PSA study: reader beware?
Two new controlled studies—one in the U.S. and one in Europe, both reported in the New England Journal of Medicine —suggest that over a 10-year period, PSA testing does not reduce mortality, so they thereby challenge the value of routine screening. [“Prostate cancer screening leads to unneeded treatments, “ ACP InternistWeekly, March 24. The data seem valid, but the conclusions certainly deserve a closer look.
First, in the American study the 90 deaths out of 77,000 subjects occurred over six or seven years, not 10. Since prostatic cancer is slow-moving and is only occasionally fatal, a process discovered early would usually take 10-20 years to end fatally, so the six-to-seven year mortalities must have represented cases which were quite advanced to begin with. The PSA test, designed as an early detection measure, would have been irrelevant in such obvious instances.
Second, the test group was offered annual PSAs, but this doesn't mean the control group didn't get tested anyway: actually, nearly half of them had periodic PSA tests on their own. And a comment from the body of the article is revealing: “The cumulative death rate from prostate cancer at 10 years in the two groups combined was 25% lower in those who had undergone two or more PSA tests at baseline than in those who had not been tested.”
Oops. Not exactly as advertised.
And, third, there's the question of quality of life in the survivors, not dealt with in this study. Furthermore, it's essential to manage one patient at a time: would an asymptomatic elevated PSA portend different outcomes in a 75-year-old man and a 50-year-old? Undoubtedly.
Meanwhile, there are hundreds of other studies, several showing significant benefit in therapy of localized disease. And, at the other end of the spectrum, one would not question the value of the PSA in monitoring the therapy of metastatic disease. Here, as always: Caveat emptor. Or, since this is an American study: Reader, beware!
William C. Waters III, MACP
As a rheumatologist, I enjoyed reading your article on Sjögren's syndrome [“Elusive Sjögren's manageable—if diagnosed, “ ACP Internist, March 2009.] You highlighted the high prevalence of this debilitating disorder, and the challenges faced in coming to a proper diagnosis. Sjögren's syndrome may occur in isolation or secondary to other autoimmune disorders, such as lupus, rheumatoid arthritis or scleroderma. The constitutional symptoms of Sjögren's Syndrome may also mimic some symptoms of fibromyalgia.
At several points in the article and in your editorial column, fibromyalgia is referenced as an autoimmune disease, similar to lupus and rheumatoid arthritis. I find it necessary to stress that fibromyalgia is not an autoimmune disease, but rather a disorder of altered pain processing. Autoimmune diseases such as lupus, rheumatoid arthritis, Sjögren's syndrome and scleroderma are systemic diseases that often require potent immunosuppression, and may lead to end organ damage. There is no autoimmune component to fibromyalgia, and treatment paradigms include the use of anticonvulsants and antidepressants to modulate pain signals. End organ damage does not occur due to fibromyalgia.
Rheumatologists are often challenged with differentiating fibromyalgia from autoimmune diseases. I find that patients with fibromyalgia are often misinformed about the nature of their disorder, and fearful of the manifestations and damage that often occur with autoimmune diseases. It is important to explain to patients that fibromyalgia is a separate disorder with a distinct pathogenesis and prognosis.
Deborah Alpert, ACP Member, PhD
Another take on near misses
I always enjoy the thought-provoking case presentations in Mindful Medicine. The most recent presentation [“Meld intuition with deliberation to sidestep diagnostic trap, “ ACP Internist, March 2009] emphasized the importance of not giving up on a patient with a clinical impression of illness despite initially normal test results.
The emphasis could equally have been on the value of taking a complete medication history to avoid delayed diagnosis and unnecessary testing. Although billing requirements can be satisfied with a copied or hastily constructed med list, it does not take much longer to ask a patient about “all of the medications you take, even those you buy without a prescription,” and to roughly quantify the amount of each medication that the patient has actually been taking. The “near miss” of the patient described in the case might usefully lead to a review and redesign of the process for obtaining a medication history in the emergency department or upon admission.
Harrison G. Weed, FACP