Twenty-six cases of candidemia associated with a well-defined urinary tract source were retrospectively identified and reviewed. Urinary tract abnormalities were present in 23 of 26 patients (88%), 19 (73%) of whom had urinary tract obstruction. Nineteen patients had undergone urinary tract procedures before the onset of candidemia. Episodes of candidemia were brief and low-grade in intensity (median duration, 1 day; median colony count, 1.5 cfu/10 mL of blood).Only eight patients (31%) received >500 mg of amphotericin B. There were five in-hospital deaths (19%); two of these deaths were attributed to candidiasis. No late complications of candidemia were documented for the surviving patients. Patients with urologic pathology and candiduria who undergo surgery or manipulation of the urinary tract are at significant risk for candidemia, and further studies should examine the issue of administration of prophylaxis to this group.Candiduria often presents a dilemma to the clinician, as it may represent colonization or infection. The prevalence of candiduria has been estimated at 0.2%-6% [1, 2] among asymptomatic volunteers and at 6.5%-20% among hospitalized patients [2,3]. Despite the high prevalence of candiduria, it is not well known which patients will develop complications such as candidemia. For patients in the critical care setting, the mortality associated with urine cultures positive for Candida species was 50%, as opposed to 19% for such patients without cultures positive for Candida [4]. On the other hand, Schonebeck and Ansehn [5] observed 40 patients with candiduria that persisted from 1 to 12 months and found that the infection spontaneously resolved in the majority of patients, although five remained candiduric after >12 months. The goal of this retrospective study was to describe the characteristics of candidemia arising in patients with a well-defined urinary tract source of infection.
Materials and MethodsAll patients with blood cultures positive for Candida species at the Mayo Clinic from March 1985 to December 1987 were identified through a review of microbiological records. Patients with candidemia were eligible for inclusion in the study if they had concomitant candiduria and no alternative source of candidemia. Candiduria was defined as the pres- ence of Candida species in urine cultures or the presence of yeast in the urine on microscopy. Recovery of Candida species from skin, sputum, or mucous membranes did not constitute criteria for exclusion. All in-patient and out-patient records were reviewed, and the following data were recorded: age, sex, underlying diseases, potential risk factors, clinical and laboratory features, radiological study findings, pathology reports, and treatment and outcome. Blood cultures were done for patients if they had fever or if the clinical circumstances suggested the possibility of systemic infection.The onset of fungemia was defined as the first day on which blood cultures positive for Candida were obtained. Potential risk factors were defined as: (1) administra...