| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Imaging, The Hospital for Sick Children, 555
University Ave., Toronto, Ontario M5G 1X8, Canada.
2 Department of Gastroenterology, The Hospital for Sick Children, Toronto,
Ontario M5G 1X8, Canada.
Received August 6, 2002;
accepted after revision October 3, 2002.
Presented at the annual meeting of the Radiological Society of North
America, November 2001.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively reviewed 74 transjugular hepatic biopsies performed in 64 pediatric patients. The selection criteria for transjugular approach in these children included mainly coagulopathy, thrombocytopenia, or ascites. The last 37 biopsies in our series were performed with combined sonographic and fluoroscopic guidance, which improved visualization of the biopsy tract.
RESULTS. Adequate samples for histopathologic analysis were obtained in all except one case (98.6%). In most patients with fulminant hepatic failure, biopsy results allowed patients to be promptly listed for orthotopic liver transplantation; in patients with less severe hepatic failure, biopsy results helped guide medical management. In patients with liver transplants, the biopsy provided information on acute graft rejection; in patients who had undergone bone marrow transplantation, the biopsy helped to determine the status of graft-versus-host disease. One death occurred immediately after the procedure but at autopsy was deemed not to have been caused by the biopsy itself. Overall, procedural complications occurred in 8.1% of patients. Complications included neck hematomas, small subcapsular hematomas, subclavian artery puncture, and extravasation of contrast material into the retroperitoneum.
CONCLUSION. Transjugular hepatic biopsy is a relatively safe procedure that has considerable impact on patient treatment. The addition of sonographic guidance during the biopsy improves visualization, increases operator confidence, and allows the performance of biopsies in smaller patients and in children with reduced liver transplants.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Despite its greater technical difficulty and higher cost, transjugular hepatic biopsy offers an effective and safe means of obtaining sufficient liver tissue samples for pathologic analysis when percutaneous biopsies are contraindicated. This technique reduces the risk of hemorrhage because a biopsy specimen is acquired through the hepatic vein; in theory, any bleeding from the puncture site escapes into the vascular space [12].
Indications for transjugular hepatic biopsy include severe or uncorrectable coagulopathy, ascites, or both [3, 4, 12]. Less common indications include unsuccessful percutaneous biopsy of the liver, obesity, small or cirrhotic liver, and segmental liver grafts or the concurrent need for measurements of hepatic and portal venous pressures or contrast-enhanced visualization of hepatic and portal veins (Fig. 1). Relative contraindications to the transjugular approach described in the literature include hemodynamic instability, lack of upper venous access, an occluded superior vena cava, absent or thrombosed inferior vena cava, abnormal vascular anatomy due to congenital heart disease, and small patient size [9, 11].
|
We describe our experience with 74 transjugular hepatic biopsies during an 9-year period. Since 1998, our transjugular hepatic biopsy technique has been modified so that the biopsy is now done under combined sonographic and fluoroscopic guidance. During the period of our study, the spectrum of patients eligible for the transjugular approach was greatly expanded. We assessed the safety and efficacy of transjugular hepatic biopsy in pediatric patients and the impact of the information provided by the procedure on patient treatment and rates of survival.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients
Between March 1992 and July 2001, 74 transjugular biopsies of the liver
were attempted in 64 consecutive patients (28 boys, 36 girls) whose ages
ranged from 1 month to 17 years 7 months (mean age, 8.3 years). Patients
weighed 3–80 kg (mean weight, 28.9 kg; median, 23.9 kg). Cores from 73
biopsies were submitted for histopathologic evaluation; one transjugular
hepatic biopsy was aborted because of technical difficulties during the
procedure.
All referrals for a transjugular hepatic biopsy came from our institution's Division of Gastroenterology and Nutrition, Liver Transplant service, Bone Marrow Transplant service, and Division of Hematology/Oncology. Forty-three biopsies were performed in 40 gastroenterology patients, six biopsies in four orthotopic liver transplantation patients, 15 biopsies in 13 bone marrow transplantation patients, and 10 biopsies in 10 hematology or oncology patients. All patients had clinical and biochemical evidence of hepatocellular dysfunction. In this group of patients, the mean aspartate aminotransferase value was 994 U/L; alanine aminotransferase, 1041 U/L; and conjugated bilirubin, 97 µmol/L.
The major indications for transjugular hepatic biopsy were coagulopathy, thrombocytopenia, and ascites (Table 1). Elevated prothrombin time (> 13.5 sec) or increased international normalized ratio (> 1.3) were found in 46 of 74 cases biopsied (62%). Despite platelet transfusions, 29 cases (39%) had thrombocytopenia (platelets < 80 x 109/L). Of 74 biopsies, in 61 instances (82%), the patient had either coagulopathy or thrombocytopenia, or both. Ascites was noted in 16 cases (22%), four of whom had no coagulopathy or thrombocytopenia. Three patients with no coagulopathy or ascites had severe thrombocytopenia (mean platelets 14 x 109/L) that was corrected shortly before the transjugular hepatic biopsy; they were thought to be at risk of platelet consumption and unable to maintain an acceptable platelet level for a percutaneous biopsy. Six patients had transjugular hepatic biopsies performed because of suspicion of hepatic vein disease and, therefore, concurrent hepatic venography was required. In all patients, attempts were made to correct or at least optimize the coagulation profile and platelet level. Subsequently, patients who were deemed by the referring physicians and interventional radiologist to have minimal bleeding risk underwent percutaneous biopsy and were not included in this study.
|
Of the 40 gastroenterology patients, 14 presented with fulminant hepatic failure, seven with acute or subacute liver failure, and 11 with chronic hepatitis. The provisional diagnosis of three other gastroenterology patients was hepatic vein disease. The provisional diagnoses before biopsy in the remaining five patients included macrophage activation syndrome, total parenteral nutrition cholestasis, metastatic neuroblastoma, undiagnosed dysmorphic syndrome, and hepatomegaly. Four orthotopic liver transplantation patients (two with full and two with reduced-size liver grafts) were referred for transjugular hepatic biopsy to assess for acute graft rejection. Thirteen bone marrow transplantation patients were referred to assess for graft-versus-host disease. The 10 patients referred by the Hematology–Oncology division underwent transjugular hepatic biopsies to diagnose liver disease complicating their primary hematologic disease.
Technique
Before the procedure, a complete blood count and coagulation profile were
obtained in all patients, and routine antibiotics (cefazolin [Ancef; Eli
Lilly, Toronto, Ontario, Canada] 40 mg/kg to a maximum of 1 g) were given. The
procedure was always performed under general anesthesia, and after December
1998 (34 cases), muscle relaxants were always administered to control
respiration and motion. After the patient was on the table, sonography was
performed to visualize the internal jugular veins, to evaluate the liver size
and the presence of hepatic veins, and to determine the gallbladder position
and any associated masses or fluid collections. After the patient was placed
in Trendelenburg's position, the neck, chest, and upper abdomen were prepared
with Germi-Stat 0.5% solution (Germiphene, Brantford, Ontario, Canada).
Sonographic guidance was always used to access the internal jugular vein. Access was obtained with a 19-gauge BSDN needle (Cook, Bloomington, IN) using a single wall puncture method followed by introduction of a 0.035-inch Glide Wire (Terumo, Somerset, NJ) through the needle. In smaller infants, access was obtained with a 22-gauge BSDN needle, then a 0.018-inch Glide Wire was advanced, followed by a 4-French MicropuntureR Desilets Hoffman introducer set (Cook), which allowed the placement of the 0.035-inch Glide Wire. For the first 11 biopsies, we used a Colapinto needle (Cook); in all remaining patients, we used the Labs 100 set (Cook), which includes a Quick-Core needle. The wire and 5-French multipurpose catheter were directed under fluoroscopic guidance into the inferior vena cava or one of the hepatic veins. Under fluoroscopy, the 9-French polyurethane sheath was advanced into the jugular vein over the catheter and wire, followed by the 14-gauge metal stiffening cannula. Sonographic guidance of the multipurpose catheter tip was used when fluoroscopy was not successful in orienting the catheter into one of the hepatic veins. In all patients, hepatic pressures and venograms were obtained with the catheter wedged and free positions, and corrected sinusoidal pressures were calculated.
The position of the stiffening cannula, the sheath tip, and the biopsy trajectory were then determined with sonography in the last 37 biopsies in our series (Fig. 2A). The position and direction of the needle were adjusted to avoid major vessels, the gallbladder, or the capsule of the liver (Fig. 2B). The needle was introduced through the stiffening cannula and advanced to the end of the sheath; the needle-tip position was then confirmed with fluoroscopy and sonography. The biopsy was performed under sonographic guidance in the direction deemed safest to avoid the liver capsule, the gallbladder, or major vessels; a fluoroscopic image was obtained for verification (Fig. 3). After the biopsy, a venogram was acquired through the sheath in the hepatic vein to identify intrahepatic or subcapsular collections or intraperitoneal leaks. Under fluoroscopic guidance, the stiffening cannula was removed first, followed by the sheath. To reduce the chance of hematoma formation, pressure on the neck site was maintained while the patient was placed in the reverse Trendelenburg's position.
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Information obtained from the biopsies in 15 of the 40 gastroenterology patients resulted in the patients' timely listing for orthotopic liver transplantation. The 10 patients presenting with fulminant hepatic failure had greater hepatocellular dysfunction (mean values: international normalized ratio, 4.1; aspartate aminotransferase, 1892 U/L; alanine aminotransferase, 2048 U/L; and conjugated bilirubin, 189 µmol/L) than all other patients (mean values: international normalized ratio, 2.0; aspartate aminotransferase, 896 U/L; alanine aminotransferase, 843 U/L; and conjugated bilirubin, 98 µmol/L). Subsequent histopathologic studies showed significant hepatic necrosis in all these patients. The remaining patients who went on to have orthotopic liver transplantation included three patients with acute or subacute hepatitis, one patient with Budd-Chiari syndrome, and one patient whose biopsy ruled out the provisional diagnosis of metastatic neuroblastoma and instead strongly suggested a metabolic storage disease (tyrosinemia).
Among the other patients, information gained from biopsies helped to direct medical management. In the remaining gastroenterology patients, one biopsy was diagnostic for Wilson's disease, and biopsies in the other patients provided additional information to establish diagnoses of autoimmune hepatitis, Wilson's disease with giant cell hepatitis, drug toxicity, autoimmune and primary sclerosing cholangitis, viral hepatitis, noncirrhotic portal hypertension, primary sclerosing cholangitis and viral hepatitis, Shinzel-Gideon syndrome, macrophage activation syndrome, and total parenteral nutrition cholestasis.
Transjugular hepatic biopsy confirmed or excluded acute rejection in all patients referred by the Liver Transplant service. None of the orthotopic liver transplantation patients showed evidence of hepatitis or posttransplantation lymphoproliferative disorder. In patients who underwent bone marrow transplantation, the biopsy provided histopathologic evidence allowing assessment for graft-versus-host disease in all except one patient. In this group, no patient had evidence of hepatitis or malignancy.
In the hematology–oncology group, the biopsy provided unsuspected and diagnostic evidence of acute myelogenous leukemia in one patient, and in others, the biopsy provided additive diagnostic value in suggesting adenoviral hepatitis, erythrophagocytic lymphohistiocytosis, drug-induced toxicity, and resolving viral infection or inherited metabolic disease. In one patient, the biopsy ruled out graft-versus-host disease and instead suggested sepsis. In a patient with sickle-cell crisis, the biopsy helped to narrow the differential diagnosis of hepatic failure.
The transjugular hepatic biopsy was aborted in one patient in our series who did not receive a muscle relaxant. Her respiratory motion, despite general anesthesia, prevented safe cannulation of the right hepatic vein, leading to abortion of the transjugular approach. Instead, a sonographically guided percutaneous biopsy was done at the same time. The patient developed a small intraperitoneal bleed, which resolved spontaneously. All 34 of the subsequent transjugular hepatic biopsies were performed under general anesthesia with muscle relaxants and controlled ventilation, and no related complications were encountered.
Before sonography was used to guide the needle trajectory, the complication rate was 11% (4/37). Complications included a small hematoma of the liver parenchyma without extracapsular extravasation (n = 1), a small subcapsular hematoma (n = 1), a subclavian artery puncture (n = 1), and presence of contrast material in the retroperitoneal space (n = 1).
One death occurred during the transjugular hepatic biopsy procedure. A three-and-a-halfyear-old girl who had acute lymphocytic leukemia and was clinically unstable underwent a transjugular hepatic biopsy for diagnostic evaluation of fulminant hepatic failure. When the biopsy was done and the guidewire had been withdrawn, the patient developed ventricular tachycardia, progressed to cardiac arrest, and was pronounced dead after 20 min of cardiopulmonary resuscitation. Sonography was immediately performed, and it revealed small amounts of intraperitoneal and pericardial fluid and a small hematoma at the neck site. At autopsy, no blood was found in the peritoneal cavity. Other relevant autopsy findings included a perforated posterior wall of the right internal jugular vein, which was deemed to be inconsequential in terms of blood loss, and a focal contraction band necrosis in the ventricles, which might have been the origin of the arrhythmia. Because the arrhythmia was ventricular and not atrial in origin, the guidewire was not thought to have been the cause. Biopsy of the liver at autopsy showed liver necrosis of more than 95%.
After the introduction of sonography for needle trajectory guidance, only two complications (5%) were recorded in 37 biopsies: a subcapsular hematoma (n = 1) and a parenchymal hematoma without extracapsular extravasation (n = 1).
In this series of 74 biopsies, the most common minor postprocedural complication was oozing from the entry site in the neck (n = 14, 19%), followed by fever of short (< 24 hr) duration (n = 6, 8%), and a small neck hematoma (n = 1, 1.3%). The oozing was always easily controlled with a pressure dressing. In our study, no major complications were noted in the smaller patients or in those who underwent liver transplantation.
Eleven patients died within a month of their transjugular hepatic biopsies as a result of complications of their primary disease, including one who died 2 days and one, 3 days after biopsy. None of the deaths were attributable to the procedure. Seven of the 64 patients underwent multiple transjugular hepatic biopsies on separate occasions: five patients had two biopsies, one had three, and one had four. In all seven patients, the interval between biopsies was at least 1 month. The reasons for repeated biopsies included changes in the disease course and difficulty in establishing a diagnosis. No patient underwent repeated transjugular hepatic biopsy because of an inadequate specimen from the initial biopsy.
The right internal jugular vein was used as the access site in 72 of the biopsies performed. In two patients, the left internal jugular vein was used to avoid a right-sided central venous line (Fig. 4). Although using the left internal jugular vein as the access site is technically more difficult, no problems were encountered in these patients. Increased portal venous pressures were found in 21 patients, and in four patients, hepatofugal flow into the portal venous system was detected. In one patient, high portal pressure in the presence of normal liver histology helped to rule out hepatic cirrhosis and established the diagnosis of noncirrhotic portal hypertension.
|
In our study, the mean exposure for fluoroscopy was 10 min, and the mean duration of the procedure was 1 hr 30 min. The mean equipment cost per transjugular hepatic biopsy, based on depreciation and procedure duration, was $266 (Canadian). The mean material cost was $626 (Canadian), which included the routine materials (listed in the "Technique" section of this article). Labor costs were not calculated.
Additional procedures were performed in 27 patients. Central venous line insertion was the most frequent associated procedure (n = 18), followed by inferior vena cava cavography (n = 2), and nasojejunal tube placement (n = 2). Bone marrow aspiration, renal biopsy, pleural drainage, biliary stent change, central venous line removal, intracranial pressure monitor insertion, and peripherally inserted central catheter insertion and removal were each done once.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2
years; weight, 14 kg), Bergey et al.
[10] (22 children: age,
10 months; weight, 4 kg), and Hoffer
[9] (29 children: age, 7
months; weight, 7 kg) all concluded that transjugular hepatic biopsy is
safe and adds valuable diagnostic information to the management of pediatric
liver disease in patients in whom percutaneous biopsy is contraindicated. No
established standard or protocol suggests at what severity of coagulopathy or
at what platelet count the transjugular approach is indicated; therefore, the
referring physicians and interventional radiologists at our center interpreted
these findings in light of the patient's overall condition. In the last 37 of 74 biopsies in our series, sonography was used routinely to visualize the actual biopsy. This novel modification allowed better visualization of the exact needle position and the projected path of the biopsy before firing. In many cases, the needle was repositioned because the projected trajectory was unsafe. Before sonographic guidance, the right liver lobe was targeted because its bigger size makes liver capsule perforation less likely. With sonographic guidance, however, any hepatic vein (and hence, lobe) can be accessed, and the risk of capsule perforation is reduced. Sonographic guidance increased operator confidence, especially in smaller patients and in patients with variable anastomoses who had undergone orthotopic liver transplantation.
We found a lower complication rate after combined fluoroscopic and sonographic guidance began to be used for the biopsy. The accuracy of this finding may be decreased by the retrospective nature of the study (no randomization) and can also be attributed in part to greater experience of the radiologists by the time sonography was routinely used. No complications occurred in the neonatal or orthotopic liver transplantation patients, but we were unable to compare complication rates before and after the addition of sonographic guidance with statistical validity. Sonography was helpful in patients except those in whom overlying ribs or gas limited its use.
Sonographic guidance helped to expand the usefulness of transjugular hepatic biopsy into diagnosis and treatment of neonatal patients who were not safe candidates for percutaneous hepatic biopsy. The youngest and smallest patient (4 weeks old, 3 kg) in our study underwent a transjugular hepatic biopsy to diagnose the cause of his fulminant hepatic failure only after extensive case discussions. He could not have been safely biopsied without sonographic guidance, and the histopathologic diagnosis of hemochromatosis led to a timely orthotopic liver transplant. In an infant (6 weeks old, 5.9 kg) who had fulminant hepatic failure and was considered a candidate for orthotopic liver transplant, the transjugular hepatic biopsy led to the diagnosis of hemophagocytic lymphohistiocytosis and its proper management.
Treatment for pediatric liver disease has undergone important advances, but proper management still often relies on results from histopathologic studies [6, 13]. In all our patients, transjugular hepatic biopsy had a notable impact in providing the diagnosis and guiding the subsequent management. In the gastroenterology group, patients with fulminant hepatic failure collectively had the worst hepatocellular dysfunction, the extent of which was confirmed by the hepatic biopsy. This evaluation resulted in their timely listing for a liver transplant. This outcome is in keeping with the conclusions that transjugular hepatic biopsy is a safe and effective method in early assessment of prognosis in adult fulminant hepatic failure patients and that it can optimize the timing of liver transplantation in these patients [14, 15]. In the remaining gastroenterology patients whose hepatic failure was less severe, the transjugular hepatic biopsy was helpful in establishing a diagnosis and planning medical management.
Our study confirms that transjugular hepatic biopsy can provide rapid evidence for or against the presence of acute graft rejection and thereby potentially save the grafts in children with orthotopic liver transplants—as others have found to be true in the adult population [15, 16]. Similarly, our study confirms that in patients who have undergone bone marrow transplantation, transjugular hepatic biopsy can safely determine whether graft-versus-host disease is present and suggest alternative causes of hepatic failure. This finding is in keeping with the conclusions of Hoffer [9] regarding the safety, efficacy, and usefulness of transjugular hepatic biopsy in pediatric bone marrow transplantation patients. Our results also agree with those of Carreras et al. [17] in adult patients who have undergone bone marrow transplantation.
Our technique had two modifications: the routine administration of a muscle relaxant along with the general anesthesia and a change from using the Colapinto needle to use of the Quick-Core needle. After the transjugular hepatic biopsy had to be aborted in a patient who had excessive abdominal respiratory motion despite general anesthesia, all such biopsies were performed with muscle relaxants and controlled ventilation. General anesthesia and muscle relaxants were not routinely used in any of the other three pediatric studies [9, 10, 11]. We changed from the Colapinto to the Quick-Core needle because the Quick-Core is automated, easier to control, yields better cores, and decreases the risk of complications [9, 18]. The Colapinto needle was used by Furuya et al. [11] and the Quick-Core, by both Bergey et al. [10] and Hoffer [9]. Bergey et al. used metal sheaths with a greater angle to facilitate the procedure in smaller children, but in our experience such modification was not necessary.
In our study, as in that of Furuya et al. [11], no relationship was observed between the number of passes and the complications that occurred. Because we never performed more than four passes, we were unable to verify that the chance of capsule perforation increases with four or more passes, as was suggested by Bergey et al. [10]. It is our belief that using sonography to determine the needle trajectory before biopsy increases safety.
To the best of our knowledge, we are the only pediatric investigators to report procedure duration, and we found it to be longer for our pediatric patients (mean, 1.5 hr) than has been reported in studies of adults (30–60 min) [6, 7]. This difference might be explained in part by our use of general anesthesia, which is not used in adults undergoing the same procedure. An additional explanation may be that the smaller liver size in children requires more time for positioning the needle and planning its trajectory before the biopsy can be performed. The use of sonography, pressure measurements, and wedged venography also lengthened our procedure time. We found no reports of studies in adults or pediatric patients that listed the total fluoroscopic exposure time, and thus we are unable to compare our values.
We compared our transjugular hepatic biopsy complication rate to that in other studies. In adults, the mortality rate has been reported between 0.1% and 0.5% [7], and in our study the mortality rate was 1.35%, although it is thought that the single death was not directly caused by the biopsy. Nonetheless, if we include all the pediatric patients in the three studies cited [9, 10, 11] with those of our study, a crude mortality rate of 0.64% results, which is comparable to the adult mortality rate. Furuya et al. [11] reported a 23% complication rate (five small subcapsular contrast extravasations, one small intrahepatic hematoma, and one inferior vena cava perforation), Bergey et al. [10] reported two complications in 24 biopsies (one capsular perforation and one intraperitoneal hemorrhage), and Hoffer [9] reported five complications in 29 biopsies (two extensive contrast agent extravasations, two abdominal pains, one transient sinus bradycardia). Of our complications, two were documented by the presence of a small amount of retroperitoneal contrast material, and we hypothesize these are due to subcapsular or intrahepatic extravasation and tracking into the retroperitoneum along tissue planes. Overall, we conclude that our complication rate compares favorably to the rates in the other studies. We found no relationship between the complication rate and the weight or the age of the patient.
We showed that the transjugular approach offers other advantages over percutaneous biopsy in addition to its being safer in patients with increased bleeding risk or ascites. In patients with suspected hepatic vein disease, hepatic venography and pressure measurements provided physiologic and anatomic information, including the direction of portal venous flow. Performing additional procedures (n = 31) while the patient was under general anesthesia was of substantial benefit to the patient in terms of anesthetic risks, morbidity, and time, and doing so reduced the health care and hospital costs.
We think that a large pediatric quaternary care center such as ours, which provides liver and bone marrow transplantation services in addition to providing care to other patients with severe hepatic disease, can provide better patient care if it is able to offer transjugular hepatic biopsy to these patients. Furthermore, we believe that transjugular hepatic biopsy should be offered only at such centers because a minimum volume of patients is needed to perform it safely and efficiently.
The major limitation of our study is its retrospective (nonrandomized) nature because factors such as patient weight and age and operator experience could not be taken into account. However, the feasibility of a prospective study is questionable.
We believe that sonographic guidance during the biopsy itself has expanded
the number of candidates for transjugular hepatic biopsy, including neonates
(weight, 3 kg) and orthotopic liver transplantation patients. At our
institution, transjugular hepatic biopsy has become an integral diagnostic
tool in the treatment of patients with severe hepatic dysfunction, allowing
patients needing a liver transplant to be to be promptly listed and enabling
physicians to make correct medical management decisions. Finally, transjugular
hepatic biopsy is generally a safe procedure. Nevertheless, the risk of
serious complications cannot be ignored, especially in clinically unstable
patients.
Acknowledgments
We thank the Editorial Services staff of The Hospital for Sick Children,
Toronto, Ontario, Canada, for assistance in the preparation of this
article.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. M. Erturk, K. J. Mortele, C. A. Binkert, J. N. Glickman, M.-R. Oliva, P. R. Ros, and S. G. Silverman CT features of hepatic venoocclusive disease and hepatic graft-versus-host disease in patients after hematopoietic stem cell transplantation. Am. J. Roentgenol., June 1, 2006; 186(6): 1497 - 1501. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
