018 inch in diameter and 4 to 18 cm in length. This procedure was performed with coordination between 1 radiologist and 12 physicians.
Target Vessels in Hepatic TAE Normal liver generally shows hepatic arterial and portal venous branches that course in parallel in the same anatomic segment. However, in polycystic livers, we found hepatic arteries and portal veins to course in a different fashion: almost all hepatic arterial branches were well developed, whereas portal venous branches in hepatic segments replaced by multiple cysts were completely or partially obstructed. By performing TAE superselectively, targeting hepatic arterial branches supplying hepatic regions with neither an intact portal vein nor intact hepatic parenchyma, we sought to minimize damage to the remaining intact liver (Figs 1 and 2). Selection of embolized hepatic regions followed the same judgment process as for resection by
746
Takei et al
Figure 1. Representative computed tomographic images. (A) Before hepatic transcatheter arterial embolization (TAE), this patient’s total liver, intrahepatic cyst, and hepatic parenchyma volumes were calculated to be 9,659, 8,427, and 1,229 cm3, respectively. Most of the liver (87%) had been replaced by multiple cysts, with essentially intact hepatic parenchyma limited to the right posterosuperior hepatic segment. (B) The same patient 24 months after hepatic TAE. Total liver and intrahepatic cyst volumes decreased to 3,229 and 1,736 cm3, whereas hepatic parenchyma volume increased to 1,495 cm3.
means of surgical partial hepatectomy.3,4 The liver ordinarily is divided into 8 segments. Angiographically, whereas a hepatic arterial branch consistent with the caudate lobe will not be identified, each of the remaining 7 hepatic segments can be identified by its hepatic arterial branch. We therefore divided the hepatic artery target for TAE into 7 hepatic segments. Hepatic segments, including embolized regions, are listed in Table 1 and shown in Fig 3. Targets for TAE included 2 of 7 segments in 7 patients, 3 segments in 6 patients, 4 segments in 7 patients, 5 segments in 9 patients, and 6 segments in 1 patient. TAE was performed 1.7 ⫾ 0.7 times (range, 1 to 3 times) per patient, with total number of microcoils used for TAE of 35.2 ⫾ 19.9 (range, 10 to 92) per patient.
Indices of Hepatic TAE Effectiveness
Liver Volume Each image obtained at intervals of 1 cm by means of CT was analyzed using National Institutes of Health Image software (Research Services Branch of the Na-
tional Institute of Mental Health, Bethesda, MD). After the computed tomographic image was digitized using a scanner, scales and units of the digital image were adjusted to those of the computed tomographic image to show actual sizes of structures in the digital image. After a certain region or site was targeted for measurement, the content of the region specified for measurement was separated into areas with different brightness intensities indicating radiodensities of cysts and hepatic parenchyma, which were divided into 256 shades on a fullcolor display. These were shown as 256 degrees of brightness on a gray scale ranging from black to white (intensity of 0, black; intensity of 256, white). For measurements, the 256 degrees were grouped into 16 classes (intensity of 0 to 15, 16 to 31, 32 to 47, and so on); a different color was assigned to each class for display (pseudocolor). Thus, different colors in the image represented different intensity classes, facilitating differentiation of cysts from hepatic parenchyma. Areas of regions showing the target intensities were measured, then total areas with each target intensity were calculated. Precise
Intravascular Treatment of Polycystic Liver
747
Figure 2. (A) Angiographic findings for the same patient as in Fig 1. Portal vein before hepatic transcatheter arterial embolization (TAE): most of the portal vein was obstructed, with only a right posterosuperior branch spared (large arrow). (B) On portal angiography, nearly intact hepatic parenchyma was seen in only the right posterosuperior segment (large arrow). (C) On hepatic arteriography, almost all hepatic arterial branches were well developed. Except for the right posterosuperior hepatic artery (large arrow), all hepatic arterial branches (small arrows), including the right (anterior and posteroinferior), middle, and left hepatic artery (ventrolateral and dorsolateral) branches, were selected as sites for embolization. (D) Angiography after TAE shows multiple linear branches obstructed by multiple microcoils. Only the right posterosuperior hepatic artery was preserved (large arrow).
automated measurements of total liver volume followed precise outlining of the organ. Hepatic area in each contrast-enhanced computed tomographic slice was separated into 2 constituents representing hepatic parenchyma and intrahepatic cysts. By adding these areas for all 1-cm computed tomographic slices and incorporating the thick-
ness of the slices, total intrahepatic cyst volume, hepatic parenchyma volume, and liver volume (cysts plus parenchyma) were calculated (at Ultmark, Osaka, Japan; Fig 4). Liver size was determined before and 28 ⫾10 months (range, 18 to 37 months) after ablation therapy. The fraction of remaining (FR) liver volumes compared with the pretreat-
Table 1. Number of Embolized Hepatic Segments Hepatic Segment
Intrahepatic Arteries
Target No. for TAE (%)
S2 (dorsolateral segment) S3 (ventrolateral segment) S4 (medial segment) S5 (anteroinferior segment) S6 (posteroinferior segment) S7 (posterosuperior segment S8 (anterosuperior segment)
Dorsolateral branch of left hepatic artery Ventrolateral branch of left hepatic artery Middle hepatic artery Anteroinferior branch of right hepatic artery Posteroinferior branch of right hepatic artery Posterosuperior branch of right hepatic artery Anterosuperior branch of right hepatic artery
18/30 (60) 19/30 (63.3) 15/30 (50) 14/30 (46.7) 5/30 (16.7) 16/30 (53.3) 25/30 (83.3)
Abbreviation: TAE, transcatheter arterial embolization.
748
Takei et al
Figure 3. Embolized hepatic segments.
ment value ([value after hepatic TAE/value before hepatic TAE] ⫻ 100%) was calculated for total liver volume, intrahepatic cysts, and intrahepatic parenchyma.
Laboratory Data Laboratory data were collected before and 1, 3, 7, and 14 days and 12 months after TAE. Laboratory data included total bilirubin, aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, alkaline phosphatase, and ␥-glutamyl transpeptidase. Serum creatinine levels were recorded for 8 patients not on dialysis therapy.
Figure 5. Liver volume at 28 months after transcatheter arterial embolization decreased significantly (P ⬍ 0.0001) compared with preintervention. Results presented as mean ⫾ SD.
We analyzed those data for only the 30 patients followed up for 28 ⫾ 10 months (range, 18 to 37 months) after ablation therapy.
Statistical Analysis Results are expressed as mean ⫾ SD. Statistical analysis was performed using StatView, version 5.0, for Macintosh (Apple Computer Japan, Shinjuku, Tokyo, Japan). For nonnormally distributed variables, differences between groups were compared using Mann-Whitney U test. Normally distributed variables were compared using unpaired t-tests. To evaluate the relationship between variables, Pearson correlation coefficient was used for normally distributed variables and Spearman ranked correlation coefficient was used when variables were non-normally distributed. P less than 0.05 is considered to indicate significance.
RESULTS
Liver Volume
Figure 4. Each computed tomographic slice (A) obtained at 1-cm intervals was analyzed using National Institutes of Health Image software. (B) In each slice, values for hepatic parenchyma (ⴱ) and intrahepatic cysts (ⴱⴱ) were calculated separately.
Total liver volume before hepatic TAE was 7,882 ⫾ 2,916 cm3 (range, 3,336 to 13,858 cm3), representing marked hepatomegaly. Twentyeight months after TAE, liver volume had significantly decreased to 6,041 ⫾ 2,282 cm3 (range, 2,692 to 10,879 cm3; P ⬍ 0.0001; Fig 5). FR of total liver volume was 78.8% ⫾17.6% (range, 33.4% to 122%). Total volume of intrahepatic cysts decreased significantly from 6,677 ⫾ 2,978 cm3 (range, 2,388 to 12,823 cm3) to 4,625 ⫾ 2,299 cm3 (range, 1,257 to 9,823 cm3; P ⬍ 0.0001; Fig 6). FR of intrahepatic cysts was 70.4% ⫾ 20.9% (range, 20.5% to 122.9%). However, hepatic parenchyma increased from 1,205 ⫾ 250 cm3 (range, 656 to 1,700 cm3) to 1,406 ⫾ 277 cm3 (range, 950 to 2,207 cm3; P ⫽ 0.0004;
Intravascular Treatment of Polycystic Liver
749
patients began to experience relief of symptoms within 1 to 2 months. On palpation, the firm, rubbery, and irregular liver surface became softer and more pliable, suggesting that cyst fluid had started to decrease, even at this early stage. Of 30 patients, 29 reported relief of symptoms by about 12 months. No relief was achieved for the remaining patient, whose liver volume increased despite TAE. Duration of hospitalization for the 30 patients was 13 ⫾ 3 days (range, 7 to 21 days). Figure 6. Total volume of intrahepatic cysts decreased significantly from 6,677 ⫾ 2,978 cm3 (range, 2,388 to 12,823 cm3) to 4,625 ⫾ 2,299 cm3 (range, 1,257 to 9,823 cm3; P ⬍ 0.0001).
Fig 7). In 29 of 30 patients, both total liver volume and intrahepatic cyst volume decreased. In 14 patients, this represented a small decrease (FR ⬎ 80% of pretreatment); in 12 patients, a moderate decrease (FR, 60% to 80%); and in 3 patients, a marked decrease (FR ⬍ 60%). However, 1 patient showed increased total liver volume (from 5,755 to 7,069 cm3) and intrahepatic cyst volume (from 4,500 to 5,531 cm3). Laboratory Data
Hepatic-related laboratory data obtained before and after TAE are listed in Table 2. Total bilirubin, aspartate aminotransferase, and alanine aminotransferase levels all increased significantly 1 and 3 days after TAE. Lactate dehydrogenase level increased significantly 3 days after TAE. Alkaline phosphatase level increased significantly at 7 and 14 days after TAE. In contrast to these transitory acute changes, values at 12 months after intervention were not significantly different from data obtained before therapy. Before and 12 months after TAE, hemoglobin levels were 9.9 ⫾ 1.7 and 10.5 ⫾ 1.7 mg/dL (99 ⫾ 17 and 105 ⫾ 17 g/L); leukocyte count, 4,600 ⫾ 1,500 and 4,900 ⫾ 1,700/L; and platelet count, 14.9 ⫾ 4.4 and 16.1 ⫾ 5.2/L, respectively. Postintervention hematologic levels after 12 months were not significantly different from those obtained before therapy. Subjective Symptoms and Hospital Stay
Before objective confirmation of hepatic size decrease by means of diagnostic imaging, most
Complications and Outcome of Hepatic TAE
Mild to moderate pain in the hypochondrium began 1 to 2 hours after the procedure in all patients. This pain subsided within 5 days in most patients. Nonsteroidal anti-inflammatory drugs and/or pentazocine were used to control the pain. Fever occurred within 6 hours after initiation of TAE in all patients. Temperature increased to a maximum of approximately 37.5°C to 38.5°C and subsided within 5 days. All patients were treated with antibiotics for 5 days, and nonsteroidal anti-inflammatory drugs were used to control fever. Serum creatinine levels were unchanged in all 8 patients not on dialysis therapy within the first 14 days. Serum creatinine levels remained unchanged at 12 months in 6 patients with normal serum creatinine levels at baseline and increased gradually from 1.3 and 3.2 mg/dL to 1.5 and 5.0 mg/dL in 2 patients with increased serum creatinine levels at baseline. Patients were managed by means of appropriate individualized fluid therapy
Figure 7. Hepatic parenchyma increased from 1,205 ⫾ 250 cm3 (range, 656 to 1,700 cm3) to 1,406 ⫾ 277 cm3 (range, 950 to 2,207 cm3; P ⫽ 0.0004).
750
Takei et al Table 2. Hepatic Laboratory Data Before and After TAE
Total bilirubin (mg/dL) Aspartate aminotransferase (IU/L) Alanine aminotransferase (IU/L) Lactate dehydrogenase (IU/L) Alkaline phosphatase (IU/L) ␥-Glutamyltransferase (IU/L)
Before
After 1 d
0.50 ⫾ 0.31 23.3 ⫾ 16.5 14.0 ⫾ 9.3 141 ⫾ 28 356 ⫾ 205 132 ⫾ 87
0.66 ⫾ 0.33* 72.7 ⫾ 52.4* 30.6 ⫾ 21.3* 205 ⫾ 74† 368 ⫾ 267 122 ⫾ 82
After 3 d
After 7 d
After 14 d
0.58 ⫾ 0.22† 0.49 ⫾ 0.23 0.48 ⫾ 0.22 44.0 ⫾ 29.0‡ 45.0 ⫾ 17.0 39.0 ⫾ 16.0 21.2 ⫾ 16.3‡ 16.4 ⫾ 14.9 13.6 ⫾ 12.6 163 ⫾ 48 149 ⫾ 40 149 ⫾ 47 362 ⫾ 268 451 ⫾ 231† 486 ⫾ 324† 100 ⫾ 86 131 ⫾ 86 127 ⫾ 73
After 12 mo
0.46 ⫾ 0.21 21.0 ⫾ 7.9 12.4 ⫾ 6.0 152 ⫾ 51 386 ⫾ 226 124 ⫾ 88
Note: To convert total bilirubin in mg/dL to mol/L, multiply by 17.1. Abbreviation: TAE, transcatheter arterial embolization. *P ⬍ 0.001 versus before. †P ⬍ 0.01 versus before. ‡P ⬍ 0.05 versus before.
from the day before TAE to 2 to 3 days after TAE. As of this writing in September 2006, we have not experienced serious complications associated with this hepatic TAE method. In 24 patients, the postinterventional clinical course was favorable, but for reasons most likely unrelated to TAE, 6 patients died (Table 3). Two patients died of intractable bacterial infection of hepatic cysts despite antibiotic therapy and cyst aspirations performed 33 and 44 months after TAE, respectively; 1 patient died of peritonitis 48 months after TAE after undergoing multiple paracenteses to relieve massive ascites; 1 patient died of hepatic failure after 20 months; 1 patient died of acute leukemia 47 months after TAE; and 1 patient died of pelvic fracture after a fall 19 months after TAE. DISCUSSION
Patients with ADPKD show a distinctive increase in renal size as renal function deteriorates, which most other renal disorders lack. When patients with ADPKD became symptomatic because of enlarged kidneys, 5 conventional methTable 3. Cause of Death in Patients Undergoing TAE
Cause of Death After TAE
Intractable hepatic cyst infection Peritonitis after multiple paracenteses for massive ascites Hepatic failure Acute leukemia Pelvic fracture after a fall
No. of Post-TAE Patients period (mo)
2
33, 44
1 1 1 1
48 20 47 19
Abbreviation: TAE, transcatheter arterial embolization.
ods were reported to decrease renal enlargement: needle aspiration of cysts followed by injection of sclerosing agents, surgical fenestration, laparoscopic fenestration, laparoscopic nephrectomy, and surgical nephrectomy, including renal transplantation after nephrectomy. For patients desiring a different treatment, we developed a safe and effective intravascular radiological technique targeting the renal arteries supplying the cysts.1,2 This intravascular embolization therapy was effective in decreasing the size of cysts, suggesting a relationship of nephromegaly caused by cyst growth to angiogenic influence on the renal arteries. This proposed mechanism also was supported in a histomorphometric analysis by Bello-Reuss et al.10 When polycystic liver becomes symptomatic in patients with ADPKD because of severe hepatomegaly, options for conventional treatment resemble those for enlarged polycystic kidney: percutaneous cyst aspiration and sclerosis, laparoscopic fenestration, open surgical cyst fenestration and partial hepatectomy,3,4 or hepatic transplantation.5-7 Percutaneous cyst aspiration with sclerosis and laparoscopic fenestration are limited to patients with 1 or a few very large easily accessible cysts and are not indicated for those with multiple diffuse liver cysts. Open surgical cyst fenestration and partial hepatectomy were reported to be effective for treating symptomatic polycystic liver.3,4 However, this method has been limited to patients younger than 60 years with normal or mildly decreased renal function. Hepatic transplantation for symptomatic polycystic liver was reported to yield excellent results. Combined liver-kidney transplantation also was
Intravascular Treatment of Polycystic Liver
751
Figure 8. Angiographically, patients with polycystic liver show hepatic artery and portal vein coursing parallel in normal segments, whereas in cystic segments, almost all hepatic arterial branches are well developed, predominating over the portal vein branches, which are obstructed to varying degrees.
suggested for patients with ADPKD undergoing maintenance dialysis.5-7 However, in Japan, where cadaveric donors are very scarce, hepatic transplantation was performed successfully in only 7 patients with polycystic liver. Because renal TAE yielded good results for patients with large polycystic kidneys, we applied TAE to patients with ADPKD with symptomatic enlarged polycystic liver. Angiographically, patients with polycystic liver showed a parallel course for the hepatic artery and portal vein in normal hepatic segments. However, in cystic segments, almost all hepatic arterial branches are well developed, predominating over portal vein branches, which show varying degrees of obstruction (Fig 8). After hepatic TAE, cystic segments become less tense and begin to
shrink. Decompressed hepatic parenchyma tends to expand in volume and return to its original position (Fig 9). The mechanism of growth of hepatic cysts remains unknown. However, because most patients with an enlarged polycystic liver are women who have had multiple pregnancies, estrogen is suspected to contribute to cyst growth in the liver. Otherwise, little is known about why cysts enlarge. After TAE for 30 patients with ADPKD with symptomatic polycystic liver, 4 died of liverrelated causes (hepatic cyst infection, peritonitis after multiple paracenteses for massive ascites, and hepatic failure). These deaths followed TAE by 20 to 48 months. Because intrahepatic cyst volume usually continues to increase in patients with an enlarged polycystic liver, the remaining
Figure 9. Liver segments occupied by cysts are rigid, encroaching upon neighboring segments that ultimately will be replaced by cysts. Remaining intact hepatic parenchyma is compressed, and the intact hepatic volume is decreased, resulting in establishment of polycystic liver with severe hepatomegaly. After hepatic transcatheter arterial embolization, cysts become smaller. Hepatic parenchyma expands, returning to its original position.
752
intact hepatic parenchyma progressively decreases. Without success in treatment of hepatic complications, a fatal outcome was likely. We attribute this failure largely to insufficient reversal of cyst progression in patients who died. From the viewpoint of our experience, TAE can be safe only when embolized arteries are chosen to correspond to hepatic regions so that neither an intact portal vein nor intact hepatic parenchyma could be identified, paralleling the indication for partial hepatectomy. In performing TAE superselectively, we aimed to preserve remaining intact hepatic parenchyma and permit reexpansion of decompressed hepatic parenchyma. Total liver volume decreased to an average of 78% of pretreatment volume after 28 months. However, the decrease remained small in about half the patients, whereas 1 patient showed further enlargement despite treatment. Our treatment to date may be less effective in decreasing total liver volume than surgical resection. However, our treatment should be safe even when performed repeatedly or after a relapse after surgical therapy. We are preparing for additional TAE in patients with a poor initial result. Hepatic TAE also may be an option for treatment of patients with symptomatic polycystic liver in poor general medical condition who are not candidates for hepatic resection.
Takei et al REFERENCES 1. Ubara Y, Katori H, Tagami T, et al: Transcatheter renal arterial embolization therapy on a patient with polycystic kidney disease on hemodialysis. Am J Kidney Dis 34: 926-931, 1999 2. Ubara Y, Tagami T, Sawa N, et al: Renal contraction therapy for enlarged polycystic kidneys by transcatheter arterial embolization in hemodialysis patients. Am J Kidney Dis 39:571-579, 2002 3. Que F, Nagorney DM, Gross JB Jr, Torres VE: Liver resection and cyst fenestration in the treatment of severe polycystic liver disease. Gastroenterology 108:487-494, 1995 4. Soravia C, Mentha G, Giostra E, Morel P, Rohner A: Surgery for adult polycystic liver disease. Surgery 117:272275, 1995 5. Lang H, Woellwarth JV, Oldhafer KJ, et al: Liver transplantation in patients with polycystic liver disease. Transplant Proc 29:2832-2833, 1997 6. Pirenne J, Aerts R, Yoong K, et al: Liver transplantation for polycystic liver disease. Liver Transpl 7:238-245, 2001 7. Ueno T, Barri YM, Netto GJ, et al. Liver and kidney transplantation for polycystic liver and kidney-renal function and outcome. Transplantation 82:501-507, 2006 8. Ubara Y, Takei R, Hoshino J, et al: Intravascular embolization therapy in a patient with an enlarged polycystic liver. Am J Kidney Dis 43:733-738, 2004 9. Ubara Y, Higa Y, Tagami T, et al: Pelvic insufficiency fracture related to autosomal dominant polycystic kidney disease. Am J Kidney Dis 46:e103-e111, 2005 10. Bello-Reuss E, Holubec K, Rajaraman S: Angiogenesis in autosomal-dominant polycystic kidney disease. Kidney Int 60:37-45, 2001
Percutaneous Transcatheter Hepatic Artery Embolization for Liver Cysts in Autosomal Dominant Polycystic Kidney Disease Ryoji Takei, MD,1 Yoshifumi Ubara, MD,2 Junichi Hoshino, MD,2 Yasushi Higa, MD,2 Tatsuya Suwabe, MD,2 Yoko Sogawa, MD,2 Kazufumi Nomura, MD,2 Shohei Nakanishi, MD,2 Naoki Sawa, MD,2 Hideyuki Katori, MD,2 Fumi Takemoto, MD,2 Shigeko Hara, MD,2 and Kenmei Takaichi, MD2 Background: We have achieved renal contraction therapy in patients with autosomal dominant polycystic kidney disease (ADPKD) by means of renal transcatheter arterial embolization (TAE) using intravascular coils, decreasing renal size and improving quality of life in almost all patients. We presently perform hepatic TAE in patients with intractable symptomatic polycystic liver. Study Design: Uncontrolled trial. Setting & Participants: 30 patients with ADPKD referred for arteriography to an academic medical center. 22 patients had kidney failure treated by means of dialysis. Intervention: We embolized arteries supplying hepatic segments replaced by cysts that were associated with well-developed hepatic arteries, but obstructed intrahepatic portal veins. Outcomes & Measurements: Various volumes before and after TAE were compared by using computed tomography and National Institutes of Health Image software in 30 patients with follow-up computed tomography 18 to 37 months after therapy. Results: Total liver volume and total intrahepatic cyst volume decreased from 7,882 ⫾ 2,916 and 6,677 ⫾ 2,978 to 6,041 ⫾ 2,282 and 4,625 ⫾ 2,299 cm3, respectively (P ⬍ 0.0001 for both). Fractions of remaining (FR) total liver volume and FR of intrahepatic cyst volume were 78.8% ⫾ 17.6% and 70.4% ⫾ 20.9%, respectively. Hepatic parenchyma increased from 1,205 ⫾ 250 to 1,406 ⫾ 277 cm3 (P ⫽ 0.0004). In 29 of 30 patients, both total liver volume and intrahepatic cyst volume decreased; in 1 patient, total liver volume increased from 5,755 to 7,069 cm3, whereas cysts enlarged from 4,500 to 5,531 cm3. No serious complications were experienced. In 24 patients, the post-TAE course was favorable. TAE failed to benefit 6 patients because of unrelated hepatic infection, peritonitis, hepatic failure, acute leukemia, or pelvic fracture. Limitations: Absence of a control group. Conclusions: TAE may be an option for patients with ADPKD with symptomatic polycystic liver who are not candidates for surgical treatment. Am J Kidney Dis 49:744-752. © 2007 by the National Kidney Foundation, Inc. INDEX WORDS: Autosomal dominant polycystic kidney disease (ADPKD); transcatheter arterial embolization (TAE); intravascular treatment; percutaneous transcatheter hepatic artery embolization; polycystic liver disease.
K
Editorial, p. 725
From the 1Department of Radiology and 2Nephrology Center, Toranomon Hospital Kajigaya, Kanagawa, Japan. Received October 22, 2006; accepted in revised form March 2, 2007. Originally published online as doi:10.1053/j.ajkd.2007.03.018 on May 8, 2007. Support: None. Potential conflicts of interest: None. Address correspondence to Yoshifumi Ubara, MD, Nephrology Center, Toranomon Hospital Kajigaya, 1-3-1, Kajigaya, Takatu-ku, Kawasaki, Kanagawa, 213-0015, Japan. E-mail: [email protected] © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/4906-0005$32.00/0 doi:10.1053/j.ajkd.2007.03.018 744
idneys of patients with autosomal dominant polycystic kidney disease (ADPKD) typically are supplied by well-developed arteries. Accordingly, we performed renal cyst contraction therapy by means of renal transcatheter arterial embolization (TAE) in patients with renal enlargement, including those with cyst hemorrhage, beginning in 1996.1,2 Renal TAE was effective in itself for patients with organomegaly that was predominantly renal. Although renal TAE was effective in patients with similar enlargement of both kidneys and liver, additional treatment was required for hepatomegaly. Renal TAE did not benefit patients with primarily hepatic enlargement who similarly required treatment for hepatomegaly. Several conventional surgical treatments have been performed as therapy for patients with
American Journal of Kidney Diseases, Vol 49, No 6 (June), 2007: pp 744-752
Intravascular Treatment of Polycystic Liver
symptomatic polycystic liver, including percutaneous cyst aspiration and sclerosis, laparoscopic fenestration, open surgical cyst fenestration and partial hepatectomy, and hepatic transplantation.3-7 Because renal TAE was successful in treating enlarged kidneys, many symptomatic patients came to our institution seeking other intravascular treatment. In June 2001, we obtained approval from our hospital ethics committee for evaluation of a new treatment for symptomatic enlarged polycystic liver; hepatic TAE. After fully informed consent was obtained, TAE was performed for patients with ADPKD with symptomatic polycystic liver for the first time in the world, beginning in 2001. After success in 2 patients reported in the American Journal of Kidney Diseases,8,9 hepatic TAE was performed in 30 patients with polycystic liver; the last patient included in the present report underwent TAE in October 2004. METHODS Patient Characteristics Considering that very few cadaveric donors for hepatic transplantation are available in Japan and open surgical cyst fenestration and partial hepatectomy were not clinically acceptable because of concomitant renal dysfunction, hepatic TAE was performed in 30 patients with ADPKD who strongly preferred to undergo that procedure to alleviate compression symptoms related to liver enlargement, even after multiple conventional surgical treatments were proposed as alternatives. The latter included hepatic transplantation and open surgical cyst fenestration and partial hepatectomy. Percutaneous cyst aspiration followed by local sclerosing agent injection was performed for cyst mass reduction at other institutions in 6 patients, but this proved ineffective. Patients gave consent after being fully informed about the new procedure, including its frequent complications of fever and pain. From June 2001 to October 2004, hepatic TAE with microcoil was performed in 5 men and 25 women with ages ranging from 41 to 73 years (mean age, 58 ⫾ 7 [SD] years). Of 30 patients, 22 were treated with dialysis, showing serum creatinine levels (measured before hemodialysis) ranging from 8.2 to 13.6 mg/dL (725 to 1,202 mol/L; mean, 10.9 ⫾ 2.2 mg/dL [964 ⫾ 194 mol/L]); 19 patients had undergone renal TAE because of kidney enlargement according to a previously described method.1,2 Two patients had increased serum creatinine levels of 3.0 and 1.3 mg/dL (265.2 and 114.9 mol/L). Six patients had serum creatinine levels less than 1.0 mg/dL (⬍88.4 mol/L). All patients were referred for hepatic TAE from various other institutions in Japan, including dialysis clinics. Massive ascites and severe spleno-
745 megaly most likely caused by portal hypertension with hepatocellular dysfunction were observed in 9 patients. On the basis of a questionnaire survey, we determined the prevalence of several symptoms in patients before they underwent TAE. Patients reported such mass lesion symptoms as abdominal distention and discomfort (86.7%), poor fit of slacks or skirt (86.7%), inability to see the feet and cut toenails (33.3%), and inability to pick up objects from the floor because of loss of balance (36.7%); such upper digestive tract symptoms as heartburn, nausea, vomiting, loss of appetite, and early satiety (70.0%); such intestinal symptoms as constipation (56.7%); such respiratory symptoms as shortness of breath and loud snoring (56.7%); lumbar spine symptoms, including “restless legs” (33.3%) and low-back pain (53.3%); such nutritional problems as thinness of legs and chest (70%); and such disturbances possibly related to anxiety as insomnia (50.0%).
Interventional Procedure for Hepatic TAE The diagnosis of ADPKD was established by means of computed tomography, magnetic resonance imaging, and ultrasonography. We performed hepatic angiography using Seldinger’s technique. After local anesthesia was achieved, the femoral artery was cannulated using a 5-Fr long-sheath catheter (Medikit Super Sheath; Yushima, Tokyo, Japan). Selective angiography of the celiac artery and superior mesenteric artery was performed using a 4-Fr Cobra catheter (Terumo; Shibuya, Tokyo, Japan) inserted through the longsheath catheter. First, portal venography was performed as the late phase of superior mesenteric artery angiography. Second, hepatic arteriograms were obtained by means of celiac arteriography. A 2.6-Fr microcatheter (Excelsior; Boston Scientific, Boston, MA) was inserted into the smaller peripheral branches of the hepatic artery through an intraCobra catheter led by a guidewire (Transend EX; Boston Scientific). After the guidewire catheter was removed, a platinum microcoil (Tornado; Cook Group Co, Bloomington, IN and C-Stopper Coil, Solution Corp, Yokohama, Japan) was introduced using a pusher (Trupush; Johnson & Johnson, Miami Lakes, FL). Coils were 0.018 inch in diameter and 4 to 18 cm in length. This procedure was performed with coordination between 1 radiologist and 12 physicians.
Target Vessels in Hepatic TAE Normal liver generally shows hepatic arterial and portal venous branches that course in parallel in the same anatomic segment. However, in polycystic livers, we found hepatic arteries and portal veins to course in a different fashion: almost all hepatic arterial branches were well developed, whereas portal venous branches in hepatic segments replaced by multiple cysts were completely or partially obstructed. By performing TAE superselectively, targeting hepatic arterial branches supplying hepatic regions with neither an intact portal vein nor intact hepatic parenchyma, we sought to minimize damage to the remaining intact liver (Figs 1 and 2). Selection of embolized hepatic regions followed the same judgment process as for resection by
746
Takei et al
Figure 1. Representative computed tomographic images. (A) Before hepatic transcatheter arterial embolization (TAE), this patient’s total liver, intrahepatic cyst, and hepatic parenchyma volumes were calculated to be 9,659, 8,427, and 1,229 cm3, respectively. Most of the liver (87%) had been replaced by multiple cysts, with essentially intact hepatic parenchyma limited to the right posterosuperior hepatic segment. (B) The same patient 24 months after hepatic TAE. Total liver and intrahepatic cyst volumes decreased to 3,229 and 1,736 cm3, whereas hepatic parenchyma volume increased to 1,495 cm3.
means of surgical partial hepatectomy.3,4 The liver ordinarily is divided into 8 segments. Angiographically, whereas a hepatic arterial branch consistent with the caudate lobe will not be identified, each of the remaining 7 hepatic segments can be identified by its hepatic arterial branch. We therefore divided the hepatic artery target for TAE into 7 hepatic segments. Hepatic segments, including embolized regions, are listed in Table 1 and shown in Fig 3. Targets for TAE included 2 of 7 segments in 7 patients, 3 segments in 6 patients, 4 segments in 7 patients, 5 segments in 9 patients, and 6 segments in 1 patient. TAE was performed 1.7 ⫾ 0.7 times (range, 1 to 3 times) per patient, with total number of microcoils used for TAE of 35.2 ⫾ 19.9 (range, 10 to 92) per patient.
Indices of Hepatic TAE Effectiveness
Liver Volume Each image obtained at intervals of 1 cm by means of CT was analyzed using National Institutes of Health Image software (Research Services Branch of the Na-
tional Institute of Mental Health, Bethesda, MD). After the computed tomographic image was digitized using a scanner, scales and units of the digital image were adjusted to those of the computed tomographic image to show actual sizes of structures in the digital image. After a certain region or site was targeted for measurement, the content of the region specified for measurement was separated into areas with different brightness intensities indicating radiodensities of cysts and hepatic parenchyma, which were divided into 256 shades on a fullcolor display. These were shown as 256 degrees of brightness on a gray scale ranging from black to white (intensity of 0, black; intensity of 256, white). For measurements, the 256 degrees were grouped into 16 classes (intensity of 0 to 15, 16 to 31, 32 to 47, and so on); a different color was assigned to each class for display (pseudocolor). Thus, different colors in the image represented different intensity classes, facilitating differentiation of cysts from hepatic parenchyma. Areas of regions showing the target intensities were measured, then total areas with each target intensity were calculated. Precise
Intravascular Treatment of Polycystic Liver
747
Figure 2. (A) Angiographic findings for the same patient as in Fig 1. Portal vein before hepatic transcatheter arterial embolization (TAE): most of the portal vein was obstructed, with only a right posterosuperior branch spared (large arrow). (B) On portal angiography, nearly intact hepatic parenchyma was seen in only the right posterosuperior segment (large arrow). (C) On hepatic arteriography, almost all hepatic arterial branches were well developed. Except for the right posterosuperior hepatic artery (large arrow), all hepatic arterial branches (small arrows), including the right (anterior and posteroinferior), middle, and left hepatic artery (ventrolateral and dorsolateral) branches, were selected as sites for embolization. (D) Angiography after TAE shows multiple linear branches obstructed by multiple microcoils. Only the right posterosuperior hepatic artery was preserved (large arrow).
automated measurements of total liver volume followed precise outlining of the organ. Hepatic area in each contrast-enhanced computed tomographic slice was separated into 2 constituents representing hepatic parenchyma and intrahepatic cysts. By adding these areas for all 1-cm computed tomographic slices and incorporating the thick-
ness of the slices, total intrahepatic cyst volume, hepatic parenchyma volume, and liver volume (cysts plus parenchyma) were calculated (at Ultmark, Osaka, Japan; Fig 4). Liver size was determined before and 28 ⫾10 months (range, 18 to 37 months) after ablation therapy. The fraction of remaining (FR) liver volumes compared with the pretreat-
Table 1. Number of Embolized Hepatic Segments Hepatic Segment
Intrahepatic Arteries
Target No. for TAE (%)
S2 (dorsolateral segment) S3 (ventrolateral segment) S4 (medial segment) S5 (anteroinferior segment) S6 (posteroinferior segment) S7 (posterosuperior segment S8 (anterosuperior segment)
Dorsolateral branch of left hepatic artery Ventrolateral branch of left hepatic artery Middle hepatic artery Anteroinferior branch of right hepatic artery Posteroinferior branch of right hepatic artery Posterosuperior branch of right hepatic artery Anterosuperior branch of right hepatic artery
18/30 (60) 19/30 (63.3) 15/30 (50) 14/30 (46.7) 5/30 (16.7) 16/30 (53.3) 25/30 (83.3)
Abbreviation: TAE, transcatheter arterial embolization.
748
Takei et al
Figure 3. Embolized hepatic segments.
ment value ([value after hepatic TAE/value before hepatic TAE] ⫻ 100%) was calculated for total liver volume, intrahepatic cysts, and intrahepatic parenchyma.
Laboratory Data Laboratory data were collected before and 1, 3, 7, and 14 days and 12 months after TAE. Laboratory data included total bilirubin, aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, alkaline phosphatase, and ␥-glutamyl transpeptidase. Serum creatinine levels were recorded for 8 patients not on dialysis therapy.
Figure 5. Liver volume at 28 months after transcatheter arterial embolization decreased significantly (P ⬍ 0.0001) compared with preintervention. Results presented as mean ⫾ SD.
We analyzed those data for only the 30 patients followed up for 28 ⫾ 10 months (range, 18 to 37 months) after ablation therapy.
Statistical Analysis Results are expressed as mean ⫾ SD. Statistical analysis was performed using StatView, version 5.0, for Macintosh (Apple Computer Japan, Shinjuku, Tokyo, Japan). For nonnormally distributed variables, differences between groups were compared using Mann-Whitney U test. Normally distributed variables were compared using unpaired t-tests. To evaluate the relationship between variables, Pearson correlation coefficient was used for normally distributed variables and Spearman ranked correlation coefficient was used when variables were non-normally distributed. P less than 0.05 is considered to indicate significance.
RESULTS
Liver Volume
Figure 4. Each computed tomographic slice (A) obtained at 1-cm intervals was analyzed using National Institutes of Health Image software. (B) In each slice, values for hepatic parenchyma (ⴱ) and intrahepatic cysts (ⴱⴱ) were calculated separately.
Total liver volume before hepatic TAE was 7,882 ⫾ 2,916 cm3 (range, 3,336 to 13,858 cm3), representing marked hepatomegaly. Twentyeight months after TAE, liver volume had significantly decreased to 6,041 ⫾ 2,282 cm3 (range, 2,692 to 10,879 cm3; P ⬍ 0.0001; Fig 5). FR of total liver volume was 78.8% ⫾17.6% (range, 33.4% to 122%). Total volume of intrahepatic cysts decreased significantly from 6,677 ⫾ 2,978 cm3 (range, 2,388 to 12,823 cm3) to 4,625 ⫾ 2,299 cm3 (range, 1,257 to 9,823 cm3; P ⬍ 0.0001; Fig 6). FR of intrahepatic cysts was 70.4% ⫾ 20.9% (range, 20.5% to 122.9%). However, hepatic parenchyma increased from 1,205 ⫾ 250 cm3 (range, 656 to 1,700 cm3) to 1,406 ⫾ 277 cm3 (range, 950 to 2,207 cm3; P ⫽ 0.0004;
Intravascular Treatment of Polycystic Liver
749
patients began to experience relief of symptoms within 1 to 2 months. On palpation, the firm, rubbery, and irregular liver surface became softer and more pliable, suggesting that cyst fluid had started to decrease, even at this early stage. Of 30 patients, 29 reported relief of symptoms by about 12 months. No relief was achieved for the remaining patient, whose liver volume increased despite TAE. Duration of hospitalization for the 30 patients was 13 ⫾ 3 days (range, 7 to 21 days). Figure 6. Total volume of intrahepatic cysts decreased significantly from 6,677 ⫾ 2,978 cm3 (range, 2,388 to 12,823 cm3) to 4,625 ⫾ 2,299 cm3 (range, 1,257 to 9,823 cm3; P ⬍ 0.0001).
Fig 7). In 29 of 30 patients, both total liver volume and intrahepatic cyst volume decreased. In 14 patients, this represented a small decrease (FR ⬎ 80% of pretreatment); in 12 patients, a moderate decrease (FR, 60% to 80%); and in 3 patients, a marked decrease (FR ⬍ 60%). However, 1 patient showed increased total liver volume (from 5,755 to 7,069 cm3) and intrahepatic cyst volume (from 4,500 to 5,531 cm3). Laboratory Data
Hepatic-related laboratory data obtained before and after TAE are listed in Table 2. Total bilirubin, aspartate aminotransferase, and alanine aminotransferase levels all increased significantly 1 and 3 days after TAE. Lactate dehydrogenase level increased significantly 3 days after TAE. Alkaline phosphatase level increased significantly at 7 and 14 days after TAE. In contrast to these transitory acute changes, values at 12 months after intervention were not significantly different from data obtained before therapy. Before and 12 months after TAE, hemoglobin levels were 9.9 ⫾ 1.7 and 10.5 ⫾ 1.7 mg/dL (99 ⫾ 17 and 105 ⫾ 17 g/L); leukocyte count, 4,600 ⫾ 1,500 and 4,900 ⫾ 1,700/L; and platelet count, 14.9 ⫾ 4.4 and 16.1 ⫾ 5.2/L, respectively. Postintervention hematologic levels after 12 months were not significantly different from those obtained before therapy. Subjective Symptoms and Hospital Stay
Before objective confirmation of hepatic size decrease by means of diagnostic imaging, most
Complications and Outcome of Hepatic TAE
Mild to moderate pain in the hypochondrium began 1 to 2 hours after the procedure in all patients. This pain subsided within 5 days in most patients. Nonsteroidal anti-inflammatory drugs and/or pentazocine were used to control the pain. Fever occurred within 6 hours after initiation of TAE in all patients. Temperature increased to a maximum of approximately 37.5°C to 38.5°C and subsided within 5 days. All patients were treated with antibiotics for 5 days, and nonsteroidal anti-inflammatory drugs were used to control fever. Serum creatinine levels were unchanged in all 8 patients not on dialysis therapy within the first 14 days. Serum creatinine levels remained unchanged at 12 months in 6 patients with normal serum creatinine levels at baseline and increased gradually from 1.3 and 3.2 mg/dL to 1.5 and 5.0 mg/dL in 2 patients with increased serum creatinine levels at baseline. Patients were managed by means of appropriate individualized fluid therapy
Figure 7. Hepatic parenchyma increased from 1,205 ⫾ 250 cm3 (range, 656 to 1,700 cm3) to 1,406 ⫾ 277 cm3 (range, 950 to 2,207 cm3; P ⫽ 0.0004).
750
Takei et al Table 2. Hepatic Laboratory Data Before and After TAE
Total bilirubin (mg/dL) Aspartate aminotransferase (IU/L) Alanine aminotransferase (IU/L) Lactate dehydrogenase (IU/L) Alkaline phosphatase (IU/L) ␥-Glutamyltransferase (IU/L)
Before
After 1 d
0.50 ⫾ 0.31 23.3 ⫾ 16.5 14.0 ⫾ 9.3 141 ⫾ 28 356 ⫾ 205 132 ⫾ 87
0.66 ⫾ 0.33* 72.7 ⫾ 52.4* 30.6 ⫾ 21.3* 205 ⫾ 74† 368 ⫾ 267 122 ⫾ 82
After 3 d
After 7 d
After 14 d
0.58 ⫾ 0.22† 0.49 ⫾ 0.23 0.48 ⫾ 0.22 44.0 ⫾ 29.0‡ 45.0 ⫾ 17.0 39.0 ⫾ 16.0 21.2 ⫾ 16.3‡ 16.4 ⫾ 14.9 13.6 ⫾ 12.6 163 ⫾ 48 149 ⫾ 40 149 ⫾ 47 362 ⫾ 268 451 ⫾ 231† 486 ⫾ 324† 100 ⫾ 86 131 ⫾ 86 127 ⫾ 73
After 12 mo
0.46 ⫾ 0.21 21.0 ⫾ 7.9 12.4 ⫾ 6.0 152 ⫾ 51 386 ⫾ 226 124 ⫾ 88
Note: To convert total bilirubin in mg/dL to mol/L, multiply by 17.1. Abbreviation: TAE, transcatheter arterial embolization. *P ⬍ 0.001 versus before. †P ⬍ 0.01 versus before. ‡P ⬍ 0.05 versus before.
from the day before TAE to 2 to 3 days after TAE. As of this writing in September 2006, we have not experienced serious complications associated with this hepatic TAE method. In 24 patients, the postinterventional clinical course was favorable, but for reasons most likely unrelated to TAE, 6 patients died (Table 3). Two patients died of intractable bacterial infection of hepatic cysts despite antibiotic therapy and cyst aspirations performed 33 and 44 months after TAE, respectively; 1 patient died of peritonitis 48 months after TAE after undergoing multiple paracenteses to relieve massive ascites; 1 patient died of hepatic failure after 20 months; 1 patient died of acute leukemia 47 months after TAE; and 1 patient died of pelvic fracture after a fall 19 months after TAE. DISCUSSION
Patients with ADPKD show a distinctive increase in renal size as renal function deteriorates, which most other renal disorders lack. When patients with ADPKD became symptomatic because of enlarged kidneys, 5 conventional methTable 3. Cause of Death in Patients Undergoing TAE
Cause of Death After TAE
Intractable hepatic cyst infection Peritonitis after multiple paracenteses for massive ascites Hepatic failure Acute leukemia Pelvic fracture after a fall
No. of Post-TAE Patients period (mo)
2
33, 44
1 1 1 1
48 20 47 19
Abbreviation: TAE, transcatheter arterial embolization.
ods were reported to decrease renal enlargement: needle aspiration of cysts followed by injection of sclerosing agents, surgical fenestration, laparoscopic fenestration, laparoscopic nephrectomy, and surgical nephrectomy, including renal transplantation after nephrectomy. For patients desiring a different treatment, we developed a safe and effective intravascular radiological technique targeting the renal arteries supplying the cysts.1,2 This intravascular embolization therapy was effective in decreasing the size of cysts, suggesting a relationship of nephromegaly caused by cyst growth to angiogenic influence on the renal arteries. This proposed mechanism also was supported in a histomorphometric analysis by Bello-Reuss et al.10 When polycystic liver becomes symptomatic in patients with ADPKD because of severe hepatomegaly, options for conventional treatment resemble those for enlarged polycystic kidney: percutaneous cyst aspiration and sclerosis, laparoscopic fenestration, open surgical cyst fenestration and partial hepatectomy,3,4 or hepatic transplantation.5-7 Percutaneous cyst aspiration with sclerosis and laparoscopic fenestration are limited to patients with 1 or a few very large easily accessible cysts and are not indicated for those with multiple diffuse liver cysts. Open surgical cyst fenestration and partial hepatectomy were reported to be effective for treating symptomatic polycystic liver.3,4 However, this method has been limited to patients younger than 60 years with normal or mildly decreased renal function. Hepatic transplantation for symptomatic polycystic liver was reported to yield excellent results. Combined liver-kidney transplantation also was
Intravascular Treatment of Polycystic Liver
751
Figure 8. Angiographically, patients with polycystic liver show hepatic artery and portal vein coursing parallel in normal segments, whereas in cystic segments, almost all hepatic arterial branches are well developed, predominating over the portal vein branches, which are obstructed to varying degrees.
suggested for patients with ADPKD undergoing maintenance dialysis.5-7 However, in Japan, where cadaveric donors are very scarce, hepatic transplantation was performed successfully in only 7 patients with polycystic liver. Because renal TAE yielded good results for patients with large polycystic kidneys, we applied TAE to patients with ADPKD with symptomatic enlarged polycystic liver. Angiographically, patients with polycystic liver showed a parallel course for the hepatic artery and portal vein in normal hepatic segments. However, in cystic segments, almost all hepatic arterial branches are well developed, predominating over portal vein branches, which show varying degrees of obstruction (Fig 8). After hepatic TAE, cystic segments become less tense and begin to
shrink. Decompressed hepatic parenchyma tends to expand in volume and return to its original position (Fig 9). The mechanism of growth of hepatic cysts remains unknown. However, because most patients with an enlarged polycystic liver are women who have had multiple pregnancies, estrogen is suspected to contribute to cyst growth in the liver. Otherwise, little is known about why cysts enlarge. After TAE for 30 patients with ADPKD with symptomatic polycystic liver, 4 died of liverrelated causes (hepatic cyst infection, peritonitis after multiple paracenteses for massive ascites, and hepatic failure). These deaths followed TAE by 20 to 48 months. Because intrahepatic cyst volume usually continues to increase in patients with an enlarged polycystic liver, the remaining
Figure 9. Liver segments occupied by cysts are rigid, encroaching upon neighboring segments that ultimately will be replaced by cysts. Remaining intact hepatic parenchyma is compressed, and the intact hepatic volume is decreased, resulting in establishment of polycystic liver with severe hepatomegaly. After hepatic transcatheter arterial embolization, cysts become smaller. Hepatic parenchyma expands, returning to its original position.
752
intact hepatic parenchyma progressively decreases. Without success in treatment of hepatic complications, a fatal outcome was likely. We attribute this failure largely to insufficient reversal of cyst progression in patients who died. From the viewpoint of our experience, TAE can be safe only when embolized arteries are chosen to correspond to hepatic regions so that neither an intact portal vein nor intact hepatic parenchyma could be identified, paralleling the indication for partial hepatectomy. In performing TAE superselectively, we aimed to preserve remaining intact hepatic parenchyma and permit reexpansion of decompressed hepatic parenchyma. Total liver volume decreased to an average of 78% of pretreatment volume after 28 months. However, the decrease remained small in about half the patients, whereas 1 patient showed further enlargement despite treatment. Our treatment to date may be less effective in decreasing total liver volume than surgical resection. However, our treatment should be safe even when performed repeatedly or after a relapse after surgical therapy. We are preparing for additional TAE in patients with a poor initial result. Hepatic TAE also may be an option for treatment of patients with symptomatic polycystic liver in poor general medical condition who are not candidates for hepatic resection.
Takei et al REFERENCES 1. Ubara Y, Katori H, Tagami T, et al: Transcatheter renal arterial embolization therapy on a patient with polycystic kidney disease on hemodialysis. Am J Kidney Dis 34: 926-931, 1999 2. Ubara Y, Tagami T, Sawa N, et al: Renal contraction therapy for enlarged polycystic kidneys by transcatheter arterial embolization in hemodialysis patients. Am J Kidney Dis 39:571-579, 2002 3. Que F, Nagorney DM, Gross JB Jr, Torres VE: Liver resection and cyst fenestration in the treatment of severe polycystic liver disease. Gastroenterology 108:487-494, 1995 4. Soravia C, Mentha G, Giostra E, Morel P, Rohner A: Surgery for adult polycystic liver disease. Surgery 117:272275, 1995 5. Lang H, Woellwarth JV, Oldhafer KJ, et al: Liver transplantation in patients with polycystic liver disease. Transplant Proc 29:2832-2833, 1997 6. Pirenne J, Aerts R, Yoong K, et al: Liver transplantation for polycystic liver disease. Liver Transpl 7:238-245, 2001 7. Ueno T, Barri YM, Netto GJ, et al. Liver and kidney transplantation for polycystic liver and kidney-renal function and outcome. Transplantation 82:501-507, 2006 8. Ubara Y, Takei R, Hoshino J, et al: Intravascular embolization therapy in a patient with an enlarged polycystic liver. Am J Kidney Dis 43:733-738, 2004 9. Ubara Y, Higa Y, Tagami T, et al: Pelvic insufficiency fracture related to autosomal dominant polycystic kidney disease. Am J Kidney Dis 46:e103-e111, 2005 10. Bello-Reuss E, Holubec K, Rajaraman S: Angiogenesis in autosomal-dominant polycystic kidney disease. Kidney Int 60:37-45, 2001