Machine learning for clinical decision support in infectious diseases: a narrative review of current applications.

BACKGROUND: Machine learning (ML) is a growing field in medicine. This narrative review describes the current body of literature on ML for clinical decision support in infectious diseases (ID). OBJECTIVES: We aim to inform clinicians about the use of ML for diagnosis, classification, outcome prediction and antimicrobial management in ID. SOURCES: References for this review were identified through searches of MEDLINE/PubMed, EMBASE, Google Scholar, biorXiv, ACM Digital Library, arXiV and IEEE Xplore Digital Library up to July 2019. CONTENT: We found 60 unique ML-clinical decision support systems (ML-CDSS) aiming to assist ID clinicians. Overall, 37 (62%) focused on bacterial infections, 10 (17%) on viral infections, nine (15%) on tuberculosis and four (7%) on any kind of infection. Among them, 20 (33%) addressed the diagnosis of infection, 18 (30%) the prediction, early detection or stratification of sepsis, 13 (22%) the prediction of treatment response, four (7%) the prediction of antibiotic resistance, three (5%) the choice of antibiotic regimen and two (3%) the choice of a combination antiretroviral therapy. The ML-CDSS were developed for intensive care units (n = 24, 40%), ID consultation (n = 15, 25%), medical or surgical wards (n = 13, 20%), emergency department (n = 4, 7%), primary care (n = 3, 5%) and antimicrobial stewardship (n = 1, 2%). Fifty-three ML-CDSS (88%) were developed using data from high-income countries and seven (12%) with data from low- and middle-income countries (LMIC). The evaluation of ML-CDSS was limited to measures of performance (e.g. sensitivity, specificity) for 57 ML-CDSS (95%) and included data in clinical practice for three (5%). IMPLICATIONS: Considering comprehensive patient data from socioeconomically diverse healthcare settings, including primary care and LMICs, may improve the ability of ML-CDSS to suggest decisions adapted to various clinical contexts. Currents gaps identified in the evaluation of ML-CDSS must also be addressed in order to know the potential impact of such tools for clinicians and patients.

Individual pharmacokinetic variation leads to underdosing of ciprofloxacin in some cystic fibrosis patients.

BACKGROUND: Ciprofloxacin (CIP) is frequently used when treating cystic fibrose (CF) patients with intermittent Pseudomonas aeruginosa (P. aeruginosa) lung colonization. However, approximately 20% of the patients progress to chronic infection despite early intervention. The aim of this study, was to investigate the pharmacokinetics of CIP, to evaluate if CYP3A4-related metabolism is involved and to find the optimal dose needed to eradicate intermittently colonizing bacteria in the lungs of CF patients. Methods An open-label, prospective pharmacokinetic study was performed. Twenty-two adult CF-patients were each given 500 mg CIP orally. One blood sample was taken at t = 0, and the following 12 hr, nine blood samples were collected. The optimal dose and interval was then calculated by Monte Carlo simulation. CYP3A4-activity was mesured using the Erythromycin Breath Test (ERMBT). Results A 14-fold variation in AUC for the 500 mg CIP (median 473.5 µg/ml × min), and a 30-fold variation in Cmax for CIP (median 2 µg/ml) was found. For CYP3A4-activity the variation was 8-fold. No correlation was found between the CYP3A4-activity and CIP-concentrations. The probability of eradicating intermittent P. aeruginosa colonization in the lungs of CF patients was found to be 57% (3 doses/day), when 500 mg CIP was given. It was calculated to be 89% (2 doses/day) and 94% (3 doses/day), respectivly if 750 mg CIP had been given. Conclusion A large pharmacokinetic difference of CIP in CF patiens was found, not explained by CYP3A4 variation. CIP should be given at 750 mg two or three times daily to adult CF patients with intermittently colonization. Pediatr Pulmonol. 2017;52:319-323. © 2016 Wiley Periodicals, Inc.

Non-uniform phenotyping of D12S391 resolved by second generation sequencing.

Non-uniform phenotyping of five case work samples were observed in the D12S391 locus. The samples were typed at least twice with the AmpFℓSTR NGM SElect PCR Amplification Kit and different alleles were called with GeneMapper ID-X in the different experiments. Detailed analyses of the electropherograms suggested that the individuals were heterozygous with two alleles that differed in size by one nucleotide. This was confirmed by amplifying the samples with the PowerPlex ESX 17 system. D12S391 is a complex STR with variable numbers of AGAT and AGAC repeats. Second generation sequencing revealed that separation of two alleles differing by one nucleotide in length was poor if the number of AGAT repeats in the short allele was higher than in the long allele. A total of 45 individuals with microvariants or off-ladder alleles in D12S391 were sequenced. Thirty different alleles were detected and sixteen of these were not previously reported.

Pharmacokinetic variability of clarithromycin and differences in CYP3A4 activity in patients with cystic fibrosis.

BACKGROUND: To investigate the correlation between CYP3A4/5 activity and clarithromycin metabolism, and between CYP3A activity and CYP3A genotype. METHODS: This is an open-label, prospective pharmacokinetic study evaluating CYP3A activity using The Erythromycin Breath Test. Eight blood samples were collected within 12h after clarithromycin 500 mg was administered orally. The clarithromycin concentrations were measured by liquid chromatography-tandem mass spectrometry. AUC, Tmax and Cmax were calculated. Selected Single Nucleotide polymorphisms in CYP3A4/5 genes were assessed by PCR and single base extension. RESULTS: Twenty-one chronically infected patients were included. An 8-fold variation in the CYP3A4 activity, 10-fold variation in AUC for clarithromycin (median 881 µg/mL × min), and a 16-fold variation in Cmax for clarithromycin (median 3.4 µg/mL) were found. A linear correlation between the CYP3A4-activity and clarithromycin metabolism was demonstrated (P < 0.05). CONCLUSION: The large variation in the clarithromycin pharmacokinetics in cystic fibrosis patients may cause treatment failure. The Erythromycin Breath Test could be valuable in identifying cystic fibrosis patients in risk of treatment failure/drug toxicity.