Predictive value of surveillance cultures for systemic infection due to Candida species

Updates in Laboratory Diagnostics for Invasive Fungal Infections

Appropriate diagnosis of invasive fungal infections (IFIs) is critical due to the high rates of morbidity and mortality, as well as the substantial economic burden, associated with the management of these diseases. The recognition of IFI and differentiation from other infections with similar clinical presentations can be challenging, which can lead to diagnostic error that not only has an impact on individual patient health outcomes but also on antimicrobial drug usage and the growing threat of antimicrobial resistance in bacteria. Therefore, there is a significant need for improved stewardship related to diagnostic testing for and treatment of IFIs. The purpose of this review is to highlight recent advances related to current fungal diagnostics, as well as explore some of the most innovative technology that has emerged with the potential to shift the paradigm of clinical mycology. In general, this review will discuss research related to enhanced fungal culture utilization and identification techniques, expanded applications of fungal antigen testing, and recently developed molecular assays and other novel nonculture fungal diagnostic approaches. Specifically, the application of mass spectrometry, novel glycobiomarker detection, and detection of fungal-specific volatile organic compounds will be reviewed, along with other key updates, to provide the reader with an updated review that extends beyond the basics of IFI laboratory diagnostics. Where appropriate, the reader will be directed to more comprehensive reviews of certain aspects of clinical mycology laboratory testing to provide a broader context for the critical consideration of these updates.

Nosocomial Candidiasis: Antifungal Stewardship and the Importance of Rapid Diagnosis

Candidemia and other forms of candidiasis are associated with considerable excess mortality and costs. Despite the addition of several new antifungal agents with improved spectrum and potency, the frequency of Candida infection and associated mortality have not decreased in the past two decades. The lack of rapid and sensitive diagnostic tests has led to considerable overuse of antifungal agents resulting in increased costs, selection pressure for resistance, unnecessary drug toxicity, and adverse drug interactions. Both the lack of timely diagnostic tests and emergence of antifungal resistance pose considerable problems for antifungal stewardship. Whereas antifungal stewardship with a focus on nosocomial candidiasis should be able to improve the administration of antifungal therapy in terms of drug selection, proper dose and duration, source control and de-escalation therapy, an important parameter, timeliness of antifungal therapy, remains a victim of slow and insensitive diagnostic tests. Fortunately, new proteomic and molecular diagnostic tools are improving the time to species identification and detection. In this review we will describe the potential impact that rapid diagnostic testing and antifungal stewardship can have on the management of nosocomial candidiasis.

Surveillance cultures in pediatric allogeneic hematopoietic stem cell transplantation

The value of surveillance cultures in predicting systemic infections and in guiding antimicrobial treatment is controversial. We investigated 57 pediatric allo-SCTs between 2007 and 2009. ALL (34), AML (5), and severe aplastic anemia (4) were the largest patient groups. Conditioning was TBI-based in 87% and 54% developed GVHD (21% grade III-IV). Of the 2594 weekly colonization samples, 24% were positive (fecal bacteria 86%, fecal fungi 16%, Clostridium difficile 16%; throat bacteria 17% and throat fungi 4%). Enterobacteria and enterococci were the most common fecal findings, staphylococci and streptococci in the throat. Of the bacterial stool samples pretransplant, 74% (mostly enterococci) were resistant to our first-line antibiotics (ceftazidime and cloxacillin). Candida species accounted for the majority of the fungal findings: 62% of the fecal and 78% in the throat. A total of 170 clinical infection episodes were recorded, and in 12 of these, the bacterial blood culture was positive. In 4/12 cases, the pathogen was detected in surveillance culture previously, leading to sensitivity and specificity of 33.3 and 47.4%, respectively. Positive predictive value of bacterial surveillance cultures was 0.9%. The antimicrobial treatment was changed in only five cases based on the surveillance culture results. Weekly surveillance cultures seldom provided clinical benefit and were not cost-effective.

Yield of fungal surveillance cultures in pediatric hematopoietic stem cell transplant patients: a retrospective analysis and survey of current practice

BACKGROUND

Fungal surveillance cultures (FSCs) have been proposed as predictors for development of invasive fungal disease (IFD) and identifiers of the causative organism, although data supporting these are limited and predate universal initiation of antifungal prophylaxis. We aimed to define the epidemiology of fungal colonization and investigate the utility of FSCs for predicting IFD in recipients of pediatric hematopoietic stem cell transplantation (HSCT).

METHODS

FSCs performed from 2007 to 2011 on HSCT patients and laboratory and clinical data were reviewed, and incidence of IFD was determined. Descriptive analyses of culture results were performed to determine the yield of FSCs and their utility. A Web-based survey of national pediatric HSCT providers was undertaken to evaluate current practice and the relevance of our results.

RESULTS

Five thousand six hundred eighteen FSCs from nares, throat, and stool from 360 patients were processed. Of these, 14.8% were positive: 30.3% from stool, 13.2% from throat, and 0.9% from nares; 64.4% of patients had >1 positive FSCs. Thirty (8.3%) patients had IFD. IFD occurred in 7.9% and 10.1% of patients with positive and negative FSCs, respectively (P = .25). Antifungal coverage was changed in 69 patients (29.9%) after positive FSC; 8.6% developed IFD (n = 2 of 6 pathogen concordance with FSC) compared with 6.7% (P = .59) who had no treatment change (n = 3 of 11 concordance). The response rate to the survey was 70.8%; 40% of institutions reported performing routine FSC. Twenty-five percent of providers would not change management based on FSC results; overall rating of usefulness of FSCs was low.

CONCLUSIONS

Although FSCs are commonly performed for pediatric HSCT patients, they have limited utility for predicting IFD.

High incidence of antifungal drug resistance in Candida tropicalis

Drug resistance among yeasts is an increasing problem. Isolates of Candida krusei and Candida glabrata are recognized as having reduced susceptibility to fluconazole and resistance to this drug has also arisen in Candida albicans isolated from AIDS patients on long term azole therapy. Candida tropicalis (CT) is being increasingly isolated from human disease and is associated with invasive infection, however, data regarding this organism's drug susceptibility is limited. We report our findings on 60 isolates of CT isolated from patients with serious infection in the North West of England. Over 60% of isolates were from adult Intensive Care Unit (ICU) patients, and almost half were from the respiratory tract. Susceptibility to fluconazole, flucytosine, itraconazole and ketoconazole were tested by standardised methods - 48% of the isolates were resistant to fluconazole (MIC > 12.5 mg/l), and 10% had intermediate susceptibility (MIC 6.25-12.5 mg/l). For flucytosine 17% of isolates were resistant (MIC > 8 mg/l) and 22% had intermediate susceptibility (MIC 2-8 mg/l). Three isolates were resistant to both drugs. For itraconazole 17% of isolates were resistant (MIC > 1 mg/l), and 12% showed intermediate susceptibility (MIC 0.5-1 mg/l). Resistance to ketoconazole was seen in 33% of isolates (MIC > 1 mg/l) and 10% showed intermediate susceptibility (MIC 0.5-1 mg/l). Differences in the degree of cross resistance between the azole drugs was observed. Candida tropicalis should be added to the list of yeasts in which drug resistance is commonly found. Given the high invasiveness of Candida tropicalis, its affinity for patients on ICU and the high incidence of drug resistance in this species, identification and susceptibility tests should be performed on all yeast isolates from patients on ICU.

Epidemiology of invasive mycoses in North America

The incidence of invasive mycoses is increasing, especially among patients who are immunocompromised or hospitalized with serious underlying diseases. Such infections may be broken into two broad categories: opportunistic and endemic. The most important agents of the opportunistic mycoses are Candida spp., Cryptococcus neoformans, Pneumocystis jirovecii, and Aspergillus spp. (although the list of potential pathogens is ever expanding); while the most commonly encountered endemic mycoses are due to Histoplasma capsulatum, Coccidioides immitis/posadasii, and Blastomyces dermatitidis. This review discusses the epidemiologic profiles of these invasive mycoses in North America, as well as risk factors for infection, and the pathogens' antifungal susceptibility.

Use of CHROMagar Candida for the presumptive identification of Candida species directly from clinical specimens in resource-limited settings

Introduction

Identification of yeast isolated from clinical specimens to the species level has become increasingly important. Ever-increasing numbers of immuno-suppressed patients, a widening range of recognized pathogens, and the discovery of resistance to antifungal drugs are contributing factors to this necessity.

Material and methods

A total of 487 yeast strains were studied for the primary isolation and presumptive identification, directly from clinical specimen. Efficacy of CHROMagar Candida has been evaluated with conventional methods including morphology on Corn meal–tween 80 agar and biochemical methods by using API 20 C AUX.

Results

The result of this study shows that CHROMagar Candida can easily identify three species of Candida on the basis of colonial color and morphology, and accurately differentiate between them i.e. Candida albicans, Candida tropicalis, and Candida krusei. The specificity and sensitivity of CHROMagar Candida for C. albicans calculated as 99%, for C. tropicalis calculated as 98%, and C. krusei it is 100%.

Conclusion

The data presented supports the use of CHROMagar Candida for the rapid identification of Candida species directly from clinical specimens in resource-limited settings, which could be very helpful in developing appropriate therapeutic strategy and management of patients.

Candida krusei, a multidrug-resistant opportunistic fungal pathogen: geographic and temporal trends from the ARTEMIS DISK Antifungal Surveillance Program, 2001 to 2005

Candida krusei is well known as a fungal pathogen for patients with hematologic malignancies and for transplant recipients. Using the ARTEMIS Antifungal Surveillance Program database, we describe geographic and temporal trends in the isolation of C. krusei from clinical specimens and the in vitro susceptibilities of 3,448 isolates to voriconazole as determined by CLSI (formerly NCCLS) disk diffusion testing. In addition, we report the in vitro susceptibilities of bloodstream infection isolates of C. krusei to amphotericin B (304 isolates), flucytosine (254 isolates), anidulafungin (121 isolates), caspofungin (300 isolates), and micafungin (102 isolates) as determined by CLSI broth microdilution methods. Geographic differences in isolation were apparent; the highest frequency of isolation was seen for the Czech Republic (7.6%) and the lowest for Indonesia, South Korea, and Thailand (0 to 0.3%). Overall, 83% of isolates were susceptible to voriconazole, ranging from 74.8% in Latin America to 92.3% in North America. C. krusei was most commonly isolated from hematology-oncology services, where only 76.7% of isolates were susceptible to voriconazole. There was no evidence of increasing resistance of C. krusei to voriconazole from 2001 to 2005. Decreased susceptibilities to amphotericin B (MIC at which 90% of isolates were inhibited [MIC(90)], 4 microg/ml) and flucytosine (MIC(90), 16 microg/ml) were noted, whereas 100% of isolates were inhibited by < or =2 microg/ml of anidulafungin (MIC(90), 0.06 microg/ml), micafungin (MIC(90), 0.12 microg/ml) or caspofungin (MIC(90), 0.25 microg/ml). C. krusei is an uncommon but multidrug-resistant fungal pathogen. Among the systemically active antifungal agents, the echinocandins appear to be the most active against this important pathogen.

Epidemiology of invasive candidiasis: a persistent public health problem

Invasive candidiasis (IC) is a leading cause of mycosis-associated mortality in the United States. We examined data from the National Center for Health Statistics and reviewed recent literature in order to update the epidemiology of IC. IC-associated mortality has remained stable, at approximately 0.4 deaths per 100,000 population, since 1997, while mortality associated with invasive aspergillosis has continued to decline. Candida albicans remains the predominant cause of IC, accounting for over half of all cases, but Candida glabrata has emerged as the second most common cause of IC in the United States, and several less common Candida species may be emerging, some of which can exhibit resistance to triazoles and/or amphotericin B. Crude and attributable rates of mortality due to IC remain unacceptably high and unchanged for the past 2 decades. Nonpharmacologic preventive strategies should be emphasized, including hand hygiene; appropriate use, placement, and care of central venous catheters; and prudent use of antimicrobial therapy. Given that delays in appropriate antifungal therapy are associated with increased mortality, improved use of early empirical, preemptive, and prophylactic therapies should also help reduce IC-associated mortality. Several studies have now identified important variables that can be used to predict risk of IC and to help guide preventive strategies such as antifungal prophylaxis and early empirical therapy. However, improved non-culture-based diagnostics are needed to expand the potential for preemptive (or early directed) therapy. Further research to improve diagnostic, preventive, and therapeutic strategies is necessary to reduce the considerable morbidity and mortality associated with IC.

Invasive Candidiasis

Invasive candidiasis remains an important nosocomial infection that continues to present major diagnostic and therapeutic challenges to the clinician. Changes in the epidemiology of this disorder have occurred for many reasons, and included especially the extensive use of prophylactic antifungal agents, broad-spectrum antibacterial agents, and medical devices (eg, chronic indwelling intravascular catheters). The diagnosis of IC remains elusive in many patients, and there is a critical need for improved diagnostics that will provide clinicians the opportunity to intervene earlier in the course of disease. Newer antifungal agents offer promise in the treatment of candidemia and other forms of IC, but the optimal use of these agents, particularly in the approach to non-albicans Candida infections, needs to be explored in more detail. Furthermore, despite an overwhelming amount of data concerning risk factors and excess mortality associated with the development of IC, there is no consistent approach to treatment and primary prevention among individuals who are deemed to be at highest risk for this complication. Research that focuses on these important clinical areas could provide valuable insights into the diagnosis and management of this common and evolving infection.

The association between anatomic site of Candida colonization, invasive candidiasis, and mortality in critically ill surgical patients

We evaluated whether the likelihood of developing invasive candidiasis (IC) differed depending upon the anatomic site of Candida colonization in 182 surgical intensive care unit (SICU) patients who participated in a randomized trial of fluconazole to prevent candidiasis. We also determined the impact of Candida colonization of different anatomic sites on all-cause SICU and hospital mortality. A total of 2851 surveillance fungal cultures collected from 5 anatomic sites were analyzed. There was a statistically significant difference in the frequency of IC comparing patients with and without urinary (13.2% versus 2.8%, P = .02), respiratory (8.0% versus 1.2%, P = .04), and rectum/ostomy (8.4% versus 0%, P = .01) colonization. Patients with negative rectum/ostomy cultures and patients with both negative urine and respiratory tract cultures did not develop IC. Candiduria detected at any time in the SICU was independently associated with SICU mortality (odds ratio, 2.86; 95% confidence interval, 1.05-7.74). Surveillance fungal cultures of particular anatomic sites may help differentiate patients at higher risk of developing IC from those at low risk.

Primary intestinal aspergillosis after high-dose chemotherapy and autologous stem cell rescue

Primary invasive aspergillosis of the gut is a rare event and is associated with high mortality. We report for the first time on a patient who had isolated aspergillosis of the small bowel after autologous stem cell transplantation. Diagnosis of invasive aspergillosis of the gut was based on abdominal pain, galactomannan antigenemia and isolation of Aspergillus fumigatus from the stool and was later confirmed by pathohistologic examination. No other site of invasive aspergillosis was evident. The patient was successfully treated with early surgery and combination antifungal therapy.

Impact of surveillance stool culture guided selection of antibiotics in the management of pediatric small bowel transplant recipients

Surveillance stool cultures (SSC) have been used in immunocompromised populations to predict the organisms associated with invasive infections and aid in the selection of empiric antibiotic regimens. To evaluate the utility of this approach in pediatric small bowel transplant (SBT) recipients, we conducted a retrospective review of 33 patients who underwent SBT, 16 of whom had SSC done. In no case was the same organism isolated from SSC and subsequent blood, peritoneal fluid or wound cultures. In the first month post-transplantation, blood cultures were positive in 44% and 35% of patients that had and did not have SSC done, respectively (p = 0.73); peritoneal fluid cultures in 44% and 65% (p = 0.30); and wound cultures in 44% and 24% (p = 0.28). There were no significant differences among both groups in time to first infection, duration of ICU stay following SBT, graft survival or long-term patient survival. We conclude that SSC-guided antibiotic selection does not have a significant impact on the incidence of invasive infections in the first month following SBT or on specific indicators of patient outcome. This suggests that empiric antibiotic regimens should be selected based on clinical presentation and hospital flora and susceptibility patterns.

Invasive Candidiasis in Immunocompromised Hospitalized Patients

The frequency of infections by Candida species is increasing worldwide, with candidemia representing the fourth most common bloodstream infection in the U.S. The risk of infection is especially high in the immunocompromised, hospitalized patient. The treatment of and prophylaxis for Candida infection have led to the emergence of resistant species and the acquisition of resistance in previously susceptible species. Current therapeutic options include amphotericin B and its lipid compounds, fluconazole, itraconazole, voriconazole, and caspofungin. Research is focusing on better diagnostics and the evaluation of strategies such as prophylaxis in high-risk hosts and pre-emptive therapy.

Genetic relatedness among Candida tropicalis isolates from sporadic cases of fungemia in two university hospitals in Korea

OBJECTIVE

To compare the epidemiology and genetic relatedness of Candida tropicalis isolates causing bloodstream infection (BSI) in two hospitals.

SETTING

Two tertiary-care hospitals in Korea.

METHODS

A retrospective molecular epidemiologic analysis using pulsed-field gel electrophoresis (PFGE) was performed with 49 C. tropicalis isolates from sporadic cases of BSI. The isolates were collected from 27 patients at Chonnam National University Hospital (CUH) during a 6-year period and 22 patients at Asan Medical Center (AMC) during a 2-year period.

RESULTS

Based on the PFGE patterns, the average similarity value (S AB) for the 27 isolates from CUH was 0.84 +/- 0.08, which was significantly higher than that for the 22 isolates from AMC (0.78 +/- 0.06; P < .001). Of the 49 strains from patients at the 2 hospitals, 9 isolates were placed into 3 subtypes with S AB values of 1.0, which indicated that they were identical. All 9 of these strains were isolated from CUH patients, and each type strain was isolated sporadically during a period ranging from 4 months to 3 years. On comparison of the clinical characteristics of the patients of the 2 hospitals, the CUH strains were isolated more frequently from non-neutropenic patients and patients with central venous catheter-related fungemia; cases from CUH had a better outcome than those from AMC (P < .05).

CONCLUSIONS

These data show that the clinical and epidemiologic characteristics of C. tropicalis fungemia may differ markedly among hospitals and that some cases of C. tropicalis fungemia may be caused by endemic strains within a hospital.

How to Improve the Design of Trials of Antifungal Prophylaxis among Neutropenic Adults with Acute Leukemia

The risk for invasive fungal infections in patients with acute leukemia is generally low (4%-8%), and the routine use of fungal prophylaxis is not warranted except in specific high-risk groups that should be identified among this population. In a prophylactic study with a new agent, fluconazole or itraconazole oral solution represent good choices for the comparator because they are proven better than placebo or oral nonabsorbable antifungal agents in reducing the risk of invasive fungal infections in patients with acute leukemia. Because prophylaxis is most valuable when the risk of infection is high, patients with well-understood risk factors (severe mucosal disruption caused by chemotherapy, impaired cell-mediated immunity caused by steroids or fludarabine, use of a central venous catheter, and colonization by Candida species) should be selected. The end points for antifungal prophylactic trials should focus on proven and probable invasive fungal infections. Superficial and mucosal fungal infections do not represent a primary end point for these studies. Poor compliance should be considered as an interruption of treatment due to side effects and should be included in the criteria for failure. Fungus-related mortality should be evaluated as a failure of prophylaxis, whereas overall mortality may be influenced by many other cofactors. Differences in gastrointestinal toxicity of antifungal agents may limit the use of double-blind designs in some situations.

Candidemia: The Impact of Antifungal Prophylaxis in a Surgical Intensive Care Unit

BACKGROUND

Candidemia is fourfold more common in 1990 compared to 1980. In addition, a shift to non-albicans species has occurred in some institutions. Antifungal prophylaxis (AP) is effective in high-risk patients including critically ill surgical patients, but its use has been attributed to a resultant shift to non-albicans candida species. We hypothesized that the use of fluconazole prophylaxis would lead to a decreased incidence of candidemia but a possible increased incidence of resistant species of Candida, especially Candida glabrata (CG).

METHODS

From 1990 to 2002, all patients with candidemia (C) in the surgical intensive care unit (SICU) of a large tertiary care hospital were identified and reviewed retrospectively. Antifungal prophylaxis began in 2000 for high-risk patients. The periods were separated into PRE (1990-2000), and POST prophylaxis (2000-2002).

RESULTS

Excluding the year of the trial studying prophylaxis, (1998; five cases of C) a total of 83 patients developed candidemia: 69 PRE (83%) (1.94/1000 patient days) and 14 POST (17%) (0.76/1000 patient days) (OR 0.44; 95% CI 0.25, 0.78; p = 0.004). In the PRE period C. albicans (45%) and CG (30%) were predominant, whereas in the POST period, CG (9/14, 64%) (p = 0.05), and C. albicans (3/14, 21%) were common. Non-albicans species were 38/69 (55%) PRE and 11/14 (79%) POST, p = 0.14. Mortality in the group was 43/83 (52%) and did not differ PRE/POST or based on treatment. Predictors of SICU mortality (model r2 = 0.61) included hospital length of stay (LOS) (OR 1.14, CI 1.04, 1.25), fever (OR 51.2, CI 2.46, 1064), and broad-spectrum antibiotics (OR 69.7, CI 2.08, 2351), whereas post-transplantation status (OR 0.005, CI 0.00, 0.56), blood sugar <180 mg/dL (OR 0.03, CI 0.01, 0.81), and fungal prophylaxis (OR 0.03, CI 0.01, 0.58) were associated with a decreased risk of mortality.

CONCLUSIONS

Unfortunately, the mortality of candidemia remains high in SICU patients (52%). In the SICU, risk factors for candidemia and mortality are common. However, antifungal prophylaxis has significantly decreased the annual incidence of candidemia without a statistically significant shift to non-albicans pathogens.

Oral candidiasis in hematopoietic cell transplantation patients: an outcome-based analysis

OBJECTIVES

Despite aggressive antifungal prophylaxis, the increased risk for systemic fungal infection in recipients of hematopoietic cell transplants (HCT) continues to be a significant concern because Candida infection can cause morbidity and mortality in these patients. The objectives of this study were to examine the relationship of oral colonization by Candida species to systemic infection, mortality, and the impact of antifungal treatment on a population of recipients of HCT.

STUDY DESIGN

One hundred and fifteen consecutive patients undergoing hematopoietic cell transplantation were evaluated. Oral examinations and cultures for Candida were completed before transplantation and on a weekly basis until discharge. The oral complications were assessed, and the level of mucositis was scored by using the National Cancer Institute grade. Systemic antifungal prophylaxis was provided to all patients. Chlorhexidine oral rinses were also routinely provided.

RESULTS

Colonization by Candida species was identified in 31% of patients. Fifty-six percent of patients with colonization had clinical evidence of oral candidiasis. Significantly decreased Candida colonization was seen in patients using chlorhexidine alone compared with those using chlorhexidine and nystatin together (P <.046). Twenty-five patients died in the immediate posttransplantation period, 17 of whom were Candida-positive. The length of hospital stay ranged from 15 to 153 days; increased stay was also associated with Candida colonization (P =.04). Seventy-four percent of all patients developed ulcerative mucositis. More severe mucositis was seen in patients undergoing chemotherapy and radiation therapy. There was no significant difference between Candida colonization and the presence or severity of mucositis.

CONCLUSIONS

Despite systemic and topical antifungal prophylaxis, oropharyngeal colonization by Candida species was common in patients who had received HCT. Candidiasis was commonly present in those who did not survive the early transplant period. Of the 25 patients who died early after the transplantation, 92% had ulcerative mucositis in comparison with 70% of those who survived, reflecting the association of oral mucositis with the toxicity of HCT. There was a significant relationship among allogeneic and autologous HCT, length of stay, and colonization of Candida. In patients undergoing systemic antifungal prophylaxis, chlorhexidine rinse was statistically more effective in reducing colonization by Candida than chlorhexidine and nystatin combined (P =.046).

Diagnosis of systemic Candida infections in patients of the intensive care unit. Significance of serum antigens and antibodies

The combined detection of Candida antigen and antibody for the determination of systemic Candida infections (SCI) was investigated. One hundred and four patients from the intensive care unit (ICU) were analysed. Seventeen of the patients were suspected of having SCI, based on clinical and laboratory criteria. In these patients, Candida antigens and antibodies were analysed extensively. Ten patients had a positive Candida antigen (titre >1:16) determined by the latex agglutination assay Cand-Tec(R) and their median antibody titre was 1:160 in the indirect haemagglutination test (HAT). Seven antigen-negative patients had a median titre of 1:1280 (HAT). Forty-one of 42 colonized control patients had negative antigen titres and a median antibody titre of 1:160. The sensitivities and specificities were 58.8% and 97.6% for antigenemia, and 52.9% and 85.7% for antibody detection. These values reached 100.0% and 83.3%, respectively, when the results of both tests were combined. This indicates a high degree of concordance between serological results (Candida antigen and/or antibodies) and clinical presentation. We conclude, that the combined investigation of antigen and antibody titres might be a helpful tool in the characterization of SCI in ICU patients, if antigen titres are >or=1:16 or antibody titres (HAT) are >or=1:640.

Usefulness of bacteriological surveillance cultures for monitoring infection in hospitalized patients: a critical reappraisal

Untargeted bacteriological surveillance of superficial and deep body sites is frequently performed routinely in various clinical settings. This practice is based on the assumption that early identification of surface microbial flora might be predictive of organisms that will later cause invasive disease and that it may consequently assist in guiding empirical antibiotic therapy. A comprehensive review of the literature however indicates that the clinical value and cost-effectiveness of such practices still remain debated and appear largely unproven in most conditions and situations where they are routinely advocated. The present article reviews and critically discusses the available body of evidence supporting or disproving the use of bacteriological surveillance cultures. It is also aimed to issue general recommendations, strategies and methodologies that could be applied in different hospital care settings including the neonatal or adult intensive care as well as the hematology-oncology units.

Fungal colonization of haematological patients receiving cytotoxic chemotherapy: emergence of azole-resistant Saccharomyces cerevisiae

Fungal colonization during cytotoxic chemotherapy was studied in 42 patients with a recent diagnosis of a haematological malignancy. In total, 2759 surveillance cultures were taken from the nostrils, throat, urine, stool and perineal region. Seven hundred and ninety-six positive surveillance cultures (28.9%) yielded 968 fungal isolates. The rate of fungal colonization did not differ between patients with acute leukaemia, patients with other haematological malignancies and control patients in the same ward at admission (71% vs. 67% vs. 80%). Patients with acute leukaemia were colonized at a significantly lower rate in samples from the throat (32%), urine (10%), stool (45%) and perineum (29%) taken during hospitalization when compared with other haematological patients (respective values 58%, 21%, 67% and 45%; P-values 0.001). This could be attributed to differences in the use of antifungal drugs. Although 21/42 (50%) of our patients had multiple-site fungal colonization at the end of follow-up, only one systemic Candida infection was diagnosed. Extensive use of antifungal treatment may have influenced the low incidence of systemic fungal infections during the follow-up. In addition to Candida species, Malassezia furfur, Geotrichum candidum and Saccharomyces cerevisiae were frequently isolated. The rate of S. cerevisiae isolation increased significantly over time after admission (1%, vs. 18% of isolates, P<0.001), suggesting hospital-acquired transmission. These isolates were highly resistant to azole antifungals (MIC90 128 microg/mL for fluconazole and 16 microg/ml, for itraconazole), and caused persistent multiple site colonization in 12 patients. Extensive use of antifungal agents in a haematological ward may keep the incidence of invasive fungal infections low in spite of heavy fungal colonization. However, there may be a risk of emergence of resistant fungal strains.

Fungal surveillance cultures during antifungal prophylaxis with itraconazole in neutropenic patients with acute leukaemia

Fungal colonization has been associated with an increased rate of invasive fungal infections in neutropenic patients. This study evaluates weekly fungal surveillance cultures from the oropharyngeal and perianal space as well as other suspected sites in 219 courses of myelosuppressive chemotherapy with itraconazole antifungal prophylaxis in 116 neutropenic patients with acute leukaemia. Itraconazole was given from the start of chemotherapy in one of six different dosing regimens. Fungal colonization occurred in 68 (31%) of courses, which was lower than in a historical control group without prophylaxis (53%, P = 0.004). Twenty-six per cent of these 116 isolates had a growth rate of more than 50 colony forming units (CFU) per culture. Candida glabrata (51%), Candida albicans (18%) and Candida krusei (4%) were the most frequently isolated species. Higher median itraconazole trough concentrations were associated with a lower growth rate in the cultures (< or = 50 CFU/culture versus > 50 CFU/culture): 710 (430-1180) ng ml-1 versus 900 (560-1650) ng ml-1 (P = 0.015). The use of itraconazole solution--compared with capsules--led to a reduced growth rate (P = 0.035). In conclusion, compared with historical controls itraconazole antifungal prophylaxis reduces the incidence and the extent of fungal colonization during neutropenia in patients with acute leukaemia.

Elucidating the origins of nosocomial infections with Candida albicans by DNA fingerprinting with the complex probe Ca3

Computer-assisted DNA fingerprinting with the complex probe Ca3 has been used to analyze the relatedness of isolates collected from individuals with nosocomial bloodstream infections (BSIs) and hospital care workers (HCWs) in the surgical and neonatal intensive care units (ICUs) of four hospitals. The results demonstrate that for the majority of patients (90%), isolates collected from commensal sites before and after collection of a BSI isolate were highly similar or identical to the BSI isolate. In addition, the average similarity coefficient for BSI isolates was similar to that for unrelated control isolates. However, the cluster characteristics of BSI isolates in dendrograms generated for each hospital compared to those of unrelated control isolates in a dendrogram demonstrated a higher degree of clustering of the former. In addition, a higher degree of clustering was observed in mixed dendrograms for HCV isolates and BSI isolates for each of the four test hospitals. In most cases, HCW isolates from an ICU were collected after the related BSI isolate, but in a few cases, the reverse was true. Although the results demonstrate that single, dominant endemic strains are not responsible for nosocomial BSIs in neonatal ICUs and surgical ICUs, they suggest that multiple endemic strains may be responsible for a significant number of cases. The results also suggest that cross-contamination occurs between patients and HCWs and between HCWs in the same ICU and in different ICUs. The temporal sequence of isolation also suggests that in the majority of cases HCWs are contaminated by isolates from colonized patients, but in a significant minority, the reverse is true. The results of this study provide the framework for a strategy for more definitive testing of the origins of Candida albicans strains responsible for nosocomial infections.

National epidemiology of mycoses survey: a multicenter study of strain variation and antifungal susceptibility among isolates of Candida species

The National Epidemiology of Mycoses Survey (NEMIS) involves six academic centers studying fungal infections in surgical and neonatal intensive care unit (ICU) patients. We studied variation in species and strain distribution and anti-fungal susceptibility of 408 isolates of Candida spp. Candida spp. were isolated from blood, other normally sterile site cultures, abscesses, wounds, catheters, and tissue biopsies of 141 patients hospitalized in the surgical (107 patients) and neonatal (34 patients) ICUs of medical centers located in Oregon, Iowa, California, Texas, Georgia, and New York. Isolates were also obtained from selected colonized patients (16 patients) and the hands of health care workers (27 individuals). DNA typing was performed using pulsed field gel electrophoresis, and antifungal susceptibility to amphotericin B, 5-fluorocytosine, fluconazole, and itraconazole was determined using National Committee for Clinical Laboratory Standards (NCCLS) methods. Important variation in susceptibility to itraconazole and fluconazole was noted: MICs of itraconazole ranged from 0.25 microgram/mL (MIC90) in Texas to 2.0 micrograms/mL (MIC90) in New York. Similarly, the MIC90 for fluconazole was higher for isolates from New York (64 micrograms/mL) compared to the other sites (8-16 micrograms/mL). In general, DNA typing revealed patient-unique strains; however, there were 13 instances of possible cross-infection noted in 5 of the medical centers. Notably, 9 of the 13 clusters involved species of Candida other than C. albicans. Potential transmission from patient-to-patient (C. albicans, C. glabrata, C. tropicalis, C. parapsilosis) and health care worker-to-patient (C. albicans, C. parapsilosis, C. krusei) was noted in both surgical ICU and neonatal ICU settings. These data provide further insight into the epidemiology of nosocomial candidiasis in the ICU setting.

Prophylaxis of fungal infections

Considering the high morbidity and mortality of deep-seated opportunistic mycoses in severely immunosuppressed patients, strategies for prophylaxis appear to be indicated. Exposure to Aspergillus spp. can be prevented by air filtration which has been shown to reduce the rate of infection. However, Candida infections are predominantly caused by colonizing fungi; therefore drug prophylaxis is more promising. Prospective randomized studies proved the effectiveness of fluconazole (FLU) to prevent infections in patients after bone marrow transplantation.

Epidemiological study of Candida spp. colonization in cardiovascular surgical patients

Candida infections involve multiple risk factors. Among the independent risk factors identified, the degree of colonization of Candida spp. allows the prediction of subsequent severe candidosis in surgical patients. The aim of this study was to assess among 13 selected variables, those that would best predict the perioperative variation of the colonization index (CI) of Candida spp. in cardiovascular surgical patients. The colonization index took into account the number of sites colonized and the density of growth. The results showed that 56.8% of our patients were colonized perioperatively. A total of 116 isolates were identified and Candida albicans accounted for 76.7% of the strains. Among the patients who developed post-surgical Candida infections, 57.1% had an increase of the CI early after the operation. By univariate analysis, three factors were significantly associated with an increase of the CI in patients after surgery; sex (female), the duration of central intravascular catheterization and the length of stay in the surgical intensive care unit (SICU). Epidemiological data could help predict those patients who are at risk of developing Candida infections.

Prevalence of Candida tropicalis in clinical specimens from patients with variable clinical syndromes over a 5-year period

In this study, we have examined our records for the isolation of Candida tropicalis from clinical specimens of patients with heterogeneous clinical presentations during the past 5 years. We have found that this species ranks third among all yeasts in frequency of isolation from clinical specimens and that the trend of recovery from the specimens is rising over the years. The isolation rate of C. tropicalis was highest from urine specimens (36%) followed by respiratory specimens (22%). The frequency of isolation of C. tropicalis from vaginal specimens was relatively high (14%), however the trend was declining over the years. In general, the high recovery of Candida tropicalis from clinical specimens of patients with variable disease supports the views of this organism being a major pathogen.

Nosocomial Candida. Epidemiology, transmission, and prevention

The NNIS and the newly established SCOPE data indicate that the relative proportion of organisms causing nosocomial bloodstream infections has changed over the last decade, with Candida species now being firmly established as one of the most frequent agents. The epidemiology of nosocomial candidemia is continually being refined, but established predisposing factors including immunosuppression and malignancies, use of broad spectrum antibiotics, and use of indwelling central catheters remain as significant risk factors. The high cost of health care and greater attention to continuous quality improvement will stimulate better and more effective ways of diagnosing and treating candida infections using combined clinical and microbiologic acumen. There is room for optimism as newer antifungal agents with reduced toxicities have impact on therapy of candidal infections. Aggressive development of still more agents and reformulations of older agents continue in earnest. Even greater consolation comes from the increased awareness of lay and medical personnel alike regarding the appropriate and judicious use of antimicrobial agents.

Infection in the bone marrow transplant recipient and role of the microbiology laboratory in clinical transplantation

Over the past quarter century, tremendous technological advances have been made in bone marrow and solid organ transplantation. Despite these advances, an enduring problem for the transplant recipient is infection. As immunosuppressive regimens have become more systematic, it is apparent that different pathogens affect the transplant recipient at different time points in the posttransplantation course, since they are influenced by multiple intrinsic and extrinsic factors. An understanding of this evolving risk for infection is essential to the management of the patient following transplantation and is a key to the early diagnosis and treatment of infection. Likewise, diagnosis of infection is dependent upon the quality of laboratory support, and services provided by the clinical microbiology laboratory play an important role in all phases of clinical transplantation. These include the prescreening of donors and recipients for evidence of active or latent infection, the timely and accurate microbiologic evaluation of the transplant patient with suspected infection, and the surveillance of asymptomatic allograft recipients for infection. Expert services in bacteriology, mycology, parasitology, virology, and serology are needed and communication between the laboratory and the transplantation team is paramount for providing clinically relevant, cost-effective diagnostic testing.

Variations in DNA subtype and antifungal susceptibility among clinical isolates of Candida tropicalis

Candida tropicalis has been known to be a major cause of invasive Candida infection. Numerous reports have documented C. tropicalis as the most common species of Candida other than C. albicans. The epidemiology and antifungal susceptibility of C. tropicalis are poorly defined. A series of 89 clinical isolates of C. tropicalis from 56 patients hospitalized at seven different U.S. medical centers were analyzed by restriction endonuclease analysis of genomic DNA (REAG) using the restriction enzymes Sfil and BssHII followed by pulsed-field gel electrophoresis (PFGE). The MICs of the isolates for amphotericin B, 5-fluorocytosine (5FC), fluconazole, itraconazole, and D0870 were determined by microbroth dilution testing. A total of 49 different DNA types were identified among the 89 isolates. Generally, each DNA type represented an individual patient, and serial isolates from the same patient were the same DNA type. Small clusters of patients infected with the same DNA type of C. tropicalis suggested possible nosocomial transmission. The MICs of the various antifungal agents were amphotericin B 0.5 to 2.0 micrograms/ml (MIC90 = 2.0 micrograms/ml), 5FC 0.25 to 1.0 microgram/ml (MIC90 = 0.5 microgram/ml), fluconazole 0.25 to 8.0 micrograms/ml (MIC90 = 1.0 microgram/ml), itraconazole 0.03 to 1.0 microgram/ml (MIC90 = 0.5 microgram/ml), and D0870 0.007 to 0.12 microgram/ml (MIC90 = 0.03 microgram/ml). These data support previous observations that infections caused by C. tropicalis frequently originate from the patient's own endogenous flora. Clusters of a single strain in individual hospitals also suggests that limited nosocomial transmission may occur.

Catheter-related complications of cancer therapy

Although the management of CVC-related infection appears complex and at times the literature seems to be contradictory, simple guidelines can direct the clinician in a stepwise fashion. Knowledge of the pathogenesis of each organism and the immune status of the host is crucial to decide whether catheter removal or retention is indicated. For example, in general, GNB bacteremia does not immediately prompt catheter removal in a neutropenic patient but does in a nonneutropenic host because of the gastrointestinal source of the former and a primary catheter source in the latter. In summary, as more CVCs are inserted in patients undergoing chemotherapeutic, antimicrobial, transfusional, and nutritional supportive care, novel approaches to prevention and treatment of the associated infectious complications inherent with such devices are needed. A multifaceted approach from impregnated catheters to local catheter-site antisepsis was reviewed. We may find, however, that as simple handwashing between patients is crucial to infection control, so too is a trained catheter-care team using total barrier precautions and ensuring proper local catheter maintenance critical to preventing CVC-related infections.

Application of CHROMagar Candida for rapid screening of clinical specimens for Candida albicans, Candida tropicalis, Candida krusei, and Candida (Torulopsis) glabrata

CHROMagar Candida is a new differential culture medium that allows selective isolation of yeasts and simultaneously identifies colonies of Candida albicans, C. tropicalis, and C. krusei. We evaluated the use of this medium with 316 yeast isolates including 247 isolated directly on CHROMagar from clinical material. Over 95% of stock and clinical isolates of C. albicans, C. tropicalis, and C. krusei were correctly identified on the basis of colony morphology and pigmentation on CHROMagar. Additionally, CHROMagar also allowed the identification of C. (Torulopsis) glabrata at a similar level of accuracy. The overall agreement between two observers in reading the CHROMagar plates was 95%. Growth of Candida sp. isolates on CHROMagar had no adverse effect on antifungal MICs or Vitek identification results. In parallel, cultures of 548 stool and rectal swab specimens set up on CHROMagar and Sabouraud glucose agar (SGA) were positive in 234 instances. CHROMagar was positive and SGA was negative for 11 specimens, and CHROMagar was negative and SGA was positive for 18 specimens. A single yeast species was isolated on both media from 162 specimens, and in 146 (90%) of these specimens the same species was detected on both CHROMagar and SGA. A total of 43 of the 234 positive cultures contained mixtures of yeast species. Twenty (47%) of these mixed cultures were detected only on CHROMagar. CHROMagar is extremely useful in making a rapid presumptive identification of common yeast species. This capability plus the ability to detect mixed cultures of Candida spp. promises to improve and streamline the work flow in the mycology and clinical microbiology laboratory.

[Laboratory diagnostic peculiarities of mycoses in immunosuppressed patients]

Domestic invasive mycoses are typically present as secondary diseases in patients definitely immunocompromised. This truth should not obscure the fact that a transient overload of the immune system, e.g. in the polytrauma patients, may likewise favour the development of mycoses. The two groups of patients show a comparable course of infection and, to some extent, diagnostic signs: surveillance cultures and monitoring of antibodies are more helpful with trauma patients and less reliable in the typically immunocompromised patients. In the latter, however, antigen tests may yield more reliable results than in the trauma patients. The different functional capacities of the immune system in the two groups of patients may also affect the appearance of fungal elements, particularly of aspergilli, in secretions and biopsies.

Variations in DNA subtype, antifungal susceptibility, and slime production among clinical isolates of Candida parapsilosis

Candida parapsilosis is an important nosocomial pathogen that can proliferate in high concentrations of glucose and form biofilms on prosthetic materials. We investigated the genotypic diversity, slime production, and antifungal susceptibility among 60 isolates of C. parapsilosis from 44 patients and 10 patient care providers from five different medical centers. Molecular typing was performed using macrorestriction digest profiles with BssHII followed by pulsed-field gel electrophoresis (REAG) and by electrophoretic karyotyping (EK). Slime production was evaluated by growing the organisms in Sabouraud broth with 8% glucose and examining the walls of the tubes for the presence of an adherent slime layer. Antifungal susceptibility to amphotericin B, 5-fluorocytosine, fluconazole, and itraconazole was determined using National Committee for Clinical Laboratory Standards proposed standard methods. Overall 28 different DNA types were identified by REAG and EK methods. MIC90 values ranged from 0.12 microgram/ml for itraconazole to 1.0 microgram/ml for fluconazole and amphotericin B. Sixty-five percent of the isolates produced slime: 37% were moderately to strongly positive, 28% were weakly positive, and 35% were negative. Overall, 83% of blood and catheter isolates were slime positive versus 53% of isolates from all other sites (P < 0.05). These data underscore the genetic diversity and susceptibility of C. parapsilosis to antifungal agents. Slime production may be important in enabling C. parapsilosis to cause catheter-related bloodstream infections.

Empiric amphotericin B therapy: the need for a reappraisal

Neutropenic patients are at high risk of developing invasive fungal diseases. A number of studies, both randomized and historical, have demonstrated that empiric therapy with amphotericin B in neutropenic patients with fever, refractory to antibiotics, results in a decrease in the frequency and mortality of deep fungal infections. Recent years have seen a number of advances in the management of neutropenic patients. Reasonably effective antifungal prophylaxis now exists and in many centres forms part of the routine care of neutropenic patients. Other centres advocate the use of selective decontamination and/or protective isolation. Furthermore the duration of neutropenia can be reduced with the use of haematopoetic growth factors. The impact of empiric amphotericin B in patients already benefiting from such treatments has not been adequately studied. The optimum dose of empiric amphotericin B is not defined. The criteria for commencing amphotericin B therapy in febrile neutropenic patients must therefore be redefined on the basis of further studies carried out in the context of these developments. We offer an approach to the use of empiric amphotericin B based on risk factors and prophylaxis.

Selective decontamination with nystatin for control of a Candida outbreak in a neonatal intensive care unit

Selective decontamination of the digestive tract (SDD) with oral nystatin was evaluated as a measure to control an outbreak of Candida infection in a neonatal intensive care unit (NICU). Seventy-six out of 106 neonates who carried Candida spp. received the main study manoeuvre (the application of oral nystatin in the throat and stomach) during the 12-month open trial. One third of the neonates weighed < 1500 g whilst about half were being ventilated. The mean stay was 33.2 d (SD +/- 46.9). Two cases with candidaemia within a fortnight were associated with a yeast carriage rate in the NICU of about 50%; more than 80% of the isolates were Candida parapsilosis. During the implementation period there were four new neonates with fungaemia caused by C. parapsilosis. Once the carriage rate dropped below 5% (P < 0.001), no new cases of systemic infection with the outbreak strain were recognized in the following 8 months. It took 3.5 months to control the outbreak. The observation that all other clinical diagnostic samples were free from Candida suggests that translocation from throat or gut into the systemic circulation occurred. SDD with oral nystatin was effective in reducing the yeast carriage index (mean index 1.93, before SDD; 0.45, after SDD; P < 0.001). A significant reduction of carriage, both in rates and indices, is thought to have contributed to the control of this candida outbreak.

Diagnosis of invasive mycoses in severely immunosuppressed patients

In compromised patients, invasive mycoses are being observed at an increasing frequency and with a broadening spectrum of causative fungi. The rapid and definitive diagnosis of such opportunistic mycoses requires the synoptical consideration of predisposing clinical conditions, of culture isolates, and of the results of nonculture techniques. At present, the latter comprise the monitoring of specific antibodies and of fungal antigens, and the microscopical examination of suspected biopsies. The validity of the individual techniques (e.g., the monitoring of antibodies or antigens) varies with individual mycoses. Alternative methods for monitoring fungal metabolites or nucleic acids are still in the developmental stage. Particularly the amplification of DNA by the polymerase chain reaction (PCR) has a high diagnostic potential. However, at present, it is uncertain whether PCR allows the necessary distinction between colonization and truly invasive infection, and whether PCR can be simplified sufficiently to allow the continuous surveillance of high-risk patients.

Value of the Hybritech ICON Candida Assay in the diagnosis of invasive candidiasis in high-risk patients

A total of 314 sera from 114 patients at risk for invasive candidiasis were analyzed for the presence of antigenemia using the Hybritech enzyme immunoassay (EIA) for detection of Candida mannan in serum (ICON Candida Assay, Hybritech Inc., San Diego, CA). Fourteen patients (12%) had invasive candidiasis documented by positive blood cultures, deep biopsy culture, and histopathology or autopsy, and five patients had probable invasive candidiasis based on a single positive blood culture and no additional signs of candidiasis. Nine patients had candiduria, 43 patients had mucous membrane colonization, 25 patients were not colonized but received empiric amphotericin B, and 18 patients were not colonized and not treated with amphotericin B. All sera were enzymatically extracted, heat treated, and reacted in a solid-phase sandwich EIA. Test results were read visually and with the ICON reader. The sensitivity and specificity of the mannan EIA in detection of documented invasive candidiasis was 86% and 92%, respectively. The positive predictive value was 60% and the negative predictive value was 98%. Among all patients with invasive candidiasis (documented plus probable), the sensitivity was 68%, the positive predictive value 62%, and the negative predictive value 94%. Specimens were positive within 3 days of the first positive culture in 11 (79%) of 14 patients with documented invasive candidiasis.

Laboratory aids in the diagnosis of invasive candidiasis

The laboratory diagnosis of candidiasis continues to be problematic; however, there have been several advances in the past decade which promise to enhance our ability to identify patients at high risk for infection and/or to document invasive candidiasis in critically ill and immunocompromised patients. The introduction of commercially available biphasic blood culture medium and subsequently the lysis-centrifugation procedure has markedly improved the ability of laboratories to detect fungemia. Although serologic methods have not been very successful in diagnosing candidiasis in immunocompromised patients, several antigen detection methods are now under investigation. In addition, detection of fungal metabolites such as D-arabinitol remains promising. Finally, application of the techniques of molecular biology for typing and detection of fungal pathogens has expanded our understanding of candidal infections and may offer the most sensitive and specific means of diagnosing invasive candidiasis.

Diagnosis of disseminated candidiasis in hospitalized patients using the Cand-Tec latex agglutination assay

A total of 1,303 sera from 202 patients at risk for disseminated candidiasis were analyzed for the presence of Candida antigen using a commercially available latex agglutination test (Cand-Tec). Twenty-three patients had disseminated candidiasis documented by positive blood cultures, deep biopsy culture and histopathology or autopsy. Six patients had transient candidemia, 15 patients had candiduria, 62 patients were not colonized yet treated empirically with amphotericin B, and 46 patients were not colonized and not treated with amphotericin B. The sensitivity and specificity of the Candida antigen test for the diagnosis of disseminated candidiasis was 87% and 36% (threshold titer of greater than or equal to 1:2), 70% and 60% (greater than or equal to 1:4), and 30% and 85% (greater than or equal to 1:8), respectively. In contrast to previous studies we were unable to demonstrate a prognostic role for the Candida antigen test in patients with documented disseminated candidiasis. The lack of sensitivity and specificity of the Cand-Tec Candida antigen test precludes its use in the diagnosis of disseminated candidiasis.

Protecting neutropenic patients from bowel-derived organisms

Prevention of infection from bowel-derived organisms in neutropenic patients requires both the appropriate use of chemoprophylaxis and close attention to the prevention of cross-colonization or cross-infection with resistant Enterobacteriaceae and pseudomonads. Control of common-source infection and control of Gram-positive infection are also important. The objectives of chemoprophylaxis should be considered and their efficacy regularly assessed. Non-absorbable antibiotics may have an important place in minimizing selection of resistant strains, but absorbed agents such as cotrimoxazole (trimethoprim/sulphamethoxazole) and 4-quinolones offer advantages over these and nalidixic acid as prophylactic agents. Ciprofloxacin prophylaxis is probably more effective at reducing Gram-negative bacteraemia than co-trimoxazole but overall mortality may be higher. Further confirmation and investigation of the reasons for this are needed. Protocols of rational antibiotic prophylaxis and treatment involving these agents can be modified to cover only the Gram-negative superinfections that are likely.

Fungemia in a university hospital 1984-1988. Clinical and mycological characteristics

118 episodes of fungemia occurring at Rigshospitalet, Copenhagen, between 1984 and 1988 were reviewed retrospectively. Underlying diseases in the patients were dominated by malignancies, primarily hematological disorders, and intraabdominal diseases requiring major abdominal surgery. Predisposing factors identified in the patients were ongoing antibacterial chemotherapy (83%), central venous catheters (72%), major abdominal surgery (39%), and neutropenia (32%). 120 fungal strains were isolated, of which 88 (73%) were Candida albicans, 23 strains representing 8 other Candida species were also isolated, as were 9 strains belonging to 7 other fungal genera. There were only 5 strains resistant to 5-fluorocytosine (MICs greater than or equal to 25 mg/l), and no strain was resistant to amphotericin B. Treatment with antifungal agents was given in 78 patients, generally a combination of amphotericin B and 5-fluorocytosine. In 14 patients (all non-hematological) the only treatment was removal of a permanent central venous catheter. The outcome was poor in patients with hematological disorders (mortality 76%), whereas patients with malignant and non-malignant intraabdominal diseases had a mortality of 35%. All patients with a permanent central venous catheter as the only risk factor recovered rapidly after removal of the catheter.

Characterization of the sequence of colonization and nosocomial candidemia using DNA fingerprinting and a DNA probe

The objective of this hospital-based study was to determine the relationship between colonizing and infecting strains of Candida species and Torulopsis glabrata. Surveillance cultures from high-risk patients were paired with subsequent bloodstream isolates. Organisms were typed by using restriction endonuclease digestion of chromosomal DNA with BstNI and EcoRI, followed by Southern hybridization with a DNA probe (pBD4) derived from Saccharomyces cerevisiae. Sixteen patients for whom documented colonization preceded documented bloodstream infection were identified. The mean time between obtainment of surveillance isolates and obtainment of bloodstream isolates was 8 days, with a range of 1 to 423 days. For 15 (94%) of 16 patients, the DNA fingerprint pattern (using BstNI) of the surveillance isolate was identical to that of the bloodstream isolate. Isolates from 13 (81%) of 16 patients were unique to those patients. Typing by Southern hybridization with the pBD4 probe was less discriminating. We conclude that for a well-defined subset of hospitalized patients who were colonized by Candida species before developing nosocomial candidemia, the colonizing and infecting strains were identical, suggesting endogenous acquisition of infection. Restriction endonuclease digestion of chromosomal DNA was shown to be a discriminating and reproducible typing method for Candida species and T. glabrata.

The use of biotyping and DNA fingerprinting in typing Candida albicans from hospitalized patients

The application of typing procedures for the purpose of strain differentiation among isolates of Candida albicans obtained from hospitalized patients has been limited. We have applied biotyping and DNA restriction fragment analysis (DNA fingerprinting) by using EcoRI to the study of C. albicans isolates obtained from hospitalized patients. A total of 68 isolates from 15 patients were studied. Thirteen subtypes were identified by biotyping, 8 by DNA fingerprinting, and 21 by a combination of the biotyping and DNA fingerprinting approaches (composite subtype). Both techniques were highly reproducible. In examining the strain variation among isolates obtained from multiple anatomic sites over time, we found that similar, if not identical, strains were recovered from the oropharynx, urine, stool, and blood in a given patient, and these strains persisted. Only rarely did two patients share the same composite subtype suggesting sporadic nosocomial transmission. The combination of biotyping and DNA fingerprinting improved strain discrimination compared to either method alone. Further investigation with these and other epidemiologic typing methods will be necessary to enhance the understanding of the epidemiology and pathogenesis of candidiasis in hospitalized patients.

Rapid, colorimetric identification of Candida albicans

A total of 706 yeast isolates were evaluated in parallel by the Candida albicans Screen (CAS; Carr-Scarborough Microbiologicals, Inc., Stone Mountain, Ga.) test and the germ tube (GT) test in comparison with the API 20C Yeast Identification System. The CAS and GT tests correctly identified 419 of the 422 isolates of C. albicans (99.3%). Two of the false-negative reactions occurring with the CAS were with GT-negative strains of C. albicans. There were two false-positive CAS reactions involving a single strain each of C. parapsilosis and C. tropicalis. Sensitivity and specificity for both tests exceeded 99%, with positive and negative predictive values of 99 and 98%, respectively.

Value of the Cand-Tec Candida antigen assay in the diagnosis and therapy of systemic candidiasis in high-risk patients

A total of 911 sera from 171 patients at risk for systemic candidiasis and 24 sera from 24 non-hospitalized control subjects were analyzed for the presence of candida antigen using a commercially available latex agglutination test (Cand-Tec). Thirty-seven (22%) patients had systemic candidiasis documented by positive blood cultures, deep biopsy culture and histopathology or autopsy. Six patients had transient candidemia, 20 patients had candiduria, 53 patients had mucous membrane colonization, 21 patients were not colonized but received empiric amphotericin B, and 34 patients were not colonized and not treated with amphotericin B. The intraobserver reproducibility was 90% for the exact titer and 100% for a deviation of one dilution. The sensitivity and specificity of the candida antigen test in detection of systemic candidiasis was 95% and 50% (greater than or equal to 1:2), 73% and 72% (greater than or equal to 1:4), and 46% and 80% (greater than or equal to 1:8) respectively. Despite the poor specificity, serial antigen determinations in patients with documented systemic candidiasis demonstrated both an early diagnostic and prognostic role for the candida antigen test. Seventy-one percent of patients whose antigen titer increased during the course of amphotericin B therapy of documented infection died versus only 13% of those whose titer decreased while on therapy (p = 0.01). The candida antigen test has a limited yet potentially useful role in the diagnosis and management of systemic candidiasis in high-risk patients.