Machine learning model to predict hypotension after starting continuous renal replacement therapy.
Ontology highlight
ABSTRACT:
SUBMITTER: Kang MW
PROVIDER: S-EPMC8387375 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BACKGROUND:Previous scoring models such as the Acute Physiologic Assessment and Chronic Health Evaluation II (APACHE II) and the Sequential Organ Failure Assessment (SOFA) scoring systems do not adequately predict mortality of patients undergoing continuous renal replacement therapy (CRRT) for severe acute kidney injury. Accordingly, the present study applies machine learning algorithms to improve prediction accuracy for this patient subset. METHODS:We randomly divided a total of 1571 adult patients who started CRRT for acute kidney injury into training (70%, n = 1094) and test (30%, n = 477) sets. The primary output consisted of the probability of mortality during admission to the intensive care unit (ICU) or hospital. We compared the area under the receiver operating characteristic curves (AUCs) of several machine learning algorithms with that of the APACHE II, SOFA, and the new abbreviated mortality scoring system for acute kidney injury with CRRT (MOSAIC model) results. RESULTS:For the ICU mortality, the random forest model showed the highest AUC (0.784 [0.744-0.825]), and the artificial neural network and extreme gradient boost models demonstrated the next best results (0.776 [0.735-0.818]). The AUC of the random forest model was higher than 0.611 (0.583-0.640), 0.677 (0.651-0.703), and 0.722 (0.677-0.767), as achieved by APACHE II, SOFA, and MOSAIC, respectively. The machine learning models also predicted in-hospital mortality better than APACHE II, SOFA, and MOSAIC. CONCLUSION:Machine learning algorithms increase the accuracy of mortality prediction for patients undergoing CRRT for acute kidney injury compared with previous scoring models.
| S-EPMC7006166 | biostudies-literature
Project description:BackgroundTransition from continuous renal replacement therapy (CRRT) to intermittent renal replacement therapy (IRRT) can be associated with intra-dialytic hypotension (IDH) although data to inform the definition of IDH, its incidence and clinical implications, are lacking. We aimed to describe the incidence and factors associated with IDH during the first IRRT session following transition from CRRT and its association with hospital mortality. This was a retrospective single-center cohort study in patients with acute kidney injury for whom at least one CRRT-to-IRRT transition occurred while in intensive care. We assessed associations between multiple candidate definitions of IDH and hospital mortality. We then evaluated the factors associated with IDH.ResultsWe evaluated 231 CRRT-to-IRRT transitions in 213 critically ill patients with AKI. Hospital mortality was 43.7% (n = 93). We defined IDH during the first IRRT session as 1) discontinuation of IRRT for hemodynamic instability; 2) any initiation or increase in vasopressor/inotropic agents or 3) a nadir systolic blood pressure of < 90 mmHg. IDH during the first IRRT session occurred in 50.2% of CRRT-to-IRRT transitions and was independently associated with hospital mortality (adjusted odds ratio [OR]: 2.71; CI 1.51-4.84, p < 0.001). Clinical variables at the time of CRRT discontinuation associated with IDH included vasopressor use, higher cumulative fluid balance, and lower urine output.ConclusionsIDH events during CRRT-to-IRRT transition occurred in nearly half of patients and were independently associated with hospital mortality. We identified several characteristics that anticipate the development of IDH following the initiation of IRRT.
| S-EPMC8214642 | biostudies-literature
Project description:BackgroundPredicting time to renal replacement therapy (RRT) is important in patients at high risk for end-stage kidney disease. We developed and validated machine learning models for predicting the time to RRT and compared its accuracy with conventional prediction methods that uses the rate of estimated glomerular filtration rate (eGFR) decline.MethodsData of adult chronic kidney disease (CKD) patients who underwent hemodialysis at Oita University Hospital from April 2016 to March 2021 were extracted from electronic medical records (N = 135). A new machine learning predictor was compared with the established prediction method that uses the eGFR decline rate and the accuracy of the prediction models was determined using the coefficient of determination (R2). The data were preprocessed and split into training and validation datasets. We created multiple machine learning models using the training data and evaluated their accuracy using validation data. Furthermore, we predicted the time to RRT using a conventional prediction method that uses the eGFR decline rate for patients who had measured eGFR three or more times in two years and evaluated its accuracy.ResultsThe least absolute shrinkage and selection operator regression model exhibited moderate accuracy with an R2 of 0.60. By contrast, the conventional prediction method was found to be extremely low with an R2 of -17.1.ConclusionsThe significance of this study is that it shows that machine learning can predict time to RRT moderately well with continuous values from data at a single time point. This approach outperforms the conventional prediction method that uses eGFR time series data and presents new avenues for CKD treatment.
| S-EPMC10943785 | biostudies-literature
Project description:BACKGROUND:Studies of the timing of end-stage renal disease (ESRD) have primarily defined "early" versus "late" initiation of dialysis using estimated glomerular filtration rate (eGFR)-based criteria. Our objective was to determine the theoretical time that could be spent in chronic kidney disease (CKD) stage 5 prior to reaching a conservative eGFR threshold of 5 mL/min/1.73 m2 compared to the actual time spent in CKD stage 5 by risk factors of interest. METHODS:Eight-hundred and seventy Chronic Renal Insufficiency Cohort participants with CKD stage 5 who started renal replacement therapy (RRT) were included for retrospective study. We used mixed models to estimate the person-specific trajectory of renal function. We then used these individual trajectories to estimate the amount of time that would be spent in CKD stage 5 (between eGFR of 15 and 5 mL/min/1.73 m2) and compared this estimate to the actual time spent in CKD stage 5 prior to ESRD (between eGFR of 15 mL/min/1.73 m2 and ESRD). RESULTS:We found the median observed time between eGFR of 15 mL/min/1.73 m2 to RRT was 9.6 months, but the median predicted time between eGFR of 15 mL/min/1.73 m2 to eGFR of 5 mL/min/1.73 m2 was 17.7 months. Some of the largest differences between the predicted and actual amount of time spent in CKD stage 5 were noted among those with systolic blood pressure <140 mm Hg (9.7 months longer predicted compared to actual), proteinuria <1 g/g (9.1 months), and serum albumin ≥3.5 g/dL (9.0 months). CONCLUSION:We found marked differences between the actual and predicted time spent in CKD stage 5 based on risk factors of interest. We believe that placing timing of dialysis initiation in the perspective of time is novel and may identify subgroups of patients who may derive particular benefit from a more concerted effort to delay RRT.
| S-EPMC6667296 | biostudies-literature
Project description:Starting renal replacement therapy (RRT) for patients with chronic kidney disease (CKD) at an optimal time, either with hemodialysis or kidney transplantation, is crucial for patient's well-being and for successful management of the condition. In this paper, we explore the possibilities of creating forecasting models to predict the onset of RRT 3, 6, and 12 months from the time of the patient's first diagnosis with CKD, using only the comorbidities data from National Health Insurance from Taiwan. The goal of this study was to see whether a limited amount of data (including comorbidities but not considering laboratory values which are expensive to obtain in low- and medium-income countries) can provide a good basis for such predictive models. On the other hand, in developed countries, such models could allow policy-makers better planning and allocation of resources for treatment. Using data from 8,492 patients, we obtained the area under the receiver operating characteristic curve (AUC) of 0.773 for predicting RRT within 12 months from the time of CKD diagnosis. The results also show that there is no additional advantage in focusing only on patients with diabetes in terms of prediction performance. Although these results are not as such suitable for adoption into clinical practice, the study provides a strong basis and a variety of approaches for future studies of forecasting models in healthcare.
| S-EPMC7274378 | biostudies-literature
Project description:BackgroundTo construct and evaluate a predictive model for in-hospital mortality among critically ill patients with acute kidney injury (AKI) undergoing continuous renal replacement therapy (CRRT), based on nine machine learning (ML) algorithm.MethodsThe study retrospectively included patients with AKI who underwent CRRT during their initial hospitalization in the United States using the medical information mart for intensive care (MIMIC) database IV (version 2.0), as well as in the intensive care unit (ICU) of Huzhou Central Hospital. Patients from the MIMIC database were used as the training cohort to construct the models (from 2008 to 2019, n = 1068). Patients from Huzhou Central Hospital were utilized as the external validation cohort to evaluate the models (from June 2019 to December 2022, n = 327). In the training cohort, least absolute shrinkage and selection operator (LASSO) regression with cross-validation was employed to select features for constructing the model and subsequently established nine ML predictive models. The performance of these nine models on the external validation cohort dataset was comprehensively evaluated based on the area under the receiver operating characteristic curve (AUROC) and the optimal model was selected. A static nomogram and a web-based dynamic nomogram were presented, with a comprehensive evaluation from the perspectives of discrimination (AUROC), calibration (calibration curve) and clinical practicability (DCA curves).ResultsFinally, 1395 eligible patients were enrolled, including 1068 patients in the training cohort and 327 patients in the external validation cohort. In the training cohort, LASSO regression with cross-validation was employed to select features and nine models were individually constructed. Compared to the other eight models, the Lasso regularized logistic regression (Lasso-LR) model exhibited the highest AUROC (0.756) and the optimal calibration curve. The DCA curve suggested a certain clinical utility in predicting in-hospital mortality among critically ill patients with AKI undergoing CRRT. Consequently, the Lasso-LR model was the optimal model and it was visualized as a common nomogram (static nomogram) and a web-based dynamic nomogram (https://chsyh2006.shinyapps.io/dynnomapp/). Discrimination, calibration and DCA curves were employed to assess the performance of the nomogram. The AUROC for the training and external validation cohorts in the nomogram model was 0.771 (95%CI: 0.743, 0.799) and 0.756 (95%CI: 0.702, 0.809), respectively. The calibration slope and Brier score for the training cohort were 1.000 and 0.195, while for the external validation cohort, they were 0.849 and 0.197, respectively. The DCA indicated that the model had a certain clinical application value.ConclusionsOur study selected the optimal model and visualized it as a static and dynamic nomogram integrating clinical predictors, so that clinicians can personalized predict the in-hospital outcome of critically ill patients with AKI undergoing CRRT upon ICU admission.
| S-EPMC11334739 | biostudies-literature
Project description:To report circuit characteristics and survival analysis in children weighing ≤10 kg enrolled in the Prospective Pediatric Continuous Renal Replacement Therapy (ppCRRT) Registry.We conducted prospective cohort analysis of the ppCRRT Registry to: (1) evaluate survival differences in children ≤10 kg compared with other children; (2) determine demographic and clinical differences between surviving and non-surviving children ≤10 kg; and (3) describe continuous renal replacement therapy (CRRT) circuit characteristics differences in children ≤5 kg versus 5-10 kg.The ppCRRT enrolled 84 children ≤10 kg between January 2001 and August 2005 from 13 US tertiary centers. Children ≤10 kg had lower survival rates than children >10 kg (36/84 [43%] versus 166/260 [64%]; P < .001). In children ≤10 kg, survivors were more likely to have fewer days in intensive care unit prior to CRRT, lower Pediatric Risk of Mortality 2 scores at intensive care unit admission and lower mean airway pressure (P(aw)), higher urine output, and lower percent fluid overload (FO) at CRRT initiation. Adjusted regression analysis revealed that Pediatric Risk of Mortality 2 scores, FO, and decreased urine output were associated with mortality. Compared with circuits from children 5-10 kg at CRRT initiation, circuits from children ≤5 kg more commonly used blood priming for initiation, heparin anticoagulation, and higher blood flows/effluent flows for body weight.Mortality is more common in children who are ≤10 kg at the time of CRRT initiation. Like other CRRT populations, urine output and FO at CRRT initiation are independently associated with mortality. CRRT prescription differs in small children.
| S-EPMC5545826 | biostudies-other
Project description:Doripenem is a broad-spectrum parenteral carbapenem with enhanced activity against Pseudomonas aeruginosa. While the initial dosing recommendation for renally competent patients and patients undergoing continuous renal replacement therapy (cRRT) was 500 mg every 8 h (q8h), the dose for renally competent patients was updated to 1 g q8h in June 2012. There are no updated data for the dosing of patients on continuous renal replacement therapy. The original dosing regimen for cRRT patients was based on nonseptic patients, while newer publications chose comparatively low target concentrations for a carbapenem. Thus, there is an urgent need for updated recommendations for dosing during cRRT. In the trial presented here, we included 13 oliguric septic patients undergoing cRRT in an intensive care setting. Five patients each were treated with hemodiafiltration or hemodialysis, while three patients received hemofiltration treatment. All patients received 1 g doripenem every 8 h. Doripenem concentrations in the plasma and ultrafiltrate were measured over 48 h. The mean hemofilter clearance was 36.53 ml/min, and the mean volume of distribution was 59.26 liters. The steady-state trough levels were found at 8.5 mg/liter, with no considerable accumulation. Based on pharmacokinetic and pharmacodynamic considerations, we propose a regimen of 1 g q8h, which may be combined with a loading dose of 1.5 to 2 g for critically ill patients. (This study has been registered with EudraCT under registration no. 2009-018010-18 and at ClinicalTrials.gov under registration no. NCT02018939.).
| S-EPMC4775919 | biostudies-literature
Project description:BackgroundRhabdomyolysis is a condition that is characterised by the breakdown of skeletal muscle tissue and leakage of intracellular myocyte contents into circulating blood. Rhabdomyolysis can lead to acute kidney injury (AKI) and is a potentially life-threatening condition. Studies have indicated that continuous renal replacement therapy (CRRT) may provide benefits for people with rhabdomyolysis by removing potentially damaging myoglobin and stabilising haemodynamic and metabolic status.ObjectivesWe aimed to: i) assess the efficacy of CRRT in removing myoglobin; ii) investigate the influence of CRRT on mortality and kidney-related outcomes; and iii) evaluate the safety of CRRT for the treatment of people with rhabdomyolysis.Search methodsWe searched the Cochrane Renal Group's Specialised Register to 6 January 2014 through contact with the Trials' Search Co-ordinator using search terms relevant to this review. We also searched China National Knowledge Infrastructure (from 1 January 1979 to 16 April 2013) and the Chinese Clinical Trials Register (to 16 April 2013).Selection criteriaAll randomised controlled trials (RCTs) and quasi-RCTs that investigated clinical outcomes of CRRT for people with rhabdomyolysis were included.Data collection and analysisTwo authors independently assessed studies for inclusion and extracted data. We derived risk ratios (RR) for dichotomous data and mean differences (MD) for continuous data with 95% confidence intervals (CI). Methodological risk of bias was assessed using the Cochrane risk of bias tool.Main resultsOf the three included studies (101 participants), one evaluated continuous arteriovenous haemodialysis and two investigated continuous venovenous haemofiltration; all included conventional therapy as control.We found significant decreases in myoglobin in patients among whom CRRT therapy was initiated on days four, eight, and 10 (day 4: MD -11.00 (μg/L), 95% CI -20.65 to -1.35; Day 8: MD -23.00 (μg/L), 95% CI -30.92 to -15.08; day 10: MD -341.87 (μg/L), 95% CI -626.15 to -57.59) compared with those who underwent conventional therapy.Although CRRT was associated with improved serum creatinine, blood urea nitrogen, and potassium levels; reduced duration of the oliguria phase; and was associated with reduced time in hospital, no significant differences were found in mortality rates compared with conventional therapy (RR 0.17, 95% CI 0.02 to 1.37). The included studies did not report on long-term outcomes or prevention of AKI.Overall, we found that study quality was suboptimal: blinding and randomisation allocation were not reported by any of the included studies, leading to the possibility of selection, performance and detection bias.Authors' conclusionsAlthough CRRT may provide some benefits for people with rhabdomyolysis, the poor methodological quality of the included studies and lack of data relating to clinically important outcomes limited our findings about the effectiveness of CRRT for people with rhabdomyolysis.There was insufficient evidence to discern any likely benefits of CRRT over conventional therapy for people with rhabdomyolysis and prevention of rhabdomyolysis-induced AKI.
| S-EPMC10634660 | biostudies-literature
Project description:BackgroundSeveral studies suggest improved outcomes for patients with kidney disease who consult a nephrologist. However, it remains undetermined whether a consultation with a nephrologist is related to a survival benefit after starting continuous renal replacement therapy (CRRT) due to acute kidney injury (AKI).MethodsData from 2,397 patients who started CRRT due to severe AKI at Seoul National University Hospital, Korea between 2010 and 2020 were retrospectively collected. The patients were divided into two groups according to whether they underwent a nephrology consultation regarding the initiation and maintenance of CRRT. The Cox proportional hazards model was used to calculate the hazard ratio (HR) of mortality during admission to the intensive care unit after adjusting for multiple variables.ResultsA total of 2,153 patients (89.8%) were referred to nephrologists when starting CRRT. The patients who underwent a nephrology consultation had a lower mortality rate than those who did not have a consultation (HR = 0.47 [0.40-0.56]; P < 0.001). Subsequently, patients who had nephrology consultations were divided into two groups (i.e., early and late) according to the timing of the consultation. Both patients with early and late consultation had lower mortality rates than patients without consultations, with HRs of 0.45 (0.37-0.54) and 0.51 (0.42-0.61), respectively.ConclusionsConsultation with a nephrologist may contribute to a survival benefit after starting CRRT for AKI.
| S-EPMC9931119 | biostudies-literature