A new core-biopsy instrument with an end-cut technique provides prostate biopsies with increased tissue yield

Transcription

1 BJU International (2002), 90, A new core-biopsy instrument with an end-cut technique provides prostate biopsies with increased tissue yield L. HÄGGARTH, P. EKMAN and L. EGEVAD* Departments of Urology and *Pathology and Cytology, Karolinska Hospital, Stockholm, Sweden Objective To evaluate the tissue yield of a new core-biopsy needle with an end-cut mechanism when used for transrectal prostate biopsies. Patients and methods The end-cut instrument has an adjustable stroke length (13, 23 or 33 mm) and an inner and an outer cannula, but no trocar. It was compared with the conventional side-notch needle (Biopty, Bard, Covington, GA, USA) with trocar and fixed stroke length (22 mm). In 60 men, octant biopsies were taken including the apex, mid-medial, mid-lateral and basal positions. At random, the left side of the prostate was biopsied with one of the instruments and the right side with the other. The length and weight of the biopsy specimens were measured. In 40 men, the stroke length of the end-cut instrument was set to 23 mm, and in 20 men it was set to 33 mm. Results The weight and the weight per length of the biopsies provided by the end-cut instrument with a stroke length of 23 mm were 18.4% and 13.7%, respectively, greater than in the biopsies from the side-notch instrument (P<0.001). In 21.9% of the attempts no tissue was obtained by the end-cut needle, compared with 1.9% with the side-notch needle. With a 33 mm stroke length, the length and weight of the end-cut biopsies were 38.4% and 33.0%, respectively, greater than in biopsies from the side-notch instrument. Conclusion The end-cut instrument provided a greater tissue yield than the side-notch needle, but a significant number of biopsies were lost. Keywords prostatic neoplasms, pathology, instrumentation, biopsy, ultrasonography, male, human Introduction Of clinically diagnosed prostate cancers, 70 80% are located in the peripheral zone of the prostate in the posterior or posterolateral part of the gland [1]. Most cancers are, thereby, accessible by transrectal biopsies, the most common method to obtain material for a morphological diagnosis of prostate cancer. Transrectal fine needle aspiration cytology guided by palpation is still used but during the last decades TRUS-guided core biopsies have become the dominant method in most countries. The most common biopsy strategy is the sextant protocol [2], but several other biopsy protocols have also been described [3 5]. A widely used device for taking core biopsies is the Biopty instrument (Bard, Covington, GA, USA), a spring-loaded automated instrument constructed by P.G. Lindgren 20 years ago [6]. Briefly, an inner needle (trocar) is advanced when the gun is released (Fig. 1). A side-notch of the inner needle is filled with tissue and an outer cannula is then immediately fired, and cuts a core biopsy from the tissue in the notch. The technique Accepted for publication 6 March 2002 has been evaluated in several studies covering both the prostate and other organs [7 11]. However, there are two major disadvantages with the side-notch needle. First, because of the position of the base of the notch, the full diameter of the outer cannula is not used. Second, the first few millimetres of the needle cuts without being filled with tissue. Several other biopsy instruments have been introduced using essentially the same technique [7 9,11,12]. Recently, a new instrument for core biopsies (BioPince, Amedic, Sollentuna, Sweden) has been developed. The mechanism is fundamentally different from the side-notch technique. Equipped with an end-cut needle with no side-notch trocar, this instrument is claimed to obtain more tissue than the side-notch needle. The aim of this study was to compare the tissue yield of this novel instrument with the side-notch needle when used for transrectal prostate biopsies. Patients and methods The study included 60 men (mean age 64 years, range 29 83) referred to one of the authors (L.H.) for TRUSguided core biopsies of the prostate. In all men but one (a 29-year-old-man biopsied for unclear pelvic pain # 2002 BJU International 51

2 52 L. HÄGGARTH et al. Fig. 1. The upper panel shows the side-notch needle and the lower panel shows the end-cut needle. Longitudinal sections of the needles with biopsy specimens (grey) are seen to the left and cross sections are seen to the right. symptoms), there was a suspicion of cancer, either because the serum PSA level was high or because of a palpable nodule on DRE. Biopsies were taken according to an octant protocol including eight standardized positions (apex, mid-lateral, mid-medial and base) and additional biopsies from TRUS-detected lesions outside these positions [13]. All men were biopsied as outpatients and received prophylactic antibiotic treatment (400 mg norfloxacin twice daily for a week, with the first dose immediately before the examination). TRUS was conducted using the Leopard machine (Bruel & Kjaer, Denmark) with a 8538 transducer and needle holder. The volume of the prostate was estimated using the ellipsoid formula. For the side-notch needle, the Biopty Magnum gun was used, equipped with an 18 G needle with an outer diameter of 1.2 mm and a notch length of 17 mm; the stroke length was 22 mm. The BioPince is a sterile, single-use biopsy gun with two cylindrical cannulae. First, the inner cannula cuts a tissue core (Fig. 1) and immediately thereafter the outer cannula is fired. The tip of the outer cannula pinches off the specimen at the tip of the cutting cannula. The outer diameter of the end-cut needle is 1.2 mm; the stroke length can be adjusted to 13, 23 or 33 mm. In the first 40 men, the stroke length was set to 23 mm. The mean (range) prostate volume in these patients as estimated by TRUS was 52.7 (21 125) ml. Thereafter, a stroke length of 33 mm was used if the prostate was large enough. Among 22 consecutive patients, the prostate was considered sufficiently large in 20 and they were included in the study; their mean prostate volume was 49.1 (26 88) ml. The volumes of the two prostates that were considered too small for the 33 mm stroke length were 19 and 22 ml. In each patient, both the side-notch needle and the endcut needle were used. According to a series of random numbers, the left side of the prostate was biopsied with one of the instruments and the right side with the other. Biopsies from other locations than the eight standardized positions were excluded from the analysis. Each biopsy specimen was immediately fixed in 4% buffered formaldehyde solution in separate containers. After fixation, the biopsies were measured and weighed by one of the authors (L.E.) unaware of which instrument had been used on which side. In most cases, the biopsies were measured on the same day. The biopsy specimens were put on a filter paper for a few seconds to absorb surplus formalin and then weighed (balance range g). The biopsies were processed according to the local routine, and stained with haematoxylin and eosin. The length of each biopsy was measured on the histological section using a ruler. Differences between means were analysed with a paired or unpaired Student s t-test where appropriate. Differences between proportions were analysed using the chi-squared test. P<0.05 was taken to indicate a significant difference. Results In the first 40 patients (end-cut needle, stroke length 23 mm) the biopsies were taken from 320 standardized positions (40r8), i.e. 160 positions biopsied with each instrument. In 35 of the attempts no tissue was obtained with the end-cut needle, giving a failure rate of 21.9% (35/160). In 12 of these a biopsy specimen was obtained after a second biopsy. With the side-notch needle the failure rate was 1.9% (three of 160), two of which were successfully repeated. This difference was highly significant (P<0.001). Of the 40 patients 23 (58%) had cancer in any biopsy. A nearly equal proportion of the biopsies from the end-cut and side-notch instrument contained cancer, i.e. 24.1% (33/137) and 23.9% (38/159), respectively. To analyse length and weight the biopsy specimens were compared in pairs; pairs with one biopsy missing were omitted from the analysis. The length, weight and weight per length of the biopsies are shown in Table 1. The length of the biopsy specimens from the two instruments did not differ significantly (P=0.42) but when measured on the glass slides, the biopsies from the end-cut needle were a mean of 1.3 mm (9.5%) longer than those from the side-notch needle (P=0.042). By using the end-cut needle, the length of the specimen that is not seen in the histological section was decreased from 18.5% (3.2/17.3) to 13.5% (2.4/17.8). The weight and the weight/length of the biopsies from the end-cut needle were 18.4% and 13.7%, respectively, more than those from the side-notch needle (P<0.001). The mean length, weight and weight/length of biopsies containing cancer and benign biopsies are also shown in Table 1. The specimen length did not differ significantly (P=0.69), but the weight and weight/length of biopsies positive for cancer were greater than in benign biopsies

3 EVALUATION OF A NEW INSTRUMENT FOR PROSTATE BIOPSIES 53 Table 1 The length, weight and weight per length of biopsy specimens, length of biopsies as measured on the glass slide (23 mm stroke length of the end-cut needle), for those containing cancer and benign tissue in the first 40 patients, with position, and for the 33 mm stroke length Mean (range) specimen Group length, mm weight, mg weight/length, mg/mm On slide length, mm End-cut, 23 mm 17.8 (1 36) 6.5 ( ) ( ) 15.4 (1 27) Side-notch 17.3 (6 27) 5.50 (1.7 10) ( ) 14.1 (5 22) P 0.4 <0.001 < Histology Cancer Benign P 0.69 <0.001 <0.001 Position* Apex Mid lateral Mid medial Base Stroke length End-cut, 33 mm 19.9 (1 34) 6.53 ( ) 16.0 Side-notch 14.4 (8 27) 4.91 ( ) 11.8 P <0.001 <0.001 <0.001 * The length and weight of the apex and mid lateral biopsies were significantly less than for the base and mid medial biopsies (P=0.034 and 0.002, respectively). There was a trend towards higher weight/length of the base biopsies than the other biopsies, although this was not significant (P=0.052). (19.8% and 17.2%, respectively, P<0.001). The same was true when specimens from the two biopsy instruments were analysed separately. The mean length, weight and weight/length of biopsies with position are also shown in Table 1. The biopsy specimens from the apex and mid lateral positions were smaller than the other biopsies, both in length and weight (P=0.034 and 0.002, respectively). The base biopsies had a greater weight/length than the other biopsies but not significantly so (P=0.052). In the last 20 patients (end-cut needle stroke length 33 mm), the length and weight of the end-cut biopsies were 38.4% and 33.0%, respectively, greater than in biopsies from the side-notch instrument with a stroke length of 22 mm (P<0.001; Table 1). When measured on the glass slides, the biopsies from the end-cut needle in this set of patients were 35.7% longer than those from the side-notch needle (Table 1). With the end-cut needle, the biopsy specimen was lost in 22.5% of the attempts, compared with none of the attempts with the side-notch needle. Only two patients reported complications; one had prolonged antibiotic treatment and an indwelling catheter for a UTI with haematuria, and the other had urinary retention that resolved spontaneously. Discussion The main purpose of taking transrectal prostatic biopsies is to detect prostate cancer. Evidently, it is desirable to obtain as much tissue as possible with a minimum of trauma. Many studies have been published on the benefit of different biopsy protocols, but less has been done to improve the needle biopsy technique. The dominating needle type has a trocar with a side-notch and provides a reliable tissue yield [6]. However, because of the trocar, the full diameter of the cannula cannot be filled. As an alternative, circular needles with no trocar, similar to bone marrow biopsy needles, have been suggested. However, a problem with this technique is how to cut the inner end of the biopsy specimen before the instrument is withdrawn. Recently, the new circular end-cut needle introduced solved the problem of cutting the inner end of the biopsies by adding a thin outer cannula with a sharp edge that pinches off the biopsies. In the present study the biopsies were compared pairwise to ensure equal conditions of prostate anatomy and tissue texture. The biopsies obtained with the end-cut instrument were a mean of 18.4% heavier than those provided by the side-notch needle with comparable stroke length. Furthermore, the mean weight/length was 13.7% greater in end-cut biopsies, indicating that these biopsies were thicker than the conventional biopsies. This is potentially useful because it is important to have sufficient material in the paraffin blocks for re-sectioning and for immunohistochemistry. In this small series, the proportion of biopsies containing cancer was essentially the same in specimens obtained with the two instruments. However, theoretically, an increased tissue yield also improves the likelihood of detecting cancer. A major

4 54 L. HÄGGARTH et al. concern with the end-cut instrument is that no tissue was obtained in a significant proportion of attempts, despite the outer cannula, the failure rate being 21.9% with the end-cut needle and 1.9% with the side-notch needle. Another result from the biopsies provided by the endcut instrument is that the thicker biopsies with a circular cross-section (Fig. 1) are less prone to curve, and hence flat embedding in the paraffin block is easier. Despite the essentially equal mean length of the specimens from the two instruments, the biopsies from the end-cut needle were a mean of 1.3 mm longer than those from the sidenotch needle, when measured on the slides. The most likely explanation for this discrepancy is that more tissue can be displayed in the same section because of a flatter embedding of the end-cut specimens. Flat embedding in the paraffin block is important to prevent unnecessary loss of material when the blocks are cut [14,15]. Examining a section from one level misses 23.4% of the total biopsy length [16]; this can be compensated for by cutting three levels at 25%, 50% and 75% through the biopsy block [16]. Cutting several slides from each block also increases the diagnostic yield [17,18], but unfortunately, much tissue is thereby lost. If immunohistochemistry is necessary, e.g. to confirm a malignant diagnosis, there is an obvious risk that the suspicious area no longer remains in a block that has been levelled. In some laboratories, this is compensated for by routinely taking an intervening section between the levels. According to Green et al. [19], the intervening section was necessary to establish a diagnosis in 2.8% of cases, but this is a time-consuming and relatively expensive routine in a busy practice. By using the endcut needle, the length of the specimen that is not seen in the histological section is decreased from 18.5% to 13.5%. The potential length of the notch of the side-notch needle is limited; if the notch and the stroke are too long, there is a risk that the outer cannula hits the notch and curves the needle, so that it will become stuck in the tissue. One of the advantages compared to the side-notch instrument is that longer biopsies can be obtained when the anatomy allows the use of the 33 mm stroke length. When the stroke length was set to 33 mm, the end-cut biopsies were 38.4% longer than the biopsies from the side-notch instrument. This is obviously useful for detecting cancers in the anterior prostate. Interestingly, some of the present specimens were actually longer than the notch length. There are two possible explanations for this. First, the prostate tissue is elastic and may be stretched when measured. Second, when the needle is pushed through the prostate, the tissue is compressed into the needle, and when the specimen is collected from the biopsy instrument, it expands and regains its original size. Although not the main purpose of this study, the biopsies positive for cancer were compared with benign biopsies. Interestingly, biopsies containing cancer were on average heavier than benign biopsies, possibly reflecting differences in tissue composition. Prostate tissue containing prostate cancer contains a greater proportion of epithelium. Furthermore, particularly if the cancer is poorly differentiated, there are more solid epithelial masses with no lumina. The biopsies from the apex and from the mid lateral position were on average shorter than the other biopsies. The reason for this is unclear but a possible explanation is that the sagittal diameter of the prostate narrows in the apical and far lateral part of the gland. The approximate price of the Biopty gun and needle (January 2002) is 1060 and 13 Euro, respectively, excluding taxes; that of the BioPince instrument is 28 Euro. Hence, with one needle per patient, the BioPince instrument will be more expensive than the Biopty instrument after <70 patients. In conclusion, the end-cut instrument provides thicker biopsies and flatter embedding of the specimens in the paraffin block. In addition, the greater potential stroke length provides longer biopsies. However, the single-use end-cut instrument is more expensive than the sidenotch needle. More importantly, there remains a need to improve the mechanism for pinching off the specimens. At present we recommend the new instrument mainly for situations where there is a suspicion of transition zone cancer (elevated serum PSA, negative DRE and previous negative core biopsies) or cancer in large prostates. The end-cut instrument can also be considered when it is particularly important to obtain sufficient material for immunohistochemistry. Acknowledgement This study did not receive any financial support but the BioPince needles were supplied free of charge by Amedic. The study was approved by The Independent Ethics Committee of The Karolinska Institute. References 1 McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol 1988; 12: Hodge KK, McNeal JE, Terris MK, Stamey TA. Random systematic versus directed ultrasound guided transrectal core biopsies of the prostate. J Urol 1989; 142: Eskew LA, Bare RL, McCullough DL. Systematic 5 region prostate biopsy is superior to sextant method for diagnosing carcinoma of the prostate. J Urol 1997; 157:

5 EVALUATION OF A NEW INSTRUMENT FOR PROSTATE BIOPSIES 55 4 Terris MK, Pham TQ, Issa MM, Kabalin JN. Routine transition zone and seminal vesicle biopsies in all patients undergoing transrectal ultrasound guided prostate biopsies are not indicated. J Urol 1997; 157: Babaian RJ, Toi A, Kamoi K et al. A comparative analysis of sextant and an extended 11-core multisite directed biopsy strategy. J Urol 2000; 163: Lindgren PG. Percutaneous needle biopsy. A new technique. Acta Radiol Diagn (Stockh) 1982; 23: Braeckman J, Corujeira-Figueira F, Goossens A, Keuppens F. Ultrasonically guided prostatic biopsy: technical improvements. Eur Urol 1990; 17: Helbich TH, Rudas M, Bohm G et al. Randomized in vitro and in vivo evaluation of different biopsy needles and devices for breast biopsy. Clin Radiol 1999; 54: Hopper KD, Baird DE, Reddy VV et al. Efficacy of automated biopsy guns versus conventional biopsy needles in the pygmy pig. Radiology 1990; 176: Ragde H, Aldape HC, Bagley CM. Ultrasound-guided prostate biopsy. Biopty gun superior to aspiration. Urology 1988; 32: Wagner HJ, Barth P, Schade-Brittinger C, Plein S, Klose KJ. Postmortem evaluation of four randomly selected automated biopsy devices for transthoracic lung biopsy. Cardiovasc Intervent Radiol 1995; 18: Hopper KD, Abendroth CS, Sturtz KW, Matthews YL, Stevens LA, Shirk SJ. Automated biopsy devices: a blinded evaluation. Radiology 1993; 187: Norberg M, Egevad L, Holmberg L, Sparen P, Norlen BJ, Busch C. The sextant protocol for ultrasound-guided core biopsies of the prostate underestimates the presence of cancer. Urology 1997; 50: Rogatsch H, Moser P, Volgger H et al. Diagnostic effect of an improved preembedding method of prostate needle biopsy specimens. Hum Pathol 2000; 31: Rogatsch H, Mairinger T, Horninger W, Gschwendtner A, Bartsch G, Mikuz G. Optimized preembedding method improves the histologic yield of prostatic core needle biopsies. Prostate 2000; 42: Lane RB, Lane CG, Mangold KA, Johnson MH, Allsbrook WC. Needle biopsies of the prostate: what constitutes adequate histologic sampling? Arch Pathol Lab Med 1998; 122: Renshaw AA. Adequate tissue sampling of prostate core needle biopsies. Am J Clin Pathol 1997; 107: Reyes AO, Humphrey PA. Diagnostic effect of complete histologic sampling of prostate needle biopsy specimens. Am J Clin Pathol 1998; 109: Green R, Epstein JI. Use of intervening unstained slides for immunohistochemical stains for high molecular weight cytokeratin on prostate needle biopsies. Am J Surg Pathol 1999; 23: Authors L. Häggarth, MD, FEBU, Consultant. P. Ekman, MD, PhD, Professor, Consultant. L. Egevad, MD, PhD, Associate Professor, Consultant. Correspondence: L. Egevad, Department of Pathology and Cytology, Karolinska Hospital, SE Stockholm, Sweden. Lars.Egevad@onkpat.ki.se