Liver, 1990: 10, 217-220 Key words: cirrhosis; liver biopsy; Surecur" needle
[email protected] mm liver biopsy in the diagnosis of cirrhosis S0REN TORP-PEDERSEN', MOGENS VYBERG', ELSE SMITH3, LISELOTTE H0JGAARD4,ULLA HANSEN', CARSTEN STADEAGER4,POUL SCHLICHTING6,NIELS JUUL' AND CHRISTIAN GLUUD4,6 'Ultrasound Department, 'Pathology Department, and 6Departmentof Medical Gastroenterology, Herlev Hospital, University of Copenhagen and 'Pathology Department, 3Department of Hepatology, and 4Department of Clinical Physiology, Hvidovre Hospital, University of Copenhagen, Denmark
ABSTRACT - Liver biopsy with the 0.6 mm (23 gauge) Surecut@needle was compared to conventional Menghini biopsy in the diagnosis of cirrhosis. Seventy-seven consecutive patients (mainly alcoholics) with a clinical indication for liver biopsy had both biopsies performed simultaneously. In 71 patients sufficient material for a morphological diagnosis concerning liver architecture was obtained with both biopsy techniques (Surecut insufficient in 5 cases and Menghini insufficient in 2 cases). The biopsies were classified as cirrhosis or non-cirrhosis. There was agreement in 69 cases (97%, confidence limits 9CrlOO%). Using the result of the Menghini biopsy as the final diagnosis, the predictive values for a positive and negative diagnosis for the Surecut needle were 96% and 98%, respectively. There were no complications to either of the biopsies. It is suggested that the 0.6 mm Surecut biopsy may be used in the diagnosis of cirrhosis in cases where conventional Menghini needle biopsy is contraindicated.
Accepted for publication 7 February 1990
Needle biopsy of the liver in the diagnosis of parenchymal changes is carried out world-wide as a routine procedure. Usually the one-second method described by Menghini is applied (I). The needles vary in design and range from 1.4 mm to 1.9 rnm in outer diameter. The Menghini method is associated with a complication rate of approximately 5% on in-patients with no coagulopathy (2). The mortality rate is small but present (0.005% 4. 00 17 %) ( 2 4 ) . Aspiration biopsy of the liver using needles with an outer diameter of 0.6 mm or 0.7 mtn may be carried out with practically no complications and is therefore performed on out-patients with few or no precautions ( 5 , 6). The aspiration material, however, usually does not allow histological
evaluation and is not suitable in the diagnosis of parenchymal liver disease (7). By using a 0.6 mrn modified Menghini needle (Surecut@needle - SN), the advantage of gross needle biopsy (histological diagnosis) may be combined with the advantage of fine needle biopsy (few or no contraindications and probably fewer complications). The aim of the present study was to evaluate the accuracy of the diagnosis of liver cirrhosis using a 0.6 mm SN compared to a conventional Menghini needle.
Material and methods Seventy-seven consecutive patients with a clinical indication for liver biopsy were included in the study. Forty-
218
TORP-PEDERSEN ET AL.
Table I Main morphological diagnoses in Menghini biopsies of 71 patients Alcoholic liver disease (cirrhosis 24, non-cirrhosis 28)
Table 2 Comparison of the morphological diagnosis of cirrhosis obtained by Menghini needle and Surecut@needle
52
Surecute needle
Primary biliary cirrhosis (cirrhosis 1, non-cirrhosis 1)
2
NonCirrhosis cirrhosis
Carcinoma (cirrhosis 0, non-cirrhosis 5 )
5
Hepatitides (cirrhosis 3, non-cirrhosis 7)
10
Venous congestion (cirrhosis 0, non-cirrhosis 2)
2
Total
71
two were females and 35 males. The patients had a median age of 54 years (range 23-82 years). The study was performed in accordance with the Helsinki Declaration 11 and was approved by the local ethics committees. After informed consent had been obtained, the Menghini biopsy and SN biopsy were carried out in one setting. The patients were prepared according to the routine for Menghini biopsy: they were in-patients with normal coagulation time, bleeding time < 5 min, coagulation factors 11, VII, and X > 40 arbitrary units (normal limits 7Cb-130).and thrombocyte count above 50 x 109/ I, (normal limits 150400). Five hundred cc blood was kept in the bank.
Cirrhosis Menghini needle Non-cirrhosis
27 1
I 42
PV-positive 96% (29-100) PV-negative 98% (89-100) 95% confidence limits in brackets
The Menghini biopsy was taken intercostally from the right lobe of the liver with a 1.6 mm needle, giving a biopsy diameter of about I mm. No imaging technique was applied. Usually one pass with the needle was sufficient. The 0.6 mm SN biopsy was taken from the left lobe of the liver, guided by ultrasound. Five passes were allowed using the technique described by Torp-Pedersen et al. (8). Usually 3 4 passes were sufficient. The 0.6 mm SN gives biopsies with a diameter of 0.24.3 mm. The Menghini and SN biopsies were reviewed separately and blindly by two pathologists in cooperation (MV and UH). The criteria for cirrhosis were those described by Anthony et al. (9). All biopsies were classified as cirrhosis or non-cirrhosis. The results of the Menghini biopsy served as the final diagnosis for the SN biopsy, allowing calculation of the predictive value of a positive and negative diagnosis (10). Six patients were
Fig. I . Surecut@biopsy from a case of cirrhosis (Gordon and Sweets’ reticulin, original magnification x 100).
SURE CUT^ NEEDLE IN DIAGNOSIS OF CIRRHOSIS
excluded: four due to insufficient SN biopsy, one due to insufficient Menghini biopsy matcrial, and one because only tumour tissue was present in both biopsies. Thus, in 71 patients sufficient material for a morphological diagnosis was obtained with both biopsy techniques.
Results The median length of the Menghini biopsies was 17 mm (range 3-60 mm). For the SN biopsies the median length was 46 mm (range 5-114 mm), obtained in 3-5 passes. The fragmentation of the SN biopsies was more pronounced than that of the Menghini biopsies. The main morphological diagnoses of the 71 patients are listed in Table 1. According to the Menghini biopsies, 28 were classified as cirrhosis and 43 as non-cirrhosis. The comparison between the two biopsies is listed in Table 2. In 69 cases there was agreement between the results of the two biopsies (97%, confidence limits 90-100%). In one case the Menghini biopsy showed cirrhosis and the SN biopsy non-cirrhosis, and in one case vice versa. Fig. 1 illustrates an SN biopsy from a case of cirrhosis. There were no complications to any of the biopsies.
219
SN may be used in out-patients, when the physical appearance of the patient does not suggest coagulation disorders. Costs of blood tests, bank blood, and nursing capacity may be reduced as well as time for the patient who can leave the hospital immediately after the biopsy. We advocate, however, that the fine needle biopsy is monitored with ultrasound since the safety of fine needle biopsy probably is in part due to major blood vessels and bile ducts being avoided (4). The present study was designed only to deal with liver architecture, which we suspected would cause most difficulty in SN biopsies. In the SN biopsies parenchymal changes identical to those of the Menghini biopsies (e.g. steatosis, Mallory bodies) could often be identified. However, to what extent the 0.6 mm SN may be used in the diagnosis of macronodular cirrhosis as well as a broad spectrum of parenchymal liver diseases remains to be settled.
Acknowledgements The authors are indebted to Professor Hemming Poulsen for his assistance in the interpretation of the biopsies. The present study was supported by a grant from the Carl F. Pedersen Foundation.
Discussion Using a 0.6 mm SN, the diagnosis of cirrhosis was made with a predictive value of a positive answer of 96% and a predictive value of a negative answer of 98%, implying that interpretation of liver architecture is possible in spite of the small diameter of these biopsies. One reason for the high score in identification of cirrhosis is probably the large number of alcoholic liver diseases in which the cirrhosis most often reveals small nodules. The two cases in which disagreement was found between the Menghini and SN biopsies were cases with severe pericellular and septa1 fibrosis obtained in alcoholic patients. The disagreement may be attributed to different sample sites (1 I). We regard the pursuit of a fine needle method in the diagnosis of liver cirrhosis as appropriate in order to overcome some of the relative or absolute contraindications to conventional liver biopsy. The SN biopsy may here be an alternative to transjugular liver biopsy (1 2). Furthermore, the
References I . MENGHINI G. One-second needle biopsy of the liver. Gastroenterology 1958: 35: 190. F. SAGNELLI E, PASQUALE G. GUISTIG 2. PICCININO et al. Complications following percutaneous liver biopsy. J Heparol 1986: 2: 165-173. 3. THALER H. Uber Vorteil und Risiko der Leberbiopsimethode nach Menghini. Wien Klin Wochenschr 1964: 76: 553-558. H. Grenzen und Gefahren der perkutanen 4. LINDNER Leberbiopsie mit der Menghini-Nadel Erfahrungen bei 80.000 Leberbiopsien. Dtsch Med Wochenschr 1958: 35: 190-199. J L, PLAFKER J . Fine needle aspiration of 5 . NOSHER the liver with ultrasound guidance. Radiology 1980: 136: 177- 180. 6. JENSEN F. Procedure and principles in ultrasonic J K guided puncture. In: HOLMH H, KRISTENSEN eds. Interventional Ulstrasound. Copenhagen: Munksgaard, 1985. S, DICK R, VAN LEEUWEN D J. Liver 7 SHERLOCK biopsy today. The Royal Free Hospital Experience. J Hepalol 1984: 1: 75-86. -
220
TORP-PEDERSEN ET AL.
8. TORP-F’EDERSEN S, JUULN, VYBERG M. Histological
sampling with a 23 gauge modified Menghini needle. Br J Radio1 1984: 57: I5 I . 9. ANTHONY P P, ISHAK K G, NAYAKN C, POULSEN P J, SOBINL H. The morphology of H E, SCHEUER cirrhosis. J Clin Parhol 1978: 31: 395414. 10. VECCHIOT J. Predictive value of a single diagnostic test in unselected populations. N Engl J Med 1966: 274 1171-1 173. 1 1 , MAHARAJ B, LEARYW P, NARANA D et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancer 1986: i: 523-525.
R, DOTTERC T. 12. ROSCHJ , LAKINP C, ANTONOVIC Transjugular approach to liver biopsy and transhepatic cholangiography. N Engl J Med 1973: 289: 221-23 I .
Address: Soren Torp-Pedersen. M.D. Ultrasound Department Herlev Hospital DK-2730 Herlev Denmark
[email protected] mm liver biopsy in the diagnosis of cirrhosis S0REN TORP-PEDERSEN', MOGENS VYBERG', ELSE SMITH3, LISELOTTE H0JGAARD4,ULLA HANSEN', CARSTEN STADEAGER4,POUL SCHLICHTING6,NIELS JUUL' AND CHRISTIAN GLUUD4,6 'Ultrasound Department, 'Pathology Department, and 6Departmentof Medical Gastroenterology, Herlev Hospital, University of Copenhagen and 'Pathology Department, 3Department of Hepatology, and 4Department of Clinical Physiology, Hvidovre Hospital, University of Copenhagen, Denmark
ABSTRACT - Liver biopsy with the 0.6 mm (23 gauge) Surecut@needle was compared to conventional Menghini biopsy in the diagnosis of cirrhosis. Seventy-seven consecutive patients (mainly alcoholics) with a clinical indication for liver biopsy had both biopsies performed simultaneously. In 71 patients sufficient material for a morphological diagnosis concerning liver architecture was obtained with both biopsy techniques (Surecut insufficient in 5 cases and Menghini insufficient in 2 cases). The biopsies were classified as cirrhosis or non-cirrhosis. There was agreement in 69 cases (97%, confidence limits 9CrlOO%). Using the result of the Menghini biopsy as the final diagnosis, the predictive values for a positive and negative diagnosis for the Surecut needle were 96% and 98%, respectively. There were no complications to either of the biopsies. It is suggested that the 0.6 mm Surecut biopsy may be used in the diagnosis of cirrhosis in cases where conventional Menghini needle biopsy is contraindicated.
Accepted for publication 7 February 1990
Needle biopsy of the liver in the diagnosis of parenchymal changes is carried out world-wide as a routine procedure. Usually the one-second method described by Menghini is applied (I). The needles vary in design and range from 1.4 mm to 1.9 rnm in outer diameter. The Menghini method is associated with a complication rate of approximately 5% on in-patients with no coagulopathy (2). The mortality rate is small but present (0.005% 4. 00 17 %) ( 2 4 ) . Aspiration biopsy of the liver using needles with an outer diameter of 0.6 mm or 0.7 mtn may be carried out with practically no complications and is therefore performed on out-patients with few or no precautions ( 5 , 6). The aspiration material, however, usually does not allow histological
evaluation and is not suitable in the diagnosis of parenchymal liver disease (7). By using a 0.6 mrn modified Menghini needle (Surecut@needle - SN), the advantage of gross needle biopsy (histological diagnosis) may be combined with the advantage of fine needle biopsy (few or no contraindications and probably fewer complications). The aim of the present study was to evaluate the accuracy of the diagnosis of liver cirrhosis using a 0.6 mm SN compared to a conventional Menghini needle.
Material and methods Seventy-seven consecutive patients with a clinical indication for liver biopsy were included in the study. Forty-
218
TORP-PEDERSEN ET AL.
Table I Main morphological diagnoses in Menghini biopsies of 71 patients Alcoholic liver disease (cirrhosis 24, non-cirrhosis 28)
Table 2 Comparison of the morphological diagnosis of cirrhosis obtained by Menghini needle and Surecut@needle
52
Surecute needle
Primary biliary cirrhosis (cirrhosis 1, non-cirrhosis 1)
2
NonCirrhosis cirrhosis
Carcinoma (cirrhosis 0, non-cirrhosis 5 )
5
Hepatitides (cirrhosis 3, non-cirrhosis 7)
10
Venous congestion (cirrhosis 0, non-cirrhosis 2)
2
Total
71
two were females and 35 males. The patients had a median age of 54 years (range 23-82 years). The study was performed in accordance with the Helsinki Declaration 11 and was approved by the local ethics committees. After informed consent had been obtained, the Menghini biopsy and SN biopsy were carried out in one setting. The patients were prepared according to the routine for Menghini biopsy: they were in-patients with normal coagulation time, bleeding time < 5 min, coagulation factors 11, VII, and X > 40 arbitrary units (normal limits 7Cb-130).and thrombocyte count above 50 x 109/ I, (normal limits 150400). Five hundred cc blood was kept in the bank.
Cirrhosis Menghini needle Non-cirrhosis
27 1
I 42
PV-positive 96% (29-100) PV-negative 98% (89-100) 95% confidence limits in brackets
The Menghini biopsy was taken intercostally from the right lobe of the liver with a 1.6 mm needle, giving a biopsy diameter of about I mm. No imaging technique was applied. Usually one pass with the needle was sufficient. The 0.6 mm SN biopsy was taken from the left lobe of the liver, guided by ultrasound. Five passes were allowed using the technique described by Torp-Pedersen et al. (8). Usually 3 4 passes were sufficient. The 0.6 mm SN gives biopsies with a diameter of 0.24.3 mm. The Menghini and SN biopsies were reviewed separately and blindly by two pathologists in cooperation (MV and UH). The criteria for cirrhosis were those described by Anthony et al. (9). All biopsies were classified as cirrhosis or non-cirrhosis. The results of the Menghini biopsy served as the final diagnosis for the SN biopsy, allowing calculation of the predictive value of a positive and negative diagnosis (10). Six patients were
Fig. I . Surecut@biopsy from a case of cirrhosis (Gordon and Sweets’ reticulin, original magnification x 100).
SURE CUT^ NEEDLE IN DIAGNOSIS OF CIRRHOSIS
excluded: four due to insufficient SN biopsy, one due to insufficient Menghini biopsy matcrial, and one because only tumour tissue was present in both biopsies. Thus, in 71 patients sufficient material for a morphological diagnosis was obtained with both biopsy techniques.
Results The median length of the Menghini biopsies was 17 mm (range 3-60 mm). For the SN biopsies the median length was 46 mm (range 5-114 mm), obtained in 3-5 passes. The fragmentation of the SN biopsies was more pronounced than that of the Menghini biopsies. The main morphological diagnoses of the 71 patients are listed in Table 1. According to the Menghini biopsies, 28 were classified as cirrhosis and 43 as non-cirrhosis. The comparison between the two biopsies is listed in Table 2. In 69 cases there was agreement between the results of the two biopsies (97%, confidence limits 90-100%). In one case the Menghini biopsy showed cirrhosis and the SN biopsy non-cirrhosis, and in one case vice versa. Fig. 1 illustrates an SN biopsy from a case of cirrhosis. There were no complications to any of the biopsies.
219
SN may be used in out-patients, when the physical appearance of the patient does not suggest coagulation disorders. Costs of blood tests, bank blood, and nursing capacity may be reduced as well as time for the patient who can leave the hospital immediately after the biopsy. We advocate, however, that the fine needle biopsy is monitored with ultrasound since the safety of fine needle biopsy probably is in part due to major blood vessels and bile ducts being avoided (4). The present study was designed only to deal with liver architecture, which we suspected would cause most difficulty in SN biopsies. In the SN biopsies parenchymal changes identical to those of the Menghini biopsies (e.g. steatosis, Mallory bodies) could often be identified. However, to what extent the 0.6 mm SN may be used in the diagnosis of macronodular cirrhosis as well as a broad spectrum of parenchymal liver diseases remains to be settled.
Acknowledgements The authors are indebted to Professor Hemming Poulsen for his assistance in the interpretation of the biopsies. The present study was supported by a grant from the Carl F. Pedersen Foundation.
Discussion Using a 0.6 mm SN, the diagnosis of cirrhosis was made with a predictive value of a positive answer of 96% and a predictive value of a negative answer of 98%, implying that interpretation of liver architecture is possible in spite of the small diameter of these biopsies. One reason for the high score in identification of cirrhosis is probably the large number of alcoholic liver diseases in which the cirrhosis most often reveals small nodules. The two cases in which disagreement was found between the Menghini and SN biopsies were cases with severe pericellular and septa1 fibrosis obtained in alcoholic patients. The disagreement may be attributed to different sample sites (1 I). We regard the pursuit of a fine needle method in the diagnosis of liver cirrhosis as appropriate in order to overcome some of the relative or absolute contraindications to conventional liver biopsy. The SN biopsy may here be an alternative to transjugular liver biopsy (1 2). Furthermore, the
References I . MENGHINI G. One-second needle biopsy of the liver. Gastroenterology 1958: 35: 190. F. SAGNELLI E, PASQUALE G. GUISTIG 2. PICCININO et al. Complications following percutaneous liver biopsy. J Heparol 1986: 2: 165-173. 3. THALER H. Uber Vorteil und Risiko der Leberbiopsimethode nach Menghini. Wien Klin Wochenschr 1964: 76: 553-558. H. Grenzen und Gefahren der perkutanen 4. LINDNER Leberbiopsie mit der Menghini-Nadel Erfahrungen bei 80.000 Leberbiopsien. Dtsch Med Wochenschr 1958: 35: 190-199. J L, PLAFKER J . Fine needle aspiration of 5 . NOSHER the liver with ultrasound guidance. Radiology 1980: 136: 177- 180. 6. JENSEN F. Procedure and principles in ultrasonic J K guided puncture. In: HOLMH H, KRISTENSEN eds. Interventional Ulstrasound. Copenhagen: Munksgaard, 1985. S, DICK R, VAN LEEUWEN D J. Liver 7 SHERLOCK biopsy today. The Royal Free Hospital Experience. J Hepalol 1984: 1: 75-86. -
220
TORP-PEDERSEN ET AL.
8. TORP-F’EDERSEN S, JUULN, VYBERG M. Histological
sampling with a 23 gauge modified Menghini needle. Br J Radio1 1984: 57: I5 I . 9. ANTHONY P P, ISHAK K G, NAYAKN C, POULSEN P J, SOBINL H. The morphology of H E, SCHEUER cirrhosis. J Clin Parhol 1978: 31: 395414. 10. VECCHIOT J. Predictive value of a single diagnostic test in unselected populations. N Engl J Med 1966: 274 1171-1 173. 1 1 , MAHARAJ B, LEARYW P, NARANA D et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancer 1986: i: 523-525.
R, DOTTERC T. 12. ROSCHJ , LAKINP C, ANTONOVIC Transjugular approach to liver biopsy and transhepatic cholangiography. N Engl J Med 1973: 289: 221-23 I .
Address: Soren Torp-Pedersen. M.D. Ultrasound Department Herlev Hospital DK-2730 Herlev Denmark
