The over-use of imaging studies in low-risk prostate cancer patients
Posted Oct 30 2010 12:00am
Every set of evidence-based best practice guidelines that we are aware of discourage routine imaging studies following diagnosis of localized prostate cancer in low-risk prostate cancer patients. There are exceptional patients for whom such imaging may be appropriate, but they are most certainly the exception and not the routine.
Despite such evidence-based guidelines, the use of imaging studies (supposedly to increase the accuracy of clinical staging) is commonplace in low-risk patients, and Lavery et al. have now documented just how prevalent this practice still is today, based on data from one specific institutional database.
In this study, the authors identified all previously diagnosed, low-risk, prostate cancer patients scheduled to undergo robot-assisted laparoscopic prostatectomy (RALP) between May 2005 and January 2010. They defined “low-risk” patients to be any patient with a PSA level ≤ 10 ng/ml and a Gleason score ≤ 6. Imaging studies considered to be potentially inappropriate for staging in such low-risk prostate cancer patients included radioisotope-based bone scans, computed tomography of the pelvis (CT scans), and endorectal magnetic resonance imaging (eMRI scans) all performed after the diagnosis of prostate cancer and before RALP.
The results of this study showed the following:
677 patients met the criteria for low-risk prostate cancer patients scheduled for RALP.
328/677 patients (48 percent) underwent one or more preoperative imaging procedure.
202/677 patients (30 percent) underwent two or more preoperative imaging procedures.
18/677 patients (3 percent) received all three preoperative imaging procedures.
“Suspicious” results were observed in 7/265 patients (2.7 percent) on CT scan and in 21/241 patients (8.7 percent) on bone scan.
27 patients underwent additional radiographic imaging as a consequence of “suspicious” results.
None of the initially “suspicious” results led to a subsequent need for intervention or additional biopsy.
Guidance from opinion-leading urologists has discouraged the excessive use of such scans in low-risk prostate cancer patients since at least the early 1990s i.e., for nearly 20 years. Despite this guidance, we continue to see large numbers of such patients subjected to tests that are unlikely to even suggest evidence of possible extraprostatic disease (only 28/328 patients or 8.5 percent in the present study, of whom not one was deemed to have a problem after further assessment).
It is hard to resist the conclusion that such over-use of imaging is being carried out for legal reasons (to protect the doctor from a law suit in the one in 1,000 cases when a patient may be found to have extraprostatic disease subsequent to surgery). However, every such additional imaging test places the patient at increased risk from radiation or contrast exposure. It also represents an additional financial burden on the patient and the health care system.
The “New” Prostate Cancer InfoLink considers that it is high time we “got real” about this type of medical and legal risk and recognized that physicians should not be “found guilty” of practicing good medicine according to well established guidelines. Medicine is not a science in which absolutely perfect outcomes should be expected by patients or demanded by lawyers. All we can ask of even the very best physicians is that they do their best according to the guidelines set by their profession.
Who in the above series of 628 patients might one justifiably give some form of scan to prior to surgery? Possibly
A patient with a very clearly positive finding of suspicion for cancer on DRE or
A patient with a very high tumor load based on biopsy findings (e.g., 10/12 positive biopsy cores with at least a couple > 50 percent positive)
Both such patients would argubly be at risk for extraprostatic extension necessitating non-nerve-sparing radical surgery, and so a 3.0 T eMRI scan might be justified (if such equipment was available). However, a 1.5 T eMRI may well not have sufficient accuracy to provide definitive clinical information about either of these patient types, and neither of these patients would appear to be in need of a pelvic CT scan or a bone scan on the basis of the available information. A bone scan would be justifiable in a patient with clear symptoms of pain in the lower back or hips as well as his prior diagnosis of prostate cancer.