Ultrasound Guided Transrectal Core Biopsies of the Palpably Abnormal Prostate

0022-534 7 /89/1421-0066$02.00/0 Vol. 142, July

THE JOURNAL OF UROLOGY

Printed in U.S.A.

Copyright© 1989 by Williams & Wilkins

ULTRASOUND GUIDED TRANSRECTAL CORE BIOPSIES OF THE PALPABLY ABNORMAL PROSTATE KATHRYN K. HODGE, JOHN E. McNEAL

AND

THOMAS A. STAMEY*

From the Division of Urology, Stanford University Medical Center, Stanford, California

ABSTRACT

Ultrasound imaging and ultrasound guided transrectal core biopsies were performed in 251 consecutive men with abnormal prostates on digital rectal examination. A hypoechoic defect on ultrasound was identified in 227 of 251 patients (90 per cent) corresponding to the area of palpable nodularity or abnormal firmness. A mean of 6.25 biopsies were obtained per patient using a commercially available spring-loaded gun. Biopsies were positive for cancer in 165 of the 251 prostates (66 per cent). Palpable nodules more often were hypoechoic and more often contained cancer than less distinct areas of abnormal firmness on digital examination. Among the clinical stages Bl, B2 and B3 nodules 70, 76 and 88 per cent, respectively, were positive for cancer, as were 100 per cent of the clinical stage C prostates. Of 77 abnormally firm, nonnodular prostates 36 per cent were positive for cancer. Random biopsy of the contralateral normal lobe in 56 men with clinical stage Bl or B2 nodules showed cancer present contralaterally in 42 and 60 per cent, respectively; 20 per cent had positive biopsies despite a contralateral isoechoic ultrasound. In 78 patients with prior digitally guided biopsies, ultrasound guided biopsies confirmed previously diagnosed cancers in 94 per cent. However, in 23 of 43 patients (53 per cent) with previous negative digitally guided biopsies, ultrasound guided biopsies made the new diagnosis of cancer. Complications, including post-biopsy fever and bleeding, occurred in 6 of 251 patients (2.4 per cent). The combination of the new spring-loaded biopsy guns and transrectal ultrasound guidance of biopsies provides the urologist with a tool that allows multiple prostate cores to be obtained safely and painlessly, reducing the sampling error and increasing the accuracy in diagnosing prostate cancer in the-man with a palpable abnormality of the prostate. (J. Ural., 142: 66-70, 1989) defects prove to be cancer on biopsy? When palpable abnormalities are separated into those with well defined nodular induration as opposed to those with less distinct but definitely abnormal firmness, what percentage of each are hypoechoic on ultrasound and what is the incidence of biopsies positive for cancer in each category? Of all biopsy proved cancers what percentage correspond to hypoechoic defects? What percentage correspond to the area of the palpable abnormality? What is the incidence of cancer in the palpably normal prostate lobe, contralateral to the stage Bl or B2 nodule? What is the cancer detection rate in patients with a previously negative, digitally guided, perineal or transrectal core biopsy at repeat biopsy with ultrasound guidance?

Modern ultrasound technology has greatly expanded our potential capability for diagnostic biopsy of prostate cancer transperineally'- 5 and transrectally. 6 • 7 However, the exact role of this new technology in cancer diagnosis has not yet been clearly established. Ultrasound findings must be evaluated more precisely in cases with different palpable abnormalities on digital rectal examination, and these results must be confirmed with histological information of the best possible quality. Two independent and unrelated developments have combined to give the urologist an unprecedented opportunity to obtain transrectally 1.5 cm. needle core biopsies of the prostate with a safe and highly accurate technique. Spring driven, 18 gauge needlest fire so rapidly that they are essentially painless transrectally and the 1.5 cm. prostate tissue cores are taken with minimal crush artifact. Newer transrectal ultrasound probes currently allow biopsy of the prostate in the sagittal plane with or without an intervening balloon. Combining these new technologies with digital rectal examination with the patient in the knee-chest position provides the urologist a powerful tool for the diagnosis of prostate cancer. By combining ultrasound evaluation of palpable prostate abnormalities with precisely targeted biopsies and multiple random biopsies, we have sought to answer several questions regarding the optimum role of digital rectal examination and transrectal, ultrasound guided biopsies in prostate cancer diagnosis. What percentage of prostates with palpable abnormalities on digital rectal examination with the patient in the kneechest position are associated with hypoechoic (echopenic) defects on transrectal ultrasound? How many of these hypoechoic

METHODS

A total of 251 consecutive patients with a mean age of 67 years (range 41 to 84 years) had abnormal prostates on digital rectal examination performed with the patient in the kneechest position. All 251 patients underwent transrectal biopsy with ultrasound guidance. Of the 251 patients 156 had distinct nodules on digital rectal examination. The nodules were classified as stage Bl, B2 or B3.8 A stage Bl nodule was defined as compressible on at least 2 sides and extending for a distance equal to or less than half of 1 prostate lobe. A stage B2 nodule was larger than a stage Bl tumor but it occupied not more than 1 full lobe. Stage B3 disease either involved an entire lobe with extension across the midline or represented bilaterally palpable disease. A total of 18 patients with clinical stage C disease exhibited palpable extension above the prostate, presumably into 1 or both seminal vesicles. There were 77 patients who had no palpable nodules on digital rectal examination but whose prostates were abnormally firm. We have divided these further into those with symmetrical

Accepted for publication October 26, 1988. Supported in part by the Richard M. Lucas Cancer Foundation. *Requests for reprints: Division of Urology (S287), Stanford University Medical Center, 300 Pasteur Dr., Stanford, California 94305-5118. t Biopty, C.R. Bard, Inc., Covington, Georgia. 66

67

ULTRASOUND GUIDED TRANSRECTAL CORE BIOPSIES OF ABNORMAL PROSTATE

malities on digital rectal examination a hypoechoic defect corresponding to the area of the nodule or abnormal firmness was seen in 227 patients (90 per cent) (table 1). Of the 251 abnormalities 165 (66 per cent) were positive for cancer on biopsy. Among the 156 patients with clinical stage B nodules cancer was found in 76 per cent (119 of 156). Equally important is the observation that almost all palpable nodules were seen as hypoechoic defects on ultrasound (91 and 92 per cent of the stages Bl and B2 nodules, and 100 per cent of the stage B3 nodules). Moreover, every nodule positive on biopsy was found to correspond with a hypoechoic defect on ultrasound. Among the 76 per cent of biopsy positive clinical stage B nodules, all but 3 were positive for cancer on biopsy in the specific area of the palpable nodule (table 1, last column). Two stage Bl and 1 stage B2 nodules were negative on biopsy in the precise area of the nodule but biopsies elsewhere within the gland were positive for cancer, accounting for the small disparity among stages Bl and B2 nodules between the number of positive biopsies and the number of positive biopsies from the specific area of the palpable nodule. Palpable nodules more often were hypoechoic (94 per cent) and positive on biopsy (76 per cent) than less distinct abnormally firm areas (82 and 36 per cent, respectively, table 1). As might be expected, the asymmetrically firm prostates, whether the firmness occurred in a prostate judged to be of normal size or enlarged, had a higher incidence of positive biopsies (20 of 50, or 40 per cent) than did symmetrically firm prostates (8 of 27, or 30 per cent) although the difference was statistically insignificant (p >0.18). Over-all, 28 of 77 abnormally firm prostates (36 per cent) were positive for cancer. Of the 28 positive biopsies 24 (86 per cent) corresponded to the area of firmness on palpation (table 1, last column), while only 20 of the 28 cancers (71 per cent) corresponded to a hypoechoic defect. Of the 28 cancers 8 (29 per cent) wne detected by random biopsies of normal-appearing, isoechoic tissues. The positive biopsy rate for all hypoechoic defects was 32 per cent in abnormally firm prostates, 82 per cent in stage B nodules and 100 per cent in clinical stage C disease. All but 5 of 41 stage B3 and all 18 clinical stage C prostates had biopsies positive for cancer, as well as obvious hypoechoic defects (table 1). The hypoechogenicity, especially in stage C prostates, often was diffuse and patchy in comparison to the distinct defects seen with smaller nodules. Biopsy of the contralateral lobe in clinical stages Bl and B2 nodules. Of 56 stages Bl and B2 nodules positive for cancer in which the contralateral palpably normal lobe was randomly biopsied from apex to base (table 2) 17 patients (30 per cent) had positive biopsies associated with a hypoechoic defect. However, 11 additional patients (20 per cent) had positive contra-

or asymmetrical firmness and whether the prostate was judged to be normal in size or enlarged. Of these 251 patients 78 had previously undergone trans rectal or transperineal biopsy with only the index finger placed transrectally as a guide for the biopsy. Of these 78 digitally guided biopsies 35 (45 per cent) had been positive for cancer, while 43 (55 per cent) were negative for cancer. All 78 patients were biopsied again with ultrasound guidance. Among the 115 patients with stages Bl and B2 nodules 56 had biopsies of the contralateral palpably normal lobe. These biopsies were performed at the apex, middle and base of the contralateral prostate lobe (in the mid lobe of the parasagittal plane) regardless of the ultrasound findings. A total of 28 patients with clinical stage B2, 21 with stage B3 and 11 with clinical stage C disease had biopsies of the seminal vesicles. Transrectal ultrasound imaging of the prostate was performed with the Brue! & Kjaer 1846 console. Each patient was scanned initially with the 7.0 MHz. transverse (radial) 1850 transducer. Subsequent sagittal scanning and biopsies were performed with the 7.0 MHz. 8537 (side-viewing) sector scanner in the majority of cases but sagittal and transverse biopsies were accomplished with the 8538 (forward viewing) sector scanner in some instances. An hour before biopsy all patients received a Fleets biphosphate enema and a 400 mg. tablet of norfloxacin for antimicrobial prophylaxis. A second norfloxacin tablet was given on the evening of the biopsy. All biopsies were performed with 18 gauge "tru-cut" type biopsy needles driven by the spring-loaded gun. An average of 6.2 biopsies were taken per patient; while we focused on areas containing a palpable abnormality or a suspicious hypoechoic area on scanning, most patients had some random (but not systematic) biopsies taken from isoechoic areas of the peripheral and central zones. All biopsy specimens were placed in 10 per cent formalin and processed with hematoxylin and eosin staining. When cancer or dysplasia was identified the extent of involvement along the tissue core was estimated in millimeters. All cancers were graded with the Gleason system;" dysplasia, a pre-malignant lesion, was defined in accord with previously reported criteria.'" Tissue cores that did not contain cancer or dysplasia were categorized as normal, post-inflammatory atrophy, cystic atrophy and acute or chronic inflammation'' provided that at least 30 per cent of the core was involved by 1 of these tissue patterns. In addition, the absence of glands or ducts in the biopsy core was noted. RESULTS

Correlation among palpable abnormalities, hypoechoic defects and positive biopsies. Of the 251 prostates with palpable abnorTABLE

l. Hypoechoic defects and positive biopsies in patients with abnormal prostates on digital rectal examination No. Pts.

No. Hypoechoic Defects(%)

No. Pos. Biopsies (%)

No. Cancers Corresponding to Hypoechoic Defects/Total

(%) Symmetrical firmness in a normal sized prostate Symmetrical firmness in an enlarged prostate Asymmetrical firmness in an enlarged prostate Asymmetrical firmness in a normal sized prostate Subtotals

Stage of nodule: Bl

B2 B:l Subtotals C

Totals

Pos. Biopsy Rate for all Hypoechoic Defects No./Total (%)

No. Cancers Corresponding to Firm or Nodule Area/Total (%)

16

11

(69)

4 (25)

4/4

(100)

4/11

(36)

4/4

(100)

11

8

(7:3)

4

(:l6)

:3/4

(75)

3/8

(38)

4/4

(100)

2:3

20

(87)

9

(39)

7/9

(78)

7/20

(35)

8/9

(89)

27

24

(89)

6/11

(55)

6/24

(25)

8/11

(7:l)

77

6:3

(82)

28

(36)

20/28

(71)

20/63

(32)

24/28

(86)

60 (91) 45 (92) 41 (100) 146 (94) 18 (100) 227 (90)

46

(70)

46/46 (100) 37 /37 (100)

46/60

(77)

37/45 36/41 119/146 18/18 157 /227

(82) (88) (82)

44/46 36/:17 36/:36 116/119 18/18 158/165

(96) (97) (100) (97) (100) (96)

66

49 41 156 18 251

~

37 (76) :J6 (88) 119 (76) 18 (100) 165 (66)

:36/36 119/119 18/18 157 /165

(100)

(100) (100) (95)

(100)

(69)

68

HODGE, MCNEAL AND STAMEY

lateral biopsies despite a normal-appearing ultrasound on that side. Thus, 42 per cent (13 of 31) of the stage Bl nodules had concomitant contralateral cancer, compared to 60 per cent (15 of 25) of the stage B2 nodules. Seminal vesicle biopsies. Seminal vesicles were defined as abnormal on transrectal ultrasound either because they were grossly displaced anteriorly or superiorly, or because they were thickened and poorly defined or asymmetrical. Twenty abnormal and 40 normal-appearing pairs of seminal vesicles in clinical stages B2, B3 and C cases were biopsied. Of 20 abnormalappearing seminal vesicles on transrectal ultrasound 14 (70 per cent) were positive for cancer (table 3). Of 40 patients with normal-appearing seminal vesicles 7 (18 per cent) also were positive for cancer on biopsy of seminal vesicles; 4 of these 7 were patients with clinical stage C disease. Noncancerous hypoechoic defects. In the combined categories of firm and nodular prostates shown in table 1 a total of 83 hypoechoic defects showed no evidence of cancer (some patients had more than 1 hypoechoic defect in the same prostate). These benign areas, occupying at least 30 per cent of each core, represented post-inflammatory atrophy in 51 per cent (42 of 83), normal tissue in 30 per cent (25 of 83), no glandular tissue in 10 per cent (8 of 23), dysplasia in 4 per cent (3 of 83), acute or chronic inflammation in 4 per cent (3 of 83) and cystic atrophy in 2 per cent (2 of 83). These relative percentages did not vary significantly among the clinical subcategories based on digital rectal examination of the prostate. Digitally guided versus ultrasound guided prostate biopsies. Of 78 patients who had undergone previous transrectal or, rarely, transperineal digitally guided biopsies 35 (45 per cent) were positive for cancer. Ultrasound guided, transrectal prostate biopsies were able to confirm a positive result in 33 of the 35 patients (94 per cent) with prior biopsies positive for cancer (table 4). However, of the 43 patients with negative digitally guided biopsies 23 (53 per cent) were positive when biopsied with ultrasound guidance (table 4). Complications of ultrasound guided needle biopsies. Of the 251 patients biopsied 2 (0.8 per cent) experienced chills and fever despite norfloxacin prophylaxis. Both patients were hospitalized as emergencies and placed on intravenous cefoxitin and gentamicin, and both were discharged from the hospital the following day. Three patients (1.2 per cent) had rectal mucosal bleeding within 2 hours of the biopsy that required anoscopy with suturing of a single arterial bleeder in the outpatient department. All 3 patients were taking aspirin or another TABLE 2.

nonsteroidal anti-inflammatory agent. Only 1 of the 251 patients had urinary retention after biopsy that required overnight catheterization. Therefore, the over-all observed complication rate was 2.4 per cent (6 of 251) with no major morbidity or mortality. DISCUSSION

Eighty-two per cent (63 of 77) of abnormally firm prostates and 94 per cent (146 of 156) of palpable nodules were seen as relative hypoechoic defects in peripheral or central zone tissue on transrectal ultrasound (table 1). These data mean that when the urologist identifies a palpable abnormality on digital rectal examination with the patient in the knee-chest position he can, with few exceptions, be confident of finding it on ultrasound. Of the clinical stage B nodules and the abnormally firm prostates 76 and 36 per cent, respectively, were positive for cancer on ultrasound guided core biopsies; all but 3 of the 119 positive clinical stage B nodules were positive in the precise area of the palpable nodule, while 24 of the 28 abnormally firm prostates positive for cancer were positive on biopsy in the area of the abnormal firmness. Random but nonsystematic biopsies of isoechoic peripheral and central zones of the prostate account for the detection of those cancers that could not be seen on ultrasound scanning. While every biopsy positive clinical stage B nodule had a corresponding hypoechoic defect on ultrasound, it is significant that 8 of the 28 cancers (29 per cent) discovered in abnormally firm but nonnodular prostates were found only by random biopsies of normal-appearing, isoechoic tissues. Shinohara and associates reported an even higher incidence (39 per cent) of isoechoic cancers of significant volume in their radical prostatectomy series. 12 A positive biopsy rate of 36 per cent in abnormally firm prostates is an important observation because many of these prostates might be watched rather than biopsied. Interestingly, among the 4 categories of abnormal firmness in table 1, asymmetrical induration in a normal sized prostate showed the highest positive biopsy rate (41 per cent), while symmetrical firmness in a normal sized prostate had the lowest positive biopsy rate (25 per cent). Although the difference in these categories is not statistically significant (p >0.13) because of the small numbers, the fact that the highest positive biopsy rate was found in normal sized prostates with asymmetrical firmness or induration seems logical. Moreover, irrespective of whether the prostate is normal in size or enlarged as estimated by digital rectal examination, asymmetrical firmness carries a

Contralateral lobe cancer in prostates with stages Bl and B2 nodules

------------------Stage

Bl contralat. B2 contralat. Totals

Contralat. Lobe Findings

No. Pts.

Ultrasound Pos.,* Biopsies Pos. t No.(%)

Ultrasound Neg.,:j: Biopsies Pos. t No.(%)

Ultrasound Pos.,* Biopsies Neg.§ No.(%)

Ultrasound Neg.,:j: Biopsies Neg.§ No.(%)

31 25

8 (26) 9 (36) 17 (30)

5 (16) 6 (24) 11 (20)

6 (19) 3 (12) 9 (16)

12 (39) 7 (28) 19 (34)

56

* Hypoechoic defect seen on ultrasound. t Biopsies positive for cancer. :j: Normal ultrasound appearance (isoechoic in peripheral and central zone). § Biopsies negative for cancer. TABLE

3. Seminal vesicle biopsies in 60 patients with clinical stages B2, B3 and C disease Seminal Vesicle Findings

Clinical Stage B2 B3 C Totals

No. Pts.

Ultrasound Pas.,* Biopsies Pos. t

Ultrasound Pos.,* Biopsies Neg.§

Ultrasound Neg.,:j: Biopsies Pos.t

Ultrasound Neg.,:j: Biopsies Neg.§

28 21 11

2

2

7 5

3 1

3 0 4

11 1

60

14

6

7

33

* Abnormal (see text) seminal vesicles seen on ultrasound. t Biopsies positive for cancer. :j: Normal ultrasound appearance. § Biopsies negative for cancer.

21

ULTRASOUND GUIDED TRANSRECTAL CORE BIOPSIES OF A:SNORivJAL PROSTATE TABLE 4.

Digitally guided versus ultrasound guided prostate biopsies in 78 patients Ultrasound Guided Biopsy Results

Digitally Guided Biopsy Results

No. Pos.

Pos., 35 pts. Neg., 43 pts.

No. Neg.

(%)

(%)

33 (94) 23 (53)

2 (6)

20 (47)

Projected cancer detection rates based on the palpably abnormal versus normal or benign enlarged prostates

TABLE 5.

Digital Rectal Examination

No.

No. With Hypoechoic Defects(%)

No. With Biopsy Proved Ca(%)

Pos. Biopsy Rate(%)

Normal Benign prostatic hyperplasia Normal+ benign prostatic hyperplasia

75 54 129

27* (36) 24* (44) 51* (40)

~

13 12

Abnormal

251 *

227 (90)

165 (66)

66

3 (4) 3 (6)

11

* Number of patients biopsied for calculation of positive biopsy rates.

higher positive biopsy rate (40 per cent) than symmetrical firmness (30 per cent), perhaps because the category of symmetrical firmness includes some prostates with benign prostatic hyperplasia that are unusually firm. Our positive biopsy rate of 76 per cent in clinical stage B nodules and 100 per cent in clinical stage C disease is the highest ever reported for core biopsies, whether obtained either the perinea! or transrectal route. 1 " 16 We believe this is due primarily to our demonstration that 90 per cent of the palpable abnormalities in the prostate can be identified on the ultrasound scan as a hypoechoic area and, therefore, they are amenable to the accuracy of ultrasound guided, transrectal core biopsies. Equally important may be our ability to take multiple random cores safely and painlessly. Some of our success also may be due to careful recording of the precise area of the abnormality by palpating the prostate with the patient in the knee-chest position immediately before introducing the ultrasound transducer. The knee-chest position places the prostate well above the abdominal venous system, allowing the periprostatic veins to empty-a fact easily demonstrable by ultrasound. Ultrasound scanning of the prostatic lobe contralateral to stages Bl and B2 nodules showed that 46 per cent had contralateral hypoechoic defects and that 65 per cent (17 of 26) of these defects were positive for cancer on biopsy (table 2). More importantly, 11 of 30 contralateral lobes (37 per cent) that appeared to be normal on ultrasound scanning of the peripheral and central zones also had cancer on random biopsies of that lobe. This not only confirms the fact that not all peripheral zone prostate cancers can be palpated on digital rectal examination" but it also provides concrete evidence that ultrasound scanning with transverse and sagittal 7.0 MHz. transducers cannot detect all cancers in the prostate, even in the peripheral zone. Of 56 stage Bl or B2 nodules 50 per cent had simultaneous cancer in the contralateral lobe that was palpably normal on digital rectal examination; in 30 per cent cancer was detected as a hypoechoic defect on ultrasound but in 20 per cent cancer was found only by random biopsies of prostatic tissue that was isoechoic and appeared to be normal on ultrasound. The significance of this information for the individual patient must await correlation with our histological reconstructions of radical prostatectomy specimens but it clearly has serious implications for those who advocate ultrasound screening to detect prostate cancer. We believe that these data support the use of these combined technologies as the most accurate means yet available to biopsy the prostate that has been determined to be abnormal on digital

69

rectal examination. In addition to the accuracy nrnuu,,,r1 transrectal ultrasound guidance of needle biopsies, the ability the spring-driven needle to take multiple cores provided biopsy systems has the advantage of further reducing sampling error, including the risk of missing a cancer altogether or, more subtly, obtaining a cancer sample with an unrepresentative Gleason grade in a single biopsy. Moreover, linear measurements in millimeters of how far the cancer extends along the 1.5 cm. core, as well as the spatial distribution of the positive cores within the prostate, yield valuable information on the volume of intraprostatic cancer, a major determinant of prognosis and, therefore, therapeutic decisions. 8 · 18 Whatever other advantages these technologies may have over digitally guided biopsy cores or even needle aspiration, it is clear that in the presence of negative core biopsies of palpable abnormalities it is mandatory to repeat the biopsies under ultrasound guidance (table 4). Of 43 patients with previously negative digitally guided biopsies 53 per cent had cancer under transrectal ultrasound guidance. Many of these patients had nodules that had been biopsied on several occasions by different urologists with and without general or regional anesthesia. The key is being able to see the prostate on the ultrasound screen and to place the needle accurately in the hypoechoic defect representing the palpable abnormality. Many of these defects are close to the capsule, extending less than 1 cm. in the anterior-posterior diameter, and they can be missed easily by blindly passing biopsy core or aspiration needles through the cancer into noncancerous tissue anterior to the palpable abnormality. We frequently biopsy the seminal vesicles not only to confirm clinical stage C disease but also to determine pathological stage C disease in clinical stages B2 and B3 nodules. Transrectal ultrasound cannot detect microscopic invasion of the seminal vesicles. When the seminal vesicles are abnormal on ultra sound, they usually are grossly invaded by prostate cancer, even though such extensive invasion may not be detected on digital rectal examination. We have shown that abnormal-appearing seminal vesicles on transrectal ultrasound can be biopsied safely and have a high likelihood (70 per cent) of being positive on biopsy in the presence of clinical stages B2, B3 and C cancer (table 3). Equally important, however, is the observation that 1 in 5 normal-appearing seminal vesicles on transrectal ultrasound has microscopic cancer invasion detectable by biopsy. Our practice is to take 2 or 3 biopsies from each seminal vesicle if the information will change our management of the patient. As would be expected, the incidence of seminal vesicle invasion (table 3) increases precipitously with increasing clinical stage18 per cent in stage B2, 33 per cent in stage B3 and 82 per cent in clinical stage C disease. Admittedly, until we are certain that seminal vesicle invasion in radical prostatectomy specimens means that the patient is destined for failure in the presence of negative pelvic lymph nodes, even with adjunctive postprostatectomy radiation treatment, positive seminal vesicle invasion in clinical stages A2, B2 and B3 disease may not deter us from an attempt at surgical cure. However, the available data on this question are not encouraging. Middleton and associates reported 50 per cent positive bone scans at 5 years after radical prostatectomy in patients with negative pelvic lymph nodes but seminal vesicle invasion 19 and Jewett had no cancer cures after any radical prostatectomy for stages Bl and B2 disease in which there was seminal vesicle invasion."" We also need to know the sensitivity of ultrasound guided, transrectal biopsies in detecting microscopic invasion of the seminal vesicles, a question that also must await correlation with histological reconstruction of our radical prostatectomy specimens. Despite the fact that we averaged 6.25 biopsies per prostate in these 251 patients (range 1 to 14 biopsies) our complications of post-biopsy fever and chills, regardless of proper prophylaxis 1 hour before biopsy to ensure high tissue antibacterial levels

70

r

HODGE, MCNEAL AND STAMEY

at the time bacteria are introduced from the rectum into the prostate, were minimal (less than 1 per cent). Nevertheless, we treat these rare instances of fever and chills aggressively by admitting each patient to the hospital for intravenous cefoxitin and gentamicin. Many patients are taking aspirin or other nonsteroidal anti-inflammatory medication and although we try to stop these platelet inactivating drugs before biopsy we are not always successful. Lastly, although we believe the current evidence does not support the use of these technologies for screening asymptomatic populations-especially those with normal or only benignly enlarged prostates on digital rectal examination-we were unable to prevent performing ultrasounds on these 2 groups of patients in our first year of using this technology until we had educated our local community and referral sources. We performed ultrasounds on 75 patients in whom we believed the digital rectal examination showed a normal prostate. Of these patients 48 had normal transrectal ultrasound studies and they were not biopsied. Of the 27 patients who had hypoechoic defects in the peripheral zone and who were all biopsied only 3 were positive for cancer. Thus, cancer was found in 4 per cent of the 75 normal prostates and the positive biopsy rate was only 11 per cent. Similarly, in 54 additional patients whom we believed had benign enlargement of the prostate 24 had peripheral zone hypoechoic defects, all of which were biopsied. Of these patients 3 had cancer. Therefore, cancer was found in 5.5 per cent (3 of 54) of patients with benign prostatic hyperplasia, with a positive biopsy rate of 12.5 per cent (3 of 24). Screening asymptomatic populations is a complex public health issue that encompasses serious unanswered questions regarding the natural history of prostate cancer, the accidental detection of cancers too small ever to be of clinical significance and the failure of transrectal ultrasound to detect the 50 per cent of cancers located in the anterior half of the prostate. 21 This is an issue that we could not begin to discuss in detail in this study but we do believe a comment is in order. It absolutely is clear that to diagnose the presence of cancer in 66 per cent of 251 patients based on a palpable abnormality of the prostate is infinitely more cost-effective than to diagnose cancer in 5 per cent of 129 patients with palpably benign prostates on digital rectal examination. It is even less cost-effective when we realize that 51 of the 129 patients had to be biopsied to detect 6 cancers. Stated another way, 40 per cent of the entire population with palpably normal prostates or benign prostatic hyperplasia would have to be biopsied to detect the 5 per cent with cancer (table 5). Surely, it is more reasonable to screen for palpably abnormal prostates with a simple digital rectal examination and know that 2 of 3 (66 per cent) will prove on ultrasound guided, transrectal core biopsies to have prostate cancer. Drs. John K. Kabalin and Martha K. Terris reviewed this paper. REFERENCES 1. Holm, H. H. and Gammelgaard, J.: Ultrasonically guided precise needle placement in the prostate and the seminal vesicles. J.

Urol., 125: 385, 1981. 2. Fornage, B. D., Touche, D. H., Deglaire, M., Faroux, M. C. and Simatos, A.: Real-time ultrasound-guided prostatic biopsy using a new transrectal linear-array probe. Radiology, 146: 547, 1983. 3. Rifkin, M. D., Kurtz, A. B. and Goldberg, B. D.: Sonographically guided transperineal prostatic biopsy: preliminary experience with a longitudinal linear-array transducer. Amer. J. Roentgen., 140: 745, 1983. 4. Lee, F., Littrup, P. J., McLeary, R. D., Kumasaka, G. H., Borlarza, • G. S., McHugh, T. A., Soiderer, M. H. and Roi, L. D.: Needle aspiration and core biopsy of prostate cancer: comparative evaluation with biplanar transrectal US guidance. Radiology, 163: 515, 1987. 5. Liddell, H. T., McDougal, W. S., Burks, D. D. and Fleischer, A. C.: Ultrasound versus digitally directed prostatic needle biopsy. J. Urol., 135: 716, 1986. 6. Cooner, W. H., Eggers, G. W. and Lichtenstein, P.: Prostate cancer: new hope for early diagnosis. Alabama Med., 56: 13, 1987. 7. Cooner, W. H., Mosley, B. R., Rutherford, C. L., Jr., Beard, J. H., .Pond, H. S., Bass, R. B., Jr. and Terry, W. J.: Clinical application of transrectal ultrasonography and prostate specific antigen in the search for prostate cancer. J. Urol., 139: 758, 1988. 8. Stamey, T. A., McNeal, J. E., Freiha, F. S. and Redwine, E.: Morphometric and clinical studies on 68 consecutive radical prostatectomies. J. Urol., 139: 1235, 1988. 9. Gleason, D. F.: Histologic grading and clinical staging of prostatic carcinoma. In: Urologic Pathology: The Prostate. Edited by M. Tannenbaum. Philadelphia: Lea & Febiger, part II, chapt. 9, pp. 171-198, 1977. 10. McNeal, J.E. and Bostwick, D. G.: lntraductal dysplasia: a premalignant lesion of the prostate. Hum. Path., 17: 64, 1986. 11. McNeal, J. E.: Normal histology of the prostate. Amer. J. Surg. Path., 12: 619, 1988. 12. Shinohara, K., Wheeler, T. M. and Scardino, P. T.: The appearance of prostate cancer on transrectal ultrasonography: correlation of imaging and pathological examinations. J. Urol., 142: 76, 1989. 13. Kaufman, J. J., Rosenthal, M. and Goodwin, W. E.: Methods of diagnosis of carcinoma of the prostate: a comparison of clinical impression, prostate smear, needle biopsy, open perinea! biopsy and transurethral biopsy. J. Urol., 72: 450, 1954. 14. Kaufman, J. J. and Schultz, J. I.: Needle biopsy of the prostate: a re-evaluation. J. Urol., 87: 164, 1962. 15. Zincke, H., Campbell, J. T., Utz, D. C., Farrow, G. M. and Anderson, M.: Confidence in the negative transrectal needle biopsy. Surg., Gynec. & Obst., 136: 78, 1973. 16. Bissada, N. K., Rountree, G. A. and Sulieman, J. S.: Factors affecting accuracy and morbidity in transrectal biopsy of the prostate. Surg., Gynec. & Obst., 145: 869, 1977. 17. Spigelman, S. S., McNeal, J. E., Freiha, F. S. and Stamey, T. A.: Rectal examination in volume determination of carcinoma of the prostate: clinical and anatomical correlations. J. Urol., 136: 1228, 1986. 18. McNeal, J.E., Bostwick, D. G., Kindrachuk, R. A., Redwine, E. A., Freiha, F. S. and Stamey, T. A.: Patterns of progression in prostate cancer. Lancet, 1: 60, 1986. 19. Middleton, R. G., Smith, J. A., Jr., Melzer, R. B. and Hamilton, P. E.: Patient survival and local recurrence rate following radical prostatectomy for prostatic carcinoma. J. Urol., 136: 422, 1986. 20. Jewett, H.J.: The case for radical perinea! prostatectomy. J. Urol., 103: 195, 1970. 21. McNeal, J.E., Price, H., Redwine, E. A., Freiha, F. S. and Stamey, T. A.: Stage A versus stage B adenocarcinoma of the prostate: morphological comparison and biological significance. J. Urol., 139: 61, 1988.