Risk factors of septic shock development and thirty-day mortality with a predictive model in adult candidemia patients in intensive care units

Mortality of patients with sepsis in intensive care units at tertiary hospitals in Jordan: Prospective cohort study

The aim of this study was to describe the 30-day mortality rate of adult patients with sepsis and septic shock in 6 intensive care units of 2 tertiary hospitals in Jordan. A prospective cohort design was used. Patients with sepsis and septic shock admitted to the medical and surgical intensive care units at 2 tertiary hospitals were followed up during the period between February 2022 and June 2022 (N = 148). Data were analyzed using SPSS, version 23. Moreover, descriptive statistics, chi-square, and binary logistic regression were used. Notably, 52.7% of patients with sepsis and septic shock died within 30 days of diagnosis of sepsis and septic shock. Sequential Organ Failure Assessment score and the history of having solid tumors significantly predicted the 30-day mortality rate. Moreover, 43 (29.0%) patients with sepsis and septic shock had positive blood cultures, and 46 (31.0%) had positive urine cultures. Patients with sepsis and septic shock have a notable mortality rate that can be predicted from total Sequential Organ Failure Assessment scores and from the history of having solid tumors. Early assessment and initiation of treatment for sepsis essentially would reduce the likelihood of progression of sepsis to septic shock and would reduce associated patients' mortality.

Biofilm formation and increased mortality among cancer patients with candidemia in a Peruvian reference center

BACKGROUND

Candidemia is an invasive mycosis with an increasing global incidence and high mortality rates in cancer patients. The production of biofilms by some strains of Candida constitutes a mechanism that limits the action of antifungal agents; however, there is limited and conflicting evidence about its role in the risk of death. This study aimed to determine whether biofilm formation is associated with mortality in cancer patients with candidemia.

METHODS

This retrospective cohort study included patients treated at Peru's oncologic reference center between June 2015 and October 2017. Data were collected by monitoring patients for 30 days from the diagnosis of candidemia until the date of death or hospital discharge. Statistical analyses evaluated the association between biofilm production determined by XTT reduction and mortality, adjusting for demographic, clinical, and microbiological factors assessed by the hospital routinary activities. Survival analysis and bivariate and multivariate Cox regression were used, estimating the hazard ratio (HR) as a measure of association with a significance level of p < 0.05.

RESULTS

A total of 140 patients with candidemia were included in the study. The high mortality observed on the first day of post-diagnosis follow-up (81.0%) among 21 patients who were not treated with either antifungal or antimicrobial drugs led to stratification of the analyses according to whether they received treatment. In untreated patients, there was a mortality gradient in patients infected with non-biofilm-forming strains vs. low/medium and high-level biofilm-forming strains (25.0%, 66.7% and 82.3%, respectively, p = 0.049). In treated patients, a high level of biofilm formation was associated with increased mortality (HR, 3.92; 95% p = 0.022), and this association persisted after adjusting for age, comorbidities, and hospital emergency admission (HR, 6.59; CI: 1.87-23.24, p = 0.003).

CONCLUSIONS

The association between candidemia with in vitro biofilm formation and an increased risk of death consistently observed both in patients with and without treatment, provides another level of evidence for a possible causal association. The presence of comorbidities and the origin of the hospital emergency, which reflect the fragile clinical condition of the patients, and increasing age above 15 years were associated with a higher risk of death.

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.

Enhancing prognostic prediction of invasive candidiasis among cancer patients with a serum C5a-based scoring model

PURPOSE

Invasive candidiasis poses a life-threatening risk, and early prognosis assessment is vital for timely interventions to reduce mortality. Serum C5a levels have recently been linked to prognosis, but confirmation in cancer patients is pending.

METHODS

We detected the concentrations of serum C5a in hospitalized cancer patients with invasive candidiasis from 2020 to 2023, and retrospectively analyzed the clinical data.

RESULTS

372 cases were included in this study, with a 90-day mortality rate of 21.8%. Candida albicans (48.7%) remained the predominant pathogen, followed by Candida glabrata (25.5%), Candida tropicalis (12.4%), and Candida parapsilosis (8.3%). Gastrointestinal cancer was the most diagnosed pathology type (37.6%). Serum C5a demonstrated a noteworthy correlation with 90-day mortality, and employing a cutoff value of 36.7 ng/ml revealed significantly higher 90-day mortality in low-C5a patients (41.2%) compared to high-C5a patients (6.3%) (p < 0.001). We also identified no source control, no surgery, metastasis, or chronic renal failure independently correlated with the 90-day mortality. Based on this, a prognostic model combining C5a and clinical parameters was constructed, which performed better than models built solely on C5a or clinical parameters. Furthermore, we weighted scores to each parameter in the model and presented diagnostic sensitivity and specificity corresponding to different score points calculated by the model.

CONCLUSION

We constructed a prognostic scoring model including serum C5a and clinical parameters, which would contribute to precise prognosis assessment and benefit the outcome among cancer patients.

A dynamic nomogram for predicting 28-day mortality in septic shock: a Chinese retrospective cohort study

Background

Septic shock is a severe life-threatening disease, and the mortality of septic shock in China was approximately 37.3% that lacks prognostic prediction model. This study aimed to develop and validate a prediction model to predict 28-day mortality for Chinese patients with septic shock.

Methods

This retrospective cohort study enrolled patients from Intensive Care Unit (ICU) of the Second Affiliated Hospital, School of Medicine, Zhejiang University between December 2020 and September 2021. We collected patients' clinical data: demographic data and physical condition data on admission, laboratory data on admission and treatment method. Patients were randomly divided into training and testing sets in a ratio of 7:3. Univariate logistic regression was adopted to screen for potential predictors, and stepwise regression was further used to screen for predictors in the training set. Prediction model was constructed based on these predictors. A dynamic nomogram was performed based on the results of prediction model. Using receiver operator characteristic (ROC) curve to assess predicting performance of dynamic nomogram, which were compared with Sepsis Organ Failure Assessment (SOFA) and Acute Physiology and Chronic Health Evaluation II (APACHE II) systems.

Results

A total of 304 patients with septic shock were included, with a 28-day mortality of 25.66%. Systolic blood pressure, cerebrovascular disease, Na, oxygenation index (PaO2/FiO2), prothrombin time, glucocorticoids, and hemodialysis were identified as predictors for 28-day mortality in septic shock patients, which were combined to construct the predictive model. A dynamic nomogram (https://zhijunxu.shinyapps.io/DynNomapp/) was developed. The dynamic nomogram model showed a good discrimination with area under the ROC curve of 0.829 in the training set and 0.825 in the testing set. Additionally, the study suggested that the dynamic nomogram has a good predictive value than SOFA and APACHE II.

Conclusion

The dynamic nomogram for predicting 28-day mortality in Chinese patients with septic shock may help physicians to assess patient survival and optimize personalized treatment strategies for septic shock.

Mortality Caused by Candida auris Bloodstream Infections in Comparison with Other Candida Species, a Multicentre Retrospective Cohort

Candida auris is an emerging pathogen considered to be critical in the World Health Organization fungal organisms list. The study aims to determine the mortality and hospital stays attributed to Candida auris (C. auris) compared to other Candida species in adult patients with candidemia. A retrospective cohort of adults with candidemia was examined from seven centres in Colombia between 2016 and 2021. The primary outcome was 30-day mortality, and the secondary outcome was the length of hospital stay among survivors. Adjustment of the confounding variables was performed using inverse probability weights of exposure propensity score (candidemia by C. auris), survival regression models (Weibull distribution), and a counting model (negative binomial distribution). A value of 244 (47.6%) of the 512 patients with candidemia died within the first 30 days. The crude mortality in C. auris was 38.1% vs. 51.1% in Candida non-auris (CNA). In the Weibull model, mortality in the C. auris group was lower (adjusted HR: aHR- 0.69, 95% CI: 0.53-0.90). Antifungal treatment also decreased mortality, with an aHR of 0.36 (95% CI 0.27-0.47), while the presence of septic shock on patient progression increased it, with an aHR of 1.73 (95% CI 1.41-2.13). Among the patients who survived, no differences in the length of hospital stay were observed between the C. auris and the CNA groups, with an incidence rate ratio of 0.92 (95% CI: 0.68-1.22). Mortality in patients with C. auris bloodstream infections appears lower when adjusted for numerous confounding variables regarding treatment and the presence of septic shock in patient progression. We identified no significant effect of C. auris on the length of hospital stay in surviving patients.

The Clinical Characteristics of Bloodstream Infections Due to Candida spp. in Patients Hospitalized in Intensive Care Units during the SARS-CoV-2 Pandemic: The Results of a Multicenter Study

Candidemia is a serious health threat. Whether this infection has a greater incidence and a higher mortality rate in patients with COVID-19 is still debated. In this multicenter, retrospective, observational study, we aimed to identify the clinical characteristics associated with the 30-day mortality in critically ill patients with candidemia and to define the differences in candidemic patients with and without COVID-19. Over a three-year period (2019-2021), we identified 53 critically ill patients with candidemia, 18 of whom (34%) had COVID-19 and were hospitalized in four ICUs. The most frequent comorbidities were cardiovascular (42%), neurological (17%), chronic pulmonary diseases, chronic kidney failure, and solid tumors (13% each). A significantly higher proportion of COVID-19 patients had pneumonia, ARDS, septic shock, and were undergoing an ECMO procedure. On the contrary, non-COVID-19 patients had undergone previous surgeries and had used TPN more frequently. The mortality rate in the overall population was 43%: 39% and 46% in the COVID-19 and non-COVID-19 patients, respectively. The independent risk factors associated with a higher mortality were CVVH (HR 29.08 [CI 95% 3.37-250]) and a Charlson's score of > 3 (HR 9.346 [CI 95% 1.054-82.861]). In conclusion, we demonstrated that candidemia still has a high mortality rate in patients admitted to ICUs, irrespective of infection due to SARS-CoV-2.