| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Imaging, Image Guided Therapy Centre, The Hospital
for Sick Children, 555 University Ave., Toronto, ON M5G 1X8, Canada.
2 Department of Pathology and Laboratory Medicine, The Hospital for Sick
Children, Toronto, ON M5G 1X8, Canada.
Received August 30, 2005;
accepted after revision November 12, 2005.
Address correspondence to J. G. Amaral
(joao.amaral{at}sickkids.ca).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Retrospective analysis of 65 biopsies performed in 61 infants between January 1999 and December 2003 was conducted. Data collected included patient demographics; details of the biopsy procedure including indication, needle type and size, number of passes, and samples; pathology results; and procedure-related complications.
RESULTS. The 61 infants studied included 37 males and 24 females with a mean age of 119 days (age range, 7-348 days; median age, 83 days) and a mean weight of 4.5 kg (1.9-8.3 kg). A total of 65 biopsies were performed in these 61 infants. General anesthesia was used in 66.1% of procedures. An 18-gauge needle was used in 47 (72.3%) procedures. Coaxial technique was used in seven procedures, and five biopsy tracts were embolized. In 63 of 65 procedures, the mean number of passes was 1.8. In two procedures, using a coaxial technique, 11 and 12 passes were made. One biopsy was considered technically unsuccessful, and 64 of 65 (98.5%) of the biopsies provided adequate tissue for pathologic analysis. There were three (4.6%) major complications related to bleeding: one requiring a blood transfusion, one requiring surgery, and one arteriobiliary fistula requiring transarterial embolization. Three (4.6%) minor complications also occurred. There were no deaths.
CONCLUSION. Sonographically guided percutaneous liver biopsy in infants is a good and effective diagnostic tool. The complication rate, however, even when performed by an experienced physician, is not insignificant in this age group of patients.
Keywords: biopsy • interventional radiology • liver • pediatric imaging
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To reduce the risk of complications, image guidance has been advocated in recent years. Some studies have questioned the value of sonographically guided percutaneous liver biopsies in children [5]. However, it has been shown that sonographic guidance is effective in selecting the biopsy site and guiding the needle [8]. In addition, a lower frequency of complications such as pneumothorax, bowel transgression, and gallbladder perforation have been reported when sonographic guidance is used [2, 9, 10]. Economic analyses comparing the cost of sonography use during biopsy to the costs of treatment of biopsy-related complications have also been performed, suggesting that sonographic guidance is cost-saving [11, 12].
Despite several publications in the English-language literature related to sonographically guided liver biopsies in children [5, 8], none specifically addresses this procedure in infants. In 1987, Lichtman et al. [7] reported their experience on percutaneous needle biopsy in infants 1 year old or younger. However, the authors used a Menghini needle and a technique without sonographic guidance. The purpose of this study is to review the technical success and complication rate of sonographically guided percutaneous liver biopsies in infants less than 1 year old performed in a tertiary pediatric center using an automated biopsy needle and to compare the results with current literature.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Information was retrospectively collected from patients' charts, laboratory results, imaging studies, and pathology reports. Data gathered for analysis in this case series included demographics (age, sex, and weight), reason for the biopsy, referring team requesting the biopsy, type and size of needle used, number of passes performed, and number of cores obtained. Preoperative coagulation studies (hemoglobin; partial thromboplastin time [PTT]; platelet count; and prothrombin time, international normalized ratio [INR]) were also reviewed.
Biopsies were considered "technically successful" if liver tissue or target tissue (in focal lesions) was provided for analysis. Adequacy of the amount of tissue was assessed by reviewing the pathology reports. Maximum length of the longest biopsy sample was also recorded.
The incidence of complications related to the biopsy and the management of those complications were evaluated. Complications were divided into minor or major according to the guidelines for imaging-guided percutaneous biopsy from the Society of Interventional Radiology [13] (Appendix 1). Special attention was given to bleeding complications and their management because bleeding was the most important complication reported in previous studies. The findings were then compared with current literature.
Biopsy Technique
All biopsies were done in the Image Guided Therapy Centre by a pediatric
interventional radiologist using sonographic guidance and free-hand technique.
Informed consent was obtained from the parents, and the coagulation profile
was checked before the procedure. If required, coagulation and hematologic
abnormalities were corrected (parameters accepted: INR, 
1.3; PTT
corrected for age and platelets, > 50,000). The patient was then sedated or
anesthetized. A limited mapping sonogram was obtained to determine a safe
needle pathway, avoiding interposing bowel, gallbladder, or major
intraparenchymal hepatic vessels and ducts. For routine biopsies, the left
lobe of the liver was sampled through a subxiphoid approach using a sterile
technique. Modifications to the approach were made for targeted biopsies. The
skin and subcutaneous tissues were infiltrated with lidocaine 1%, 0.5 mL/kg
(Xylocaine, AstraZeneca), and a small incision (5 mm) with a number 11 scalpel
blade was made. The biopsy needle was advanced under sonographic guidance to
the edge of the liver (kept outside of the liver), fired during end
inspiration or end expiration, and removed. Samples were obtained and sent for
analysis.
The number of passes performed through the liver and the choice to use a coaxial technique was at the discretion of the interventional radiologist performing the biopsy. Compression was applied to the skin overlying the biopsy site to achieve satisfactory hemostasis. Four different types of needle were used including 18-gauge and 21-gauge Autovac Angiomed biopsy needles (Angiomed GmbH), 18-gauge and 20-gauge MAGNUM Biopty needles (Bard), 18-gauge Biopince needles (MDTECH), and 17-gauge ASAP Pinpoint guiding introducer needles (Medi-tech). An absorbable gelatin foam (Gelfoam, Pharmacia & Upjohn) slurry was the embolic agent used in all cases of tract embolization. Our practice is to cut the Gelfoam in small pieces, load it into a 10-mL syringe, fill a second syringe with 10 mL of contrast material and swish back and forth between both syringes through a stopcock, forming the slurry (Figs. 1A and 1B). All patients were kept in the hospital for observation at least 24 hours after the procedure. During this period, vital signs (pulse, blood pressure, temperature, and respiratory rate) were monitored.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thirty-seven (61%) of the infants were boys and 24 (39%) were girls. The
mean age was 119 days (age range, 7-348 days; median age, 83 days) and the
mean weight was 4.5 kg (1.9-8.3 kg). The majority of biopsies (43/65) were
performed with the patient under general anesthesia. Patients had a normal or
corrected coagulation profile (INR, 1.3; PTT corrected for age and
platelets, > 50,000) before the procedure. However, records of preoperative
laboratory values were unavailable from the charts in four patients for
retrospective review. The mean hemoglobin level was 106 g/L before the biopsy.
The lowest hemoglobin level at the time of biopsy was 69 g/L in a patient with
Budd-Chiari syndrome. The biopsy was performed in this patient without
complications using a coaxial technique and tract embolization.
The most common clinical indication for biopsy at this age, representing more than half of the cases (n = 39), was to differentiate among biliary atresia, neonatal hepatitis, and total parenteral nutrition (TPN) cholestasis (Table 1). Patients were referred for biopsy from a variety of services (Table 2).
|
|
Forty biopsies were performed using an 18-gauge Autovac Angiomed biopsy needle (Fig. 2). The mean number of passes was 1.8 (Table 3). Two patients with focal lesions (hepatoblastomas) had 11 and 12 biopsies, respectively, through a coaxial approach to provide enough tissue for special pathology studies. These patients were not included in the calculation for the mean number of passes. The biopsy core mean length was 1.8 cm (minimum, 0.5 cm; maximum, 4 cm). In seven patients biopsies were performed through a coaxial approach (17-gauge ASAP Pinpoint guiding introducer needle) and five of these needle tracts were embolized with Gelfoam made into slurry. None of the patients with an embolized tract had bleeding complications.
|
|
All biopsies, except one, provided enough liver or target tissue for analysis, giving a technical success rate of 98.5%. One biopsy was considered unsuccessful because, although liver tissue was obtained, the tissue was most likely tissue adjacent to the lesion, as shown on pathologic analysis. A diagnosis of hemangioendothelioma was subsequently established by imaging and required catheter embolization for heart failure. No further biopsies were performed in this patient. Only one patient had a biopsy repeated because of inconclusive pathology results even though there was sufficient tissue. In all the remaining patients enough tissue was provided and the pathologist was able to reach a conclusion.
Three patients (4.6%) had minor and three (4.6%) had major complications. In the minor complication group, one patient presented with oozing on the skin site of the biopsy that was treated with topical thrombostatin powder (Thrombostat, Pfizer [product discontinued]). The second patient presented with transient hypotension immediately after the biopsy and subsequently stabilized after administration of IV fluids. A limited abdominal sonography and an enhanced CT showed no evidence of free fluid or active bleeding. The third patient experienced mild tachycardia associated with a drop in the hemoglobin count and on sonography had free peritoneal fluid in the abdominal cavity. The patient was managed conservatively and received IV crystalloid fluids.
In the major complication group, one patient presented with marked hypotension and free echogenic fluid in the abdominal cavity 2 hours after biopsy (Fig. 3). This patient was unresponsive to medical support, including crystalloids and blood products, and required surgery with suturing of the liver to control hemorrhage. ARC syndrome (arthrogryposis, renal tubular acidosis, and cholestasis) was diagnosed in this patient; however, this was not suspected before the biopsy. Patients with ARC syndrome have a normal coagulation profile and platelet count but abnormally large platelets that do not function properly. Our patient stabilized after surgery with no sequelae related to the biopsy. The second patient had a significant drop in hemoglobin level (before biopsy, 104 g/L; after biopsy, 73 g/L) requiring blood transfusion and close observation. The third child presented with hemobilia, echogenic material in the gallbladder (Fig. 4A), and melena 24 hours after the liver biopsy. Despite 1 week of medical support and multiple blood transfusions, the melena continued. Angiography showed an arteriobiliary fistula that was embolized endovascularly with Gelfoam (Figs. 4B and 4C). The patient improved clinically after embolization and the melena resolved. No deaths occurred in our series.
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We noticed an increased incidence of bleeding in our series (9.2%, including 4.6% minor and 4.6% major) in comparison with the current English-language literature (2.8-5%) [1], and we speculate that this may be related to the age of our population. Scheimann et al. [5] published an overall complication rate of 6.8% and a major complication rate of 2.4% after 249 sonographically guided percutaneous liver biopsies performed in children of all ages. Cohen et al. [4] found a major complication rate of 4.5% in a retrospective review of unguided percutaneous liver biopsies in 483 children of all ages. These data, although not representing a definitive prevalence of complications after liver biopsy in the pediatric population, do suggest a slightly higher complication rate in children than in adults.
Because hemorrhage is the most common complication after guided liver biopsy, evaluating the patient's bleeding history, blood parameters, and coagulation profile is paramount. The patient's bleeding history may be the most important factor to elicit because some hematologic dyscrasias present with normal quantitative laboratory parameters, such as von Willebrand's disease. One example in our study was the patient with ARC syndrome who had significant bleeding despite a normal platelet count [14].
We think that the use of sonography is an important factor in reducing the inadvertent puncture of major vascular and biliary structures as well as bowel. In our series, all complications were bleeding-related episodes. No pneumothorax, hemothorax, bile leak, bowel perforation, peritonitis, sepsis, pneumoperitoneum, or pneumoscrotum were seen. This most likely is related to the ability of the interventionalist to directly visualize the trajectory of the needle inside the target organ or lesion and to avoid major structures such as the main bile ducts, bowel, gallbladder, and thorax. Lindor et al. [9] published results of a randomized trial comparing unguided versus sonographically guided percutaneous liver biopsy and showed a decreased rate of hospitalization, hypotension, and bleeding in the patients when sonography was used. Nobili et al. [15], in a retrospective review of percutaneous liver biopsies in children, also reported a significantly higher incidence of hemorrhagic complications and technical failure in unguided biopsies in comparison with sonographically guided biopsies. Furthermore, several authors in the adult literature reported major complication rates between 0.5-1.8% when sonographic guidance was used, which is less than the complication rate reported for unguided liver biopsies [9, 16, 17].
In our series, 98.5% of the samples provided sufficient liver tissue for analysis using slight variation in needle type and size (18- to 21-gauge). Similar adequacy was shown by Rossi et al. [2] for pathologic diagnosis in 99.3% of 142 sonographically guided percutaneous liver biopsies. Although sample volume and size may be adequate for pathologic analysis, occasionally a pathologic diagnosis still can be difficult. In a biopsy specimen, several of the liver diseases of infants (biliary atresia, TPN cholestasis, extrinsic bile duct compression) present with similar pathologic features to those of large biliary duct obstruction, requiring close clinical correlation to establish a final definitive diagnosis.
Some authors advocate the use of a coaxial technique in patients with a moderate risk of bleeding, in oncology patients, or in patients with focal lesions [3, 18]. We agree that this technique may reduce the risk of bleeding after liver biopsy and may increase the safety of the procedure. In our series, none of the five patients whose biopsy tract was embolized developed any hemorrhagic complication. However, no definite evidence-based data are yet available in the literature to support the coaxial technique as a universal practice in all patients. Furthermore, the coaxial technique requires the use of a larger outer needle, which potentially increases the risk of bleeding. It also requires the use of embolic agents, which is not free of risks (paradoxical embolization, abscess formation, allergic reaction to the embolic agent, pulmonary embolism). Future randomized trials and cost-effective analysis studies comparing liver biopsies done with and without a coaxial technique and tract embolization may clarify the use of this technique.
All patients in this case series were admitted for at least 24 hours of observation. This policy is in agreement with the position statement on outpatient percutaneous liver biopsy provided by the North American Society for Pediatric Gastroenterology and Nutrition (NASPGN) from 1996. In this review, NASPGN suggests that patients at early infancy are at significantly higher risk of a complication or poor outcome and are not considered candidates for outpatient percutaneous liver biopsy [1].
Regarding the financial cost related to the addition of sonography to the procedure, Younossi et al. [11] and Pasha et al. [12] showed that the cumulative cost for all patients is less than the potential cost for treatment of major complications if the added cost of the sonography is less than approximately $100 [11, 12]. Furthermore, sonography minimizes the chance that large liver vessels or large biliary ducts will be damaged by the needle and ensures that the biopsy sample is taken from the liver and not from any adjacent structure.
Some of the limitations of this study include that it was a retrospective review of a limited number of patients from a single institution. Although our series represents a large number of infants undergoing percutaneous liver biopsy, it does not provide enough information to statistically determine risk factors for complications during percutaneous liver biopsies for this age group. It also does not allow the authors to offer recommendations on which approach is best to perform liver biopsies for this age group. Another limitation is that all the biopsies were performed by highly skilled pediatric interventional radiologists, which may have biased the results. Finally, minor complications such as pain or clinically insignificant bleeds were definitely underestimated, partly due to this study's retrospective nature and partly due to difficulties in clinical assessment of pain in this age group.
In conclusion, in infants less than 1 year old, percutaneous liver biopsy using automated biopsy guns provides adequate samples for accurate diagnosis and is clinically effective. However, the procedure is not free of risk and complications may be more frequent than in older children or adults. Further, and probably multicenter, prospective studies are necessary to better assess the risk of biopsy for this age group.
| APPENDIX 1: Classification of Complications by Outcome Minor Complications
Major Complications
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
