Candidemia on presentation to the hospital: development and validation of a risk score

Predictive Modeling Using Artificial Intelligence and Machine Learning Algorithms on Electronic Health Record Data: Advantages and Challenges

The rapid adoption of electronic health record (EHR) systems in US hospitals from 2008 to 2014 produced novel data elements for analysis. Concurrent innovations in computing architecture and machine learning (ML) algorithms have made rapid consumption of health data feasible and a powerful engine for clinical innovation. In critical care research, the net convergence of these trends has resulted in an exponential increase in outcome prediction research. In the following article, we explore the history of outcome prediction in the intensive care unit (ICU), the growing use of EHR data, and the rise of artificial intelligence and ML (AI) in critical care.

Development and Validation of a Risk Score for Predicting Invasive Candidiasis in Intensive Care Unit Patients by Incorporating Clinical Risk Factors and Lymphocyte Subtyping

Objective

To develop and validate a rapid invasive candidiasis (IC)-predictive risk score in intensive care unit (ICU) patients by incorporating clinical risk factors and parameters of lymphocyte subtyping.

Methods

A prospective cohort study of 1054 consecutive patients admitted to ICU was performed. We assessed the clinical characteristics and parameters of lymphocyte subtyping at the onset of clinical signs of infection and their potential influence on IC diagnosis. A risk score for early diagnosis of IC was developed and validated based on a logistic regression model.

Results

Sixty-nine patients (6.5%) had IC. Patients in the cohort (N=1054) were randomly divided into a development (n=703) or validation (n=351) cohorts. Multivariate logistic regression identified that CD8+ T-cell count ≤143 cells/mm³, receipt of high-dose corticosteroids (dose ≥50 mg prednisolone equivalent), receipt of carbapenem/tigecycline, APACHE II score≥15, (1,3)-β-D-glucan (BDG) positivity and emergency gastrointestinal/hepatobiliary (GIT/HPB) surgery were significantly related with IC. IC risk score was calculated using the following formula: CD8+ T-cell count ≤143 cells/mm³ + receipt of high-dose corticosteroids + receipt of carbapenem/tigecycline + APACHE II score ≥15 + BDG positivity + emergency GIT/HPB surgery ×2. The risk scoring system had good discrimination and calibration with area under the receiver operating characteristic (AUROC) curve of 0.820 and 0.807, and a non-significant Hosmer-Lemeshow test P=0.356 and P=0.531 in the development and validation cohorts, respectively. We categorized patients into three groups according to risk score: low risk (0-2 points), moderate risk (3-4 points) and high risk (5-7 points). IC risk was highly and positively associated with risk score (Pearson contingency coefficient=0.852, P for trend=0.007). Candida score had a moderate predicting efficacy for early IC diagnosis. The AUROC of the risk score was significantly larger than that of Candida score (0.820 versus 0.711, Z=2.013, P=0.044).

Conclusions

The predictive scoring system, which used both clinical factors and CD8+ T cell count, served as a clinically useful predictive model for rapid IC diagnosis in this cohort of ICU patients.

Clinical Trial Registration

chictr.org.cn, identifier ChiCTR-ROC-17010750.

Clinical predictive models of invasive Candida infection: a systematic literature review

Clinical predictive models (CPM) serve to identify and categorize patients into risk categories to assist in treatment and intervention recommendations. Predictive accuracy and practicality of models varies depending on methods used for their development, and should be evaluated.The aim of this study was to summarize currently available CPM for invasive candidiasis, analyze their performance, and assess their suitability for use in clinical decision making.We identified studies that described the construction of a CPM for invasive candidiasis from PubMed/MEDLINE, EMBASE, SCOPUS, Web of Science, Cochrane Library databases and Clinicaltrials.gov. Data extracted included: author, data source, study design, recruitment period, characteristics of study population, outcome types, predictor types, number of study participants and outcome events, modelling method and list of predictors used in the final model. Calibration and discrimination in the derivative datasets were used to assess the performance of each model.Ten articles were identified in our search and included for full text review. Five models were developed using data from ICUs, and five models included all hospitalized patients. The findings of this review highlight the limitations of currently available models to predict invasive candidiasis, including lack of generalizability, difficulty in everyday clinical use, and overly optimistic performance.There are significant concerns regarding predictive performance and usability in every day practice of existing CPM to predict invasive candidiasis.

LAY SUMMARY

Clinical predictive models may assist in early identification of patients at risk for invasive candidiasis to initiate appropriate treatment. The findings of this systematic review highlight the limitations of currently available models to predict invasive candidiasis.

Disease-based antimicrobial stewardship: a review of active and passive approaches to patient management

Although new antimicrobial stewardship programmes (ASPs) often begin by targeting the reduction of antimicrobial use, an increasing focus of ASPs is to improve the management of specific infectious diseases. Disease-based antimicrobial stewardship emphasizes improving patient outcomes by optimizing antimicrobial use and increasing compliance with performance measures. Directing efforts towards the comprehensive management of specific infections allows ASPs to promote the shift in healthcare towards improving quality, safety and patient outcome metrics for specific diseases. This review evaluates published active and passive disease-based antimicrobial stewardship interventions and their impact on antimicrobial use and associated patient outcomes for patients with pneumonia, acute bacterial skin and skin structure infections, bloodstream infections, urinary tract infections, asymptomatic bacteriuria, Clostridium difficile infection and intra-abdominal infections. Current literature suggests that disease-based antimicrobial stewardship effects on medical management and patient outcomes vary based on infectious disease syndrome, resource availability and intervention type.

Open Abdomen Management and Candida Infections: A Very Likely Link

OBJECTIVE

Laparostomy can be applied in trauma, abdominal sepsis, intra-abdominal hypertension, or compartment syndrome. Systemic infections, especially if complicated by Candida, are associated with a high risk of mortality.

METHODS

This is a single-centre retrospective case series of 47 cases admitted to our Department, which required laparostomy procedure; we analyzed the type of surgery, temporary abdominal closure, duration of open abdomen, complications, SOFA score, mortality with Candida infections, and empirical or targeted antifungal therapy.

RESULTS

We found that patients with Candida infection were related with a statistically significant difference (p < 0.05) with a complication after OA closure, total complications, time elapsed after OA application, time spent on the first surgical OA application, type of temporary abdominal closure that is used, and duration of the open abdomen. The use of empirical and targeted antifungal therapy is related to the duration of open abdomen too.

CONCLUSIONS

Management of the OA is often burdened by sepsis or septic shock, especially when complicated by Candida infection. Candida score is a validated tool to identify patients who can be treated empirically, but every situation must be considered on an individual basis.

Evaluation of Candida scoring systems to predict early candidemia: A prospective and observational study at a tertiary care hospital, Uttarakhand

Background:

Candidemia in critically ill patients is usually a severe and life-threatening condition. Furthermore, due to its nonspecific presentation, it is difficult to diagnose leading to delayed treatment, prolonged hospitalization, and increased health-care costs with increase in morbidity and mortality.

Objectives:

In view of lack of data on “Candida scoring systems,” this study was designed to evaluate the effectiveness of these scoring systems in predicting the development of candidemia among the Intensive Care Unit patients.

Materials and Methods:

The “Candida score” was calculated at the onset of systemic inflammatory response syndrome, sepsis, or shock. Various scoring systems were compared using the area under the receiver operating characteristic curve.

Results:

Among all three bedside risk scoring systems to predict candidemia both Leon score and Wenzel score offered significant discrimination between candidemic and noncandidemic patients with P = 0.000 and 0.001, respectively. The area under the curve for the scoring systems was 0.946 (95% confidence interval [CI] = 0.89–1) and 0.818 (95% CI = 0.687–0.949).

Conclusion:

Leon scoring system was found to have highest specificity, diagnostic accuracy, and positive likelihood ratio among all. Thus, we might conclude that a Leon score of ≥2.5 was most suitable for diagnosis of candidemia with significant accuracy and shortening of turnaround time when compared to the gold standard of blood culture. To the best of our knowledge, this is the first report on the subject.

Invasive Fungal Infections in the Intensive Care Unit

Patients in the intensive care unit are exposed to multiple stressors that predispose them to invasive fungal infections (IFIs), which carry high morbidity and mortality. Getting acquainted with the diagnostic methods and therapies is imperative for patient safety and for providing high-quality health care. This article focuses on the most frequent IFIs: invasive candidiasis and invasive aspergillosis.

Diagnosis and management of invasive candidiasis in the ICU: an updated approach to an old enemy

Invasive fungal infections, particularly those caused by Candida species, are not uncommon in critically ill patients and are associated with considerable morbidity and mortality. Diagnosis and management of these infections can be challenging. In this review, we will briefly discuss recent epidemiological data on invasive candidiasis and current diagnostic approaches before concentrating on antifungal treatments.

Risk factors for candidemia in non-neutropenic critical patients in Colombia

OBJECTIVES

Due to the increase in isolation of Candida spp. in critically ill patients, and the high mortality and economic costs which this infection entails, a study was made of the risk factors associated to candidemia in critically ill patients from 7 intensive care units in Colombia.

MATERIALS AND METHODS

A multicenter matched case-control study was conducted in 7 intensive care units of 3 university hospitals. Data on overall length of hospital stay (including both general wards and the intensive care unit) were recorded.

RESULTS

A total of 243 subjects (81 cases and 162 controls) between January 2008 and December 2012 were included. In order of frequency, C. albicans, C. tropicalis and C. parapsilosis were isolated. The main identified risk factors were: overall length of hospital stay>25 days (OR 5.33, 95% CI 2.6-10.9), use of meropenem (OR 3.75, 95% CI 1.86-7.5), abdominal surgery (OR 2.9, 95% CI 1.39-6.06) and hemodialysis (OR 3.35, 95% CI 1.5-7.7). No differences in mortality between patients with candidemia and controls were found (39.5 vs. 36.5%, respectively, P=.66) were found.

CONCLUSIONS

In Colombia, a long hospital stay, abdominal surgery, the use of meropenem and hemodialysis were identified as risk factors for candidemia.

Relevance of Candida and other mycoses for morbidity and mortality in severe sepsis and septic shock due to peritonitis

This single-centre retrospective cohort study evaluated the incidence and outcome of mycoses in critical ill patients (n = 283) with sepsis due to peritonitis. Overall mortality was 41.3%, and the 28-day mortality was 29.3%. Fungal pathogens were found in 51.9%. The common first location was the respiratory tract (66.6%), followed by the abdominal site (19.7%). Candida colonisation was found in 64.6%, and invasive Candida infection in 34.0%. Identified fungi were Candida spp. in 98.6% and Aspergillus spp. in 6.1%. Patients with fungal pathogens showed a higher rate of postoperative peritonitis, APACHE II and tracheotomy. In comparison to patients without fungal pathogens, these patients showed a longer duration on mechanical ventilation, and a higher overall mortality. Patients with Candida-positive swabs from abdominal sites had more fascia dehiscence and anastomosis leakage. Seventy-two patients (48.9%) received antifungal therapy, 26 patients were treated empirically. Antifungal therapy was not associated with a decrease in mortality. Age and renal replacement therapy were associated with mortality. In conclusion, fungi are common pathogens in critically ill patients with peritonitis, and detection of fungi is associated with an increase in overall mortality. Particularly, Candida-positive abdominal swabs are associated with an increase in morbidity. However, we were not able to demonstrate a survival benefit for antifungal therapy in peritonitis patients.

Development and validation of a clinical prediction rule for candidemia in hospitalized patients with severe sepsis and septic shock

OBJECTIVE

To develop and internally validate a prediction rule for the presence of candidemia in patients with severe sepsis and septic shock (candidemia rule) that will fill the gap left by previous rules. To compare the accuracy of the available Candida prediction models.

DESIGN

Retrospective cohort study.

SETTING

Barnes-Jewish Hospital, St. Louis, Missouri.

PATIENTS/SUBJECTS

Two thousand five hundred ninety-seven consecutive patients with a positive blood culture and severe sepsis or septic shock.

INTERVENTIONS

Logistic regression and a bootstrap resampling procedure were employed for model development and internal validation.

MEASUREMENTS AND MAIN RESULTS

Two hundred sixty-six (10.2%) had blood cultures positive for Candida spp. Mortality was significantly higher in patients with candidemia than in patients with bacteremia (47.0% versus 28.4%; P<.001). Administration of total parenteral nutrition, prior antibiotic exposure, transfer from an outside hospital or admission from a nursing home, mechanical ventilation and presence of a central vein catheter were independent predictors of candidemia while the lung as a source for infection was protective. The prediction rule had an area under the receiver operating characteristic curve of 0.798 (95% CI 0.77-0.82). Internal validation using bootstrapping technique with 1000 repetitions produced a similar area under the receiver operating characteristic curve of 0.797 (bias, -0.037; root mean square error 0.039). Our prediction rule outperformed previous rules with a better calibration slope of 0.96 and Brier score of 0.08.

CONCLUSIONS

We developed and internally validated a prediction rule for candidemia in hospitalized patients with severe sepsis and septic shock that outperformed previous prediction rules. Our study suggests that locally derived prediction models may be superior by accounting for local case mix and risk factor distribution.

Risk prediction for invasive candidiasis

Over past few years, treatment of invasive candidiasis (IC) has evolved from targeted therapy to prophylaxis, pre-emptive and empirical therapy. Numerous predisposing factors for IC have been grouped together in various combinations to design risk prediction models. These models in general have shown good negative predictive value, but poor positive predictive value. They are useful in selecting the population which is less likely to benefit from empirical antifungal therapy and thus prevent overuse of antifungal agents. Current article deals with various risk prediction models for IC and their external validation studies.

Nosocomial candidemia; risk factors and prognosis revisited; 11 years experience from a Norwegian secondary hospital

The aim of the study was to review the epidemiology and prognosis of candidemia in a secondary hospital, and to examine the intra-hospital distribution of candidemia patients. Study design is a retrospective cohort study. Trough 2002–2012, 110 cases of candidemia were diagnosed, giving an incidence of 2, 6/100000 citizens/year. Overall prognosis of candidemia was dismal, with a 30 days case fatality rate of 49% and one year case fatality rate of 64%. Candidemia was a terminal event in 55% of 30 days non-survivors, defined as Candida blood cultures reported positive on the day of death or thereafter (39%), or treatment refrained due to hopeless short-term prognosis (16%). In terminal event candidemias, advanced or incurable cancer was present in 29%. Non-survivors at 30 days were 9 years (median) older than survivors. In 30 days survivors, candidemia was not recognised before discharge in 13% of cases. No treatment were given and no deaths or complications were observed in this group. Candidemia patients were grouped into 8 patient categories: Abdominal surgery (35%), urology (13%), other surgery (11%), pneumonia (13%), haematological malignancy (7%), intravenous drug abuse (4%), other medical (15%), and new-borns (3%). Candidemia was diagnosed while admitted in the ICU in 46% of patients. Urology related cases were all diagnosed in the general ward. Multiple surgical procedures were done in 60% of abdominal surgery patients. Antibiotics were administered prior to candidemia in 87% of patients, with median duration 17 (1–108) days. Neutropenia was less common than expected in patients with candidemia (8/105) and closely associated to haematological malignancy (6/8). Compared with previous national figures the epidemiology of invasive candidiasis seems not to have changed over the last decade.

Candida albicans Niche Specialization: Features That Distinguish Biofilm Cells from Commensal Cells

The fungus Candida albicans is a frequent commensal colonizer of the human gastrointestinal (GI) tract, but is also an opportunistic pathogen. This review explores features that distinguish the colonizing and pathogenic forms of C. albicans. Candida albicans in a biofilm is used as an example of a pathogenic form of the organism, because biofilms are a common feature of device-associated C. albicans infections. Biofilms (complex, sessile communities of cells) have been the subject of several large-scale gene expression studies. Biofilms and commensal C. albicans colonizing the murine GI tract show a variety of differentially expressed genes. Cell surface proteins encoded by these differentially expressed genes are especially attractive as targets for new clinical prevention, diagnosis, or treatment tools that are specific for C. albicans in its pathogenic biofilm state.

Invasive candidiasis as a cause of sepsis in the critically ill patient

Invasive fungal infections are an increasingly frequent etiology of sepsis in critically ill patients causing substantial morbidity and mortality. Candida species are by far the predominant agent of fungal sepsis accounting for 10% to 15% of health-care associated infections, about 5% of all cases of severe sepsis and septic shock and are the fourth most common bloodstream isolates in the United States. One-third of all episodes of candidemia occur in the intensive care setting. Early diagnosis of invasive candidiasis is critical in order to initiate antifungal agents promptly. Delay in the administration of appropriate therapy increases mortality. Unfortunately, risk factors, clinical and radiological manifestations are quite unspecific and conventional culture methods are suboptimal. Non-culture based methods (such as mannan, anti-mannan, β-d-glucan, and polymerase chain reaction) have emerged but remain investigational or require additional testing in the ICU setting. Few prophylactic or pre-emptive studies have been performed in critically ill patients. They tended to be underpowered and their clinical usefulness remains to be established under most circumstances. The antifungal armamentarium has expanded considerably with the advent of lipid formulations of amphotericin B, the newest triazoles and the echinocandins. Clinical trials have shown that the triazoles and echinocandins are efficacious and well tolerated antifungal therapies. Clinical practice guidelines for the management of invasive candidiasis have been published by the European Society for Clinical Microbiology and Infectious Diseases and the Infectious Diseases Society of North America.

A review of Candida species causing blood stream infection

The incidence of candidemia has been on a rise worldwide. The epidemiology of invasive fungal infections in general and of candidemia in particular has changed in the past three decades because of a variety of factors like the AIDS epidemic, increased number of patients receiving immunosuppressive therapy for transplantation and the increasing use of antimicrobials in the hospital setups and even in the community. The important risk factors for candidemia include use of broad-spectrum antimicrobials, cancer chemotherapy, mucosal colonization by Candida species, indwelling vascular catheters like central venous catheters, etc. More than 90% of the invasive infections due to Candida species are attributed to five species-Candida albicans, Candida glabrata, Candida parapsilosis, Candida tropicalis and Candida krusei. However, the list of new species of Candida isolated from clinical specimens continues to grow every year. Early diagnosis and proper treatment is the key for management of candidemia cases.

Efficacy and safety of anidulafungin in elderly, critically ill patients with invasive Candida infections: a post hoc analysis

Post hoc analysis of a non-comparative, prospective, multicentre, phase IIIb study was performed to compare efficacy and safety of anidulafungin in elderly (≥65 years) versus non-elderly (<65 years) Intensive Care Unit (ICU) patients with candidaemia/invasive candidiasis (C/IC). Adult ICU patients with confirmed C/IC meeting ≥1 of the following criteria were enrolled: post-abdominal surgery; solid tumour; renal/hepatic insufficiency; solid organ transplantation; neutropenia; age ≥65 years. Patients received anidulafungin (200 mg on Day 1, 100 mg/day thereafter) for ≥10 days followed by optional azole step-down therapy for a total treatment duration of 14-56 days. The primary efficacy endpoint was global (clinical and microbiological) response at the end of all therapy (EOT). Primary efficacy analysis was performed in the modified intent-to-treat (mITT) population (n=170), excluding unknown and missing responses. In total, 80 patients (47.1%) were aged ≥65 years and 90 (52.9%) were aged <65 years; the mean age difference between the two groups was 21.9 years. Global success at EOT in mITT patients was similar in elderly (68.1%) and non-elderly (70.7%) patients (P=0.719). However, global success rates were significantly lower in elderly versus non-elderly patients at 2 and 6 weeks after EOT (P=0.045 and P=0.016, respectively). Ninety-day survival was significantly lower (P=0.006) for elderly (42.8%) versus non-elderly patients (63.3%). The incidence and profile of adverse events were similar in elderly and non-elderly patients. Anidulafungin was effective and safe for treatment of C/IC in elderly ICU patients, despite higher baseline severity of illness scores.

Septic shock attributed to Candida infection: importance of empiric therapy and source control

BACKGROUND

Delayed treatment of candidemia has previously been shown to be an important determinant of patient outcome. However, septic shock attributed to Candida infection and its determinants of outcome have not been previously evaluated in a large patient population.

METHODS

A retrospective cohort study of hospitalized patients with septic shock and blood cultures positive for Candida species was conducted at Barnes-Jewish Hospital, a 1250-bed urban teaching hospital (January 2002-December 2010).

RESULTS

Two hundred twenty-four consecutive patients with septic shock and a positive blood culture for Candida species were identified. Death during hospitalization occurred among 155 (63.5%) patients. The hospital mortality rate for patients having adequate source control and antifungal therapy administered within 24 hours of the onset of shock was 52.8% (n = 142), compared to a mortality rate of 97.6% (n = 82) in patients who did not have these goals attained (P < .001). Multivariate logistic regression analysis demonstrated that delayed antifungal treatment (adjusted odds ratio [AOR], 33.75; 95% confidence interval [CI], 9.65-118.04; P = .005) and failure to achieve timely source control (AOR, 77.40; 95% CI, 21.52-278.38; P = .001) were independently associated with a greater risk of hospital mortality.

CONCLUSIONS

The risk of death is exceptionally high among patients with septic shock attributed to Candida infection. Efforts aimed at timely source control and antifungal treatment are likely to be associated with improved clinical outcomes.

Evaluation of "Candida score" in critically ill patients: a prospective, multicenter, observational, cohort study

Introduction

Although prompt initiation of appropriate antifungal therapy is essential for the control of invasive Candida infections and an improvement of prognosis, early diagnosis of invasive candidiasis remains a challenge and criteria for starting empirical antifungal therapy in ICU patients are poorly defined. Some scoring systems, such as the "Candida score" could help physicians to differentiate patients who could benefit from early antifungal treatment from those for whom invasive candidiasis is highly improbable. This study evaluated the performance of this score in a cohort of critically ill patients.

Methods

A prospective, observational, multicenter, cohort study was conducted from January 2010 to March 2011 in five intensive care units in Nord-Pas de Calais, an area from North of France. All patients exhibiting, on ICU admission or during their ICU stay, a hospital-acquired severe sepsis or septic shock could be included in this study. The data collected included patient characteristics on ICU admission and at the onset of severe sepsis or septic shock. The "Candida score" was calculated at the onset of sepsis or shock. The incidence of invasive candidiasis was determined and its relationship with the value of the "Candida score" was studied.

Results

Ninety-four patients were studied. When severe sepsis or shock occurred, 44 patients had a score = 2, 29 patients had a score = 3, 17 patients had a score = 4, and 4 patients had a score = 5. Invasive candidiasis was observed in five (5.3%) patients. One patient had candidemia, three patients had peritonitis, and one patient had pleural infection. The rates of invasive candidiasis was 0% in patients with score = 2 or 3, 17.6% in patients with score = 4, and 50% in patients with score = 5 (p < 0.0001).

Conclusions

Our results confirm that the "Candida score" is an interesting tool to differentiate among ICU patients who exhibit hospital-acquired severe sepsis or septic shock those would benefit from early antifungal treatment (score > 3) from those for whom invasive candidiasis is highly improbable (score ≤ 3).

Insights into Candida tropicalis nosocomial infections and virulence factors

Candida tropicalis is considered the first or the second non-Candida albicans Candida (NCAC) species most frequently isolated from candidosis, mainly in patients admitted in intensive care units (ICUs), especially with cancer, requiring prolonged catheterization, or receiving broad-spectrum antibiotics. The proportion of candiduria and candidemia caused by C. tropicalis varies widely with geographical area and patient group. Actually, in certain countries, C. tropicalis is more prevalent, even compared with C. albicans or other NCAC species. Although prophylactic treatments with fluconazole cause a decrease in the frequency of candidosis caused by C. tropicalis, it is increasingly showing a moderate level of fluconazole resistance. The propensity of C. tropicalis for dissemination and the high mortality associated with its infections might be strongly related to the potential of virulence factors exhibited by this species, such as adhesion to different host surfaces, biofilm formation, infection and dissemination, and enzymes secretion. Therefore, the aim of this review is to outline the present knowledge on all the above-mentioned C. tropicalis virulence traits.

Cross-species discovery of syncretic drug combinations that potentiate the antifungal fluconazole

The authors screen for compounds that show synergistic antifungal activity when combined with the widely-used fungistatic drug fluconazole. Chemogenomic profiling explains the mode of action of synergistic drugs and allows the prediction of additional drug synergies.

The authors screen for compounds that show synergistic antifungal activity when combined with the widely-used fungistatic drug fluconazole. Chemogenomic profiling explains the mode of action of synergistic drugs and allows the prediction of additional drug synergies.

Chemical screens with a library enriched for known drugs identified a diverse set of 148 compounds that potentiated the action of the antifungal drug fluconazole against the fungal pathogens Cryptococcus neoformans, Cryptococcus gattii and Candida albicans, and the model yeast Saccharomyces cerevisiae, often in a species-specific manner.

Chemogenomic profiles of six confirmed hits in S. cerevisiae revealed different modes of action and enabled the prediction of additional synergistic combinations; three-way synergistic interactions exhibited even stronger synergies at low doses of fluconazole.

The synergistic combination of fluconazole and the antidepressant sertraline was active against fluconazole-resistant clinical fungal isolates and in an in vivo model of Cryptococcal infection.

Rising fungal infection rates, especially among immune-suppressed individuals, represent a serious clinical challenge (Gullo, 2009). Cancer, organ transplant and HIV patients, for example, often succumb to opportunistic fungal pathogens. The limited repertoire of approved antifungal agents and emerging drug resistance in the clinic further complicate the effective treatment of systemic fungal infections. At the molecular level, the paucity of fungal-specific essential targets arises from the conserved nature of cellular functions from yeast to humans, as well as from the fact that many essential yeast genes can confer viability at a fraction of wild-type dosage (Yan et al, 2009). Although only ∼1100 of the ∼6000 genes in yeast are essential, almost all genes become essential in specific genetic backgrounds in which another non-essential gene has been deleted or otherwise attenuated, an effect termed synthetic lethality (Tong et al, 2001). Genome-scale surveys suggest that over 200 000 binary synthetic lethal gene combinations dominate the yeast genetic landscape (Costanzo et al, 2010). The genetic buffering phenomenon is also manifest as a plethora of differential chemical–genetic interactions in the presence of sublethal doses of bioactive compounds (Hillenmeyer et al, 2008). These observations frame the difficulty of interdicting network functions in eukaryotic pathogens with single agent therapeutics. At the same time, however, this genetic network organization suggests that judicious combinations of small molecule inhibitors of both essential and non-essential targets may elicit additive or synergistic effects on cell growth (Sharom et al, 2004; Lehar et al, 2008). Unbiased screens for drugs that synergistically enhance a specific bioactive effect, but which are not themselves individually active—termed a syncretic combination—are one means to substantially elaborate chemical space (Keith et al, 2005). Indeed, compounds that enhance the activity of known agents in model yeast and cancer cell line systems have been identified both by focused small molecule library screens and by computational methods (Borisy et al, 2003; Lehar et al, 2007; Nelander et al, 2008; Jansen et al, 2009; Zinner et al, 2009).

To extend the stratagem of chemical synthetic lethality to clinically relevant fungal pathogens, we screened a bioactive library of known drugs for synergistic enhancers of the widely used fungistatic drug fluconazole against the clinically relevant pathogens C. albicans, C. neoformans and C. gattii, as well as the genetically tractable budding yeast S. cerevisiae. Fluconazole is an azole drug that inhibits lanosterol 14α-demethylase, the gene product of ERG11, an essential cytochrome P450 enzyme in the ergosterol biosynthetic pathway (Groll et al, 1998). We identified 148 drugs that potentiate the antifungal action of fluconazole against the four species. These syncretic compounds had not been previously recognized in the clinic as antifungal agents, and many acted in a species-specific manner, often in a potent fungicidal manner.

To understand the mechanisms of synergism, we interrogated six syncretic drugs—trifluoperazine, tamoxifen, clomiphene, sertraline, suloctidil and L-cycloserine—in genome-wide chemogenomic profiles of the S. cerevisiae deletion strain collection (Giaever et al, 1999). These profiles revealed that membrane, vesicle trafficking and lipid biosynthesis pathways are targeted by five of the synergizers, whereas the sphingolipid biosynthesis pathway is targeted by L-cycloserine. Cell biological assays confirmed the predicted membrane disruption effects of the former group of compounds, which may perturb ergosterol metabolism, impair fluconazole export by drug efflux pumps and/or affect active import of fluconazole (Kuo et al, 2010; Mansfield et al, 2010). Based on the integration of chemical–genetic and genetic interaction space, a signature set of deletion strains that are sensitive to the membrane active synergizers correctly predicted additional drug synergies with fluconazole. Similarly, the L-cycloserine chemogenomic profile correctly predicted a synergistic interaction between fluconazole and myriocin, another inhibitor of sphingolipid biosynthesis. The structure of genetic networks suggests that it should be possible to devise higher order drug combinations with even greater selectivity and potency (Sharom et al, 2004). In an initial test of this concept, we found that the combination of a non-synergistic pair drawn from the membrane active and sphingolipid target classes exhibited potent three-way synergism with a low dose of fluconazole. Finally, the combination of sertraline and fluconazole was active in a G. mellonella model of Cryptococcal infection, and was also efficacious against fluconazole-resistant clinical isolates of C. albicans and C. glabrata.

Collectively, these results demonstrate that the combinatorial redeployment of known drugs defines a powerful antifungal strategy and establish a number of potential lead combinations for future clinical assessment.

Resistance to widely used fungistatic drugs, particularly to the ergosterol biosynthesis inhibitor fluconazole, threatens millions of immunocompromised patients susceptible to invasive fungal infections. The dense network structure of synthetic lethal genetic interactions in yeast suggests that combinatorial network inhibition may afford increased drug efficacy and specificity. We carried out systematic screens with a bioactive library enriched for off-patent drugs to identify compounds that potentiate fluconazole action in pathogenic Candida and Cryptococcus strains and the model yeast Saccharomyces. Many compounds exhibited species- or genus-specific synergism, and often improved fluconazole from fungistatic to fungicidal activity. Mode of action studies revealed two classes of synergistic compound, which either perturbed membrane permeability or inhibited sphingolipid biosynthesis. Synergistic drug interactions were rationalized by global genetic interaction networks and, notably, higher order drug combinations further potentiated the activity of fluconazole. Synergistic combinations were active against fluconazole-resistant clinical isolates and an in vivo model of Cryptococcus infection. The systematic repurposing of approved drugs against a spectrum of pathogens thus identifies network vulnerabilities that may be exploited to increase the activity and repertoire of antifungal agents.

Design and use of Candida scores at the intensive care unit

Invasive Candida infections are recognised as a cause of increased morbidity and mortality in intensive care patients, particularly those with recent extensive gastroabdominal surgery. Due to the difficulties of diagnosis, several authors have analysed risk factors suggestive of invasive candidiasis to identify patients at highest risk. Such patients may be potential candidates for preemptive antifungal therapy before becoming seriously ill. The extent of body site colonisation due to Candida species was recognised to be related with consequent invasive disease. The quantification of the colonisation was expressed as the Candida colonisation index. Based on the evaluation of independent risk factors predictive of invasive Candida infections, clinically relevant scores were evaluated in the last decade. Particularly, the Candida score that combines the clinical risk factors preceding surgery, total parenteral nutrition and severe sepsis with Candida multi-site colonisation can be considered a useful bedside scoring system to discern patients with mere Candida colonisation from patients with the risk of invasive candidiasis in non-neutropaenic critically ill patient population.

Anidulafungin for the treatment of invasive candidiasis

Candidaemia/invasive candidiasis (C/IC) is the most frequently occurring invasive fungal infection worldwide, with a particularly strong impact and high incidence in the intensive-care unit, where there is a need for new treatment options and strategies. The echinocandin anidulafungin has broad in vitro activity against a wide range of Candida species, along with favourable pharmacokinetics that allow administration in hepatic and renal impairment and with any comedication without the need for dose adjustments. The efficacy and safety of anidulafungin for the treatment of C/IC were demonstrated in a number of clinical studies and by some limited data from clinical practice. In a randomized comparative trial for the treatment of C/IC in adults, 76% of patients receiving anidulafungin and 60% of those given fluconazole were treated successfully (95% CI for difference: 4-27; p 0.01). Post hoc analyses suggest that anidulafungin is significantly more effective than standard-dose fluconazole for the treatment of candidaemia in critically ill patients. Anidulafungin is generally well tolerated, with commonly reported side effects including headache, hypokalaemia, gastrointestinal symptoms, abnormal liver function test results, and rash. In pharmaco-economic analyses, anidulafungin compared favourably with fluconazole (in terms of overall costs and hospital resource use) as well as with other echinocandins. Echinocandins, including anidulafungin, are now generally recommended as first-line therapy in moderately to severely ill patients, those with prior azole exposure, and patients with C/IC caused by Candida glabrata or Candida krusei.

The value of a risk model for early-onset candidemia

Bloodstream infections from Candida species are associated with an increased length of stay, increased hospital costs, and higher mortality when compared with bacterial bloodstream infections. Delayed or inappropriate therapy in candidemia leads to increased mortality, thus early recognition becomes paramount. With biomarkers showing promise, blood cultures still remain the gold standard but require 24 to 72 hours for growth. The reliance on epidemiologic risk factors for the initiation of empiric antifungal therapy therefore provides the best method for early appropriate therapy. Shorr and colleagues have devised a risk score to identify patients with early-onset candidemia as defined by positive blood cultures within 2 days of admission, thus allowing for the initiation of early appropriate antifungal therapy.