Using artificial intelligence to predict mortality in AKI patients: a systematic review/meta-analysis.

Background: Acute kidney injury (AKI) is associated with increased morbidity/mortality. With artificial intelligence (AI), more dynamic models for mortality prediction in AKI patients have been developed using machine learning (ML) algorithms. The performance of various ML models was reviewed in terms of their ability to predict in-hospital mortality for AKI patients. Methods: A literature search was conducted through PubMed, Embase and Web of Science databases. Included studies contained variables regarding the efficacy of the AI model [the AUC, accuracy, sensitivity, specificity, negative predictive value and positive predictive value]. Only original studies that consisted of cross-sectional studies, prospective and retrospective studies were included, while reviews and self-reported outcomes were excluded. There was no restriction on time and geographic location. Results: Eight studies with 37 032 AKI patients were included, with a mean age of 65.3 years. The in-hospital mortality was 18.0% in the derivation and 15.8% in the validation cohorts. The pooled [95% confidence interval (CI)] AUC was observed to be highest for the broad learning system (BLS) model [0.852 (0.820-0.883)] and elastic net final (ENF) model [0.852 (0.813-0.891)], and lowest for proposed clinical model (PCM) [0.765 (0.716-0.814)]. The pooled (95% CI) AUC of BLS and ENF did not differ significantly from other models except PCM [Delong's test P = .022]. PCM exhibited the highest negative predictive value, which supports this model's use as a possible rule-out tool. Conclusion: Our results show that BLS and ENF models are equally effective as other ML models in predicting in-hospital mortality, with variability across all models. Additional studies are needed.

An Overview of Rickets in Children.

Rickets is a common bone disease worldwide that is associated with disturbances in calcium and phosphate homeostasis and can lead to short stature and joint deformities. Rickets can be diagnosed based on history and physical examination, radiological features, and biochemical tests. It can be classified into 2 major groups based on phosphate or calcium levels: phosphopenic and calcipenic. Knowledge of categorization of the type of rickets is essential for prompt diagnosis and proper management. Nutritional rickets is a preventable disease through adequate intake of vitamin D through both dietary and sunlight exposure. There are other subtypes of rickets, such as vitamin D-dependent type 1 rickets and vitamin D-dependent type 2 rickets (due to defects in vitamin D metabolism), renal rickets (due to poor kidney function), and hypophosphatemic rickets (vitamin D-resistant rickets secondary to renal phosphate wasting wherein fibroblast growth factor-23 (FGF-23) often plays a major role), which requires closer monitoring and supplementation with activated vitamin D with or without phosphate supplements. An important development has been the introduction of burosumab, a human monoclonal antibody to FGF-23, which is approved for the treatment of X-linked hypophosphatemia among children 1 year and older.

Antibiotic Dosing in Sustained Low-Efficiency Dialysis in Critically Ill Patients.

PURPOSE OF REVIEW: Sustained low-efficiency dialysis (SLED) is increasingly used as a renal replacement modality in critically ill patients with acute kidney injury (AKI) and hemodynamic instability. There is, therefore, a greater need for the understanding of the antibiotic dosage and pharmacokinetics in these patients, to provide them with optimal therapy. SOURCES OF INFORMATION: PubMed/Medline, Embase, and Google Scholar. METHODS: PubMed/Medline, Embase, and Google Scholar databases were searched using a combination of key words: dialysis, end stage renal disease, renal failure, sustained low efficiency dialysis, extended daily dialysis, prolonged intermittent renal replacement therapy (PIRRT), and antibiotic dosing. Studies that investigated antibiotic dosing and pharmacokinetics during SLED/extended daily dialysis/PIRRT were selected for this review. KEY FINDINGS: Eleven studies met inclusion criteria and selected for data extraction. The data with regard to dialysis specifications, type of antibiotic including dosages, drug clearances, and dosage recommendations are summarized in Table 1. It is a challenge to find therapeutic doses for antibiotics during SLED therapy because, in general, only aminoglycosides and vancomycin can be assayed in clinical laboratories. LIMITATIONS: Although current studies on antibiotic dosing in SLED are limited due to diverse and undersized patient populations, antibiotic dosage adjustments for patients receiving SLED discussed here will serve as a valuable guide. Future large-scale research should focus on establishing guidelines for antibiotic dosage in SLED. IMPLICATIONS: Pharmacokinetic principles should be taken into consideration for the appropriate dosing of drugs during SLED, yet it is vital to monitor response to drug to make sure therapeutic goals are achieved. Antibiotic dosing and timing relative to the initiation of SLED may be important to maximize either the time above the minimum inhibitory concentration (MIC) (time-dependent) or the peak to MIC ratio (concentration-dependent), balancing efficacy and toxicity concerns. Critical care physicians should liaise with nephrologists to make decisions regarding appropriate antibiotic dosing in patients undergoing SLED.

JUSTIFICATION: On recourt de plus en plus à l'hémodialyse prolongée à faible efficacité (SLED ­ sustained low-efficiency dialysis) comme thérapie de remplacement rénal chez les patients gravement malades présentant une insuffisance rénale aiguë (IRA) et une instabilité hémodynamique. Dès lors, il devient impératif de se pencher sur la posologie des antibiotiques et leur pharmacocinétique chez ces patients de façon à leur offrir le meilleur traitement possible. SOURCES: Les bases de données PubMed/Medline, Embase et Google Scholar. MÉTHODOLOGIE: Nous avons épluché les bases de données PubMed/Medline, Embase et Google Scholar à l'aide d'une combinaison de mots-clés : dialysis (dialyse), end stage renal disease (insuffisance rénale terminale), renal failure (insuffisance rénale), sustained low efficiency dialysis (hémodialyse prolongée à faible efficacité), extended daily dialysis (dialyse quotidienne prolongée), prolonged intermittent renal replacement therapy ou PIRRT (thérapie de remplacement rénal intermittente prolongée) et antibiotic dosing (posologie des antibiotiques). Ont été retenues pour cette revue les études qui exploraient la posologie des antibiotiques et leur pharmacocinétique dans les contextes de la SLED, de la dialyse quotidienne prolongée et de la PIRRT. CONSTATS: Onze études répondaient à nos critères d'inclusion. Les données obtenues sur les caractéristiques de la dialyse, de même que sur le type d'antibiotique, sa posologie, sa clairance et les doses recommandées sont résumées dans le Tableau 1. L'établissement de la dose thérapeutique d'antibiotique durant la SLED pose un défi puisque, généralement, seuls les aminoglycosides et la vancomycine peuvent être dosés en laboratoire clinique. LIMITES: Bien que la fiabilité d'études traitant de la posologie des antibiotiques utilisés durant la SLED soit limitée, en raison notamment d'échantillons de sujets faibles et hétérogènes, les ajustements posologiques pour les patients traités par SLED discutés ci-après constitueront des balises utiles. De futurs essais à plus grande échelle devraient se concentrer sur l'établissement de lignes directrices concernant les doses thérapeutiques d'antibiotiques à administrer pendant la SLED. CONCLUSION: Les principes pharmacocinétiques devraient être pris en compte pour établir la posologie d'antibiotique appropriée pendant la SLED. Il demeure toutefois crucial de surveiller la réponse au médicament pour s'assurer que les objectifs thérapeutiques sont atteints. La posologie d'antibiotiques et le moment d'initiation de la SLED pourraient s'avérer importants pour maximiser soit le temps au-dessus de la concentration minimale inhibitrice (en fonction du temps), soit le rapport entre le pic et cette même concentration (en fonction de la concentration), de façon à établir un équilibre entre les préoccupations liées à l'efficacité et celles relatives à la toxicité. Les médecins des unités de soins intensifs et les néphrologues devraient coopérer pour prendre des décisions conjointes quant à la posologie d'antibiotique appropriée pour les patients traités par SLED.