Browsing by Author "Srivastava, Rachana"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Continuous Kidney Replacement Therapy and Survival in Children and Young Adults: Findings From the Multinational WE-ROCK Collaborative
    (Elsevier, 2024) Starr, Michelle C.; Gist, Katja M.; Zang, Huaiyu; Ollberding, Nicholas J.; Balani, Shanthi; Cappoli, Andrea; Ciccia, Eileen; Joseph, Catherine; Kakajiwala, Aadil; Kessel, Aaron; Muff-Luett, Melissa; Santiago Lozano, María J.; Pinto, Matthew; Reynaud, Stephanie; Solomon, Sonia; Slagle, Cara; Srivastava, Rachana; Shih, Weiwen V.; Webb, Tennille; Menon, Shina; Worldwide Exploration of Renal Replacement Outcomes Collaborative in Kidney Disease (WE-ROCK) Collaborative; Pediatrics, School of Medicine
    Rationale & objective: There are limited studies describing the epidemiology and outcomes in children and young adults receiving continuous kidney replacement therapy (CKRT). We aimed to describe associations between patient characteristics, CKRT prescription, and survival. Study design: Retrospective multicenter cohort study. Setting & participants: 980 patients aged from birth to 25 years who received CKRT between 2015 and 2021 at 1 of 32 centers in 7 countries participating in WE-ROCK (Worldwide Exploration of Renal Replacement Outcomes Collaborative in Kidney Diseases). Exposure: CKRT for acute kidney injury or volume overload. Outcomes: Death before intensive care unit (ICU) discharge. Analytical approach: Descriptive statistics. Results: Median age was 8.8 years (IQR, 1.6-15.0), and median weight was 26.8 (IQR, 11.6-55.0) kg. CKRT was initiated a median of 2 (IQR, 1-6) days after ICU admission and lasted a median of 6 (IQR, 3-14) days. The most common CKRT modality was continuous venovenous hemodiafiltration. Citrate anticoagulation was used in 62%, and the internal jugular vein was the most common catheter placement location (66%). 629 participants (64.1%) survived at least until ICU discharge. CKRT dose, filter type, and anticoagulation were similar in those who did and did not survive to ICU discharge. There were apparent practice variations by institutional ICU size. Limitations: Retrospective design; limited representation from centers outside the United States. Conclusions: In this study of children and young adults receiving CKRT, approximately two thirds survived at least until ICU discharge. Although variations in dialysis mode and dose, catheter size and location, and anticoagulation were observed, survival was not detected to be associated with these parameters. Plain-language summary: In this large contemporary epidemiological study of children and young adults receiving continuous kidney replacement therapy in the intensive care unit, we observed that two thirds of patients survived at least until ICU discharge. However, patients with comorbidities appeared to have worse outcomes. Compared with previously published reports on continuous kidney replacement therapy practice, we observed greater use of continuous venovenous hemodiafiltration with regional citrate anticoagulation.
  • Thumbnail Image
    Item
    Machine Learning-Based Prediction Model for ICU Mortality After Continuous Renal Replacement Therapy Initiation in Children
    (Wolters Kluwer, 2024-12-17) Thadani, Sameer; Wu, Tzu-Chun; Wu, Danny T. Y; Kakajiwala, Aadil; Soranno, Danielle E.; Cortina, Gerard; Srivastava, Rachana; Gist, Katja M.; Menon, Shina; Pediatrics, School of Medicine
    Background: Continuous renal replacement therapy (CRRT) is the favored renal replacement therapy in critically ill patients. Predicting clinical outcomes for CRRT patients is difficult due to population heterogeneity, varying clinical practices, and limited sample sizes. Objective: We aimed to predict survival to ICUs and hospital discharge in children and young adults receiving CRRT using machine learning (ML) techniques. Derivation cohort: Patients less than 25 years of age receiving CRRT for acute kidney injury and/or volume overload from 2015 to 2021 (80%). Validation cohort: Internal validation occurred in a testing group of patients from the dataset (20%). Prediction model: Retrospective international multicenter study utilizing an 80/20 training and testing cohort split, and logistic regression with L2 regularization (LR), decision tree, random forest (RF), gradient boosting machine, and support vector machine with linear kernel to predict ICU and hospital survival. Model performance was determined by the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) due to the imbalance in the dataset. Results: Of the 933 patients included in this study, 538 (54%) were male with a median age of 8.97 years and interquartile range (1.81-15.0 yr). The ICU mortality was 35% and hospital mortality was 37%. The RF had the best performance for predicting ICU mortality (AUROC, 0.791 and AUPRC, 0.878) and LR for hospital mortality (AUROC, 0.777 and AUPRC, 0.859). The top two predictors of ICU survival were Pediatric Logistic Organ Dysfunction-2 score at CRRT initiation and admission diagnosis of respiratory failure. Conclusions: These are the first ML models to predict survival at ICU and hospital discharge in children and young adults receiving CRRT. RF outperformed other models for predicting ICU mortality. Future studies should expand the input variables, conduct a more sophisticated feature selection, and use deep learning algorithms to generate more precise models.