The purpose of this guideline is to provide guidance about liver biopsy at Great Ormond Street Hospital (GOSH).
The liver, which is located in the upper right quadrant of the abdomen, just behind the lower portion of the ribs (which protects it from injury), is the largest and most metabolically complex organ in the body.
The liver carries out hundreds of different functions designed to maintain a favourable internal environment in the body (metabolic homeostasis), and helps to protect against infection.
Position of the liver within the body
Anatomy of the liver
Liver biopsy is sometimes used in the investigation of suspected liver disease. The benefits and risks of a liver biopsy must be assessed carefully for each patient and take into account the results of other investigations.
Liver biopsy samples can be obtained by percutaneous core needle biopsy sampling, by a transjugular core needle approach, or by laparascopic or open surgical techniques.
Percutaneous liver biopsy is associated with small but definite risks to the patient, even in the most experienced hands, and it should therefore only be performed when the potential benefits of the test outweigh the risks (Rationale 1).
The potential benefit of knowing what the liver looks like under the microscope (or how it functions biochemically or whether it is the site of infection) is that this will lead to specific, effective treatment or at least define the likely disease outcome.
This benefit should be continually re-evaluated as alternative diagnostic tests (eg DNA analysis) become available and also as new treatment options become available, such as has occurred with the new antiviral therapies in viral hepatitis and in liver transplantation.
Indications for liver biopsy
Liver biopsy (in combination with the patient’s clinical history, physical examination and data from imaging and laboratory tests) is a powerful clinical tool for diagnosing, treating and monitoring liver disease.
At GOSH, all liver biopsies are performed by the Interventional Radiology (IR) department, usually under general anaesthetic.
It is assumed that all patients referred to IR for a liver biopsy will have had previous imaging of the liver and biliary tree. This imaging will be reviewed by IR and discussed with the referring team prior to accepting the biopsy request.
Indications for a liver biopsy:
Investigation of suspected diffuse liver disease, such as infective, autoimmune, cholestatic and congenital forms of hepatitis, Langerhan’s Cell Histiocytosis (LCH), metabolic liver disease such as Wilson’s disease and glycogen-storage disorders, and some types of neuroblastoma (diffuse forms of stage 4 and stage 4S) (Roebuck 2008).
Investigation of focal liver disease, such as mesenchymal hamartoma, teratoma, hepatoblastoma, rhabdoid tumour, sarcoma, non-Hodgkin lymphoma, hepatocellular carcinoma and hepatic adenoma.
Management of liver transplant.
Management of drug therapies that affect the liver parenchyma.
- a patient who is too unstable or critically unwell to undergo this procedure (Rationale 2)
- significant coagulopathy (Rationales 3 and 4)
- significant thrombocytopaenia (usually taken to be less than 80,000/mm3)
- significant ascites
Relative contraindications to liver biopsy include:
- Extra-hepatic biliary obstruction (Grant et al 1999)
- Acute cholangitis
A patient in whom any of these factors are present should be discussed with the IR team on a consultant to consultant basis, in order to decide whether the potential benefits of biopsy outweigh the risks.
Some of these factors may be only relative contraindications. In some cases, transjugular biopsy route may be appropriate, as the risks of bleeding are greatly reduced using this technique.
Clotting abnormalities must be discussed with a consultant haematologist, and corrected, as far as possible, prior to the procedure.
Patients on anticoagulant medication or non-steroidals should have their medication discussed with their clinical team and the haematology team, prior to a decision to proceed with biopsy.
- An ultrasound machine with a standard 13-6 MHz linear transducer. Occasionally a small parts 13-6 MHz linear probe is useful in infants.
- Co-axial core needle biopsy system. Semi-automated systems are available from various manufacturers (examples include Coaxial Temno: Allegiance Healthcare, McGaw Park IL; Meditech, Watertown, MA) (Rationale 5).
The appropriate medium for the biopsy should be obtained from the relevant laboratory prior to the procedure (Rationale 6).
The medium is usually a very small amount of sterile normal saline in a sterile specimen pot. When saline is used, the sample must be transported immediately to the laboratory so that it can be processed correctly, eg some fixed in formalin, some in glutaraldehyde, some frozen etc.
Some investigations require that the liver is snap frozen immediately and this will require solid carbon dioxide (card ice) or liquid nitrogen at the venue of the procedure.
Percutaneous core needle biopsy technique
Careful patient preparation prior to undertaking a liver biopsy is essential. Ideally, patient/parent education should begin at an outpatient visit, prior to the biopsy.
Explain the procedure in sufficient detail, with careful attention to anxiety and pain management issues. Explain the potential benefits and (at least to the parent/guardian) the risks involved.
The patient will usually be asked to complete a six-hour fast (Rationale 7). Some children (those at risk of a low blood sugar) will require an intravenous infusion during this time.
In most instances, the biopsy procedure is undertaken on an inpatient basis, following the recommendations outlined below:
On the day before the biopsy, the clinical team review recent laboratory evaluation of prothrombin time, activated partial thromboplastin time and complete blood count, including platelets (Rationale 8).
Normal coagulation parameters and platelets >80,000 (sampled within the last month) are acceptable if the child’s clinical condition is stable and they do not have cholestatic jaundice (Rationale 9 and 10).
Explain the procedure to the parents and child as appropriate. The consent should include adequate understanding of the risks and benefits of the procedure, complications, post-procedural care and home care (Rationale 11).
A pre-medication can be administered to the child prior to the patient attending theatre, to make the experience less frightening or stressful (Rationale 12). This decision will be made by the anaesthetist in charge of the case (Rationale 13).
An intravenous cannula may need to be inserted prior to the procedure.
Children with an underlying metabolic disorder or a history of hypoglycaemia should commence an intravenous infusion of glucose at the commencement of fasting (Rationale 14):
5 per cent glucose: 5-10mg/kg/min, 0.1-0.2ml/kg/min
10 per cent glucose: 5-10mg/kg/min, 0.05-0.1ml/kg/min
Record the child’s blood glucose 1-2 hourly depending upon the child’s underlying condition.
The blood glucose should be maintained at 5-8mmol/l.
Medical staff should provide clear, written instructions of action to be taken in the event of hypoglycaemia during fasting (Rationale 15).
Standard precautions and personal protective equipment (PPE) is standard practice within an intra-operative theatre environment (Rationale 16).
The patient should be placed in a supine position (Rationale 17).
Consider moving and handling risks when positioning and moving the child (Rationale 18).
- A suitable biopsy site is identified by the radiologist, using ultrasound (Rationale 19).
- Prepare the field with alcohol-based solution (povidone-iodine) and place sterile drapes over the patient (Rationale 20).
- Administer local anaesthetic with 0.25 per cent levobupivacaine 2.5mg/ml in both superficial and deep planes (Rationales 21, 22 and 23).
A small nick in the skin is made with a surgical blade, usually using a subcostal approach, to allow introduction of the biopsy needle.
The radiologist will then advance the co-axial biopsy needle) through the skin and into the liver, under direct ultrasound guidance. While this is done, the anaesthetist usually ensures that the patient is breath-holding, so that the passage of the needle through the liver capsule is accurate (Rationales 24 and 25) .
If the biopsy is non-targeted (for investigation of diffuse liver disease), the semi-automated inner needle of the co-axial system is fired twice, in slightly different areas of the liver, to ensure two good cores are obtained. This is done under real time ultrasound guidance. The number of cores may vary depending on the clinical situation.
In targeted biopsies (for focal liver lesions) the outer needle is usually advanced through normal liver parenchyma into the abnormal liver lesion, and then the inner needle is used to sample from the area of abnormality. Again, this is performed using real time ultrasound guidance. The number of cores obtained may be as many as 15, so that the pathologist has enough tissue to ensure a diagnosis can be made.
Once the biopsy samples have been taken, the inner needle of the system is removed and the track is filled with several plugs of a haemostatic substance such as gelatine foam sheets (Gelfoam®, Pharmacia and Upjohn, Kalamazoo, MI), as the outer needle is withdrawn (Rationale 26).
The biopsy sample is obtained and placed in the correct medium. The appropriate medium for the biopsy should be obtained from the relevant laboratory prior to the procedure. The medium is usually a very small amount of sterile normal saline in a sterile specimen container.
The patient is rolled onto the right side and an instruction is given for them to remain in this position for one hour to help prevent bleeding or bile leakage complications (Rationale 28).
The practitioner undertaking the procedure must ensure that the sample is clearly labelled with the patient’s details, sample type date and time, prior to sending it to the appropriate laboratory (Rationale 29).
A request form must accompany the sample and include accurate patient details and clear instructions for the analysis of the sample.
The operator performing the biopsy should record the procedure in the child’s health care record (Rationale 30) including:
how many passes were made
any medication administered
any apparent complications
specific post-operative care requirements
Vital signs are measured and recorded (Rationale 31):
- every 15 minutes for the first two hours
- every 30 minutes for two hours thereafter
- hourly thereafter until eight hours post biopsy
- at eight hours post-biopsy, the child should be reviewed by a senior member of the medical team to ensure the child is well enough to discontinue specialist nursing care where appropriate (see Discharge planning section)
Any alteration in the child early warning score (CEWS) such as a rise in pulse rate, fall in blood pressure or respiratory distress, should be notified immediately to the clinical team. They will assess and determine the possible need for blood transfusion, clotting factors, platelets or other products/interventions.
If significant bleeding is suspected they will need to inform the person who undertook the procedure and the consultant in overall charge of the child. A surgeon should be informed at this early stage. Urgent imaging, embolisation or surgery may be required.
The actual site of the biopsy must be observed for any signs of bleeding, at the same time as vital signs are recorded, for the first 12-hour period after the procedure (Rationale 32). An intravenous cannula must remain in situ for the 24-hour period post biopsy (Rationale 33).
It is important to remember that sick children deteriorate for reasons unrelated to the liver biopsy (Rationale 34).
For example, a child with mitochondrial disease who appears slow to recover from anaesthetic/regained consciousness may have undetected severe hypoglycaemia.
Assess the child’s need for analgesics postoperatively and administer as appropriate (Rationale 35).
The patient may eat and drink two hours post procedure, once awake and orientated (Rationales 36 and 37).
Patients should remain on bed rest for eight hours post procedure (Rationale 38). However, the child can be allowed brief toilet visits under supervision by a nurse or a responsible family member during this period, provided their vital signs are stable.
The position that the patient should be nursed in following liver biopsy is not significant. There is no consensus in the literature and no controlled trials have been carried out to assess the various possible positions. The child should be encouraged to adopt a position that is comfortable and reassuring for him/her (Grant et al 1999, Perraul et al 1978).
Ensure that the child has passed urine post anaesthetic (Rationales 39 and 40).
Most children will remain in hospital for 24 hours post biopsy. Children who are otherwise well, with no clotting abnormalities, may be discharged eight hours post procedure, if the biopsy was performed early on the morning theatre list, but this decision must be discussed and agreed with all clinical teams prior to biopsy.
The child must then be reviewed by a senior doctor on the clinical team prior to discharge. The patient should have a responsible person to stay with them on the first post-biopsy night at home and should be able to return to hospital in a timely manner should the need arise.
The liver biopsy fact sheet should be given to the family as part of the discharge process:
Complications of liver biopsy
Paediatric studies suggest an overall complication rate for percutaneous biopsy of 0-6.83 per cent (2.4-4.5 per cent major complications) (Cohen et al 1992, Lachaux et al 1995, Nobili et al 2003, Schiemann et al 2000, Gonzalez-Vallina et al 1993). Mortality rates are quoted at 0-0.6 per cent (Cohen et al 1992, Lachaux et al 1995, Lichtman et al 1987). The data is difficult to compare between series as authors vary in their definition of major and minor complications, and in the exclusion criteria they employ.
Complications include intraperitoneal haemorrhage, biliary peritonitis, haemobilia and injury to the duodenum, colon or lung. The risk of significant bleeding after an image-guided percutaneous liver biopsy, as measured by a decrease in haematocrit, is reported to be 1.2-2 per cent (Lachaux et al 1995, Rivera-Sanfeliz et al 2005).
McGill et al 1990 showed that the risk of significant bleeding is increased in malignant versus non-malignant liver disease. Paediatric studies have shown that the risk of complications is greater in patients with malignant disease, cirrhosis or recent bone marrow transplant (Cohen et al 1992, Lachaux et al 1995). It may be that these children should always be considered for a transjugular biopsy approach.
Tumour seeding along the biopsy track remains a significant concern but unlike in hepatocellular carcinoma in the adult population, is rare in the commoner paediatric lesions such as hepatoblastoma.
Smaller risks include lack of an adequate sample. With good operative technique, post-operative pain should be minimal, and post-operative infection rare.
The majority of complications occur in the first three hours after liver biopsy (Grant et al 1999).
It is important to remember that sick children may deteriorate for clinical reasons unrelated to the liver biopsy itself.
Advice for families post-discharge
- The parents should be supplied with the family factsheet liver biopsy (percutaneous), if they do not already have one:
- The parents should be advised that the histology results are usually available no earlier than two days after the procedure being performed (Rationale 41). The family must be made aware of how they will receive the results. For microbiology and metabolic analysis, these results will take longer. Results will usually be available at the next outpatient appointment for discussion with the child’s consultant.
- When children go home they can resume a degree of normal activity. There should be an interval of somewhere between two weeks and three months where the child may have restrictions (Rationale 42). Children may have to avoid contact sports and swimming for six to eight weeks. Advice from their clinician will be given as relevant. However, parents should be aware of siblings and other children indulging in rough and tumble play.
- If the dressing and wound closure strip, eg steri-strip®, is in place it may be removed after three days or soaked off in the bath (Louise Autio et al 2002).
- If the skin looks red and angry and/or the child develops pyrexia, parents should seek medical advice from their GP (Rationale 43).
- It has been shown that delayed haemorrhage can occur up to 15 days after percutaneous liver biopsy in patients who develop a post biopsy coagulopathy. The parents need to look out for signs of a high temperature, unexpected lethargy and generally being unwell/irritable.
- The occurrence of delayed haemorrhage is also documented after the reinstatement of warfarin therapy several days after percutaneous liver biopsy. The child’s consultant must give the family clear advice regarding re-commencing a child’s anti-coagulant drugs (see Contraindications section).
Rationale 1: In the majority of cases, percutaneous core needle sampling is the modality of choice, due to the relative technical ease of the procedure, and its high diagnostic accuracy (Cohen et al 1992; Lachaux et al 1995; Lichtman et al 1987; Nobili et al 2003; Rivera-Sanfeliz et al 2005; Hoffer 2000).
Rationale 2: There is an increased risk of mortality for an acutely sick child.
Rationale 3: There is no published safe lower limit for platelet count, though studies usually consider levels below 50-100,000/mm3 to be unsafe for a percutaneous approach (Lichtman et al 1987; Nobili et al 2003; Rivera-Sanfeliz et al 2005; Lindor et al 1996; Schiemann et al 2000; Hatfield et al 2008; Little et al 1996; Sharma et al 1982). Sharma et al (1982) have shown that the risk of bleeding increases if the platelet count is below 60,000/mm3. It makes sense that the risk of haemorrhage increases in the presence of abnormal clotting parameters (Lachaux et al 1995) but Cohen et al (1992) has noted that pre-biopsy coagulation parameters were not predictive of subsequent complications. Importantly, operators should be wary of temporary correction of significant coagulopathies, as the abnormality may recur in the hours or days following the procedure, leading to delayed bleeding. In such circumstances, there should be a low threshold for considering a transjugular approach.
Rationale 4: A small volume of ascites can probably be tolerated by most operators, particularly if the percutaneous biopsy track is to be plugged, though significant volumes of ascites should prompt consideration of a transjugular approach. Published studies vary in their definition of the volume of ascites, making interpretation of the data difficult. It may be that image-guided percutaneous procedures are less risky in cases of mild-moderate ascites than previously thought (Little et al 1996).
Rationale 5: Older series describe the use of aspiration needles, but more recent series have shown that automated devices are safe and accurate (Rivera-Sanfeliz et al 2005; Hoffer 2000). There is no strong evidence to recommend a particular biopsy needle size, although the literature suggests that an 18G core needle sample is adequate for a diagnostic sample in the majority of cases (Lichtman et al; Hoffer 2000).
Rationale 6: To ensure optimal condition of the biopsy arriving in the laboratory for analysis.
Rationale 7: An empty stomach may decrease the likelihood of post-procedure nausea and vomiting and reduces the risk of inadvertent peri-operative aspiration of stomach contents.
Rationale 8: To check the child’s clotting status and to minimise the risk of haemorrhage.
Rationale 9: If the patient is unwell, the results should be from a sample collected within 24 hours of the biopsy. In an emergency situation, the biopsy may proceed with fresh frozen plasma (FFP) cover to minimise the risk of haemorrhage, but this should be agreed by all parties prior to commencing the procedure.
Rationale 10: If there is abnormal coagulation or thrombocytopaenia that cannot be corrected, a transjugular biopsy may be more appropriate.
Rationale 11: To obtain informed and written consent (Department of Health 2009).
Rationale 12: A play specialist may be able to assist in preparing the child for the procedure.
Rationale 13: To facilitate the anaesthetic induction and to manage fluid balance. The decision to place a cannula prior to induction should be made by the anaesthetist in charge of the case.
Rationale 14: To prevent hypoglycaemia and minimise the risk of metabolic acidosis.
Rationale 15: To ensure the child’s safety.
Rationale 16: To safeguard the practitioner and patient.
Rationale 17: To allow ultrasound examination of the abdomen.
Rationale 18: To ensure staff safety.
Rationale 19: To ensure the safest approach to the liver parenchyma is used.
Rationale 20: To provide the optimum operative field.
Rationale 21: Local anaesthetic minimises post procedural pain.
Rationale 22: Levobupivacaine has a long duration of action and ensures patient comfort post operatively.
Rationale 23: Lidocaine 1 per cent buffered with sodium bicarbonate may be used for non-GA cases, as it is less painful, but most children have this procedure performed under general anaesthetic.
Rationale 24: This allows the time spent in the liver to be kept to a minimum.
Rationale 25: This technique means you can collect many cores with only one hole in the liver capsule.
Rationale 26: The gelatine plugs act to minimise the risks of both tumour spill and of bleeding (Smith et al 1996). In one paediatric series, detectable bleeding occurred in three patients (14 per cent) despite biopsy track embolisation, two with focal disease and one with diffuse disease (Rivera-Sanfeliz et al 2005). In another series, there was no statistical difference in complication rates between patients that did or did not undergo track embolisation (Hatfield et al 2008).
Rationale 27: To ensure optimal condition of the biopsy arriving in the laboratory for analysis.
Rationale 28: When the patient is positioned with their right side down, pressure is being applied to the liver to encourage haemostasis.
Rationale 29: To ensure that the sample is analysed appropriately.
Rationale 30: To maintain an accurate record and ensure patient safety.
Rationale 31: To monitor vital signs and to act promptly to restore circulating blood volume and stop bleeding.
Rationale 32: Complications usually occur within the first three hours of liver biopsy as the blood pressure returns to normal parameters for the child (Grant and Neuberger 1999). Note that most children who bleed post-liver biopsy do so into the peritoneal cavity and therefore there is often no external sign of bleeding.
Rationale 33: For fluid management, blood and platelet transfusion if indicated.
Rationale 34: Junior staff must inform a more senior nurse/doctor should this occur for advice/appropriate medical intervention.
Rationale 35: To ensure the child’s comfort and minimise the risk of complications.
Rationale 36: To prevent aspiration and vomiting.
Rationale 37: To ensure adequate nutrition which will promote wound healing.
Rationale 38: It is difficult to ensure that children remain on bed rest for this period. Creative thinking may need to be used, and the family and play specialist could be involved in keeping the child occupied while on bed rest.
Rationale 39: To ensure they are not dehydrated and showing signs of postoperative urine retention.
Rationale 40: To detect signs of postoperative metabolic instability.
Rationale 41: This may be due to the complexity of the analysis being performed on the biopsy specimen.
Rationale 42: This is to allow the liver to heal following the biopsy and to minimise the risk of internal haemorrhage.
Rationale 43: This may be due to infection and the child will need to visit their GP who will contact the hospital to advise whether they need to return to the hospital for blood test analysis and appropriate treatment.
Cohen MB, A-Kader HH, Lambers D, Heubi JE (1992) Complications of percutaneous liver biopsy in children. Gastroenterology 102 (2): 629-32.
Department of Health (2009) Reference guide to consent for examination or treatment (2nd Ed). www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/digitalas.... Viewed on: 10/06/10
Gonzalez-Vallina R, Alonso EA, Rand E, Black DB, Whitington PF (1993) Outpatient percutaneous liver biopsy in children. J Pediatr Gastroenterol Nutr 17: 370-5.
Hatfield MK, Beres RA, Sane SS, Zaleski GX (2008) Percutaneous imaging-guided solid organ core needle biopsy: coaxial versus noncoaxial method. AJR Am J Roentgenol 190 (2): 413-7.
Grant A, Neuberger J (1999) Guidelines on the use of liver biopsy in clinical practice. British Society of Gastroenterology. Gut 45 Suppl 4: IV1-IV11.
Hoffer FA (2000) Liver biopsy methods for pediatric oncology patients. Pediatr Radiol 30 (7): 481-8.
Lachaux A, Le Gall C, Chambon M, Regnier F, Loras-Duclaux I, Bouvier R, Pinzaru M, Stamm D, Hermier M (1995) Complications of percutaneous liver biopsy in infants and children. Eur J Pediatr 154 (8): 621-3.
Lichtman S, Guzman C, Moore D, Weber JL, Roberts EA (1987) Morbidity after percutaneous liver biopsy. Arch Dis Child 62 (9): 901-4.
Lindor KD, Bru C, Jorgensen RA, Rakela J, Bordas JM, Gross JB, Rodes J, McGill DB, Reading CC, James EM, Charboneau JW, Ludwig J, Batts KP, Zinsmeister AR (1996) The role of ultrasonography and automatic-needle biopsy in outpatient percutaneous liver biopsy. Hepatology 23 (5): 1079-83.
Little AF, Ferris JV, Dodd GD 3rd, Baron RL (1996) Image-guided percutaneous hepatic biopsy: effect of ascites on the complication rate. Radiology 199 (1): 79-83.
Louise Autio K, Kouzer O (2002) The Four S's of Wound Management: Staples, Sutures, Steri-strips and Sticky Stuff. Holistic Nurse Practice 16(2): 80-8.
McGill DB, Rakela J, Zinsmeister AR, Ott BJ (1990) A 21-year experience with major hemorrhage after percutaneous liver biopsy. Gastroenterology 99 (5): 1396-400.
Nobili V, Comparcola D, Sartorelli MR, Natali G, Monti L, Falappa P, Marcellini M (2003) Blind and ultrasound-guided percutaneous liver biopsy in children. Pediatr Radiol 33 (11): 772-5.
Perraul J, McGill DB, Ott BJ, Taylor WF (1978) Liver biopsy: complications in 1,000 in-patients and outpatients. Gastroenterology 74: 103-6.
Rivera-Sanfeliz G, Kinney TB, Rose SC, Agha AK, Valji K, Miller FJ, Roberts AC (2005) Single-pass percutaneous liver biopsy for diffuse liver disease using an automated device: experience in 154 procedures. Cardiovasc Intervent Radiol 28 (5): 584-8.
Schiemann AO, Barrios JM, Al-Tawil YS, Gray KM, Gilger MA (2000) Percutaneous liver biopsy in children: impact of ultrasonography and spring-loaded biopsy needles. J Pediatr Gastroenterol Nutr 31 (5): 536-9.
Sharma P, McDonald GB, Banaji M (1982) The risk of bleeding after percutaneous liver biopsy: relation to platelet count. J Clin Gastroenterol 4 (5): 451-3.
Smith TP, McDermott VG, Ayoub DM, Suhocki PV, Stackhouse DJ (1996) Percutaneous transhepatic liver biopsy with tract embolization. Radiology 198 (3): 769-74.