Medication patterns and dosing guidance in pediatric patients supported with intermittent hemodialysis or continuous kidney replacement therapy.

BACKGROUND: Hemodialysis is a life-saving technology used during periods of acute or chronic kidney failure to remove toxins, and maintain fluid, electrolyte and metabolic balance. While this technology plays an important role for pediatric patients with kidney dysfunction, it can alter the pharmacokinetic behavior of medications placing patients at risk for suboptimal dosing and drug toxicity. The ability to directly translate pharmacokinetic alterations into dosing recommendations has thus far been limited and dosing guidance specific to pediatric hemodialysis patients is rare. Despite differences in dialysis prescription and patient populations, intermittent (iHD) and continuous kidney replacement therapy (CKRT) patients are often pooled together. In order to develop evidence-based dosing guidelines, it is important to first prioritize drugs for study in each modality. METHODS: Here we aim to identify priority drugs in two hemodialysis modalities, through: 1) Identification of hospitalized, pediatric patients who received CKRT or intermittent hemodialysis (iHD) using a machine learning-based predictive model based on medications; 2) Identification of medication administration patterns in these patient cohorts; and 3) Identification of the most commonly prescribed drugs that lack published dosing guidance. RESULTS: Notable differences were found in the pattern of medications and drug dosing guidance between iHD and CKRT patients. Antibiotics, diuretics and sedatives were more common in CKRT patients. Out of the 50 most commonly administered medications in the two modalities, only 34% and 28% had dosing guidance present for iHD and CKRT, respectively. CONCLUSIONS: Our results add to the understanding of the differences between iHD and CKRT patient populations by identifying commonly used medications that lack dosing guidance for each hemodialysis modality, helping to pinpoint priority medications for further study. Overall, this study provides an overview of the current limitations in medication use in this at-risk population, and provides a framework for future studies by identifying commonly used medications in pediatric CKRT and iHD patients.

Maximum likelihood estimation of renal transporter ontogeny profiles for pediatric PBPK modeling.

Optimal treatment of infants with many renally cleared drugs must account for maturational differences in renal transporter (RT) activity. Pediatric physiologically-based pharmacokinetic (PBPK) models may incorporate RT activity, but this requires ontogeny profiles for RT activity in children, especially neonates, to predict drug disposition. Therefore, RT expression measurements from human kidney postmortem cortical tissue samples were normalized to represent a fraction of mature RT activity. Using these data, maximum likelihood estimated the distributions of RT activity across the pediatric age spectrum, including preterm and term neonates. PBPK models of four RT substrates (acyclovir, ciprofloxacin, furosemide, and meropenem) were evaluated with and without ontogeny profiles using average fold error (AFE), absolute average fold error (AAFE), and proportion of observations within the 5-95% prediction interval. Novel maximum likelihood profiles estimated ontogeny distributions for the following RT: OAT1, OAT3, OCT2, P-gp, URAT1, BCRP, MATE1, MRP2, MRP4, and MATE-2 K. Profiles for OAT3, P-gp, and MATE1 improved infant furosemide and neonate meropenem PBPK model AFE from 0.08 to 0.70 and 0.53 to 1.34 and model AAFE from 12.08 to 1.44 and 2.09 to 1.36, respectively, and improved the percent of data within the 5-95% prediction interval from 48% to 98% for neonatal ciprofloxacin simulations, respectively. Even after accounting for other critical population-specific maturational differences, novel RT ontogeny profiles substantially improved neonatal PBPK model performance, providing validated estimates of maturational differences in RT activity for optimal dosing in children.

Anakinra Removal by Continuous Renal Replacement Therapy: An Ex Vivo Analysis.

OBJECTIVES: Patients with sepsis are at significant risk for multiple organ dysfunction, including the lungs and kidneys. To manage the morbidity associated with kidney impairment, continuous renal replacement therapy (CRRT) may be required. The extent of anakinra pharmacokinetics in CRRT remains unknown. The objectives of this study were to investigate the anakinra-circuit interaction and quantify the rate of removal from plasma. DESIGN: The anakinra-circuit interaction was evaluated using a closed-loop ex vivo CRRT circuit. CRRT was performed in three phases based on the method of solute removal: 1) hemofiltration, 2) hemodialysis, and 3) hemodiafiltration. Standard control samples of anakinra were included to assess drug degradation. SETTING: University research laboratory. PATIENTS: None. INTERVENTIONS: Anakinra was administered to the CRRT circuit and serial prefilter blood samples were collected along with time-matched control and hemofiltrate samples. Each circuit was run in triplicate to assess inter-run variability. Concentrations of anakinra in each reference fluid were measured by enzyme-linked immunosorbent assay. Transmembrane filter clearance was estimated by the product of the sieving coefficient/dialysate saturation constant and circuit flow rates. MEASUREMENTS AND MAIN RESULTS: Removal of anakinra from plasma occurred within minutes for each CRRT modality. Average drug remaining (%) in plasma following anakinra administration was lowest with hemodiafiltration (34.9%). The average sieving coefficient was 0.34, 0.37, and 0.41 for hemodiafiltration, hemofiltration, and hemodialysis, respectively. Transmembrane clearance was fairly consistent across each modality with the highest during hemodialysis (5.53 mL/min), followed by hemodiafiltration (4.99 mL/min), and hemofiltration (3.94 mL/min). Percent drug remaining within the control samples (93.1%) remained consistent across each experiment, indicating negligible degradation within the blood. CONCLUSIONS: The results of this analysis are the first to demonstrate that large molecule therapeutic proteins such as anakinra, are removed from plasma with modern CRRT technology. Current dosing recommendations for patients with severe renal impairment may result in subtherapeutic anakinra concentrations in those receiving CRRT.

Interaction of ceftazidime and clindamycin with extracorporeal life support.

BACKGROUND: Ceftazidime and clindamycin are commonly prescribed to critically ill patients who require extracorporeal life support such as ECMO and CRRT. The effect of ECMO and CRRT on the disposition of ceftazidime and clindamycin is currently unknown. METHODS: Ceftazidime and clindamycin extraction were studied with ex vivo ECMO and CRRT circuits primed with human blood. The percent recovery of these drugs over time was calculated to determine the degree of interaction between these drugs and circuit components. RESULTS: Neither ceftazidime nor clindamycin exhibited measurable interactions with the ECMO circuit. In contrast, CRRT cleared 100% of ceftazidime from the experimental circuit within the first 2 h. Clearance of clindamycin from the CRRT circuit was slower, with about 20% removed after 6 h. CONCLUSION: Clindamycin and ceftazidime dosing adjustments are likely required in patients who are supported with CRRT, and future studies to quantify these adjustments should consider the pathophysiology of the patient in combination with the clearance due to CRRT. Dosing adjustments to account for adsorption to ECMO circuit components are likely unnecessary and should focus instead on the pathophysiology of the patient and changes in volume of distribution. These results will help improve the safety and efficacy of ceftazidime and clindamycin in patients requiring ECMO and CRRT.

Drug exposure during pregnancy: Current understanding and approaches to measure maternal-fetal drug exposure.

Prescription drug use is prevalent during pregnancy, yet there is limited knowledge about maternal-fetal safety and efficacy of this drug use because pregnant individuals have historically been excluded from clinical trials. Underrepresentation has resulted in a lack of data available to estimate or predict fetal drug exposure. Approaches to study fetal drug pharmacology are limited and must be evaluated for feasibility and accuracy. Anatomic and physiological changes throughout pregnancy fluctuate based on gestational age and can affect drug pharmacokinetics (PK) for both mother and fetus. Drug concentrations have been studied throughout different stages of gestation and at or following delivery in tissue and fluid biospecimens. Sampling amniotic fluid, umbilical cord blood, placental tissue, meconium, umbilical cord tissue, and neonatal hair present surrogate options to quantify and characterize fetal drug exposure. These sampling methods can be applied to all therapeutics including small molecule drugs, large molecule drugs, conjugated nanoparticles, and chemical exposures. Alternative approaches to determine PK have been explored, including physiologically based PK modeling, in vitro methods, and traditional animal models. These alternative approaches along with convenience sampling of tissue or fluid biospecimens can address challenges in studying maternal-fetal pharmacology. In this narrative review, we 1) present an overview of the current understanding of maternal-fetal drug exposure; 2) discuss biospecimen-guided sampling design and methods for measuring fetal drug concentrations throughout gestation; and 3) propose methods for advancing pharmacology research in the maternal-fetal population.

Medication based machine learning to identify subpopulations of pediatric hemodialysis patients in an electronic health record database.

Electronic health records (EHRs) have given rise to large and complex databases of medical information that have the potential to become powerful tools for clinical research. However, differences in coding systems and the detail and accuracy of the information within EHRs can vary across institutions. This makes it challenging to identify subpopulations of patients and limits the widespread use of multi-institutional databases. In this study, we leveraged machine learning to identify patterns in medication usage among hospitalized pediatric patients receiving renal replacement therapy and created a predictive model that successfully differentiated between intermittent (iHD) and continuous renal replacement therapy (CRRT) hemodialysis patients. We trained six machine learning algorithms (logistical regression, Naïve Bayes, k-nearest neighbor, support vector machine, random forest, and gradient boosted trees) using patient records from a multi-center database (n = 533) and prescribed medication ingredients (n = 228) as features to discriminate between the two hemodialysis types. Predictive skill was assessed using a 5-fold cross-validation, and the algorithms showed a range of performance from 0.7 balanced accuracy (logistical regression) to 0.86 (random forest). The two best performing models were further tested using an independent single-center dataset and achieved 84-87% balanced accuracy. This model overcomes issues inherent within large databases and will allow us to utilize and combine historical records, significantly increasing population size and diversity within both iHD and CRRT populations for future clinical studies. Our work demonstrates the utility of using medications alone to accurately differentiate subpopulations of patients in large datasets, allowing codes to be transferred between different coding systems. This framework has the potential to be used to distinguish other subpopulations of patients where discriminatory ICD codes are not available, permitting more detailed insights and new lines of research.

Remdesivir and GS-441524 Extraction by Ex Vivo Extracorporeal Life Support Circuits.

Patients with severe, COVID-related multi-organ failure often require extracorporeal life support (ECLS) such as extracorporeal membrane oxygenation (ECMO) or continuous renal replacement therapy (CRRT). An ECLS can alter drug exposure via multiple mechanisms. Remdesivir (RDV) and its active metabolite GS-441524 are likely to interact with ECLS circuits, resulting in lower than expected exposures. We evaluated circuit-drug interactions in closed loop, ex vivo ECMO and CRRT circuits. We found that mean (standard deviation) recovery of RDV at 6 hours after dosing was low in both the ECMO (33.3% [2.0]) and CRRT (3.5% [0.4]) circuits. This drug loss appears to be due primarily to drug adsorption by the circuit materials and potentially due to metabolism in the blood. GS-441524 recovery at 6 hours was high in the ECMO circuit 75.8% (16.5); however, was not detectable at 6 hours in the CRRT circuit. Loss in the CRRT circuit appears to be due primarily to efficient hemodiafiltration. The extent of loss for both molecules, especially in CRRT, suggests that in patients supported with ECMO and CRRT, RDV dosing adjustments are needed.

Diet modulates cecum bacterial diversity and physiological phenotypes across the BXD mouse genetic reference population.

The BXD family has become one of the preeminent genetic reference populations to understand the genetic and environmental control of phenotypic variation. Here we evaluate the responses to different levels of fat in the diet using both chow diet (CD, 13-18% fat) and a high-fat diet (HFD, 45-60% fat). We studied cohorts of BXD strains, both inbred parents C57BL/6J and DBA/2J (commonly known as B6 and D2, respectively), as well as B6D2 and D2B6 reciprocal F1 hybrids. The comparative impact of genetic and dietary factors was analyzed by profiling a range of phenotypes, most prominently their cecum bacterial composition. The parents of the BXDs and F1 hybrids express limited differences in terms of weight and body fat gain on CD. In contrast, the strain differences on HFD are substantial for percent body fat, with DBA/2J accumulating 12.5% more fat than C57BL/6J (P < 0.0001). The F1 hybrids born to DBA/2J dams (D2B6F1) have 10.6% more body fat (P < 0.001) than those born to C57BL/6J dams. Sequence analysis of the cecum microbiota reveals important differences in bacterial composition among BXD family members with a substantial shift in composition caused by HFD. Relative to CD, the HFD induces a decline in diversity at the phylum level with a substantial increase in Firmicutes (+13.8%) and a reduction in Actinobacteria (-7.9%). In the majority of BXD strains, the HFD also increases cecal sIgA (P < 0.0001)-an important component of the adaptive immunity response against microbial pathogens. Host genetics modulates variation in cecum bacterial composition at the genus level in CD, with significant quantitative trait loci (QTLs) for Oscillibacter mapped to Chr 3 (18.7-19.2 Mb, LRS = 21.4) and for Bifidobacterium mapped to Chr 6 (89.21-89.37 Mb, LRS = 19.4). Introduction of HFD served as an environmental suppressor of these QTLs due to a reduction in the contribution of both genera (P < 0.001). Relations among liver metabolites and cecum bacterial composition were predominant in CD cohort, but these correlations do not persist following the shift to HFD. Overall, these findings demonstrate the important impact of environmental/dietary manipulation on the relationships between host genetics, gastrointestinal bacterial composition, immunological parameters, and metabolites-knowledge that will help in the understanding of the causal sources of metabolic disorders.

Variation in time and magnitude of immune response and viremia in experimental challenges with Porcine circovirus 2b.

BACKGROUND: Porcine circovirus 2 is the primary agent responsible for inducing a group of associated diseases known as Porcine Circovirus Associated Diseases (PCVAD), which can have detrimental effects on production efficiency as well as causing significant mortality. The objective of this study was to evaluate variation in viral replication, immune response and growth across pigs (n = 974) from different crossbred lines. The approach used in this study was experimental infection with a PCV2b strain of pigs at an average of 43 days of age. RESULTS: The sequence of the PCV2b isolate used in the challenge was similar with a cluster of PCV2b isolates known to induce PCVAD and increased mortality rates. The swine leukocyte antigen class II (SLAII) profile of the population was diverse, with nine DQB1 haplotypes being present. Individual viremia and antibody profiles during challenge demonstrate variation in magnitude and time of viral surge and immune response. The correlations between PCV2 specific antibodies and average daily gain (ADG) were relatively low and varied between - 0.14 to 0.08 for IgM and -0.02 and 0.11 for IgG. In contrast, PCV2 viremia was an important driver of ADG decline following infection; a moderate negative correlation was observed between viral load and overall ADG (r = - 0.35, P < 0.001). The pigs with the lowest 10% level of viral load maintained a steady increase in weekly ADG (P < 0.0001) compared to the pigs that had the 10% greatest viral load (P < 0.55). In addition, the highly viremic group expressed higher IgM and IgG starting with d 14 and d 21 respectively, and higher tumor necrosis factor - alpha (TNF-α) at d 21 (P < 0.005), compared to low viremic group. CONCLUSIONS: Molecular sources of the observed differences in viremia and immune response could provide a better understanding of the host factors that influence the development of PCVAD and lead to improved knowledge of swine immunity.

Murine gut microbiota is defined by host genetics and modulates variation of metabolic traits.

The gastrointestinal tract harbors a complex and diverse microbiota that has an important role in host metabolism. Microbial diversity is influenced by a combination of environmental and host genetic factors and is associated with several polygenic diseases. In this study we combined next-generation sequencing, genetic mapping, and a set of physiological traits of the BXD mouse population to explore genetic factors that explain differences in gut microbiota and its impact on metabolic traits. Molecular profiling of the gut microbiota revealed important quantitative differences in microbial composition among BXD strains. These differences in gut microbial composition are influenced by host-genetics, which is complex and involves many loci. Linkage analysis defined Quantitative Trait Loci (QTLs) restricted to a particular taxon, branch or that influenced the variation of taxa across phyla. Gene expression within the gastrointestinal tract and sequence analysis of the parental genomes in the QTL regions uncovered candidate genes with potential to alter gut immunological profiles and impact the balance between gut microbial communities. A QTL region on Chr 4 that overlaps several interferon genes modulates the population of Bacteroides, and potentially Bacteroidetes and Firmicutes-the predominant BXD gut phyla. Irak4, a signaling molecule in the Toll-like receptor pathways is a candidate for the QTL on Chr15 that modulates Rikenellaceae, whereas Tgfb3, a cytokine modulating the barrier function of the intestine and tolerance to commensal bacteria, overlaps a QTL on Chr 12 that influence Prevotellaceae. Relationships between gut microflora, morphological and metabolic traits were uncovered, some potentially a result of common genetic sources of variation.

Genotypes of Brassica rapa respond differently to plant-induced variation in air CO2 concentration in growth chambers with standard and enhanced venting.

Growth chambers allow measurement of phenotypic differences among genotypes under controlled environment conditions. However, unintended variation in growth chamber air CO2 concentration ([CO2]) may affect the expression of diverse phenotypic traits, and genotypes may differ in their response to variation in [CO2]. We monitored [CO2] and quantified phenotypic responses of 22 Brassica rapa genotypes in growth chambers with either standard or enhanced venting. [CO2] in chambers with standard venting dropped to 280 micromol mol(-1) during the period of maximum canopy development, approximately 80 micromol mol(-1) lower than in chambers with enhanced venting. The stable carbon isotope ratio of CO2 in chamber air (delta13C(air)) was negatively correlated with [CO2], suggesting that photosynthesis caused observed [CO2] decreases. Significant genotype x chamber-venting interactions were detected for 12 of 20 traits, likely due to differences in the extent to which [CO2] changed in relation to genotypes' phenology or differential sensitivity of genotypes to low [CO2]. One trait, 13C discrimination (delta13C), was particularly influenced by unaccounted-for fluctuations in delta13C(air) and [CO2]. Observed responses to [CO2] suggest that genetic variance components estimated in poorly vented growth chambers may be influenced by the expression of genes involved in CO2 stress responses; genotypic values estimated in these chambers may likewise be misleading such that some mapped quantitative trait loci may regulate responses to CO2 stress rather than a response to the environmental factor of interest. These results underscore the importance of monitoring, and where possible, controlling [CO2].