SHORT REPORT

Liver biopsy for parenchymal liver disease—Is routine realtime image guidance unnecessary?

Anil John & Saad Al Kaabi & Madiha Emran Soofi & Muneera Mohannadi &Salva Manam Kandath & Moataz Derbala & Rafie Yakoub & Esra Mohammed Al-Ahdal &Manik Sharma & Hamid Wani & Nazeeh Dweik & Anjum John & Mohammed Tariq Butt

Received: 18 May 2013 /Accepted: 20 August 2013 /Published online: 19 September 2013# Indian Society of Gastroenterology 2013

Abstract Liver biopsy even today remains the standard ofcare for grading and staging chronic hepatitis despite advancesin noninvasive markers of liver fibrosis. Literature suggests anexpanding role for real-time image guided liver biopsy anddeclining trend for blind liver biopsies. In our center, wherewe perform around 400 liver biopsies per year, we performeda prospective clinical audit of our practice of blind outpatientpercutaneous liver biopsies. Patients requiring histologicalgrading and staging of chronic hepatitis routinely undergoblind outpatient percutaneous liver biopsies in our endoscopyunit unless there is a definite indication for real-time imageguidance. All procedures were assessed for safety, and allspecimens were evaluated by a specimen quality gradingscore for adequacy for grading and staging of chronic hepati-tis. Of the 446 patients referred for histological grading andstaging of chronic hepatitis C by liver biopsy, only 42 patients(9.5 %) required real-time ultrasound for liver biopsy. Theremaining 404 patients underwent blind outpatient percutane-ous liver biopsies which were found to be extremely safe withno major complications, yielding adequate liver tissue withhigh specimen quality score allowing optimal grading andstaging of chronic hepatitis.

Keywords Complications . Percutaneous biopsy .

Quality assurance

Introduction

Despite advances in assessing liver fibrosis by noninvasivemarkers and techniques, liver biopsy even today remains thestandard of care for obtaining hepatic tissue for histopathologicalgrading and staging of chronic hepatitis. The role of liver biopsyassumesmore relevance in the current era of newer treatments forchronic hepatitis B and C and in this age of liver transplantation.

Our unit has been performing on an average around 400liver biopsies per year since the late 1990s as part of thediagnostic evaluation of patients with mostly chronic hepatitisC, who receive subsidized state funded antiviral treatment withpegylated interferon and ribavirin [1]. In view of the largevolumes of patients for liver biopsy, we have been followingthe practice of blind biopsies by gastroenterologists as anoutpatient bedside procedure reserving real time image guid-ance for specific indications and situations like morbid obesity,presence of incidental focal liver lesions, post-liver transplantstate, anatomical variations, or in the event of failure to aspirateliver tissue by blind technique or in any clinical situation wherethe point of maximum liver dullness cannot be percussed outand marked. Image guided liver biopsy is performed by inter-ventional radiologists in our institution and has a long waitingtime in our overburdened radiology suite and incurs the extracost of real time imaging by ultrasonography.

We prospectively audited the safety and histological ade-quacy of our practice of blind non-image-guided biopsies forchronic parenchymal liver disease as an outpatient bedsideprocedure with discharge after an observation period of 4 h.

Methods

Division of Gastroenterology and Hepatology had received 446patients referred for liver biopsy for grading and staging ofchronic hepatitis during the 1-year period from June 2010 to

A. John (*) : S. Al Kaabi :M. Mohannadi :M. Derbala :R. Yakoub :M. Sharma :H. Wani :N. Dweik :M. T. ButtDepartment of Gastroenterology, Hamad Medical Corporation,Doha, Qatare-mail: aniljohn44@gmail.com

M. E. Soofi : S. M. Kandath : E. M. Al-AhdalDepartment of Pathology, Hamad Medical Corporation, Doha, Qatar

A. JohnMedical Research Center, Hamad Medical Corporation, Doha, Qatar

Indian J Gastroenterol (January–February 2014) 33(1):50–54DOI 10.1007/s12664-013-0393-3

May 2011. Of these, we prospectively studied 404 patients withchronic hepatitis who underwent blind percutaneous liver biop-sies. The remaining 42 patients had definite indication for realtime image guided liver biopsy.

All patients were informed about the indication and possiblerisk of liver biopsies and written informed consent obtained. Acomplete blood count with hemoglobin total and differentialcount, platelet count, prothrombin time, liver function tests, andrenal function tests were done in all cases. All our patients wereunder work up for chronic hepatitis B or C prior to commence-ment of antiviral treatment.

All patients underwent an ultrasonography as part of workup for chronic hepatitis B or C. Patients with any incidentalfocal lesions in the liver, anatomical variations, post-livertransplant state, or those with evident cirrhotic changes withfeatures of portal hypertension were excluded from blindbiopsies. All patients had to fulfill the criteria of normalcoagulation profile with a prothrombin time–internationalizedratio (INR) less than 1.3, platelets more than 60,000, and aliver size of more than 8 cm by ultrasonography with clinicalconfirmation of liver dullness by bedside percussion.

The biopsy was performed via a transcostal approach viathe midaxillary line using Menghini technique (TSK SurecutBiopsy needle 17 G × 70 mm, TSK Laboratory, Japan).The procedure was performed always by our staffhepatogastroenterologist. The number of passes madedepended on the operator's assessment of the adequacy ofthe length of the liver tissue obtained measured using a scaleprovided in the liver biopsy kit. If the specimen was less than2 cm in length, additional passes were performed to obtainmore tissue. Biopsy was placed in a formalin-filled containeron a filter paper and sent to the Pathology Department forhistological examination.

All biopsy specimens were examined by an experiencedhepatopathologist for adequacy of the sample in terms of thelength of specimen, presence of fragmentation, adequacy ofthe tissue for making a definitive diagnosis, and presence ofcrush artifacts. The number of complete portal tracts, partialportal tracts, and central veins was documented. Each speci-men was scored according to a histopathological evaluationform (Table 1) from Farrell et al. [2].

Complete portal tract (CPT) was defined as connective tissuefoci within liver parenchyma with complete circumference andcontaining at least two luminal portal structures. Partial portaltract was defined as connective tissue foci within liver parenchy-ma not meeting the criteria of CPT. Length of specimen wasdefined as the aggregate length of all fragments obtained. Ade-quacy for tissue diagnosis was based on the ability of thehistopathologist to confidently grade and stage chronic hepatitisB or C by the Scheuer system [3]. All biopsy specimens wereprocessed by routine methods and evaluated by a blinded pathol-ogist with extensive expertise in interpretation of liver biopsyspecimens

We also evaluated post-procedure pain with a visual ana-logue scale (0–10), 10 being the worst pain experienced. Theneed for additional parenteral analgesics, hypotension requir-ing intravenous fluid administration, or the occurrence of avasovagal phenomenon requiring any therapeutic interventionwere documented as minor complications. All patients wereclosely monitored using continuous blood pressure recording,heart rate measurements, and pulse oximetry in the recoverysuite of our endoscopy unit for 4 h and were discharged homeif found to be stable at the end of 4 h with normal vital signsand were pain free with no complications. They were asked toreport to the accident and emergency unit of our hospital forany complications and were advised to avoid any form ofstrenuous physical exertion for the next 48 h.

It was ensured that all patients stayed within 1 h reach ofthe hospital for 48 h after the procedure.

Results

Of the 446 patients who were referred for grading and stagingof chronic hepatitis during the study period, only 9.5 % hadindications for image guided liver biopsy. The reasons forimage guidance was the presence of incidental hemangiomasor focal lesions in 24 patients, post-liver transplant status inseven patients and large liver cyst in one patient. It wasdifficult to percuss and mark out the point of maximum liverdullness in two patients due to obesity and in another twopatients due to hyperinflated lung fields. In five patients, wefailed to aspirate adequate length of liver tissue, and in onepatient, we had aspiration of extrahepatic tissue, all of whomwere directed for ultrasound-guided biopsies with successfuloutcomes. In total, only 42 of the 446 patients required real

Table 1 Liver biopsy specimen quality grading score

Point score

A—tissue fragmentation

No tissue 0

Multiple fragments, none >5 mm 1

Multiple fragments, at least 1>5 mm 2

At least 1 fragment >10 mm 3

B—crush artifact

No tissue 0

Severe crush destroying most of specimen 1

Some crush but does not impair interpretation 2

Limited or no crush 3

C—diagnostic quality

No tissue 0

Tissue present but stage and grade not obtained 1

Stage and grade of disease identified 2

Total quality histopathology score 8

Indian J Gastroenterol (January–February 2014) 33(1):50–54 51

time image guidance which is less than 10 % of our patientsreferred for liver biopsy.

All biopsies were performed successfully as an outpatientprocedure with 96% of the discharges being completed after 4 hof observation in the recovery room, and the rest weredischarged within a 6-h observation period. Though specificallyinstructed, none of the patients reported back to the Accidentand Emergency Department during the 48-h post-biopsy period.

The minor complications reported within the observationperiod in the recovery room was only in 24 patients, of which18 had pain at the biopsy site or right shoulder requiringadditional doses of parenteral analgesics over and above thelocal anesthetic which was used prior to biopsy. Five patientshad transient hypotension requiring intravenous fluid infu-sion, and one patient had a vasovagal phenomenon requiringatropine. Thus, the minor complications were noted in justaround 5 % of the patients and there were no major compli-cations requiring hospitalizations or interventions.

The mean number of passes was 1.162±0.43 (mean±SD),and in 85 % of patients undergoing blind biopsy, adequateliver tissue of more than 2 cm was aspirated in a single pass.The mean length of the liver tissue aspirated was 2.335±0.79 cm. The length of liver tissue aspirated was more than2 cm in 340 of the 404 patients and in the rest aggregate lengthof multiple fragments added up to 2 cm or above. The meannumber of complete portal tracts was 9.06±4.7. Patients hav-ing less than five portal tracts where tissue was deemedrelatively inadequate for optimal grading and staging wasnoted in 51 patients (12.6 %), though grading and stagingcould be carried out in this group as well.

The mean specimen quality grading score for the adequacyof the specimen based on tissue fragmentation, crush artifact,and diagnostic quality (Table 1) was calculated to be 6.98±0.879 on a scale ranging from 0 to 8 (Table 2).

Tissue was adequate for fibrosis staging with stage 0 re-ported in 81 patients, stage 1 in 128 patients, stage 2 in 120patients, stage 3 in 54 patients, and stage 4 in 9 patients.

Discussion

While blind percutaneous needle biopsy is a traditional tech-nique, the use of ultrasound guidance has increased consider-ably in recent years. Paul Ehlrich has been credited with thefirst needle biopsy in 1883. Since then, the technical aspects ofthis procedure has been continuously redefined and modifiedand debate still surrounds the quest for the best and safepractice. The American Association of the Study of LiverDiseases (AASLD) did a random survey among its memberson the practice of liver biopsy and found 62 % relied on anultrasonographer to locate andmark the biopsy sites and, 18%had biopsy performed by a radiologist under real time ultra-sound guidance [4]. Another survey from the USA study says

that 50 % of the liver biopsies are performed by radiologiststoday [5]. In the UK, 34 % of the liver biopsies are performedby radiologists, 33 % by internists, 28 % by gastroenterolo-gists, and the rest by others. One third of 1,500 percutaneousliver biopsies in England and Wales were performed underradiological guidance by radiologists [6]. In France, 89 % ofpercutaneous liver biopsies are performed by gastroenterolo-gists and 11 % by radiologists [7]. Thus, the practice of liverbiopsy is diverse and varied the world over.

Though ultrasound guided biopsy has the obvious advantageof high diagnostic yield for focal liver lesions [8], its routine usein parenchymal liver disease is still deemed unnecessary and isdebatable. The advantage of image guided liver biopsy with itsextra expenditure and waiting time over conventional blindbedside liver biopsies in parenchymal liver disease remain

Table 2 Results

Variables Values

Number of chronic hepatitisreferred for liver biopsy

446

Patients requiring imageguidance

42 (9.5 %)

Number of patients whounderwent blind outpatientbiopsy

404 (male 360, female 44)

Indications for liver biopsyin blind group

HCV−390, HBV–14

Age in years 41.82±9.7

Weight in kg 84.05±13.37

Average liver span in cm 14.772±1.82

INR 1.01±0.83

Platelet count (×1,000/μL) 194.47±55.85

Failure to aspirate liver tissuesby blind technique or extrahepatictissue aspiration

6 patients (1.4 %)

Patients (%) from whomadequate liver tissue wasaspirated after single pass

85

Number of passes for biopsies 1.162±0.4378

Length of liver tissuesaspirated (cm)

2.335±0.99

Number of complete portaltracts

9.06±4.7

Less than five portal tracts 48 (12.06 %)

Five to ten portal tracts 200 (50.25 %)

More than ten portal tracts 150 (37.6 %)

Number of partial portal tracts 2.35±1.8

Specimen quality grading 6.98±0.879

Minor complications (%) 24 (5.9)

Major complications (%) 0

Discharges within 4 h 96 %

Discharges within 6 h 100 %

Values shown are mean±SD unless stated otherwise

INR internationalized ratio

52 Indian J Gastroenterol (January–February 2014) 33(1):50–54

poorly defined in current clinical practice. In a recent study,ultrasound guidance did not provide any advantage over blindbiopsies as regards safety or adequacy of specimens [9].

Outpatient liver biopsies have been popular and have notbeen shown to be associated with increased complications [10].Most outpatient liver biopsy patients are observed for up to 6 hpost-biopsy, but since the majority of the complications occurwithin the first 4 h post-procedure period, many experts advo-cate reduction in the post-biopsy observation time to 4 h [11].Most guidelines for daytime percutaneous liver biopsy recom-mend an observation period for 4–6 h [12]. This is based onstudies showing that majority of complications occur within 2 hof the procedure. We could discharge 96 % of our patients after4 h of observation in a stable condition, and the rest weredischarged after 6 h of observation in the recovery room. Noneof the discharged patients reported to the Emergency sectionwithin the 48-h post-biopsy period.

Liver biopsy specimens containing less than ten completeportal tracts and less than 2 cm in length tend to underestimatethe grade and stage of chronic hepatitis C [13], though thesestandards are seldom uniformly met in routine clinical practice.In a study comparing blind and ultrasound guided biopsies byFarrell et al. [2], though the mean length of liver tissue aspi-rated by blind technique was 1.62 cm and the mean number ofportal tracts was 7.8±1.6, still the specimens were adequate forgrading and staging of chronic hepatitis. In our study, 12 % ofthe biopsies had less than five portal tracts; nevertheless, ourpathologists reported the fibrosis stage though there was aconcern of underestimating the stage as alluded to above.According to the British guidelines for liver biopsy, a biopsyspecimen containing at least six to eight portal tracts is consid-ered adequate [12]. In contrast, the AASLD position paperrecommends at least 11 portal tracts for accurate grading andstaging of chronic viral hepatitis with biopsy specimens morethan 2 cm in length obtained by 16 G needle [14].

In our study, the mean length of liver tissue aspirated was2.335 cm and mean number of portal tracts was 9.06±4.7 cm,which was sufficient for histological grading and staging ofchronic hepatitis by our pathologist. Bedossa et al. havementioned that the minimum acceptable specimen length forstaging was 2.5 cm [15]. However, another report by Bravoet al. said that 1.5 cm of liver tissue was usually adequate fordiagnosis of diffuse liver diseases [16]. In addition to thelength of the tissue, the diameter of the core is also importantin the staging of viral hepatitis. Needles of 16 to 18 G havebeen shown to be much more useful for obtaining bigger livertissue for optimal staging [17].

The higher the number of passes, the higher the morbidityfrom liver biopsy [18]. The mean number of passes in ourseries was 1.162±0.43 cm, and 85 % of patients undergoingblind biopsy required only one pass to secure an adequatelength of liver tissue, contributing to lower procedure-relatedmorbidity.

Specimen quality grading based on an aggregate numericalscore for tissue fragmentation, crush artifact, and diagnosticquality (Table 1) done on each specimen yielded a mean scoreof 6.8 against the maximum score of 8, thus confirming theoverall adequacy of quality of specimens obtained by blindtechnique.

Conclusion

In summary, the audit of our liver biopsy practice confirms thatreal time image guidance would be required for definite indi-cations in just around 10 % of patients referred for evaluationfor chronic parenchymal liver disease and blind techniquewould be safe and sufficient for the vast majority. It would beprudent to judiciously reserve image guidance for relevant andappropriate clinical indications in the practice of liver biopsy,thus saving significant waiting time and cost, which would beeven more pertinent and relevant in high volume liver units.

References

1. John AK, Al KS, John A, Singh R, Derbala M. Audit of state-fundedantiviral treatment for chronic hepatitis C in Qatar. East MediterrHealth J. 2010;16:1121–7.

2. Farrell RJ, Smiddy PF, Pilkington RM, et al. Guided versus blindliver biopsy for chronic hepatitis C: clinical benefits and costs. JHepatol. 1999;30:580–7.

3. Desmet VJ, Gerber M, Hoofnagle JH, Manns M, Scheuer PJ. Clas-sification of chronic hepatitis: diagnosis, grading and staging.Hepatology. 1994;19:1513–20.

4. Angtuaco TL, Lal SK, Banaad-Omiotek GD, Zaidi SS, Howden CW.Current liver biopsy practices for suspected parenchymal liver dis-eases in the United States: the evolving role of radiologists. Am JGastroenterol. 2002;97:1468–71.

5. Mayoral W, Lewis JH. Percutaneous liver biopsy: what is the currentapproach? Results of a questionnaire survey. Dig Dis Sci. 2001;46:118–27.

6. Gilmore IT, Burroughs A, Murray-Lyon IM, Williams R, Jenkins D,Hopkins A. Indications, methods, and outcomes of percutaneousliver biopsy in England and Wales: an audit by the British Societyof Gastroenterology and the Royal College of Physicians of London.Gut. 1995;36:437–41.

7. Cadranel JF, Rufat P, Degos F. Practices of liver biopsy in France:results of a prospective nationwide survey. For the Group of Epide-miology of the FrenchAssociation for the Study of the Liver (AFEF).Hepatology. 2000;32:477–81.

8. Buscarini L, Fornari F, Bolondi L, et al. Ultrasound-guided fine-needle biopsy of focal liver lesions: techniques, diagnostic accuracyand complications. A retrospective study on 2091 biopsies. JHepatol. 1990;11:344–8.

9. Akkan CZ, Sezikli M, Guzelbulut F, et al. Liver biopsy: ultrasonog-raphy guidance is not superior to the blind method. J GastrointestinLiver Dis. 2010;19:49–52.

10. Douds AC, Joseph AE, Finlayson C, Maxwell JD. Is day case liverbiopsy underutilised? Gut. 1995;37:574–5.

Indian J Gastroenterol (January–February 2014) 33(1):50–54 53

11. Firpi RJ, Soldevila-Pico C, Abdelmalek MF, Morelli G, Judah J,Nelson DR. Short recovery time after percutaneous liver biopsy:should we change our current practices? Clin Gastroenterol Hepatol.2005;3:926–9.

12. Grant A, Neuberger J. Guidelines on the use of liver biopsy in clinicalpractice. British Society of Gastroenterology. Gut. 1999;45 Suppl 4:IV1–IV11.

13. Colloredo G, Guido M, Sonzogni A, Leandro G. Impact of liverbiopsy size on histological evaluation of chronic viral hepatitis: thesmaller the sample, the milder the disease. J Hepatol. 2003;39:239–44.

14. Rockey DC, Caldwell SH, Goodman ZD, Nelson RC, Smith AD.Liver biopsy. Hepatology. 2009;49:1017–44.

15. Bedossa P, Dargere D, Paradis V. Sampling variability of liver fibro-sis in chronic hepatitis C. Hepatology. 2003;38:1449–57.

16. Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl J Med.2001;344:495–500.

17. Scheuer PJ. Liver biopsy size matters in chronic hepatitis: bigger isbetter. Hepatology. 2003;38:1356–8.

18. McGill DB, Rakela J, Zinsmeister AR, Ott BJ. A 21-year experiencewith major hemorrhage after percutaneous liver biopsy. Gastroenter-ology. 1990;99:1396–400.

54 Indian J Gastroenterol (January–February 2014) 33(1):50–54