ClinicalRadiology(1992) 46, 111 113
Renal Biopsy in Diffuse Renal Disease- Experience With a 14-Gauge Automated Biopsy Gun K. T. T U N G , M. O. D O W N E S and P. J. O ' D O N N E L L *
Department of Diagnostic Radiology, Kent and Canterbury Hospital, Canterbury and *Department of Histopathology, King's College Hospital, London The diagnostic and complication rates of 104 percutaneous renal biopsies performed for diffuse renal disease in native kidneys were retrospectively reviewed. Biopsies were performed by one radiologist using continuous ultrasound guidance and a 14-gauge biopsy needle in an automated gun (Biopty TM, Radiplast TM, Uppsala). 103 of 104 (99%) biopsies resulted in adequate tissue for a definitive histological diagnosis which improves on previously published diagnostic rates. Four patients (3.8%) experienced transient macroscopic haematuria. There were two symptomatic peri-renal haematomas, both of whom required transfusion, and one arteriovenous'fistula which was successfully embolized (total 2.9% significant complications). Our results compare favourably with results using more conventional techniques. We suggest that use of real-time ultrasound with the 14-gauge Biopty needle should be the method of choice for percutaneous renal biopsy in adults. Tung, K.T., Downes, M.O. & O'Donnell, P.J. (1992). Clinical Radiology 46, 111-113. Renal Biopsy in Diffuse Renal Disease Experience With a 14-Gauge Automated Biopsy G u n
Accepted for Publication 4 March 1992
Percutaneous renal biopsy (PRB) has been used widely in the last 40 years as a diagnostic procedure in the investigation of diffuse renal disease. Since the early descriptions of the technique [1,2], there have been several modifications and improvements to increase diagnostic yield and minimize complications. Refinements have included different methods of localizing the kidney using fluoroscopy [3], radionuclide scanning [4], ultrasound [5,6,7] or computed t o m o g r a p h y [8], and using different biopsy needles [9-12]. This study is a retrospective review of our experience of the last consecutive 104 biopsies in native kidneys over a period of 3 years 10 months using an automated biopsy device with a 14-gauge cutting needle (Biopty T M , Uppsala) under continuous ultrasound guidance. We found our diagnostic and complication rate to compare favourably with existing rates reported in the literature.
M A T E R I A L S AND M E T H O D S Between October 1987 to July 1991, 103 patients (64 male, 39 female, age range 14 to 78 years, mean 45.4 years) with suspected diffuse renal disease in native kidneys were referred to our department for PRB. There were a total of 104 attendances for PRB, one patient requiring two attendances to achieve a diagnostic result. The case records of these patients were recalled and reviewed with reference to the results of PRB together with any complications directly attributable to the biopsy procedure. Patients were referred for biopsy after review of clinical indications by two nephrologists. These included acute renal failure (13 patients), chronic renal impairment (22 patients), nephrotic syndrome or suspected glomeruloCorrespondence to: M. O. Downes, Department of Diagnostic Radiology, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent CT1 3NG.
nephritis (59 patients) and systemic disorders with suspected renal involvement (9 patients). All patients had prior ultrasound to assess renal size and to exclude anatomical abnormalities and urinary tract obstruction. Patients with abnormal coagulation screens and those with severe hypertension had the procedure deferred until the underlying disorder was controlled. Having obtained full written consent, patients were given pre-medication consisting of a small dose of an opiate (10-20 mg Omnopon, Roche Ltd) 1 h before the procedure. On arrival in the ultrasound room, patients were re-scanned in the prone position using a 3.5 M H z probe (ATL 8 or Toshiba SAL 100SSA) and the skin site overlying the lateral lower pole of one kidney (usually left) marked. Following infiltration with local anaesthetic, a 14-gauge Biopty needle loaded within the gun was advanced along a biopsy guide. When a guide was not used, visualization of the needle was achieved by aligning the ultrasound transducer parallel to the long axis of the needle. During quiet respiration, the needle was advanced into the peri-renal fat space. As the needle tip was seen to approach the renal capsule, the patient was instructed to suspend respiration in a phase appropriate to allowing the needle to traverse the lower pole cortex avoiding renal medulla and the sinus echo complex. Within this single breath-hold the needle was advanced just beyond the renal capsule under direct ultrasound visualization and the biopsy gun fired. The whole device was immediately withdrawn before commencement of respiration. Often the kidney was seen to tilt or move as the needle penetrated the capsule. If the needle tip was poorly visualized it was withdrawn into peri-renal fat before allowing respiration to commence and prior to a further attempt. The tissue core obtained was examined under lowpower light microscopy by an experienced scientific officer in attendance and an assessment of the adequacy of the specimen and estimate of number of glomeruli made.
112
CLINICAL RADIOLOGY
If the patient's clinical condition allowed further passes were made if the specimens did not contain renal tissue or were felt to contain an insufficient number of glomeruli. This was more likely to occur in patients who were obese or muscular, making ultrasound visualization of the needle tip difficult, or in patients with small mobile kidneys. Occasionally, despite adequate visualization of the needle tip and its passage into the kidney, the core obtained was small or fragmented necessitating a further pass. Specimens were divided and placed in formalin for light microscopy, glutaraldehyde for electron microscopy and phosphate-buffered saline for immunofluorescence studies. The histopathology was reviewed by a single renal pathologist. Biopsy was performed by a single consultant radiologist in 54 instances or a number of senior registrars instructed and initially supervised by the consultant on 50 occasions. Following the procedure, patients were kept on bed rest for 24 h with regular observations of the puncture site, haemodynamic status and urine analysis. Patients with no signs of complications were discharged the following day if they had attended solely for renal biopsy and were reviewed in the outpatient clinic subsequently.
RESULTS Of 104 attendances for biopsy, 103 resulted in retrieval of adequate material for examination by light microscopy, immunofluorescence and electron microscopy and for achieving a full histological diagnosis. In one case when biopsy was performed by a senior registrar insufficient tissue was obtained (containing three glomeruli) and the biopsy subsequently repeated successfully. An average of 2.6 passes were made at each procedure (range 1-5) and a mean of 13.4 glomeruli were present in each tissue core (range 3-28). There were post-procedure complications in seven instances. Four (3.8%) of these were minor, consisting of macroscopic haematuria; all settled within 48 h without requirement of transfusion or other intervention. Complications were considered to be significant in three (2.9%) cases; two had symptomatic peri-renal haematomas both of whom required transfusions of 2 and 4 units of blood respectively, but no further intervention. The third patient had an arterio-venous fistula which manifested as persistent haematuria requiring transfusion of 7 units of blood and having failed to settle was subsequently successfully embolized following a diagnostic angiogram, in our department. All three cases with significant complications did not have an excessive number of passes during biopsy with two passes each in two cases and three passes in one case. There was no clear difference in complication rate in procedures performed by the consultant (2 minor, t significant of 54 biopsies) compared to those performed by senior registrars (2 minor, 2 significant of 50 biopsies) in this small study, although the latter tended to make a larger number of passes at each procedure (average of 2.8) compared to the consultant (average of 2.4). Length of follow-up ranged from 9 days to 4 years. There were no late complications attributable to PRB.
DISCUSSION A wide variety of methods are available for obtaining renal tissue by pcrcutaneous biopsy [2,5,6,10,12]. The blind method using surface anatomical landmarks [2] was a standard technique but requires experience and has a significant failure and complication rate [13,14]. An inherent difficulty of this technique is accurate localization of the kidney. Although ultrasound is of proven value in localizing the kidney [15,16], many use this modality solely to mark the skin site for insertion of a fine needle to determine depth of kidney by respiratory motion prior to performing the biopsy [17-t9]. The use of an exploring needle followed by the biopsy needle, without direct ultrasound guidance, increases the time of the procedure and runs the risk of capsular laceration especially if the technique is faulty. More recently biopsies have been performed with continuous ultrasound guidance [6,7,9,10], allowing direct visualization of the biopsy needle as it enters the kidney and eliminating the need for an exploring needle to determine depth. In many cases, however, two operators are required; one for ultrasonic guidance and the other for operation of the biopsy needle, usually either a Vim-Silverman or a Tru-Cut needle (Travenol Laboratories, Deerfield, IL). The Biopty device has been a significant advance in procurement of tissue and since its initial description [20], has been widely used in biopsy of many organs [21,22]. It retrieves adequate tissue cores with minimal fragmentation and crush artefact and is safer as the time the needle stays within the target tissue is reduced. By combining single-handed use of this device with real-time ultrasound it is possible for a single operator to perform PRB rapidly, accurately and safely as shown in our series. This is in accordance with early favourable reports [23,24] using this method. It is essential when using this technique to ensure that the needle tip is well within the capsule as the Biopty gun is released. In a few cases when this has not occurred, particularly in patients with tough pcri-renal tissues, we have witnessed the kidney tip or bounce away as the gun is released. This may account for failure to obtain diagnostic specimens when the blind technique is used if perirenal tissues are thick and may also cause capsular laceration. It is also important to perform the whole manoeuvre of advancing the needle into the renal capsule, releasing the Biopty gun and withdrawal all within a single breath-hold. This will avoid lacerations and tears to the kidney caused by respiratory motion. In addition to biopsy technique, sample size is important in achieving a full diagnostic result. We have an experienced scientific officer/technician in attendance to examine the tissue cores as they are obtained and advise on the adequacy of specimens and hence the need or otherwise to make further passes. The probability that the sample accurately reflects the disease process in the kidney depends on both the number of glomeruli in the specimen and the actual fraction of glomeruli affected by disease. With small samples, focal disease may be missed and classification of disease severity is less likely to be accurate [25]. Retrieval of about 10 intact glomeruli is usually required to establish a definitive diagnosis, and as tissue should be examined by light and electron microscopy, and immunofluorescence, two tissue cores are required in most cases [19]. The use of a 14-gauge needle satisfies these requirements with an average of 13.4
RENAL BIOPSY IN DIFFUSE RENAL DISEASE
glomeruli per tissue core in our study. We were thus able to achieve a definitive diagnosis in 99% of our biopsies in comparison with rates ranging from 81 to 95% [5,6,9,14,17] in the literature. Clearly it would be tempting to attempt PRB using smaller needles (e.g. 18-gauge) but given the requirements for a representative histopathological specimen, the diagnostic rates are likely to be lower unless several more passes are made to obtain a large number of smaller cores. Using both 16- and 18-gauge needles, no glomeruli were obtained in 10.7 21.4% of biopsies due to sampling error in one series [9]. In paediatric practice, however, where glomeruli are more tightly packed together within the kidney a smaller specimen obtained by a 18-gauge needle is more likely to be sufficient [26]. In a study comparing 15 different biopsy devices and needles (including the 18-gauge Biopty, but not including the 14-gauge Biopty) in a controlled setting of open renal biopsy, in pigs, the 14-gauge Tru-Cut needle proved to be superior [1 l]. We believe that in PRB for diffuse disease, the 14-gauge Biopty needle combines the advantage of size with superior and safer cutting characteristics. The requirements for a larger specimen needle need to be balanced against a theoretical higher risk of complications. The incidence of complications following PRB may be relatively high if these are actively sought using routine post-procedure imaging studies [27,28]. However the majority of these are clinically silent and resolve spontaneously. As with most clinical studies, we did not perform post-biopsy renal ultrasound or CT routinely and only proceed to further imaging if clinically indicated. The clinical complication rate in our study compares favourably with existing clinical complication rates of PRB in the literature. Our minor complication rate of 3.8% compares with reported rates varying from 5 to 14% [5,8,9,14,16,17]. For significant complications, our rate of 2.9% is within the range of I to 6.4% reported in various series [5,6,8,9,14,16,17]. We believe this to be acceptable particularly in view of the high diagnostic rate in our series. In summary, the use of real-time ultrasound with the 14-gauge Biopty device has many advantages in comparison with more conventional techniques. It is highly accurate and has a low complication rate. The time of procedure is considerably reduced and only a single operator is required. Complications such as a peri-renal haematomas or arterio-venous fistulae can be detected and/or treated by the radiologist performing PRB. We recommend this method of PRB to radiologists and to nephrologists using more conventional techniques. Acknowledgements. We are grateful to Drs M. Goggin and D. Prosser for referring their patients and Mr T. Spice, MLSO, for expert interpretation and handling of specimens during the biopsy procedures. Our thanks also to J. Irons, R. L. Saunders and L. Bolt for secretarial assistance. REFERENCES
1 IversonP, BrunC. Aspirationbiopsyofkidney. AmericanJournalof Medicine 1951;11:324 330.
113
2 Kark RM, Muehrcke RC. Biopsy of kidney in prone position. Lancet 1954;1:1047 1049. 3 Lusted LB, Mortimer GE, Hopper JJ. Needle renal biopsy under image amplifier control. American Journal of Roentgenology 1956;75:953--955. 4 Telfer N, Ackroy DAE, Stock SL. Radio-isotopic localisation for renal biopsy. Lancet 1964;1:132 [33. 5 Mets T, Lameire N, Matthys E, Afschrift M. Sonically guided renal biopsy. Journal of Clinical Ultrasound 1979;7:190 191. 6 Birholz JC, Kasinath BS, Corwin HL. An improved technique for ultrasound guided percutaneous renal biopsy. Kidney International t985;27:80-82. 7 Yoshimoto M, Fujisawa A, Sudo M. Percutaneous renal biopsy well visualised by orthogonal ultrasound application using linear scanning. Clinical Nephrology 1988;30:106 110. 8 Nadel L, Baumgartner BR, Bernardino ME. Percutaneous renal biopsies: accuracy, safety and indications. Urological Radiology 1986;8:67 71. 9 Mostbeck GH, Wittich GR, Derfler K, Ulrich W, Walter RM, Herold C, et al. Optimal needle size for renal biopsy; in vitro and in vivo evaluation. Radiology 1989;173:819-822. 10 Wiseman DA, Hawkins R, Numerow LM, Taub KJ. Percutaneous renal biopsy utilizing real time ultrasonic guidance and a semiautomated biopsy device. Kidney International 1990;38:347 349. 11 Hopper KD, Baird DE, Reddy VV, Landis JR, Parker SH, Tyler HN, et al. Efficacy of automated biopsy gun versus conventional biopsy needles in the pygmy pig. Radiology 1990;176:671 676. 12 Almkuist RD, Buckalew VM. Techniques of renal biopsy. Urological Clinics of North America 1979;6:503 517. 13 Slotkin EA, Madsen PO. Complications of renal biopsy: incidence in 5,000 reported cases. Journal of Urology 1962;87:13 15. 14 Diaz-Buxo JA, Donadio JV. Complications of percutaneous renal biopsy: an analysis of t,000 consecutive biopsies. Clinical NephroIogy 1975;4:223 227. 15 Bolton WK, Tully RJ, Lewis EJ, Ranniger K. Localisation of the kidney by percutaneous biopsy. A comparative study of methods. Annals of Internal Medicine 1974;81:159-164. 16 Health and Public Policy Committee, American College of Physicians. Clinical competence in percutaneous renal biopsy. Annals of Internal Medicine 1988;108:301 303. 17 Chan JCM, Brewer WH, Still WJ. Renal biopsies under ultrasound guidance: 100 consecutive biopsies in children. Journal of Urology 1983;129:103 107. 18 Beaman M. How to perform a renal biopsy. British Journal of HospitaIMedicine 1989;41:158 160, 19 Madaio MP. Renal biopsy. Kidney International 1990;38:529-543. 20 Lindgren PG. Percutaneous needle biopsy. A new technique. Acta Radiologica 1982;23:653 656. 21 Parker SH, Hopper KD, Yakes WF, Gibson MD, Ownbey JL, Carter TE. Image-directed percutaneous biopsies with a biopsy gun. Radiology 1989; 171:663-669. 22 Bernardino ME. Automated biopsy devices: significance and safety. Radiology 1990;176:615 616. 23 Mackenzie JC, Mackay IG, Millar ND, Maclver AG, Penry JB. The Biopty cut procedure for renal biopsies (letter). British Medical Journal 1988;296:1128-1129. 24 Worth DP, Khanna SK, Ubhi CS, Irving HC. The Biopty cut procedure for renal biopsies (letter). Britis'h Medical Journal 1988;296:1603. 25 Corwin HL, Schwartz MM, Lewis EJ. The importance of sample size in the interpretation of renal biopsy. American Journal c?f Nephrology 1988;8:85 89. 26 Poster RB, Jones DB, Spirt BA. Percutaneous pediatric renal biopsy: Use of the biopsy gun. Radiology 1990; 176:725 727. 27 Jorstad S, Borander U, Berg KJ, Wideroe TE. Evaluation of complications due to percutaneous renal biopsy: a clinical and angiographic study. American Journal of Kidney Diseases 1984;4:162-165. 28 Ralls PW, Barakos JA, Kaptein EM, Friedman PE, Fouladian G, Boswell WD, et al. Renal biopsy-related hacmorrhage: frequency and comparison of CT and sonography. Journal o[" Computer Assisted Tomography 1987; t 1: 1031- 1034.
Renal Biopsy in Diffuse Renal Disease- Experience With a 14-Gauge Automated Biopsy Gun K. T. T U N G , M. O. D O W N E S and P. J. O ' D O N N E L L *
Department of Diagnostic Radiology, Kent and Canterbury Hospital, Canterbury and *Department of Histopathology, King's College Hospital, London The diagnostic and complication rates of 104 percutaneous renal biopsies performed for diffuse renal disease in native kidneys were retrospectively reviewed. Biopsies were performed by one radiologist using continuous ultrasound guidance and a 14-gauge biopsy needle in an automated gun (Biopty TM, Radiplast TM, Uppsala). 103 of 104 (99%) biopsies resulted in adequate tissue for a definitive histological diagnosis which improves on previously published diagnostic rates. Four patients (3.8%) experienced transient macroscopic haematuria. There were two symptomatic peri-renal haematomas, both of whom required transfusion, and one arteriovenous'fistula which was successfully embolized (total 2.9% significant complications). Our results compare favourably with results using more conventional techniques. We suggest that use of real-time ultrasound with the 14-gauge Biopty needle should be the method of choice for percutaneous renal biopsy in adults. Tung, K.T., Downes, M.O. & O'Donnell, P.J. (1992). Clinical Radiology 46, 111-113. Renal Biopsy in Diffuse Renal Disease Experience With a 14-Gauge Automated Biopsy G u n
Accepted for Publication 4 March 1992
Percutaneous renal biopsy (PRB) has been used widely in the last 40 years as a diagnostic procedure in the investigation of diffuse renal disease. Since the early descriptions of the technique [1,2], there have been several modifications and improvements to increase diagnostic yield and minimize complications. Refinements have included different methods of localizing the kidney using fluoroscopy [3], radionuclide scanning [4], ultrasound [5,6,7] or computed t o m o g r a p h y [8], and using different biopsy needles [9-12]. This study is a retrospective review of our experience of the last consecutive 104 biopsies in native kidneys over a period of 3 years 10 months using an automated biopsy device with a 14-gauge cutting needle (Biopty T M , Uppsala) under continuous ultrasound guidance. We found our diagnostic and complication rate to compare favourably with existing rates reported in the literature.
M A T E R I A L S AND M E T H O D S Between October 1987 to July 1991, 103 patients (64 male, 39 female, age range 14 to 78 years, mean 45.4 years) with suspected diffuse renal disease in native kidneys were referred to our department for PRB. There were a total of 104 attendances for PRB, one patient requiring two attendances to achieve a diagnostic result. The case records of these patients were recalled and reviewed with reference to the results of PRB together with any complications directly attributable to the biopsy procedure. Patients were referred for biopsy after review of clinical indications by two nephrologists. These included acute renal failure (13 patients), chronic renal impairment (22 patients), nephrotic syndrome or suspected glomeruloCorrespondence to: M. O. Downes, Department of Diagnostic Radiology, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent CT1 3NG.
nephritis (59 patients) and systemic disorders with suspected renal involvement (9 patients). All patients had prior ultrasound to assess renal size and to exclude anatomical abnormalities and urinary tract obstruction. Patients with abnormal coagulation screens and those with severe hypertension had the procedure deferred until the underlying disorder was controlled. Having obtained full written consent, patients were given pre-medication consisting of a small dose of an opiate (10-20 mg Omnopon, Roche Ltd) 1 h before the procedure. On arrival in the ultrasound room, patients were re-scanned in the prone position using a 3.5 M H z probe (ATL 8 or Toshiba SAL 100SSA) and the skin site overlying the lateral lower pole of one kidney (usually left) marked. Following infiltration with local anaesthetic, a 14-gauge Biopty needle loaded within the gun was advanced along a biopsy guide. When a guide was not used, visualization of the needle was achieved by aligning the ultrasound transducer parallel to the long axis of the needle. During quiet respiration, the needle was advanced into the peri-renal fat space. As the needle tip was seen to approach the renal capsule, the patient was instructed to suspend respiration in a phase appropriate to allowing the needle to traverse the lower pole cortex avoiding renal medulla and the sinus echo complex. Within this single breath-hold the needle was advanced just beyond the renal capsule under direct ultrasound visualization and the biopsy gun fired. The whole device was immediately withdrawn before commencement of respiration. Often the kidney was seen to tilt or move as the needle penetrated the capsule. If the needle tip was poorly visualized it was withdrawn into peri-renal fat before allowing respiration to commence and prior to a further attempt. The tissue core obtained was examined under lowpower light microscopy by an experienced scientific officer in attendance and an assessment of the adequacy of the specimen and estimate of number of glomeruli made.
112
CLINICAL RADIOLOGY
If the patient's clinical condition allowed further passes were made if the specimens did not contain renal tissue or were felt to contain an insufficient number of glomeruli. This was more likely to occur in patients who were obese or muscular, making ultrasound visualization of the needle tip difficult, or in patients with small mobile kidneys. Occasionally, despite adequate visualization of the needle tip and its passage into the kidney, the core obtained was small or fragmented necessitating a further pass. Specimens were divided and placed in formalin for light microscopy, glutaraldehyde for electron microscopy and phosphate-buffered saline for immunofluorescence studies. The histopathology was reviewed by a single renal pathologist. Biopsy was performed by a single consultant radiologist in 54 instances or a number of senior registrars instructed and initially supervised by the consultant on 50 occasions. Following the procedure, patients were kept on bed rest for 24 h with regular observations of the puncture site, haemodynamic status and urine analysis. Patients with no signs of complications were discharged the following day if they had attended solely for renal biopsy and were reviewed in the outpatient clinic subsequently.
RESULTS Of 104 attendances for biopsy, 103 resulted in retrieval of adequate material for examination by light microscopy, immunofluorescence and electron microscopy and for achieving a full histological diagnosis. In one case when biopsy was performed by a senior registrar insufficient tissue was obtained (containing three glomeruli) and the biopsy subsequently repeated successfully. An average of 2.6 passes were made at each procedure (range 1-5) and a mean of 13.4 glomeruli were present in each tissue core (range 3-28). There were post-procedure complications in seven instances. Four (3.8%) of these were minor, consisting of macroscopic haematuria; all settled within 48 h without requirement of transfusion or other intervention. Complications were considered to be significant in three (2.9%) cases; two had symptomatic peri-renal haematomas both of whom required transfusions of 2 and 4 units of blood respectively, but no further intervention. The third patient had an arterio-venous fistula which manifested as persistent haematuria requiring transfusion of 7 units of blood and having failed to settle was subsequently successfully embolized following a diagnostic angiogram, in our department. All three cases with significant complications did not have an excessive number of passes during biopsy with two passes each in two cases and three passes in one case. There was no clear difference in complication rate in procedures performed by the consultant (2 minor, t significant of 54 biopsies) compared to those performed by senior registrars (2 minor, 2 significant of 50 biopsies) in this small study, although the latter tended to make a larger number of passes at each procedure (average of 2.8) compared to the consultant (average of 2.4). Length of follow-up ranged from 9 days to 4 years. There were no late complications attributable to PRB.
DISCUSSION A wide variety of methods are available for obtaining renal tissue by pcrcutaneous biopsy [2,5,6,10,12]. The blind method using surface anatomical landmarks [2] was a standard technique but requires experience and has a significant failure and complication rate [13,14]. An inherent difficulty of this technique is accurate localization of the kidney. Although ultrasound is of proven value in localizing the kidney [15,16], many use this modality solely to mark the skin site for insertion of a fine needle to determine depth of kidney by respiratory motion prior to performing the biopsy [17-t9]. The use of an exploring needle followed by the biopsy needle, without direct ultrasound guidance, increases the time of the procedure and runs the risk of capsular laceration especially if the technique is faulty. More recently biopsies have been performed with continuous ultrasound guidance [6,7,9,10], allowing direct visualization of the biopsy needle as it enters the kidney and eliminating the need for an exploring needle to determine depth. In many cases, however, two operators are required; one for ultrasonic guidance and the other for operation of the biopsy needle, usually either a Vim-Silverman or a Tru-Cut needle (Travenol Laboratories, Deerfield, IL). The Biopty device has been a significant advance in procurement of tissue and since its initial description [20], has been widely used in biopsy of many organs [21,22]. It retrieves adequate tissue cores with minimal fragmentation and crush artefact and is safer as the time the needle stays within the target tissue is reduced. By combining single-handed use of this device with real-time ultrasound it is possible for a single operator to perform PRB rapidly, accurately and safely as shown in our series. This is in accordance with early favourable reports [23,24] using this method. It is essential when using this technique to ensure that the needle tip is well within the capsule as the Biopty gun is released. In a few cases when this has not occurred, particularly in patients with tough pcri-renal tissues, we have witnessed the kidney tip or bounce away as the gun is released. This may account for failure to obtain diagnostic specimens when the blind technique is used if perirenal tissues are thick and may also cause capsular laceration. It is also important to perform the whole manoeuvre of advancing the needle into the renal capsule, releasing the Biopty gun and withdrawal all within a single breath-hold. This will avoid lacerations and tears to the kidney caused by respiratory motion. In addition to biopsy technique, sample size is important in achieving a full diagnostic result. We have an experienced scientific officer/technician in attendance to examine the tissue cores as they are obtained and advise on the adequacy of specimens and hence the need or otherwise to make further passes. The probability that the sample accurately reflects the disease process in the kidney depends on both the number of glomeruli in the specimen and the actual fraction of glomeruli affected by disease. With small samples, focal disease may be missed and classification of disease severity is less likely to be accurate [25]. Retrieval of about 10 intact glomeruli is usually required to establish a definitive diagnosis, and as tissue should be examined by light and electron microscopy, and immunofluorescence, two tissue cores are required in most cases [19]. The use of a 14-gauge needle satisfies these requirements with an average of 13.4
RENAL BIOPSY IN DIFFUSE RENAL DISEASE
glomeruli per tissue core in our study. We were thus able to achieve a definitive diagnosis in 99% of our biopsies in comparison with rates ranging from 81 to 95% [5,6,9,14,17] in the literature. Clearly it would be tempting to attempt PRB using smaller needles (e.g. 18-gauge) but given the requirements for a representative histopathological specimen, the diagnostic rates are likely to be lower unless several more passes are made to obtain a large number of smaller cores. Using both 16- and 18-gauge needles, no glomeruli were obtained in 10.7 21.4% of biopsies due to sampling error in one series [9]. In paediatric practice, however, where glomeruli are more tightly packed together within the kidney a smaller specimen obtained by a 18-gauge needle is more likely to be sufficient [26]. In a study comparing 15 different biopsy devices and needles (including the 18-gauge Biopty, but not including the 14-gauge Biopty) in a controlled setting of open renal biopsy, in pigs, the 14-gauge Tru-Cut needle proved to be superior [1 l]. We believe that in PRB for diffuse disease, the 14-gauge Biopty needle combines the advantage of size with superior and safer cutting characteristics. The requirements for a larger specimen needle need to be balanced against a theoretical higher risk of complications. The incidence of complications following PRB may be relatively high if these are actively sought using routine post-procedure imaging studies [27,28]. However the majority of these are clinically silent and resolve spontaneously. As with most clinical studies, we did not perform post-biopsy renal ultrasound or CT routinely and only proceed to further imaging if clinically indicated. The clinical complication rate in our study compares favourably with existing clinical complication rates of PRB in the literature. Our minor complication rate of 3.8% compares with reported rates varying from 5 to 14% [5,8,9,14,16,17]. For significant complications, our rate of 2.9% is within the range of I to 6.4% reported in various series [5,6,8,9,14,16,17]. We believe this to be acceptable particularly in view of the high diagnostic rate in our series. In summary, the use of real-time ultrasound with the 14-gauge Biopty device has many advantages in comparison with more conventional techniques. It is highly accurate and has a low complication rate. The time of procedure is considerably reduced and only a single operator is required. Complications such as a peri-renal haematomas or arterio-venous fistulae can be detected and/or treated by the radiologist performing PRB. We recommend this method of PRB to radiologists and to nephrologists using more conventional techniques. Acknowledgements. We are grateful to Drs M. Goggin and D. Prosser for referring their patients and Mr T. Spice, MLSO, for expert interpretation and handling of specimens during the biopsy procedures. Our thanks also to J. Irons, R. L. Saunders and L. Bolt for secretarial assistance. REFERENCES
1 IversonP, BrunC. Aspirationbiopsyofkidney. AmericanJournalof Medicine 1951;11:324 330.
113
2 Kark RM, Muehrcke RC. Biopsy of kidney in prone position. Lancet 1954;1:1047 1049. 3 Lusted LB, Mortimer GE, Hopper JJ. Needle renal biopsy under image amplifier control. American Journal of Roentgenology 1956;75:953--955. 4 Telfer N, Ackroy DAE, Stock SL. Radio-isotopic localisation for renal biopsy. Lancet 1964;1:132 [33. 5 Mets T, Lameire N, Matthys E, Afschrift M. Sonically guided renal biopsy. Journal of Clinical Ultrasound 1979;7:190 191. 6 Birholz JC, Kasinath BS, Corwin HL. An improved technique for ultrasound guided percutaneous renal biopsy. Kidney International t985;27:80-82. 7 Yoshimoto M, Fujisawa A, Sudo M. Percutaneous renal biopsy well visualised by orthogonal ultrasound application using linear scanning. Clinical Nephrology 1988;30:106 110. 8 Nadel L, Baumgartner BR, Bernardino ME. Percutaneous renal biopsies: accuracy, safety and indications. Urological Radiology 1986;8:67 71. 9 Mostbeck GH, Wittich GR, Derfler K, Ulrich W, Walter RM, Herold C, et al. Optimal needle size for renal biopsy; in vitro and in vivo evaluation. Radiology 1989;173:819-822. 10 Wiseman DA, Hawkins R, Numerow LM, Taub KJ. Percutaneous renal biopsy utilizing real time ultrasonic guidance and a semiautomated biopsy device. Kidney International 1990;38:347 349. 11 Hopper KD, Baird DE, Reddy VV, Landis JR, Parker SH, Tyler HN, et al. Efficacy of automated biopsy gun versus conventional biopsy needles in the pygmy pig. Radiology 1990;176:671 676. 12 Almkuist RD, Buckalew VM. Techniques of renal biopsy. Urological Clinics of North America 1979;6:503 517. 13 Slotkin EA, Madsen PO. Complications of renal biopsy: incidence in 5,000 reported cases. Journal of Urology 1962;87:13 15. 14 Diaz-Buxo JA, Donadio JV. Complications of percutaneous renal biopsy: an analysis of t,000 consecutive biopsies. Clinical NephroIogy 1975;4:223 227. 15 Bolton WK, Tully RJ, Lewis EJ, Ranniger K. Localisation of the kidney by percutaneous biopsy. A comparative study of methods. Annals of Internal Medicine 1974;81:159-164. 16 Health and Public Policy Committee, American College of Physicians. Clinical competence in percutaneous renal biopsy. Annals of Internal Medicine 1988;108:301 303. 17 Chan JCM, Brewer WH, Still WJ. Renal biopsies under ultrasound guidance: 100 consecutive biopsies in children. Journal of Urology 1983;129:103 107. 18 Beaman M. How to perform a renal biopsy. British Journal of HospitaIMedicine 1989;41:158 160, 19 Madaio MP. Renal biopsy. Kidney International 1990;38:529-543. 20 Lindgren PG. Percutaneous needle biopsy. A new technique. Acta Radiologica 1982;23:653 656. 21 Parker SH, Hopper KD, Yakes WF, Gibson MD, Ownbey JL, Carter TE. Image-directed percutaneous biopsies with a biopsy gun. Radiology 1989; 171:663-669. 22 Bernardino ME. Automated biopsy devices: significance and safety. Radiology 1990;176:615 616. 23 Mackenzie JC, Mackay IG, Millar ND, Maclver AG, Penry JB. The Biopty cut procedure for renal biopsies (letter). British Medical Journal 1988;296:1128-1129. 24 Worth DP, Khanna SK, Ubhi CS, Irving HC. The Biopty cut procedure for renal biopsies (letter). Britis'h Medical Journal 1988;296:1603. 25 Corwin HL, Schwartz MM, Lewis EJ. The importance of sample size in the interpretation of renal biopsy. American Journal c?f Nephrology 1988;8:85 89. 26 Poster RB, Jones DB, Spirt BA. Percutaneous pediatric renal biopsy: Use of the biopsy gun. Radiology 1990; 176:725 727. 27 Jorstad S, Borander U, Berg KJ, Wideroe TE. Evaluation of complications due to percutaneous renal biopsy: a clinical and angiographic study. American Journal of Kidney Diseases 1984;4:162-165. 28 Ralls PW, Barakos JA, Kaptein EM, Friedman PE, Fouladian G, Boswell WD, et al. Renal biopsy-related hacmorrhage: frequency and comparison of CT and sonography. Journal o[" Computer Assisted Tomography 1987; t 1: 1031- 1034.
