The Value of Digital Rectal Examination in Men With Lower Urinary Tract Symptoms and Prostate-Specific Antigen Less Than 3 ng/mL

ABSTRACT

INTRODUCTION: Controversy remains as to the value of digital rectal examination (DRE) for detecting prostate cancer in men presenting with lower urinary tract symptoms and low prostate-specific antigen (PSA) levels. The purposes of the present retrospective investigation were to: (1) evaluate the positive predictive value (PPV) of an abnormal DRE in patients with PSA < 3 ng/mL, (2) describe the stage and grade of the cancers detected, and (3) describe the treatment modalities used for patients at each clinical stage.

METHODS: A prospectively maintained prostate biopsy service database of consecutive men undergoing prostate biopsies from April 2004 to April 2006 was reviewed. Patients with PSA < 3 ng/mL and definitely abnormal DRE were divided into 3 groups according to PSA range. The relationship between abnormal DRE and positive prostate biopsy was determined. The International Prostate Symptom Score (IPSS), clinicopathologic features of prostate cancer, and the treatments used for each clinical stage were summarized.

RESULTS: From 1235 men who underwent prostate biopsies, 59 (4.6%) had PSA < 3 ng/mL and an abnormal DRE. Their mean age was 63.4 years. Fifteen of the 59 patients had true-positive DRE; 44 patients had false-positive DRE. The PPV of an abnormal DRE for detecting prostate cancer was 25.4%. The incidence of prostate cancer was 13.3%, 33%, and 53.3% in patients with abnormal DRE and PSA levels of 0.1-1.0 ng/mL, 1.1-2.0 ng/mL, and 2.1-2.9 ng/mL, respectively. Patients with significant LUTS (IPSS > 19) were likely to have false-positive DREs (P = .019). Patients with mild LUTS (IPSS < 7) and abnormal DREs tended to have positive prostate biopsies (P = .030). Modalities used to treat the 15 patients with cancer included active monitoring, brachytherapy, and external beam radiotherapy in 3 (20%), 5 (33.3%) and 4 (26.6%) patients, respectively. Hormonal deprivation was used for 1 patient; radical prostatectomy was used for 2 patients.

CONCLUSIONS: DRE has a significant role in detecting prostate cancer in men presenting with PSA < 3 ng/mL. The higher the PSA, the greater the possibility of detecting prostate cancer in this selective cohort.

KEYWORDS: Digital rectal examination; Low PSA; Prostate cancer diagnosis

CORRESPONDENCE: Michael Nomikos MD, FEBU, Consultant Urologist, Knossou 275, Heraklion, Crete, 71409,Greece ().

CITATION: Urotoday Int J. 2010 Jun;3(3). doi:10.3834/uij.1944-5784.2010.06.07

ABBREVIATIONS AND ACRONYMS: DRE, digital rectal examination; GP, general practitioner; IPSS, International Prostate Symptom Score; LUTS, lower urinary tract symptoms; PPV, positive predictive value; TRUS, transrectal ultrasound.

INTRODUCTION

Prostate cancer is a major public health issue. It is the most common malignancy in men, accounting for more than 27000 new cancer cases in the United Kingdom in 2000 and nearly 9900 deaths in 2002 [1]. The incidence of prostate cancer has increased over the years, in part because of the natural aging of the population. Additionally, the widespread use of serum prostate-specific antigen (PSA) as a diagnostic test has resulted in an increase in the number of men diagnosed with localized prostate cancer. The number of registered new cases of prostate cancer rose from 6179 in 1971 to 17210 in 1993 [1,2,3].

The digital rectal examination (DRE) is considered mandatory in the diagnosis and staging of prostate cancer; however, studies investigating the accuracy of a DRE for these purposes have demonstrated conflicting results [4,5,6]. The DRE has low reproducibility and is dependent on examiner experience [7]. Furthermore, it requires an examination room and trained examiners, adding considerably to screening costs. Despite these limitations, 10% to 30% of prostate cancers detected in screening programs are in men with PSA ≤ 4 ng/mL and an abnormal DRE. It has been demonstrated that the use of PSA in conjunction with DRE-enhanced early prostate cancer detection [8].

Most studies have been focused on a randomly chosen group of participants that are intended to represent the general population. The importance of an abnormal DRE in detecting prostate cancer in men with PSA < 3 ng/mL presenting in urologic practice with lower urinary tract symptoms (LUTS) has not yet been determined. The purposes of the present retrospective investigation were to: (1) evaluate the positive predictive value (PPV) of an abnormal DRE in this selective cohort, (2) describe the stage and grade of the cancers detected, and (3) describe the treatment modalities used for patients at each clinical stage.

METHODS

Database

In April 2004, the authors set up a prostate biopsy service that takes direct referrals from general practitioners (GPs). Patients with elevated PSA and no contraindication are referred to this service as a matter of routine. All patients are then reevaluated prior to scheduling a biopsy. Standard physical examinations, including DRE, are performed by one of 2 experienced urologists who are members of the service. Counseling and biopsy are then performed.

The authors reviewed the tertiary centerâ€™s prospectively maintained prostate biopsy service database of consecutive men undergoing prostate biopsies between April 2004 and April 2006. All of the clinical information was concurrently recorded, which reduced the potential bias that is sometimes associated with retrospective studies.

Overall, 1235 men underwent prostate biopsies through the prostate biopsy service during this period. The authors included in the study men with PSA < 3 ng/mL who had a definitely abnormal DRE, which is defined as the presence of palpable indurations, nodularity, and irregularity. Patients who had undergone transurethral prostatic resection before the biopsy were excluded from the study.

Procedures

Transrectal ultrasound (TRUS) prostate biopsies were performed with the 10-core biopsy protocol described by Bauer et al [9], under antibiotic coverage. Prostate volume was determined by TRUS immediately before biopsy and was calculated using the prelate ellipsoid formula [10]. If the TRUS revealed abnormal findings, additional directed prostate biopsies were taken. Local anesthesia with periprostatic nerve block was performed before every biopsy.

The clinical stage was based on DRE findings, magnetic resonance imaging of the pelvis, and bone scintigraphy. The 2002 TNM system was used for the classification of prostate cancer [11]. Patients were divided into 3 groups according to PSA range: 0.1 to 1.0 ng/mL, 1.1 to 2.0 ng/mL, and 2.1 to 2.9 ng/mL.

Data Analysis

The relationship between abnormal DRE and positive prostate biopsy in men with PSA < 3 ng/mL was evaluated. The PPV was defined as the proportion of biopsy-positive patients among those whose DRE was definitely abnormal [12]:

PPV = True positive DRE/True positive DRE + False Positive DRE.

The International Prostate Symptom Score (IPSS), clinicopathologic features of prostate cancer, and the treatments used for each clinical stage were collected and reviewed. The Fisherâ€™s exact test was used to evaluate the relationship between IPSS and positive prostate biopsies in men with LUTS and abnormal DRE.

RESULTS

The final dataset included 59 patients who had PSA < 3 ng/mL. The patients had a mean age of 63.4 years (range, 45-78 years) and a definitely abnormal DRE. Table 1 contains the DRE findings, stratified according to age. Prostate cancer was found in 15 patients (25.4%).

Table 2 shows the demographic and clinicopathologic features of the 15 men with a true-positive DRE. Table 3 shows the demographic and clinicopathologic features of the 44 men with false-positive DRE. The PPV of an abnormal DRE for PSA 0.1-1.0 ng/mL, 1.1-2.0 ng/mL, and 2.1-2.9 ng/mL was 13.3%, 33.3%, and 53.3%, respectively. The diagnosis was made at the second prostate biopsy in 3 patients who where rebiopsied due to high-grade prostate intraepithelial neoplasia.

Thirteen patients (86.6%) had moderately differentiated prostate cancer (Gleason score 6-7) and 2 patients (13.4%) had poorly differentiated disease (Gleason score 9-10). Clinical staging revealed localized prostate cancer (T2A,BN0M0) in 11 patients (73.5%). Locally advanced disease (T3N0M0) was detected in 3 patients; metastatic disease (T2N1M1) was detected in 1 patient.

Patients with significant LUTS (IPSS > 19) were likely to have false-positive DREs (P = .019). Patients with mild LUTS (IPSS < 7) and abnormal DREs tended to have positive prostate biopsies (P = .030).

The modalities used to treat the 15 patients with true-positive findings are shown in Table 4. The modalities are stratified according to the clinical stage. Treatments included active monitoring, brachytherapy, and external beam radical radiotherapy in 3 patients (20%), 5 patients (33.3%), and 4 patients (26.6%), respectively. Hormonal deprivation was used for 1 patient with metastatic lung disease at the time of presentation. Open radical retropubic prostatectomy (RRP) was used for 2 patients (13.3%).

DISCUSSION

Historically, DRE served as the primary method of establishing a diagnosis of prostate cancer. However, the advent of PSA testing in the mid-1980s greatly improved the ability to detect prostate cancer, identify recurrences, and evaluate the presence of metastatic disease. In view of these developments, the role of DRE as a diagnostic tool in prostate cancer detection became the focus of debate.

Previous studies have shown that the PPV of an abnormal DRE for the detection of prostate cancer is increased with a rising PSA level, even if the serum PSA level is in the normal range [13]. However, the value of the DRE performed on patients with low PSA levels remains controversial [14,15]. Carvalhal et al [16] reported that the PPV of a DRE with a PSA <1 ng/mL, 1.1-2.5 ng/mL, and 2.6-4.0 ng/mL was 5%, 14%, and 29%, respectively. In the European Randomized Study of Screening for Prostate Cancer (ERSPC), in which the usefulness of a DRE alone as a screening test was evaluated, Schroder et al [17] found that the PPV of a DRE with a PSA <1 ng/mL, 1.0-1.9 ng/mL, 2.0-2.9 ng/mL, and 3.0-3.9 ng/mL was 4%, 10%, 11%, and 33%, respectively. Although Carvalhal et al concluded that the PPV of a suspicious DRE was appreciable in men with low PSA levels and recommended its continuous use, Schroder et al concluded that DRE had poor performance at low PSA levels and recommended that it should be replaced with a more sensitive test [16,17]. A recent study in the United States that evaluated the diagnostic value of abnormal DRE in a urologic referral practice revealed that DRE results are most informative when used in conjunction with other clinical characteristics such as IPSS and prostate volume [18]. Despite the controversy that exists regarding the accuracy of an abnormal DRE as a diagnostic tool in detecting prostate cancer, it is considered to be an absolute indication for prostate biopsy.

It is important to distinguish the use of DRE in the general population and in a clinic setting. From a urologist's point of view, a particular test performed as part of the physical examination in primary care or in a screening program may be less valid than the same test applied in a urologic practice [19].

The present authors studied the diagnostic value of a definitely abnormal DRE in a specific group of men referred for prostate assessment with PSA < 3 ng/mL and LUTS. The incidence of prostate cancer in this selective cohort was 25.4%. This result is certainly higher than the 10% and 14% reported by other authors [14,15]. The discrepancy is probably explained by the fact that the present authors excluded softer findings from the definition of an abnormal DRE, whereas several of the previous studies used less restrictive criteria. The diagnostic agreement between clinical finding and biopsy results was 22.6% (12 of 53 patients) when T2 disease was suspected. The DRE was more accurate in the diagnosis of prostate cancer in men with locally advanced disease, because all patients suspected of having T3 or T4 disease had biopsy-confirmed prostate cancer. This finding confirmed a previous report by Obek et al [20].

In the present study, most diagnosed prostate cancers (86.5%) were moderately differentiated (Gleason score ≥ 6). There were no patients with well-differentiated disease, which is in contrast with the 34% to 72% of patients with well-differentiated prostate cancer reported in screening studies [7,8]. Interestingly, 2 patients in the present study had poorly differentiated disease (Gleason score 9-10). This finding is in agreement with Gosselaar et al [21], who reported that 46% of men who underwent biopsy due to abnormal DRE alone had a Gleason score ≥ 7.

It is important to keep in mind that the present data reflect some degree of selection bias. The study population was preselected and then referred to urology by a primary care provider. The almost 26% positive biopsy rate is significantly higher than the 10% to 15% reported in the literature for screened populations. In view of the potential limitations, the present findings may not apply to primary care. This study targeted a selective cohort of patients presenting in urologic practice. Prospective randomized clinical trials are needed to give definitive guidelines to clinicians.

CONCLUSION

DRE is an important tool in guiding diagnosis and treatment for prostate cancer in men with normal PSA presenting for the evaluation of LUTS. Neglecting DRE would have harmed 26% of the present study participants, who were diagnosed with prostate cancer based solely on abnormal DRE. A substantial proportion of detected cancers had features associated with clinically aggressive tumors. The higher the PSA, the greater the possibility of detecting prostate cancer in this selective cohort. Therefore, DRE should not be omitted in the evaluation of men presenting with PSA < 3 ng/mL. Many of the cancers detected by DRE alone are potentially curable but may have worse outcomes by the time that PSA reaches a higher level. In the modern era, physical examination should remain the cornerstone of daily urological practice.

Conflict of Interest: none declared

REFERENCES

Majeed A, Babb P, Jones J, Quinn M. Trends in prostate cancer incidence mortality and survival in England and Wales 1971-1998. BJU Int. 2000;85(9):1058-1062.

Login