This report describes two cases of intravenous injection of elemental mercury. The presence of mercury in the environment is due to pollution from power plants and in food, particularly fish, and dental amalgams. One patient succumbed and the other remains asymptomatic two years after the surgical removal of all the injected mercury. Management of intravenous injection of elemental mercury (intended to be an aphrodisiac in these two cases) is discussed here and the need for surgical removal of all accessible mercury has been emphasized. In case no.1, it is difficult to come to any conclusion as we do not have information about the duration of exposure and the volume of mercury injected. In contrast, in case no.2, all the metallic elemental mercury was removed on two occasions and the patient has led an uneventful life showing no clinical signs of mercury poisoning in the follow-up period of two years.
463 Background: Urine cytology has often been reported as a highly specific but poorly sensitive test. Cytology is reported as positive, negative, atypical, or suspicious. Atypical/suspicious cytologies account for roughly a quarter of the results and present a clinical dilemma. Physicians’ risk aversion means they are typically treated as positive and result in clinical action. We test the effects of this assumption on sensitivity and specificity. Methods: After IRB approval, we queried clinical and pathology databases to identify all subjects at Duke University Medical Center who had undergone both a urine cytology and a cystoscopy from 1/2003 to 1/2012. Diagnostic test performance metrics were calculated using logistic models: (a) a generalized estimating equation (GEE) and (b) a generalized linear mixed model (GLMM). These take into account clustered/correlated test results that occur due to repeated testing within subjects. Results: A total of 990 unique subjects were identified that provided 4,733 pairs of cytology and cystoscopy for analysis. Our cohort was 61% male, 75% Caucasian, and had 54% current or former smokers. Of cytologies, 1898 (40%) were negative, 423 (9%) positive, and 2408 (51%) suspicious or atypical. When suspicious/atypical cytology results using the GLMM model were classified as positive, the specificity was 62% [95%CI: 58-66%] and the sensitivity was 41% [95% CI: 38-44%]. When these results were re-classified as negative, this had the effect of a large increase in specificity 100% [95%CI: 100-100%] with a consequent decrease in sensitivity 0% [95%CI: 0-2%]. Conclusions: In our study, the performance of urine cytology depended heavily on how the equivocal (atypical/suspicious) results were classified and dealt with. Our sensitivity was maximized when equivocal cytologies were considered positive, but at significant detriment of the specificity. Contrarily, our specificity improved greatly when the equivocal results were considered negative, but at the expense of a poor sensitivity. Furthermore, the diagnosis of an atypical/suspicious cytology was higher at our medical center than reported in the literature, and therefore significantly overestimated the performance of the urine cytology test.
