Quantitative urine cultures do not reliably detect renal candidiasis in rabbits.

The significance of quantitative urine cultures in patients at risk for hematogenous disseminated candidiasis is controversial. While various concentrations of Candida spp. in urine have been suggested as critical cutoff points in the diagnosis of renal candidiasis, other investigators consider quantitative cultures less critical in diagnosing upper tract infections. To determine the significance of quantitative urine cultures in renal candidiasis, we studied serial quantitative urinary cultures of Candida albicans in a rabbit model of hematogenous infection. Of 197 urine samples from 34 infected animals, 144 were culture positive, with a sensitivity of 73.1% for urine cultures and a lower limit of detection of 10 CFU/ml. The yield of urine cultures varied according to severity and duration of infection. The mean renal and urinary concentrations of C. albicans from rabbits with subacute candidiasis differed significantly from those from rabbits with acute candidiasis (P = 0.013 and P < or = 0.001, respectively). During the first 4 days of subacute renal candidiasis, more than one-half of all urine cultures were negative for C. albicans. Only 12 (8.1%) of 148 urine cultures in animals with subacute renal candidiasis had concentrations of > 10(3) CFU/ml, 2.7% had concentrations of > 10(4) CFU/ml, and none were > or = 10(5) CFU/ml. By comparison, all urine cultures from the animals with lethal acute renal candidiasis had higher concentrations of C. albicans and were positive throughout the course of infection. Urinary concentrations of C. albicans were not predictive of the amount of Candida in the kidney (r < or = 0.49) and did not correlate with survival (r = 0.0232). However, the renal concentration of C. albicans (in CFU/gram) inversely correlated with the duration of survival (in days) of rabbits with renal candidiasis (r = 0.76; P < 0.001). These findings indicate that a negative urine culture in rabbits does not preclude the presence of renal candidiasis. The interpretation of a urine culture positive at any concentration, on the other hand, must involve an analysis of the risk factors for renal candidiasis, for any urinary concentration of C. albicans may reflect kidney infection.

Detection of Candida casts in experimental renal candidiasis: implications for the diagnosis and pathogenesis of upper urinary tract infection.

The distinction between upper versus lower urinary tract infection in patients with candiduria is a commonly encountered and therapeutically important diagnostic dilemma. Candida casts have been reported in the urine of several individual case reports of human renal candidiasis. The specificity of Candida casts would identify unequivocally a patient with upper urinary tract disease. Little is known, however, about the sensitivity and the formation of Candida casts. We therefore studied the diagnostic yield, methods for detection and pathogenesis of Candida cast formation in serially collected urine specimens from immunologically intact and granulocytopenic rabbit models of haematogenous disseminated candidiasis. Refractile blastoconidia and pseudohyphae of Candida encased in the granular matrix were seen on wet mounts while Candida stained a brilliant red in the fuschia pink tubular matrix on periodic acid Schiff (PAS) stained cytopathology filters. Among 24 rabbits with disseminated candidiasis, 11 (46%) had Candida casts detectable by wet mount and PAS-stained urine filters in comparison to none of 10 non-infected immunologically normal controls (P = 0.014). Fifteen (70%) of 21 episodes of Candida casts were detected within the first 3 days of infection, indicating possible utility in the early diagnosis of renal candidiasis. No Candida casts were detected in the urine of granulocytopenic rabbits, possibly due to the rapid destruction of tubules and abrogation of cast formation. This absence of detectable Candida in eight infected granulocytopenic rabbits differed significantly from that of 24 non-granulocytopenic infected rabbits, in which Candida casts were detected in 11 (46%) (P = 0.029). Candida cast formation occurred predominantly in the cortex. Histopathological examination demonstrated invasion of Candida into the glomerular tufts and peritubular capillaries, followed by development of Candida casts in the proximal and distal tubules, respectively. Detection of renal Candida casts may be a useful diagnostic marker in distinguishing upper versus lower urinary tract candidiasis.

Hemodialysis-hemoperfusion in fulminant viral hepatitis.

It is theorized that charcoal hemoperfusion and hemodialysis with a highly permeable membrane should be synergistic in the management of fulminant hepatic failure. Previous studies have shown marked differences between these two treatments in their effects in the various biochemical abnormalities of this serious clinical problem. Survival rates in patients treated with charcoal hemoperfusion approximate the rates in those treated with hemodialysis with a highly permeable membrane but are only slightly better than in untreated cases. The excellent recovery of two patients with fulminant viral hepatitis in stage III and IV coma treated with this combined regimen indicates that further studies should be done to confirm this thesis.