Invasive fungal infections in acute and chronic liver impairment: A systematic review

[Infections and liver cirrhosis]

End-stage liver disease is a life-threatening clinical syndrome combined with a state of immune dysfunction. In this constellation patients are prone to bacterial, fungal and viral infections associated with markedly increased morbidity and mortality rates. Bacterial infections are the most prevalent kind of infection in patients with end-stage liver disease accounting for nearly 30%. The evolving rates of multidrug resistant organisms present enormous challenges in treatment strategies. Therefore, the urgent needs for prevention, early detection strategies and widespread treatment options are a necessity to handle the rising incidence of infection complications in end-stage liver disease.

INTERPRETABLE MACHINE LEARNING FOR PREDICTING RISK OF INVASIVE FUNGAL INFECTION IN CRITICALLY ILL PATIENTS IN THE INTENSIVE CARE UNIT: A RETROSPECTIVE COHORT STUDY BASED ON MIMIC-IV DATABASE

ABSTRACT

The delayed diagnosis of invasive fungal infection (IFI) is highly correlated with poor prognosis in patients. Early identification of high-risk patients with invasive fungal infections and timely implementation of targeted measures is beneficial for patients. The objective of this study was to develop a machine learning-based predictive model for invasive fungal infection in patients during their intensive care unit (ICU) stay. Retrospective data was extracted from adult patients in the MIMIC-IV database who spent a minimum of 48 h in the ICU. Feature selection was performed using LASSO regression, and the dataset was balanced using the BL-SMOTE approach. Predictive models were built using six machine learning algorithms. The Shapley additive explanation algorithm was used to assess the impact of various clinical features in the optimal model, enhancing interpretability. The study included 26,346 ICU patients, of whom 379 (1.44%) were diagnosed with invasive fungal infection. The predictive model was developed using 20 risk factors, and the dataset was balanced using the borderline-SMOTE (BL-SMOTE) algorithm. The BL-SMOTE random forest model demonstrated the highest predictive performance (area under curve = 0.88, 95% CI = 0.84-0.91). Shapley additive explanation analysis revealed that the three most influential clinical features in the BL-SMOTE random forest model were dialysis treatment, APSIII scores, and liver disease. The machine learning model provides a reliable tool for predicting the occurrence of IFI in ICU patients. The BL-SMOTE random forest model, based on 20 risk factors, exhibited superior predictive performance and can assist clinicians in early assessment of IFI occurrence in ICU patients. Importance: Invasive fungal infections are characterized by high incidence and high mortality rates characteristics. In this study, we developed a clinical prediction model for invasive fungal infections in critically ill patients based on machine learning algorithms. The results show that the machine learning model based on 20 clinical features has good predictive value.

Risk factors for short-term prognosis of end-stage liver disease complicated by invasive pulmonary aspergillosis

BACKGROUND AND AIM

Patients with end-stage liver disease (ESLD) are susceptible to invasive pulmonary aspergillosis (IPA). This study aimed to investigate the risk factors affecting the occurrence and short-term prognosis of ESLD complicated by IPA.

METHODS

This retrospective case-control study included 110 patients with ESLD. Of them, 27 ESLD-IPA received antifungal therapy with amphotericin B (AmB); 27 AmB-free-treated ESLD-IPA patients were enrolled through 1:1 propensity score matching. Fifty-six ESLD patients with other comorbid pulmonary infections were enrolled as controls. The basic features of groups were compared, while the possible risk factors affecting the occurrence and short-term outcomes of IPA were analyzed.

RESULTS

Data analysis revealed invasive procedures, glucocorticoid exposure, and broad-spectrum antibiotic use were independent risk factors for IPA. The 54 patients with ESLD-IPA exhibited an overall treatment effectiveness and 28-d mortality rate of 50.00% and 20.37%, respectively, in whom patients treated with AmB-containing showed higher treatment efficacy than patients treated with AmB-free antifungal regimens (66.7% vs. 33.3%, respectively, χ2 = 6.000, P = 0.014). Multivariate logistic regression analysis revealed that the treatment regimen was the only predictor affecting patient outcomes, with AmB-containing regimens were 4.893 times more effective than AmB-free regimens (95% CI, 1.367-17.515; P = 0.015). The only independent predictors affecting the 28-d mortality rate were neutrophil-to-lymphocyte ratio and IPA diagnosis (OR = 1.140 and 10.037, P = 0.046 and 0.025, respectively).

CONCLUSIONS

Glucocorticoid exposure, invasive procedures, and broad-spectrum antibiotic exposure increased the risk of IPA in ESLD patients. AmB alone or combined with other antifungals may serve as an economical, safe, and effective treatment option for ESLD-IPA.

Acute Invasive Pulmonary Aspergillosis: Clinical Presentation and Treatment

Among all clinical manifestations of pulmonary aspergillosis, invasive pulmonary aspergillosis (IPA) is the most acute presentation. IPA is caused by Aspergillus hyphae invading the pulmonary tissue, causing either tracheobronchitis and/or bronchopneumonia. The degree of fungal invasion into the respiratory tissue can be seen as a spectrum, going from colonization to deep tissue penetration with angio-invasion, and largely depends on the host's immune status. Patients with prolonged, severe neutropenia and patients with graft-versus-host disease are at particularly high risk. However, IPA also occurs in other groups of immunocompromised and nonimmunocompromised patients, like solid organ transplant recipients or critically ill patients with severe viral disease. While a diagnosis of proven IPA is challenging and often warranted by safety and feasibility, physicians must rely on a combination of clinical, radiological, and mycological features to assess the likelihood for the presence of IPA. Triazoles are the first-choice regimen, and the choice of the drug should be made on an individual basis. Adjunctive therapy such as immunomodulatory treatment should also be taken into account. Despite an improving and evolving diagnostic and therapeutic armamentarium, the burden and mortality of IPA still remains high. This review aims to give a comprehensive and didactic overview of the current knowledge and best practices regarding the epidemiology, clinical presentation, diagnosis, and treatment of acute IPA.

Enhancing voriconazole therapy in liver dysfunction: exploring administration schemes and predictive factors for trough concentration and efficacy

Introduction: The application of voriconazole in patients with liver dysfunction lacks pharmacokinetic data. In previous study, we proposed to develop voriconazole dosing regimens for these patients according to their total bilirubin, but the regimens are based on Monte Carlo simulation and has not been further verified in clinical practice. Besides, there are few reported factors that significantly affect the efficacy of voriconazole. Methods: We collected the information of patients with liver dysfunction hospitalized in our hospital from January 2018 to May 2022 retrospectively, including their baseline information and laboratory data. We mainly evaluated the efficacy of voriconazole and the target attainment of voriconazole trough concentration. Results: A total of 157 patients with liver dysfunction were included, from whom 145 initial and 139 final voriconazole trough concentrations were measured. 60.5% (95/157) of patients experienced the adjustment of dose or frequency. The initial voriconazole trough concentrations were significantly higher than the final (mean, 4.47 versus 3.90 μg/mL, p = 0.0297). Furthermore, daily dose, direct bilirubin, lymphocyte counts and percentage, platelet, blood urea nitrogen and creatinine seven covariates were identified as the factors significantly affect the voriconazole trough concentration. Binary logistic regression analysis revealed that the lymphocyte percentage significantly affected the efficacy of voriconazole (OR 1.138, 95% CI 1.016-1.273), which was further validated by the receiver operating characteristic curve. Conclusion: The significant variation in voriconazole trough concentrations observed in patients with liver dysfunction necessitates caution when prescribing this drug. Clinicians should consider the identified factors, particularly lymphocyte percentage, when dosing voriconazole in this population.

The impact of gene polymorphism and hepatic insufficiency on voriconazole dose adjustment in invasive fungal infection individuals

Voriconazole (VRZ) is a broad-spectrum antifungal medication widely used to treat invasive fungal infections (IFI). The administration dosage and blood concentration of VRZ are influenced by various factors, posing challenges for standardization and individualization of dose adjustments. On the one hand, VRZ is primarily metabolized by the liver, predominantly mediated by the cytochrome P450 (CYP) 2C19 enzyme. The genetic polymorphism of CYP2C19 significantly impacts the blood concentration of VRZ, particularly the trough concentration (Ctrough), thereby influencing the drug's efficacy and potentially causing adverse drug reactions (ADRs). Recent research has demonstrated that pharmacogenomics-based VRZ dose adjustments offer more accurate and individualized treatment strategies for individuals with hepatic insufficiency, with the possibility to enhance therapeutic outcomes and reduce ADRs. On the other hand, the security, pharmacokinetics, and dosing of VRZ in individuals with hepatic insufficiency remain unclear, making it challenging to attain optimal Ctrough in individuals with both hepatic insufficiency and IFI, resulting in suboptimal drug efficacy and severe ADRs. Therefore, when using VRZ to treat IFI, drug dosage adjustment based on individuals' genotypes and hepatic function is necessary. This review summarizes the research progress on the impact of genetic polymorphisms and hepatic insufficiency on VRZ dosage in IFI individuals, compares current international guidelines, elucidates the current application status of VRZ in individuals with hepatic insufficiency, and discusses the influence of CYP2C19, CYP3A4, CYP2C9, and ABCB1 genetic polymorphisms on VRZ dose adjustments and Ctrough at the pharmacogenomic level. Additionally, a comprehensive summary and analysis of existing studies' recommendations on VRZ dose adjustments based on CYP2C19 genetic polymorphisms and hepatic insufficiency are provided, offering a more comprehensive reference for dose selection and adjustments of VRZ in this patient population.

A review of prophylactic regimens to prevent invasive fungal infections in hematology patients undergoing chemotherapy or stem cell transplantation

INTRODUCTION

The recent introduction of targeted therapies, including monoclonal antibodies, tyrosine-kinase inhibitors, and immunotherapies has improved the cure rate of hematologic patients. The implication of personalized treatment on primary antifungal prophylaxis will be discussed.

AREAS COVERED

We reviewed the literature for clinical trials reporting the rate of invasive fungal infections during targeted and cellular therapies and stem cell transplant, and the most recent international guidelines for primary antifungal prophylaxis.

EXPERT OPINION

As the use of personalized therapies is growing, the risk of invasive fungal infection has emerged in various clinical settings. Therefore, it is possible that the use of mold-active antifungal prophylaxis would spread in the next years and the risk of breakthrough infections would increase. The introduction of new antifungal agents in the clinical armamentarium is expected to reduce clinical unmet needs concerning the management of primary antifungal prophylaxis and improve outcome of patients.