Time and (t) in months since surgery; R, radiotherapy; H, hormone therapy; *Gleason score not evaluated as the patients had hormone therapy before RP.
A digital rectal examination after radical prostatectomy is unnecessary if prostate specific antigen is undetectable
Article first published online: 23 MAY 2003
Volume 91, Issue 9, pages 887–888, June 2003
How to Cite
Saad, F. and Lattouf, J.-B. (2003), A digital rectal examination after radical prostatectomy is unnecessary if prostate specific antigen is undetectable. BJU International, 91: 887–888. doi: 10.1046/j.1464-410X.2003.04236.x
- Issue published online: 23 MAY 2003
- Article first published online: 23 MAY 2003
- Accepted for publication 12 February 2003
The DRE is essential in screening for prostate cancer and traditionally its use is routine in the follow-up of patients treated by radical prostatectomy (RP) for prostate cancer. The serum PSA level is also very useful and has been part of the follow-up of these patients since its introduction almost 15 years ago [1–3]. However, few papers have specifically addressed the value of the DRE in the follow-up of RP [4–6]. Descriptions of local recurrence detectable on DRE and no detectable PSA are limited to rare case reports [7–9]. Rare cases of systemic recurrence have also been described with no biochemical relapse . We reviewed our experience to determine whether biochemical recurrence was found in all cases of clinical recurrence.
We reviewed the charts of all patients who had undergone radical RP at our institution between April 1989 and November 2000; 21 had received hormone therapy before RP. Patients with no PSA measurements or DRE during the follow-up were excluded, as were those with lymph node metastasis on final pathology.
Analyses included the date of operation, the postoperative pathological report, PSA levels during the follow-up, the results of the DRE, anastomotic biopsies, bone scans and date of death, where applicable. Relapse was defined clinically as a clearly suspicious induration or nodularity on the DRE. A biochemical relapse was defined as a PSA level of ≥ 0.3 ng/mL on at least two consecutive visits.
In all, 423 men met the inclusion criteria (mean age 62.8 years, sd 10.8, range 47–75); 30 (7%) of the men had a positive DRE during the follow-up (Table 1). The mean (sd, range) delay between surgery and local recurrence was 27.7 (17.3, 1–36) months. All of these patients had a detectable PSA level before the positive DRE. The mean (sd) lead-time for biochemical relapse before clinical relapse was 14 (11) months.
|No.||Time of +ve DRE||Gleason of RP||Margin status||PSA at +ve DRE (t)||Treatment|
Ten anastomotic biopsies and one transurethral resection were undertaken in patients with a positive DRE, and yielded only four positive results (Gleason 5, 6, 6 and 7). Nine patients with clinical relapse (24%) developed metastatic disease, documented on bone scans, and one died from prostate cancer. Table 1 also shows the characteristics of the patients with a positive DRE.
In all, 98 (25%) patients had an isolated biochemical relapse and 63 of them received either radiotherapy, hormone therapy or both. Nine developed bone metastasis but none have as yet died from prostate cancer.
COMPARISON WITH OTHER METHODS
To our knowledge, only two recent reports have specifically addressed the question of the utility of a DRE after RP [4,5]. Pound et al. reported the largest series (1916 men), finding no patients with an undetectable PSA level who had a local recurrence on the DRE. The mean (sd) lead-time of PSA recurrence before DRE recurrence in that series was 2.3 (2) years. The second series  (501 men) still reported no DRE recurrence before PSA became detectable. Both these studies concluded that a DRE is unnecessary after RP when PSA is undetectable.
However, a few case reports have stressed that there are patients with undetectable PSA values but with local recurrence on a DRE [8,9]; such a situation is rare enough to justify a case report . Patients in these cases had moderately differentiated Gleason scores on pathology after RP, but the pathology in the recurrence was poorly differentiated in all. Hence it was concluded that there might be an association between poorly differentiated Gleason scores and tumours that do not secrete PSA. The present series included 142 patients with Gleason scores of ≥ 7 after RP; of these, 56 had biochemical relapse with or with no clinical relapse, and there was no clinical relapse before biochemical relapse. Anastomotic biopsies or TURP was used in 12 men to confirm clinical relapse, with only one of four positive biopsies having a Gleason score of 7 (the others being 6, 6 and 5). This suggests that, although the few tumours that relapse only clinically may be poorly differentiated, most poorly differentiated relapses are first detected biochemically.
The present definition of relapse was a conservative compromise among those reported previously. Amling et al. showed that biochemical relapse can vary widely depending on the method used to define PSA failure. They suggested a PSA of 0.4 ng/mL as the optimum threshold to define relapse; others use a threshold as low as 0.2 ng/mL . Although it is possible that a PSA threshold at 0.3 ng/mL with the requirement of two increasing values might have skewed the present biochemical relapse rate to a more favourable outcome, all the clinical relapses in the present series were still preceded by biochemical relapses. Even when using a PSA threshold of 0.4 ng/mL, none of the clinical relapses preceded the biochemical ones.
The report by Lightner et al. showed that investigating a benign DRE yielded completely negative results. We therefore chose not to consider benign irregularities on a DRE as clinical relapse. Such a finding did not alter the therapeutic decision and was deemed irrelevant.
In conclusion, the DRE was never positive if PSA was undetectable after RP. Although a DRE is often used routinely after RP it appears to be safe to reserve it for patients with an increasing PSA level.