Postoperative complications in children who underwent liver transplantation (Ltx) are often difficult to specify and characterize, especially in the early postoperative period. Clinical and laboratory findings may not be specific enough for the accurate identification of posttransplant complications. Elevated liver enzymes may indicate acute or chronic graft rejection, toxic liver cell damage resulting from immunosuppressive drugs, hypoperfusion of the graft, or infections. Therefore, histological analysis of the transplant may be required to determine the appropriate treatment. Liver biopsy can be performed percutaneously, even in small children, as has been shown in previous studies (1). In stable patients, ultrasound-guided biopsy should be performed using, for example, a spring-loaded biopsy device or Menghini needles. Liver biopsy–related complications may include bleeding, infection, pneumothorax, biliary fistula, pain, or sedation-related complications (2). The overall incidence of complications in children, which is higher than that observed in adult liver transplant recipients, varies from 2.8% to 9.2% (1–3). However, no data are available on the diagnostic value of histological findings from liver biopsy in confirming or ruling out the suspected posttransplant complications and, hence, in selecting the appropriate therapy.
The aim of the present study was to evaluate the incidence of liver biopsy–related complications in children who underwent Ltx and, moreover, to assess the value of this procedure in accurately detecting posttransplant complications.
PATIENTS AND METHODS
We analyzed the data of 120 consecutive liver biopsies from 67 patients during a 4-year period (2005–2008). In the majority of cases, elevated liver enzymes (n = 100) were the reason for biopsy, but a few patients showed isolated elevated bilirubin levels (n = 2) or splenomegaly (n = 6). We correlated the results with our clinical and laboratory findings. Inclusion criteria were normal plasma coagulation and no contraindication for analgesia and/or sedation. All of the parents had given written informed consent. This was a retrospective, noncontrolled study. All of the data were extracted from the patients' charts.
Technique of Ultrasound-guided Liver Biopsy and Monitoring
In each case, the optimal biopsy site was determined by an experienced pediatric radiologist using a 2- to 5-, 4- to 7-, or a 5- to 12-MHz transducer and a commercial HDI 5000 ultrasound machine (Philips Corp, Netherlands). The children primarily received midazolam (0.1 mg/kg body weight) and ketamine S (1–2 mg/kg body weight), but a few patients received other drugs such as morphine or disoprivan. In all of the children, substernal liver biopsy was performed by aspiration technique using a Menghini needle (Hepafix, B. Braun Melsungen AG, Melsungen, Germany) in a free-hand technique (4).
Oxygen saturation, heart rate, and blood pressure of the children were continuously monitored during the procedure and for up to 2 hours after the biopsy. All of the children received inpatient treatment and were clinically followed up for at least 24 hours on the ward.
Liver biopsy specimens were immediately put into formalin and analyzed by an experienced pathologist. A hematoxylin and eosin stain was performed in each case, but in cases of special interest, several other histopathological techniques were used.
There were 34 male and 33 female patients. Median age was 7.5 years (range 0.5–18 years). The diagnoses requiring Ltx were as follows: biliary atresia, n = 28; acute liver failure, n = 7; Alagille syndrome, n = 7; progressive familial intrahepatic cholestasis, n = 6; liver cirrhosis, n = 4; metabolic disease, n = 4; Crigler-Najjar syndrome, n = 3; primary sclerosing cholangitis, n = 2; autoimmune hepatitis, n = 2; hemangioendothelioma, n = 1; and others, n = 3.
The overall incidence of complications following liver biopsy was 5.0%. Details are shown in Table 1. Except for 2 cases of severe complications in our study population, all of the complications were successfully treated conservatively. One patient had a severe bile leak (without having bile duct dilatation) and, as a consequence, peritonitis, which required surgical intervention; another patient developed an intraabdominal abscess that had to be drained percutaneously. Intraabdominal bleeding occurred in 1 case following liver biopsy combined with splenoportography, but it was self-limiting.
Predictive Value of Liver Histology
Overall, in 91.7% of the cases (110/120 biopsies), the suspected complication was confirmed by histological analysis. Details are shown in Table 2. In 3.3% (4/120 biopsies), acute graft rejection was suspected, but histological analysis showed tissue damage caused by drug toxicity, and in 1 case, instead of acute graft rejection, a relapse of the primary disease that had required Ltx. In 5.0% of the biopsies (6/120 specimens), chronic graft dysfunction was suspected, but histological analysis showed tissue damage caused by drug toxicity, and in 2 cases even normal liver tissue.
There are few published studies addressing complications following liver biopsy in children who underwent Ltx. In clinical practice, determination of the underlying reason for elevated liver enzymes and increased bilirubin concentrations in the early postoperative period and in long-term follow-up historically has relied on histological diagnosis. Percutaneous liver biopsy is therefore held to be mandatory for accurate determination of the complications occurring after Ltx.
Our hypothesis was that laboratory and clinical findings mostly reveal the underlying pathology and that histological analysis merely confirms the suspected complications. Furthermore, we report the incidence of complications following ultrasound-guided liver biopsy in children after Ltx.
Acute graft rejection was suspected in the presence of significantly elevated liver enzymes (at least 3 times above the upper normal concentration), trough levels of calcineurin inhibitors in the target range, regular findings in hepatic ultrasound, and in the absence of clinical signs of infection.
Drug toxicity was suspected in the absence of clinical signs of infection, regular hepatic ultrasound, slightly elevated aspartate aminotransferase and alanine aminotransferase, or increased γ-glutamyltransferase and bilirubin. Furthermore, relatively high levels of calcineurin inhibitors and/or potentially hepatotoxic drugs such as azathioprine or mycophenolate mofetil have been used as predictors of toxic liver damage.
Chronic graft dysfunction was assumed in cases of pathological ultrasound findings such as reduced flow velocities in the portal vein or hepatic artery, or inhomogeneity of liver parenchyma. Moreover, slightly elevated liver enzymes and/or bilirubin indicate chronic graft dysfunction. In nearly 92% of the cases (110/120 biopsies), histological analysis only confirmed the suspected complication, but did not reveal any additional findings. However, 8.0% histological analyses did not confirm the suspected complication and therefore influenced the treatment of our patients.
In deciding how to handle illness after Ltx, liver biopsy is necessary for the inexperienced care team. We believe, however, on the basis of our work as set out above that it can be dispensed with more and more frequently as team experience grows. Based on our findings that in 23 of 25 instances of such illness the patient could have been treated on the basis of assessment without liver biopsy, our institution's policy now is to conduct liver biopsy only as a secondary measure, that is, only if patient response is unsatisfactory to therapy chosen without drawing on liver biopsy findings. To avoid complications, treatment errors seem to be more acceptable than complications owing to biopsy. However, we should always work hard to achieve no complications following liver biopsy. In clinical practice, every decision to act or not to act has its fixed risk and should be made on an individual basis.
The main limitation of our study is the question of the validity of liver biopsies. A single biopsy is not necessarily representative of ongoing processes in other parts of the transplanted liver. An additional bias may arise from the evaluation of the specimens by different pathologists. The present study was not designed to examine whether team experience had an effect; however, we do believe in this effect. Because we do consider performing liver biopsy as an outpatient procedure in selected children, it was important to assess the postbiopsy complications in our inpatient setting.
We observed a low incidence of complications (5.0%) and only 2 severe complications resulting from liver biopsy (severe peritonitis due to biliary leakage and intraabdominal abscess). Based on the evaluation of our data, it should also be noted that 2 complications were due to sedation (1.7%, hypotension and oxygen desaturation) and not to the biopsy procedure. The incidence of complications is comparable with those of other studies. Scheimann et al (2) found an overall incidence of complications of 6.83% following ultrasound-guided percutaneous liver biopsy in children after Ltx. The incidence of major complications was 2.4% (hemothorax or intraabdominal hemorrhage requiring abdominal exploration) (2). In another large study (5) conducted in France, a low incidence of complications was observed in 144 consecutive percutaneous liver biopsies in infants and children. The only complications reported were bleeding complications in 1.2% of all of the cases (5). In contrast, Amaral et al (1) reported an incidence of bleeding complications in 9.2%.
In our setting, bleeding complications occurred only in 1 patient who had not received any anticoagulation therapy, but it is important to note that all of the children in the first posttransplant year received low-dose aspirin (5 mg/kg body weight 3 times per week), which does not seem to be a relevant risk factor for bleeding complications after biopsy. However, subclinical bleeding, such as minor intrahepatic hemorrhages, cannot be excluded because we do not routinely perform an ultrasound examination after biopsy. We only monitor hemoglobin concentrations up to 24 hours following liver biopsy.
Our review of the literature did not show another case of severe biliary leakage following liver biopsy. Our patient had an unexpectedly severe loss of biliary fluid of >300 mL accumulating in the abdomen and, as a consequence, biliary peritonitis, requiring surgical intervention. Usually, injury to the bile duct during liver biopsy is self-limiting and should not result in such a severe complication.
Two main conflicting questions arise from our results: First, is a liver biopsy required in every case considering that the predictive value of this procedure is >90%? In cases of suspected acute graft rejection, for example, blind treatment with steroids may be preferable. The theoretical risk of blind steroid therapy should not be of clinical relevance. Liver biopsy could be postponed until failure of steroid therapy. The key question in the decision-making process should be whether findings from histological analysis are likely to influence patient management (6). Second, can a liver biopsy be safely performed as an outpatient procedure? A couple of reports clearly document that it appears to be a safe and cost-effective procedure even in children (3,7–9).
There is evidence that ultrasound-guided liver biopsy is superior to blind needle biopsy (10,11) and the only way to perform liver biopsy in children who underwent Ltx because of the individual anatomy of patients who underwent transplantation. However, serious complications as seen in our case (severe biliary leakage) may occur in spite of ultrasound-guided biopsy.
Finally, it may be possible to diagnose fibrosis or cirrhosis in liver transplants by using new noninvasive techniques such as the FibroScan (Echosens, Paris, France), which would help avoid liver biopsies in specific cases (12). However, evidence-based data for children have yet to be collected and published.
In summary, our retrospective single-center experience with ultrasound-guided liver biopsies in infants and children who underwent Ltx confirms the data of other studies showing that the incidence of complications is relatively low. However, our case of severe biliary peritonitis clearly indicates that serious complications may occur. We therefore conclude that liver biopsy can be avoided in routine cases, which could be managed based on laboratory, ultrasound, and clinical findings.
1. Amaral JG, Schwartz J, Chait P, et al
. Sonographically guided percutaneous liver biopsy
in infants: a retrospective review. Am J Radiol 2006; 187:W644–W649.
2. Scheimann AO, Barrios JM, Al-Tawil, et al. Percutaneous liver biopsy
: impact of ultrasonography and spring-loaded biopsy needles. J Pediatr Gastroenterol Nutr
3. Fox VL, Cohen MB, Whitington PF, et al
. Outpatient liver biopsy
: a medical position statement of the North American Society for Pediatric Gastroenterology and Nutrition. J Pediatr Gastroenterol Nutr 1996; 23:213–216.
4. Menghini G. One-second needle biopsy of the liver. Gastroenterology 1958; 35:190–199.
5. Lachaux A, Le Gall C, Chambon M, et al
of percutaneous liver biopsy
in infants and children
. Eur J Pediatr 1995; 154:621–623.
6. Griffiths A, Viiala CH, Olynyk JK. Liver biopsy
in the 21st century: where and why? MJA 2002; 176:52–53.
7. Gonzalez-Vallina R, Alonso EM, Rand E, et al
. Outpatient percutaneous liver biopsy
. J Pediatr Gastroenterol Nutr 1993; 17:370–375.
8. Pokorny CS, Waterland M. Short-stay, out-of-hospital, radiologically guided liver biopsy
. Med J Aust 2002; 176:67–69.
9. Rossi P, Sileri P, Gentileschi P, et al
. Percutaneous liver biopsy
using an ultrasound-guided subcostal route. Dig Dis Sci 2001; 46:128–132.
10. Al Knawy B, Shiffman M. Percutaneous liver biopsy
in clinical practice. Liver Int 2007; 27:1166–1173.
11. Nobili V, Comparcola D, Sartorelli MR, et al
. Blind and ultrasound-guided percutaneous liver biopsy
. Pediatr Radiol 2003; 33:772–775.
12. De Lédinghen V, Le Bail B, Rebouissoux L, et al
. Liver stiffness measurement in children
using FibroScan: feasibility study and comparison with Fibrotest, aspartate transaminase to platelets ratio index, and liver biopsy
. J Pediatr Gastroenterol Nutr 2007; 45:443–450.