Development of a clinical prediction model for recurrence and mortality outcomes after Clostridioides difficile infection using a machine learning approach

Ruzicka, Daniel J.; Kondo, Takayuki; Fujimoto, Go; Craig, Andrew P.; Kim, Seok-Won; Mikamo, Hiroshige

doi:10.1016/j.anaerobe.2022.102628

Cited by 7 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of these predictive R-CDI models yielded much lower AUC values than ours, including those of Ruzicka et 'Agostino et al, 2014;Larrainzar-Coghen et al, 2016;Zilberberg et al, 2016;Escobar et al, 2017;Cobo et al, 2018;Ruzicka et al, 2022).Origüen et al obtained a predictive model for R-CDI with an AUC very similar to ours, ie, 0.785. Their model also included the PCR cycles of the C. difficile toxin (as did ours), although when they included the result of toxin detection by EIA, their AUC dropped to 0.775 (Origüen et al, 2019).…”

Section: Figuresupporting

confidence: 55%

“…Of these models, only that of Cobo et al. incorporated laboratory parameters related to C. difficile ( D’Agostino et al., 2014 ; Larrainzar-Coghen et al., 2016 ; Zilberberg et al., 2016 ; Escobar et al., 2017 ; Cobo et al., 2018 ; Ruzicka et al., 2022 ).Origüen et al. obtained a predictive model for R-CDI with an AUC very similar to ours, ie, 0.785.…”

Section: Discussionmentioning

confidence: 58%

“…Hence, the data on which the model is based are easy to obtain at the time of diagnosis of the primary CDI episode, with a reasonable AUC (0.783). Although several prediction models have been developed for R-CDI ( Eyre et al., 2012 ; D’Agostino et al., 2014 ; LaBarbera et al., 2015 ; Escobar et al., 2017 ; Cobo et al., 2018 ; Ruzicka et al., 2022 ), none have been widely implemented in clinical practice. External validations of prediction tools have revealed disappointing results, probably owing to the low quality of most studies and the small effects of the proposed prognostic factors ( van Rossen et al., 2022 ).…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Microbiome profile and calprotectin levels as markers of risk of recurrent Clostridioides difficile infection

Vázquez-Cuesta,

Lozano García,

Fernández

et al. 2023

Front. Cell. Infect. Microbiol.

View full text Add to dashboard Cite

IntroductionClostridioides difficile infection (CDI) is the main cause of nosocomial diarrhoea in developed countries. Recurrent CDI (R-CDI), which affects 20%-30% of patients and significantly increases hospital stay and associated costs, is a key challenge. The main objective of this study was to explore the role of the microbiome and calprotectin levels as predictive biomarkers of R-CDI.MethodsWe prospectively (2019-2021) included patients with a primary episode of CDI. Clinical data and faecal samples were collected. The microbiome was analysed by sequencing the hypervariable V4 region of the 16S rRNA gene on an Illumina Miseq platform.ResultsWe enrolled 200 patients with primary CDI, of whom 54 developed R-CDI and 146 did not. We analysed 200 primary samples and found that Fusobacterium increased in abundance, while Collinsella, Senegalimassilia, Prevotella and Ruminococcus decreased in patients with recurrent versus non-recurrent disease. Elevated calprotectin levels correlated significantly with R-CDI (p=0.01). We built a risk index for R-CDI, including as prognostic factors age, sex, immunosuppression, toxin B amplification cycle, creatinine levels and faecal calprotectin levels (overall accuracy of 79%).DiscussionCalprotectin levels and abundance of microbial genera such as Fusobacterium and Prevotella in primary episodes could be useful as early markers of R-CDI. We propose a readily available model for prediction of R-CDI that can be applied at the initial CDI episode. The use of this tool could help to better tailor treatments according to the risk of R-CDI.

show abstract

Section: Figuresupporting

confidence: 55%

Section: Discussionmentioning

confidence: 58%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Microbiome profile and calprotectin levels as markers of risk of recurrent Clostridioides difficile infection

Vázquez-Cuesta,

Lozano García,

Fernández

et al. 2023

Front. Cell. Infect. Microbiol.

View full text Add to dashboard Cite

show abstract

“…48 Calbo noted that CDI mortality risk factors are similar to those for CDI recurrence, including age, number of comorbidities, cancer, and BMI. 49 Poor outcomes in CDI result in 14.0-25.0% recurrence, 4.0-20.0% refractory, and approximately 6.0% mortality, with in-hospital mortality ranging from 8.0-37.2%. 48 Moreover, outcome drivers include C. difficile virulence factors, host factors such as age and malignancy, and treatment.…”

Section: Understanding the Risk Factors And Outcomes Of Recurrent Clo...mentioning

confidence: 99%

“…48 Moreover, outcome drivers include C. difficile virulence factors, host factors such as age and malignancy, and treatment. 49,50 A retrospective single centre cohort study on almost 4,000 patients with CDI found that recurrent CDI is associated with an increased risk of death at 3 and 6 months; and a UK study with 6,682 patients with CDI, including 1,140 patients with recurrent CDI, found an increased risk of mortality and complication at 12 months. 29,30 Additional studies show that CDI is characterised by a high delayed and unrelated mortality rate (18% at 75 days), associated with age (>65 years), comorbidity, and faecal incontinence.…”

Section: Understanding the Risk Factors And Outcomes Of Recurrent Clo...mentioning

confidence: 99%

Clostridioides difficile Infection: Targeting an Unwelcome and Persistent Threat

Moir

2023

EMJ Microbiol Infect Dis

View full text Add to dashboard Cite

An integrated symposium, with an esteemed panel of experts, discussed the burden and treatment approaches for Clostridioides difficile infection (CDI). The session took place on 17th April 2023 as part of the 33rd European Congress of Clinical Microbiology and Infectious Diseases (ECCMID) 2023 in Copenhagen, Denmark. The session was co-chaired by Mark Wilcox, Head of Microbiology Research & Development and Infection Lead at Leeds Teaching Hospitals NHS Trust, UK, and Professor of Medical Microbiology at the University of Leeds, UK; and Anne Gonzales-Luna, Assistant Professor at the University of Houston College of Pharmacy, Texas, USA. Sarah Tschudin-Sutter, Professor and Head Division of Hospital Epidemiology at the University Hospital Basel, Switzerland, explored the changing landscape for CDI, including the epidemiology and surveillance of infection, and approaches to diagnosis. John Coia, Professor Emeritus at the Institute of Regional Health Research, University of South Denmark, and Honorary Research Fellow at Glasgow University, UK; and Esther Calbo, Professor and Head of Infectious Disease at the Hospital Universitari Mútua de Terrassa, Spain, gave an overview of the burden of recurrent CDI, dysbiosis in CDI, and the role in recurrent CDI, mortality risk, and the importance of targeting recurrence. Calbo also presented an interactive case study that allowed audience members to engage and review the patient case presented. The faculty panel explored with the audience how to approach assessment and treatment to achieve the best outcome for the patient case. Benoît Guery, Associate Professor Infectious Disease at the Centre Hospitalier Universitaire Vaudois, Switzerland, presented data from the EXTEND study, which explored extended-pulse dosing regimens of antibiotics (fidaxomicin) in CDI.

show abstract

Automatic antibiotic resistance prediction in Klebsiella pneumoniae based on MALDI-TOF mass spectra

Guerrero-López

Sevilla-Salcedo

Candela

et al. 2023

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

Development of a clinical prediction model for recurrence and mortality outcomes after Clostridioides difficile infection using a machine learning approach

Cited by 7 publications

References 33 publications

Microbiome profile and calprotectin levels as markers of risk of recurrent Clostridioides difficile infection

Microbiome profile and calprotectin levels as markers of risk of recurrent Clostridioides difficile infection

Clostridioides difficile Infection: Targeting an Unwelcome and Persistent Threat

Automatic antibiotic resistance prediction in Klebsiella pneumoniae based on MALDI-TOF mass spectra

Contact Info

Product

Resources

About