tion may have to be abandoned and other treatment options for prostate cancer considered. RE: PLASMA SELENIUM LEVEL BEFORE DIAGNOSIS AND THE RISK OF PROSTATE CANCER DEVELOPMENT J. D. Brooks, E. J. Metter, D. W. Chan, L. J. Sokoll, P. Landis, W. G. Nelson, D. Muller, R. Andres and H. B. Carter J Urol, 166: 2034 –2038, 2001 To the Editor. The authors report that the development of prostate cancer is associated with low levels of plasma selenium, and that plasma selenium decreases with patient age. Their data (plotting plasma selenium versus age) show that the decrease in plasma selenium occurs only after age 65 years. Nevertheless, in each age group the plasma selenium level was lower among patients in whom prostate cancer developed. Of particular interest were the 8 men between 65 and 77 years old without prostate cancer, who were the only patients to deviate strikingly from the regression line (above the red line) and to have plasma selenium levels higher than 14.5 g./dl. (see figure). The authors did not mention whether any or all of the patients had taken selenium supplements, which would have ele- vated their selenium levels and possibly had a different biological effect on the development of prostate cancer. Excluding these 8 cases (which alone are too few to sustain analysis free from statistical variation) from the data would make the divergence of the respective regression curves less significant, and raises 2 questions: 1) Does selenium administration produce an elevation of the plasma sele- nium level without additional biological effect on prostate cancer? 2) Conversely, does achievement of an elevated plasma selenium level protect against the development of prostate cancer? The anticipated prospective study of the effect of selenium administration on the development of prostate cancer should answer these questions. Respectfully, David T. Schwartz 5252 Dawes Ave. Alexandria, Virginia 22311 Reply by Authors. All cases and controls in our study were from the Baltimore Longitudinal Study of Aging and were diagnosed well before 1996, when Clark et al reported a positive association between selenium intake and decreased risk of subsequent prostate cancer. 1 Therefore, it is highly unlikely that any of the patients had taken supplementary selenium, although these data are not available. Nonetheless, whether the subjects in our study took supplementary selenium is immaterial to the question we sought to address, namely in men with prostate cancer, does plasma selenium level several years prior to diagnosis differ from closely matched men without prostate cancer? We found a striking association between plasma selenium level and prostate cancer risk, with subjects in the lowest quartile showing a 4 to 5-fold increased risk of diagnosis nearly 4 years later. Since we carefully matched case and control subjects prior to measuring plasma selenium, it is not clear why Schwartz would like to arbitrarily exclude 8 subjects without prostate cancer who happened to have high plasma selenium levels. Our study, as well as several other case control studies and 1 intervention trial, suggests an inverse association between plasma selenium levels and the subsequent risk of prostate cancer. We agree that the Selenium and Vitamin E Cancer Prevention Trial (SELECT) is justified by these findings, and will test whether this association translates into an effective approach to prevent prostate cancer. 1. Clark, L. C., Combs, G. F., Jr., Turnbull, B. W., Slate, E. H., Chalker, D. K., Chow, J. et al: Effects of selenium supplemen- tation for cancer prevention in patients with carcinoma of the skin. A randomized controlled trial. Nutritional Prevention of Cancer Study Group. JAMA, 276: 1957, 1996 RE: SKELETAL FRACTURE ASSOCIATED WITH ANDROGEN SUPPRESSION INDUCED OSTEOPOROSIS: THE CLINICAL INCIDENCE AND RISK FACTORS FOR PATIENTS WITH PROSTATE CANCER M. G. Oefelein, V. Ricchuiti, W. Conrad, A. Seftel, D. Bodner, H. Goldman and M. Resnick J Urol, 166: 1724 –1728, 2001 To the Editor. We appreciate the authors’ acknowledgment that their study design is open to criticism. This admission allows us to refer to valuable information that we believe is lacking in their article. Oefelein et al analyze many of the variables that may be associated with the risk of delayed skeletal fracture in patients with prostate cancer who are undergoing androgen suppression. Never- theless, no mention is made of other oncological treatment previously received by those patients in whom benign bone fractures developed. We treated a case that underscores the fact that prior radiation therapy may also increase the risk of pelvic fracture in patients with prostate cancer. A 78-year-old man with a previous history of localized prostate cancer (Gleason sum 8) presented with pelvic pain. He had been treated with radiation therapy 16 months earlier and had not re- ceived castration therapy or steroid treatment. Plain x-rays showed iliac bone fractures and lytic lesions in the sacrum and pubis char- acteristic of bone metastasis. 99m Technetium bone scan revealed a nonspecific circumscribed pattern of symmetrically increased uptake areas within the pelvic ring (the so-called “butterfly” pattern). 1 Since the serum prostate specific antigen level was undetectable, an addi- tional radiological study was elected prior to histological diagnostic procedures. Subsequent pelvic computerized tomography demon- strated iliac, pubic and sacral wing fractures, and the absence of soft tissue masses excluded the diagnosis of metastatic bone disease. Nonsteroidal anti-inflammatory therapy was then prescribed and physiotherapy was commenced, with functional and radiological pro- gressive healing. Because of factors such as early detection of prostate cancer and increasing age related bone disorders in developed countries, the incidence of the radiation induced “butterfly” pattern is expected to be increased even in those patients with no previous androgen block- ade. 2, 3 Thus, any health practitioner dedicated to the care of patients with prostate cancer should be aware of the possibility of benign skeletal fractures in elderly men previously treated with radiation therapy. Respectfully, Carlos Miro ´, Ottavio de Cobelli and Roberto Orecchia Departments of Radiation Oncology and Urology European Institute of Oncology University of Milan Milan Italy and Gemma Sancho, Humberto Villavicencio and Jordi Craven-Bartle Departments of Radiation Oncology and Urology Santa Creu i Sant Pau Hospital Autonomous University of Barcelona Barcelona Spain LETTERS TO THE EDITOR 662