of transrectal ultrasound of the prostate (TRUS) to assist in the diagnosis and ... of tissue obtained by transrectal biopsy of the prostate (TB), were all available.
1994, The British Journal of Radiology, 67, 668-671
The clinical utility of transrectal ultrasound and digital rectal examination in the diagnosis of prostate cancer 1
A SHAPIRO, MD, 2P D LEBENSART, MD, 1 D PODE, MD and 2R A BLOOM, M D
Departments o f 1 Urology and 2Radiology, Hadassah University Hospital, Jerusalem, Israel Based on a paper presented at the BMUS 25th Annual Meeting in Eastbourne, on 7-9 December 1993 Abstract
The development of high definition transrectal ultrasound probes has led to an increased interest in the ability of transrectal ultrasound of the prostate (TRUS) to assist in the diagnosis and management of prostate cancer. The present study was designed to examine the correlation of TRUS with digital rectal examination (DRE). The study group comprised 471 patients in whom the results of (a) DRE, (b) TRUS, and (c) histology of tissue obtained by transrectal biopsy of the prostate (TB), were all available. In those patients where both TRUS and DRE were negative, but prostate specific antigen (PSA) was greater than 10 /ig ml" 1, six random biopsies were performed. In all other cases the biopsies were TRUS directed to the suspicious lesion. There were 142 cases in whom both DRE and TRUS were negative or only mildly suspicious of malignancy. TB in these cases was positive for cancer in 17 cases (12%). In a further 126 cases, TRUS was positive for cancer, while DRE demonstrated no suspicious nodule. TB was positive in only 17 of these cases (13.5%). Similarly, in the 31 cases in which DRE was positive but TRUS was negative, TB was positive in only three cases (10%). In the 172 cases in whom both DRE and TRUS were positive, 99 biopsies were positive (57.5%). It was concluded from this study that DRE remains the most valuable single examination in the diagnosis of prostate cancer. TRUS increases the sensitivity of DRE if both are positive. When there is a discrepancy between the two examinations, the biopsy yield is low. When both are positive, a high cancer yield is obtained, TRUS having added value in directing the biopsy needle to the suspicious site. TRUS is thus a valuable adjunct to DRE in the diagnosis of prostate cancer.
Prostate cancer is the most common cancer in American men, and is the second most common male cancer killer [1]. Among white Americans, the annual mortality rate owing to prostate cancer during 1983-1984 was 22.7/100000 males [2]. The disease appears to be increasing in prevalence, and the number of cases in the USA has increased from 100000/year in 1989 to more than 150000/year in 1993 [1]. Occult carcinoma of the prostate in middle-aged and elderly men is even more common. It has been estimated that with a mean life expectancy in developed nations extending into the eighth decade, between 21% and 36% of men can be expected to develop such occult tumours [3]. Based on autopsy data, Franks found that 30% of men in the sixth decade and 67% of men in the ninth decade of life had microscopic evidence of prostate cancer at the time of death from other causes [4]. These findings correlate reasonably well with n • J ~>n AT L mm J J i *r i inn; Received 29 November 1993 and accepted 1 March 1994. y
Address correspondence to R A Bloom, MD, Department of Radiology, Hadassah University Hospital, Ein Kerem, Jerusalem, Israel. 668
the incidental findings of prostate cancer in 19% of patients during prostatectomy for benign prostatic hyperplasia [5]. The natural history of prostate cancer is unpredictable and only a small proportion of men with small malignant areas within the prostate later develop clinical symptoms caused by growth or metastasis of these tumours [6], the majority of which are very small and remote in time from aggressive behaviour. Consideration of only those tumours larger than 0.3 ml in size as likely to have metastatic potential, provides a rough dividing line between histological and clinically significant carcinomata [7]. This volume of 0.3 ml is contained in a tumour of diameter of approximately 0.8 cm, and it is tumours of this size that are within the limits of resolution of digital rectal examination (DRE) and transrectal ultrasound of the prostate (TRUS). The development of high definition TRUS probes has facilitated the detection of small prostatic lesions which can be biopsied. This has probably increased the detec.• . K. , . . * \ . ., , f .,tion rate of early prostate cancer, but the value of this modality in relation to the traditional method of DRE has not been finally determined. The present study is a prospective evaluation of a clinical series comprising 471 The British Journal of Radiology, July 1994
Clinical utility of TRUS and DRE in the diagnosis of prostate cancer Table I. A record of the number of patients in whom cancer was demonstrated by TB in relation to concordance or discordance between DRE and TRUS
DRE+ DRE+ DRE- DRETRUS+ TRUS- TRUS+ TRUSNo. of patients with transrectal biopsy positive for carcinoma
99
No. of patients with transrectal biopsy negative for carcinoma
73
28
17
17
109
125
patients in whom the results of TRUS, DRE, and histology obtained by transrectal biopsy of the prostate (TB) were all available. The purpose of this study is to examine the correlation of DRE and TRUS with the pathological findings. Material and methods
All patients were referred either to the Urology Department or to the Ultrasound Unit of the Department of Radiology of Hadassah University Hospital. No patients included in this series had previously proven prostate cancer. Other than that, no patient selection was made. It is to be stressed that this was a clinical population of patients. The urologist was not blinded to the results of TRUS, if this had been performed previously; and the radiologist knew the urologist's opinion, if there was one, before commencing his study. In any event the radiologist always performed DRE prior to inserting the transrectal probe. Some of the patients were referred owing to a high prostate specific antigen (PSA) level, without any clinical suspicion of prostate cancer. The origin of patient referral was diverse. Hadassah University Hospital is a tertiary referral centre and many of the patients had been examined previously by urologists from other centres. Some were referred directly to the Ultrasound Unit for TRUS; others were referred to the Urology Department and the TRUS was performed immediately prior to TRUS-directed biopsy. Only 7% of the patients derived from an ongoing
prostate cancer detection programme. All biopsies were performed jointly by a staff urologist and radiologist. The DRE findings were usually a consensus opinion of the two doctors. If there was a discrepancy in findings, the urologist's opinion was considered final. All TRUS examinations were performed with a Bruel and Kjaer (Naerum, Denmark) ultrasound unit, Type 1849, with a 7 MHz multiplanar probe, which enables scanning to be performed in the axial and saggital modes. In patients where no definite lesion was found by either DRE or TRUS, six random systematic ultrasound guided transrectal core biopsies were taken, using a spring-driven Bard Biopty gun (C R Bard, Inc., Covington, Georgia, USA), using an 18 gauge needle. The majority of these patients had clinical enlargement of the prostate, and had either been referred with a clinical suspicion of prostate cancer or had a raised PSA titre in excess of 10 mg ml"1. In cases where a nodule was felt, but not confirmed by TRUS, three TRUS directed core biopsies were taken from the quadrant of the prostate under suspicion. No biopsies were taken outside the area of interest. In all other cases several TRUS guided biopsies were taken from the peripheral zone hypoechoic focus. If several hypoechoic areas were seen, all were biopsied. Thus, up to eight biopsies could be taken from a given patient. Results
There were 471 patients in whom the results of (a) DRE, (b) TRUS and (c) TB were available. These derived from a total population of approximately 2500 patients who underwent TRUS during the period 1991-1993. Table I records the results of TB in relation to concordance or discordance between DRE and TRUS. Table II records the sensitivity, specificity, negative predictive value and positive predictive value of DRE and TRUS separately and when they were concordant. Table III shows the percentage of cases in whom positive TB pathology was found in the four following groups: (a) both TRUS and DRE free of suspicion of cancer; (b) TRUS suspicious but DRE non-suspicious; (c) TRUS non-suspicious but DRE suspicious and (d) both TRUS and DRE suspicious for malignancy. From Table II it can be seen that TRUS is slightly more sensitive in picking up carcinoma than is DRE,
Table II. A record of the sensitivity, specificity, negative predictive values and positive predictive values for DRE and TRUS separately and when they were concordant
DRE and TRUS in accordance DRE alone TRUS alone
Vol. 67, No. 799
Sensitivity (%)
Specificity (%)
Negative predictive value (%)
Positive predictive value (%)
85
63
88
58
76 85
70 46
87 88
50 39
669
A Shapiro, P D Lebensart, D Pode and R A Bloom Table III. A record of the percentage of cases in which a positive cancer biopsy was obtained when DRE and TRUS were negative, discordant, or concordant D R E - D R E - DRE+ DRE+ TRUS- TRUS+ TRUS- TRUS+ Transrectal biopsy positive for malignancy
12%
13.5%
10%
57.5%
but considerably less specific. When the two examinations were in accordance the sensitivity was the same as with TRUS alone, as was the negative predictive value. However, both the specificity and positive predictive value increased markedly. The importance of concordance between the two examinations is shown most clearly in Table III. In the 142 cases where neither DRE nor TRUS were suspicious for malignancy, most of whom, as has been mentioned, having a high PSA level, systematic random biopsies demonstrated a TB positive for malignancy in 12% of cases. In the 126 cases where TRUS was suspicious for cancer, but DRE was not, 15% had a positive TB. Similarly, in the 31 cases where DRE was suspicious, but TRUS was not, TB was positive in only three cases (10%). However, in the 172 cases where both were suspicious, TB was positive in 99 cases (57.5%). Discussion TRUS of the prostate gland has developed into an important adjunct to the diagnosis of prostate cancer. It is particularly useful in demonstrating the 70% of prostate cancers that arise in the peripheral zone of the gland [8]. Approximately 60-76% of these appear as hypoechoic areas on the sonogram. In the other 24-40% the tumour is isoechoic and cannot be distinguished clearly from the surrounding tissue [9, 10]. These hypoechoic areas are demonstrable owing to the contrast with the normally homogeneous and echogenic peripheral zone. The transitional and central zones of the prostate from which 30% of prostate cancers derive are both hypoechoic and inhomogeneous, and carcinomas in these areas cannot be diagnosed reliably by TRUS [11]. In addition, it must be realized that a large percentage of hypoechoic lesions in the peripheral zone result from benign pathology. TRUS directed biopsies of such lesions reveal cancer in 20-40% of cases [12, 13]. Thus, while the modality is useful, it cannot be expected to reveal all cases of prostate cancer. The traditional method of diagnosing carcinoma of the prostate is by DRE. Screening asymptomatic men in the cancer age group by this method has shown a detection rate of 1.3-1.7% [14, 15]. The detection rate of TRUS in screening for carcinoma of the prostate varies. Wanatabe [16] reported detection of 48 cases in 7235 apparently normal men over the age of 55 years (0.6%), while Lee et al [17] detected 20 cases in 784 men over 670
the age of 60 years (2.6%). The detection rates of the two methods are thus very similar and this rate represents only a small percentage of men with occult carcinomata. The present series in no way represents a screening programme. The large majority of our patients presented with a history of prostate symptomatology or with a raised PSA level. The purpose of our study was to examine the value of TRUS in relation to DRE. For this reason we included only those patients in whom the pathology report on the biopsy specimen was unequivocally positive or negative. The indices used to evaluate the two diagnostic methods were: sensitivity, specificity, negative predictive value and positive predictive value. Sensitivity is defined as the ratio of true positives/ (true positives + false negatives), and is the fraction of positive findings that will be diagnosed as positive by a given test. Specificity is defined as the ratio of true negatives/(true negatives + false positives), and is the fraction of negative findings that will be diagnosed as negative by a given test. Negative predictive value is the ratio of true negatives/(true negatives + false negatives), and measures the reliability of a negative finding. Positive predictive value is the ratio of true positives/ (true positives + false positives), and measures the reliability of a positive finding. It can be seen that TRUS was slightly more sensitive than DRE, but considerably less specific. These findings are similar to those of previous series, and derive from the fact that the majority of suspicious hypoechoic areas in the peripheral zone as demonstrated by TRUS result from benign causes. Our positive predictive value for TRUS of 39% is very similar to that of Lee et al [18], and almost double that of Rifkin and Choi [12]. The negative predictive values of DRE and TRUS are virtually identical, and a combination of the two methods does not change this value. Table III examines thefiguresin a different way. It is seen that when both methods are non-suspicious for cancer, or when they do not correlate, TRUS directed biopsies were positive for cancer in 10-13% of cases, but when both were suspicious for cancer 57.5% yielded tissue positive for cancer. This compares with the 50% value shown by Lee et al [17] in a small series of patients, and the 43% of Cooner et al [19]. The 10-13.5% found in the groups where both or one of the tests were negative can be compared with the incidental finding of prostate cancer in 19% of patients during prostatectomy for benign hypertrophy of the prostate (BHP) [5], and with the incidental finding of cancer in 10% of men with a clinical diagnosis of BHP who underwent multiple directed and systemic ultrasound guided biopsies of the prostate [20]. It would seem from our findings that a DRE diagnosis of cancer which is confirmed by TRUS has a high possibility of being correct. If one, or both, of these tests is negative, biopsy of the prostate is no more likely to demonstrate cancer than would random systematic biopsies of all patients with prostatic enlargement. The British Journal of Radiology, July 1994
Clinical utility of TRUS and DRE in the diagnosis of prostate cancer References
1. BARING, C C, SQUIRES, T S and TONG, T, Cancer statistics, Ca, 43, 7-26 (1993). 2. DEVESA, S S, SILVERMAN, D T, YOUNG, J L E T AL, Cancer incidence and mortality trends among Whites in the United States, J. Nat. Cancer Inst., 79, 701-778 (1978). 3. SHELDON, C A, WILLIAMS, R D and FRALEY, E E, Incidental carcinoma of the prostate: a review of the literature and critical reappraisal of classification, J. Urol., 124, 626-633 (1980). 4. FRANKS, L M, Latent carcinoma of the prostate, /. Pathol. Bad., 68, 603-616 (1954). 5. AGATSTEIN, E H, HERNANDEZ, F J, LAYFIELD, L J ET AL, Use of fine needle aspiration for detection of Stage A prostatic cancer before transurethral resection of the prostate: a clinical trial, J. Urol., 138, 551-553 (1987). 6. WHITMORE, W F, The natural history of prostate cancer, Cancer, 32, 1104-1112 (1973). 7. STAMEY, T A and McNEAL, J E, Adenocarcinoma of the prostate. In Campbell's Urology (6th edn) Ed. by P C Walsh, A B Retik, T A Stamey and E D Vaughn Jr (W B Saunders, Philadelphia) p. 1161 (1992). 8. LEE, F, TORP-PEDERSON, S T, SIDERS, D B ET AL, Transrectal ultrasound in the diagnosis and staging of prostatic carcinoma, Radiology, 170, 609-615 (1989). 9. 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, 76-82 (1989). 10. HAMPER, U M, SHETH, S, WALSH, P C ET AL, Capsular transgression of prostatic carcinoma: evaluation with transrectal US with pathological correlation, Radiology, 187, 791-795 (1991). 11. RICKARDS, D, Transrectal ultrasound, J. Urol., 69, 449^55 (1992).
Vol. 67, No. 799
12. RIFKIN, M D and CHOI, H, Implications of small, peripheral, hypoechoic lesions in endorectal US of the prostate, Radiology, 166, 619-622 (1988). 13. LEE, F, TORP-PEDERSON, S T and LITTRUP, P J, Hypoechoic lesions of the prostate: clinical relevance of tumor size, digital rectal examination, and prostate-specific antigen, Radiology, 170, 29-32 (1989). 14. GILBERTSEN, V A, Cancer of the prostate gland, JAMA, 215, 81-84 (1971). 15. CHODAK, G V and SCHOENBERG, H W, Early detection of prostate cancer by routine screening, JAMA, 252, 3261-3264 (1984). 16. WANATABE, H, Transrectal sonography: a personal review and recent advances, Scand. J. Nephrol. (Suppl.J, 137, 75-83 (1991). 17. LEE, F, LITTRUP, P J, TORP-PEDERSEN, S T ET AL, Prostate cancer: comparison of transrectal US and digital rectal examination for screening, Radiology, 168, 389-394 (1988). 18. LEE, F, TORP-PEDERSEN, S T, METTLIN, C ET AL, Hypoechoic lesions of the prostate: clinical relevance of tumor size, digital rectal examination, and prostate-specific antigen, Radiology, 170, 29-32 (1989). 19. COONER, W H, MOSLEY, B R, RUTHERFORD, C L ET AL, Prostate cancer detection in a clinical urological practice by ultrasonography, digital rectal examination, and prostate specific antigen, /. Urol., 143, 1146-1154 (1990). 20. COPLEN, D E, ANDRIOLE, G L, YUAN, J J J and CATALONA, W J, The ability of systematic transrectal ultrasound guided biopsy to detect prostate cancer in men with the clinical diagnosis of benign prostatic hypertrophy, J. Urol., 146, 75-77 (1991).
671
