| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Rm. 1502, Duke University Medical Center, Box 3808,
Durham, NC 27710.
2 Present address: Wake Radiology Consultants, 3614 Haworth Dr., Raleigh, NC
27609.
3 Department of Medicine, Duke University Medical Center, Box 3808, Durham, NC
27710.
Received June 7, 2002;
accepted after revision July 11, 2002.
Address correspondence to T. P. Smith.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A total of 371 patients underwent 410 attempted transjugular biopsies of the liver during an 80-month period. Data collected included the retrospective review of patients' computerized medical records, clinical charts, and nursing documents. Patient demographic data, indications for liver biopsy, laboratory findings of coagulation values, procedural data including number of needle passes performed, and histologic description of the specimens were noted. Indications varied and included traditional contraindications to the percutaneous approach such as coagulopathy (53%) and ascites (29%). In one patient, the hepatic veins could not be catheterized because of angulation with the inferior vena cava, and in one patient, biopsy was performed using the femoral route because of occlusion of the jugular vein. All patients were followed up for a minimum of 24 hr after the procedure to determine complications.
RESULTS. The mean number of needle passes per procedure was 3.4 (range, 0-18). Hepatic tissue was obtained in 409 procedures via the venous route (408 transjugular and one transfemoral), and a tissue diagnosis was achieved in 403 (98%). The six tissue samples were nondiagnostic because they were too small (n = 3) or too fragmented (n = 1) or because they did not contain hepatic tissue (n = 2). Ten complications (2.4%) occurred, including three intraperitoneal hemorrhages that resulted in one death.
CONCLUSION. Transjugular biopsy of the liver using an automatic core biopsy needle is safe and produces adequate tissue specimens in a high percentage of patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
Data collected included a retrospective review of patients' computerized medical records, clinical charts, and nursing documents. Laboratory values, the length of time that had elapsed between collection of samples from which laboratory values were derived and the performance of the procedure, the number of needle passes required, and histologic descriptions of the specimens were noted. In cases in which the histologic description included factors that could have affected the accuracy of the diagnosis (e.g., small sample size or sample fragmentation), the report and, if necessary, the histologic slides were reviewed with respect to the patient's clinical findings to determine adequacy of the biopsy in achieving a definitive histologic diagnosis. Patient follow-up after the biopsy took place during rounds the next morning by the interventional radiologist for hospitalized patients or over the telephone the day after the procedure for outpatients. In addition, we reviewed computerized medical records and clinical charts for additional follow-up information. All data were processed with no identifying links to patient-specific information, and therefore this study received an exemption from the institutional review board.
During the 80-month study interval from July 1, 1995 to February 28, 2002, 373 consecutive patients were brought to the interventional radiology suite for performance of transjugular biopsy of the liver. The attempted transjugular biopsy was unsuccessful in one patient because the angle of the hepatic veins in relation to the inferior vena cava precluded successful placement of the biopsy device. This patient underwent successful percutaneous biopsy with embolization of the biopsy tract (which has previously been described [16]) and was included in the data set as a transjugular failure. Another patient presented with an occluded jugular access and successfully underwent the procedure with a transfemoral venous approach and the same biopsy kit without attempted transjugular access. This patient was also included in the data set. Two patients presented with an occlusion of the right internal jugular vein that was visualized on sonography. In these patients, jugular access was not attempted, and their biopsies were successfully performed percutaneously with embolization of the biopsy tract. These two patients were not included in the data set.
Therefore, 371 consecutive patients underwent attempted transjugular biopsy of the liver—207 males and 164 females who ranged in age from 1 to 82 years (mean and median age, 48 years). Nine patients were younger than 18 years; two were older than 80 years. Twenty-four patients underwent more than one biopsy procedure as follows: 18 patients underwent two procedures each, two patients underwent three, a single patient underwent four, two patients underwent five, and one patient underwent seven. In one of these patients, the procedure was repeated because only renal tissue was found in the biopsy specimen. The transjugular biopsy was successfully repeated 2 weeks later. All other repeated biopsies were performed for evaluation and comparison of hepatic tissue on the basis of the patient's clinical needs. The time interval between the two procedures ranged from 1 week to 46 months, with a mean of 4.2 months. Therefore, 371 different patients underwent 410 attempted transjugular liver biopsies.
The indications for transjugular biopsy of the liver are given in Table 1. The most common indications were coagulopathy, ascites, or a combination of both in 334 (81%) of 410 procedures. Ascites was the sole indication for transjugular biopsy of the liver in 36 procedures. A combination of ascites and decreased coagulation factors was the indication in 81 procedures. Transjugular biopsy was also considered clinically indicated for patients who, in spite of having normal coagulation parameters, had risk factors for decreased platelet function, such as established renal failure (n = 18). Three patients had transjugular biopsy of the liver performed at the time of initial transjugular intrahepatic portosystemic shunt placement (n = 2) or during intrahepatic portosystemic shunt placement revision (n = 1). Three patients scheduled for central-line placement into the jugular vein had transjugular biopsy of the liver performed in the same sitting for the convenience of the patients. The reason for using transjugular biopsy of the liver rather than a percutaneous procedure could not be determined from records in nine patients.
|
Laboratory values obtained before biopsy procedures included coagulation factors, most notably platelet count and prothrombin time. Information on platelet count was available for 393 procedures (96%) and on prothrombin time for 395 (96%). Coagulopathy was defined by the referring clinical service as a platelet count of less than 150 x 109/L or a prothrombin time longer than 13 sec after correction with blood products or both. Of the patients in our study population, 252 patients presented with a platelet count of less than 150 x 109/L, and 312 patients had a prothrombin time longer than 13 sec. Sixty patients underwent transjugular biopsy of the liver with a platelet count of less than 50 x 109/L (mean and median, 37 x 109/L; range, 18-49 x 109/L). The ranges of prothrombin time for the patients with coagulopathy were 13.1-14.0 sec in 109 patients, 14.1-15.0 sec in 81 patients, and more than 15.1 sec in 122 patients.
The time elapsed between the determination of the coagulation parameters and the performance of the biopsy varied, depending on the particular patient's clinical situation and the indication for biopsy. This time ranged from less than 1 hr to 210 days (mean, 5.2 days). Platelet counts were determined within 48 hr of the procedure for 293 patients, 205 of whom had platelet counts of less than 150 x 109/L and 53 of whom had counts of less than 50 x 109/L. Prothrombin times were also measured within 48 hr of the procedure in 312 patients, 294 (94%) of whom had coagulopathy. The distribution was 109 patients with prothrombin times ranging from 13.1 to 14.0 sec, 77 patients with prothrombin times ranging from 14.1 to 15.0 sec, and 108 patients with prothrombin times longer than 15.1 sec. Patients whose platelet counts were initially less than 50 x 109/L were given platelet transfusions and those with prothrombin times longer than 15.1 sec were given fresh frozen plasma in an attempt to correct the coagulopathy as much as possible. Although the values presented here represent the corrected values, 13 patients were receiving blood products as the biopsy procedure began; therefore, the precise level of coagulopathy in all patients is unknown.
The 410 procedures were performed by 10 interventional radiologists (ranging from one procedure to 122 procedures per interventionist). Five of the interventional radiologists performed 390 of the procedures (95%). The transjugular biopsy needle was successfully placed in the hepatic vein and needle passes were successfully made in 370 (99.7%) of 371 patients and in 409 (99.8%) of 410 procedures (including the transfemoral access as a successful placement).
The number of needle passes required was left to the discretion of the interventional radiologist based on observation of the gross hepatic tissue specimen. The number of needle passes required ranged from 0 (percutaneous procedure) to 18, with a mean of 3.4 and a median of 3.0 passes. Distribution of needle passes was as follows: no passes in one patient, one pass in four patients, two in 74 patients, three in 188, four in 83, five in 30, six in 10, seven in one, eight and nine passes in four patients each, 10 passes in one patient, and 18 needle passes in one patient. In nine patients, the number of passes required was unknown. In five of these patients, no reference was made to the number of needle passes in the procedure report; in four patients, the number of needle passes was recorded as "multiple." Excluding these nine patients, a total of 1361 passes were made in 401 patients, resulting in an average of 3.4 passes per patient.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1-antitrypsin disease,
sarcoidosis, and schistosomiasis. The most common diagnosis was acute or
chronic hepatitis either as a single entity (102 biopsies) or coupled with
cirrhosis (20 biopsies). Cirrhosis was the dominant diagnosis in 90 biopsies.
Steatosis with or without an inflammatory component (steatohepatitis) was the
diagnosis in 26, cholestasis in 18, graft-versus-host disease in 14 patients
having undergone bone marrow transplantation and 23 liver transplant patients
with allograft rejection. Findings of 11 biopsies were normal. Ten complications occurred, resulting in an overall complication rate of 2.4%. Six capsular perforations were documented either radiographically or pathologically but were without clinical sequelae. One complication involved a capsular perforation that was noted at the time of the procedure and was successfully embolized with an absorbable gelatin sponge (Gelfoam; Pharmacia & Upjohn, Kalamazoo, MI) using the existing introducer sheath for access. Five biopsies yielded tissue other than tissue from the liver and were indicative of capsular invasion: cartilage (n = 1), renal tissue only (n = 1), renal and hepatic tissue (n = 2), hepatic tissue and a portion of a large pancreatic pseudocyst (n = 1), although the latter three patients had enough hepatic tissue for a definitive diagnosis to be determined. All of these patients remained asymptomatic.
Three patients (0.2%) experienced intraperitoneal hemorrhage after biopsy that was found clinically and seen on CT. For these three patients, platelet counts were 108 x 109/L, 37 x 109/L, and 73 x 109/L, respectively. Their prothrombin times were 14.3, 14.1, and 15.1 sec, respectively. All three patients were treated with blood transfusions after the procedure. The first patient recovered without further intervention and was discharged from the hospital 2 days after the biopsy. The second patient also recovered after the blood transfusion but died of multisystem decompensation 35 days after the procedure. In the final patient, intraperitoneal hemorrhaging could not be adequately controlled with transfusion therapy, and the patient died 48 hr after the biopsy without further intervention, honoring the family's request to withhold invasive measures. Finally, one additional patient experienced respiratory arrest at the termination of the procedure, which was believed to have been a reaction to the drug used for conscious sedation. No evidence of an intraperitoneal hemorrhage was found, and this patient had an uneventful recovery with supportive care.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gilmore et al. [20] described a modified core biopsy needle in 1978 that achieved adequate biopsy specimens in 29 (97%) of 30 patients. In 1983, Bull et al. [21] presented their results with this needle (Table 2); in that study, 193 patients underwent biopsies that yielded histologic specimens for adequate diagnosis in 188 patients (97%). These researchers noted that multiple needle passes were easily accomplished once the catheter was in place but did not specifically report the number of needle passes required. Although the complication rate was relatively high and included one death, the diagnostic yield of 97% was much improved over the yield reported in previous studies for aspiration techniques. The popularity of this style of needle for percutaneous biopsy continued to increase, and in 1996 Little et al. [6] described their experience with the semiautomated Quick-Core biopsy needle (Cook), which produced adequate biopsy specimens in 42 (98%) of 43 patients.
|
Since the publication of these studies, other researchers have studied small groups of patients using this type of needle system; their findings are summarized in Table 2. These studies have shown the device to be efficient, safe, and highly successful in obtaining diagnostic tissue in a limited number of patients. A comparison of the core biopsy needle technique with the aspiration technique has also been performed, again in small series, and has shown that less fragmentation occurs with the core biopsy needle [13]. In addition, more definitive histopathologic diagnoses have been achieved [13, 14]. However, patient and procedure numbers are limited: the largest series [9] included slightly more than 100 procedures.
The purpose of our study was to determine the safety of transjugular biopsy of the liver and the quality of the biopsy samples obtained in a large clinical population. Safety assessment is based on the complication rate. Our overall complication rate of 2.4% is well within that expected even for percutaneous biopsy [22]. Half of the complications were liver capsular perforations without clinical sequelae. The rate of complication resulting in the clinical sequelae of intraperitoneal hemorrhage (three patients) or respiratory arrest (one patient) was therefore only 1%. However, all three of the patients with intraperitoneal hemorrhage had coagulopathy, with platelet counts of less than 150 x 109/L and prothrombin times longer than 13 sec. One patient had moderate ascites that was visualized on sonography 2 days before the biopsy. In the subset of patients with coagulopathy, these patients represent a complication rate of 1.3%, again within reasonable expectations.
The transjugular technique has been criticized as not capable of producing adequate tissue samples, typically because of the small size or fragmentation of the tissue sample. Difficulties with small sample sizes were not found in our study. Although 17 samples were noted at pathology to be small, only two of the samples were nondiagnostic or too small for a confident pathologic diagnosis. Another of the major complaints concerning the transjugular technique has been the fragmentation of the tissue. Fragmentation was especially typical with the aspiration techniques, but its cause has also been ascribed to the transjugular core biopsy kits [7].
In our series, although fragmentation occurred in 32 samples, all but one of these samples were diagnostic. Although we attribute the adequate tissue samples in large part to the needle design, other factors may be at work. More needle passes—resulting in more tissue—were made in the biopsies in our study than were made in other studies (Table 2). Also, most interventional radiologists today are more facile with transjugular techniques in the liver because of having experience placing transjugular intrahepatic portosystemic shunts. This increased facility creates difficulties when one compares earlier studies with more recent ones, regardless of the design of the needle. Finally, we used the 18-gauge needle system exclusively. This same device is available in a 19-gauge model, which is considered by some [6, 9] to be superior to the 18-gauge in producing tissue with less fragmentation. However, to our knowledge, no researchers have directly compared results from the two sizes of needles.
The main limitation of our study is its retrospective nature. We did not have an adequate control population with which to compare the results of our study population. Patient selection was at the preference of the clinical teams and often did not reflect the traditional contraindications to standard percutaneous biopsy. For example, transjugular biopsy of the liver was performed in a number of obese patients in whom percutaneous biopsy with cross-sectional imaging would have been an acceptable alternative. For those patients with traditional contraindications of coagulopathy and ascites, the exact character of these conditions is unknown. The level of coagulopathy is unknown because a number of patients were receiving blood products at the time of the biopsy, and no laboratory findings were obtained to assess a patient's precise coagulation status at the time of biopsy. Nor were controlled imaging studies undertaken to quantify the degree of ascites in the patients. In spite of these problems, the value of our report lies in the size of the patient population, which is nearly four times the number of patients in the next largest series using commercially available automated core biopsy devices. Our study also reflects a busy clinical practice, and all patients referred for biopsy were reported.
Another common method of preventing hemorrhage in patients with contraindications to traditional transhepatic biopsy of the liver is performing the percutaneous biopsy while sealing the biopsy tract [16, 23]. Sawyerr et al. [24] compared an aspiration transjugular technique with tract embolization using a gelatin sponge. They concluded that the percutaneous route was quicker and produced larger tissue samples. Although they reported two hemorrhages in the percutaneous procedure group (and none in the transjugular procedure group), these occurrences did not reach statistical significance. The smaller tissue sample reported in the transjugular group is somewhat expected because an automated device was not used. In our practice, the transjugular method has replaced percutaneous biopsy with tract embolization for hepatic biopsy in patients with suspected diffuse liver disease. However, the percutaneous route with tract embolization using cross-sectional imaging remains valuable because it permits more precise localization of specific lesions.
In conclusion, our results show that transjugular biopsy of the liver using an 18-gauge automated core biopsy needle is safe and produces adequate tissue specimens in an extremely high percentage of patients.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E Cholongitas, A Quaglia, D Samonakis, M Senzolo, C Triantos, D Patch, G Leandro, A P Dhillon, and A K Burroughs Transjugular liver biopsy: how good is it for accurate histological interpretation? Gut, December 1, 2006; 55(12): 1789 - 1794. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
