scholarly journals Unsupervised Machine learning to subtype Sepsis-Associated Acute Kidney Injury

2018 ◽  
Author(s):  
Kumardeep Chaudhary ◽  
Aine Duffy ◽  
Priti Poojary ◽  
Aparna Saha ◽  
Kinsuk Chauhan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Kidney Injury ◽  
Tertiary Care ◽  
Vital Signs ◽  
Kidney Injury ◽  
Machine Learning Techniques ◽  
Care Hospital ◽  
Unsupervised Machine Learning ◽  
Laboratory Variables ◽  
Cluster 2

AbstractObjectiveAcute kidney injury (AKI) is highly prevalent in critically ill patients with sepsis. Sepsis-associated AKI is a heterogeneous clinical entity, and, like many complex syndromes, is composed of distinct subtypes. We aimed to agnostically identify AKI subphenotypes using machine learning techniques and routinely collected data in electronic health records (EHRs).DesignCohort study utilizing the MIMIC-III Database.SettingICUs from tertiary care hospital in the U.S.PatientsPatients older than 18 years with sepsis and who developed AKI within 48 hours of ICU admission.InterventionsUnsupervised machine learning utilizing all available vital signs and laboratory measurements.Measurements and Main ResultsWe identified 1,865 patients with sepsis-associated AKI. Ten vital signs and 691 unique laboratory results were identified. After data processing and feature selection, 59 features, of which 28 were measures of intra-patient variability, remained for inclusion into an unsupervised machine-learning algorithm. We utilized k-means clustering with k ranging from 2 – 10; k=2 had the highest silhouette score (0.62). Cluster 1 had 1,358 patients while Cluster 2 had 507 patients. There were no significant differences between clusters on age, race or gender. We found significant differences in comorbidities and small but significant differences in several laboratory variables (hematocrit, bicarbonate, albumin) and vital signs (systolic blood pressure and heart rate). In-hospital mortality was higher in cluster 2 patients, 25% vs. 20%, p=0.008. Features with the largest differences between clusters included variability in basophil and eosinophil counts, alanine aminotransferase levels and creatine kinase values.ConclusionsUtilizing routinely collected laboratory variables and vital signs in the EHR, we were able to identify two distinct subphenotypes of sepsis-associated AKI with different outcomes. Variability in laboratory variables, as opposed to their actual value, was more important for determination of subphenotypes. Our findings show the potential utility of unsupervised machine learning to better subtype AKI.

scholarly journals Evaluation of Aetiology and Outcome of Acute Kidney Injury in a Tertiary Care Hospital of India- A Prospective Observational Study

2020 ◽  
Author(s):  
VS Gaurav Narayan ◽  
SG Ramya ◽  
Sonal Rajesh Kumar ◽  
SK Nellaiappa Ganesan
Keyword(s):  
Acute Kidney Injury ◽  
Intensive Care ◽  
Observational Study ◽  
Prospective Observational Study ◽  
Tertiary Care ◽  
Kidney Injury ◽  
Rapid Decline ◽  
Care Hospital ◽  
Medical Intensive Care

Introduction: The Acute Kidney Injury (AKI) is a rapid decline in renal filtration function. The aetiological spectrum, prevalence of AKI and outcome is highly variable. This variation exists due to the difference in the criteria used, study population and demographic features. Huge differences are noted when AKI is compared in developing and developed countries. Hence, it is important to analyse the spectrum of AKI to facilitate earlier diagnosis and treatment which shall help in improving the outcome. Aim: To study the prevalence, aetiology and outcome of AKI in the medical intensive care. Materials and Methods: This was a prospective observational study conducted in a medical intensive care for 18 months where 1490 patients were screened and 403 patients were included as AKI by KDIGO criteria. History, examination, appropriate investigations and treatment details including dialysis were noted. The serum creatinine levels were obtained every day, to know the time of onset of AKI, at the time of death or discharge, and after one month for patients who turned up for follow-up. Patients were categorised based on outcome as survivors and nonsurvivors. Survivors were divided into as fully recovered and partially recovered and those who left the Intensive Care Unit (ICU) against medical advice were termed as lost to follow-up. Results: A total of 403 patients (27.04% of 1490) of medical intensive care admissions were found to have AKI. Sepsis was the most common cause of AKI. At the end of the month, 78.4% of AKI patients fully recovered, 1.2% partially recovered and the mortality was 14.9%. Mortality was higher in AKI associated with chronic medical conditions like cardiac failure, chronic liver disease and stroke. Conclusion: If treated early, AKI is mostly reversible. Regional differences in AKI should be studied extensively and local guidelines should be formulated by experts for prevention and early treatment, to improve the disease outcome.

A CLINICAL STUDY OF ETIOLOGY, OUTCOME AND PROGNOSTIC FACTORS OF ACUTE KIDNEY INJURY AMONG ICU ADMISSIONS IN RURAL TERTIARY CARE HOSPITAL

2021 ◽  
pp. 201-204
Author(s):  
Shashikantha Shashikantha ◽  
Sohil Sharda. ◽  
Bernice Robert ◽  
Gangurde Bhushan Daulatrao
Keyword(s):  
Acute Kidney Injury ◽  
Prognostic Factors ◽  
Renal Disease ◽  
Tertiary Care Hospital ◽  
Chronic Renal Disease ◽  
Tertiary Care ◽  
Ventilatory Support ◽  
Kidney Injury ◽  
Care Hospital ◽  
Study Results

INTRODUCTION: Acute kidney injury is a common occurrence in ICU admissions causing increased morbidity and mortality. Present study aimed to determine the causes and prognostic factors of acute kidney injury in intensive care unit. MATERIAL AND METHODS: This Hospital based Cross sectional Study was conducted at a tertiary care Hospital and Research Center, including 100 patients aged >18 years with Acute Kidney Injury admitted in ICU from the period of October 2018 to June 2020. Patients with chronic renal disease, previous renal transplantation, congenital renal disease were excluded from the study. RESULTS: Most of the patients (63%) were aged above 50 years. Diabetes was found in 55% and hypertension in 26% of AKI cases. Most common cause identied were sepsis, CLD, renal, CNS and CVD. Hypotension occurred in 48% patients, while oliguria occurred in 45% patients. Ventilatory support was required by 43% patients, while 31% patients required haemodialysis. Mortality rate in AKI was 51%. Mortality was signicantly associated with advanced age, presence of Diabetes, and RIFLE criteria. Spot urine <40 meq/L, hyperkalemia, serum creatinine >4 mg/dl, blood urea >100 mg/dl and acidosis were associated with higher mortality. CONCLUSION: Continuous monitoring parameters like Spot Fe Na, Serum Potasium and pH especially in patients at risk, like elderly patients with diabetes, those with sepsis, can help in early identication and appropiate management, thus reduce the incidence or severity of AKI.

P0643PROGNOSTIC IMPACT OF THROMBOCYTOPENIA IN THE RISK OF INTRAHOSPITAL BLEEDING IN ACUTE KIDNEY INJURY (AKI) PATIENTS

2020 ◽  
Vol 35 (Supplement_3) ◽  
Author(s):  
Zorica Dimitrijevic ◽  
Branka Mitic ◽  
Danijela Tasic ◽  
Goran Paunovic ◽  
Karolina Paunovic ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Platelet Count ◽  
Major Bleeding ◽  
Tertiary Care ◽  
Kidney Injury ◽  
Bleeding Risk ◽  
Prognostic Impact ◽  
Care Hospital ◽  
Ulcer Disease ◽  
Post Surgery

Abstract Background and Aims Low platelet count is a marker of adverse events in acute kidney injury (AKI) patients. Thrombocytopenia has often been reported as an indicator of underlying disease severity and worse patient outcomes; however, it’s role in the prediction of the risk of bleeding is not well defined. Our study aimed to assess the prognostic impact of admission thrombocytopenia in the risk of major bleeding in non-septic, non-post surgery AKI patients. Method This retrospective study enrolled patients with AKI hospitalized at tertiary care hospital during the three years. Admission thrombocytopenia was defined as a platelet count &lt; 150x103/mL. The primary endpoint was major bleeding, as defined by the International Society on Thrombosis and Haemostasis. Results Of 178 included patients (age 61.7±11.1 years; 68.3% males), 26 (14.6%) had thrombocytopenia. These patients had more comorbidities: cancer (19.4 vs 9.6%; p=0.05); previous ulcer disease (17.6 vs. 8.8; p=0.04) and bleeding history (7.4% vs. 2.0%;p=0.04). While in a hospital, there was a trend for the use of more blood transfusions (7.4% vs. 2.7%; p=0.03) and more concomitant medications (12.7 vs. 5.1; p&lt;0.05) in patients with thrombocytopenia. During a hospital stay (IQR: 7-29 days), 19 patients (10.7%) died), 22 (12.35%) had major bleeding, and 5 (2.8%) intracranial bleeding. After adjusting for age, presence of cancer, and use of oral anticoagulant medications, patients with thrombocytopenia had a higher risk of major bleeding (HR 3.34 95%CI: 1.57-7.26; p &lt; 0.001). Conclusion Thrombocytopenia is a predictor of major intrahospital bleeding in the non-septic, non-post surgery AKI patients. It should be regarded in bleeding risk estimation and therapeutic strategy decisions.

Comparing Machine Learning Algorithms for Predicting Acute Kidney Injury

2019 ◽  
Vol 85 (7) ◽  
pp. 725-729 ◽  
Author(s):  
Joshua Parreco ◽  
Hahn Soe-Lin ◽  
Jonathan J. Parks ◽  
Saskya Byerly ◽  
Matthew Chatoor ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Kidney Injury ◽  
Learning Algorithms ◽  
Vital Signs ◽  
Kidney Injury ◽  
Machine Learning Algorithms ◽  
Laboratory Measurements ◽  
Icu Patients ◽  
Laboratory Values ◽  
F Measure

Prior studies have used vital signs and laboratory measurements with conventional modeling techniques to predict acute kidney injury (AKI). The purpose of this study was to use the trend in vital signs and laboratory measurements with machine learning algorithms for predicting AKI in ICU patients. The eICU Collaborative Research Database was queried for five consecutive days of laboratory measurements per patient. Patients with AKI were identified and trends in vital signs and laboratory values were determined by calculating the slope of the least-squares-fit linear equation using three days for each value. Different machine learning classifiers (gradient boosted trees [GBT], logistic regression, and deep learning) were trained to predict AKI using the laboratory values, vital signs, and slopes. There were 151,098 ICU stays identified and the rate of AKI was 5.6 per cent. The best performing algorithm was GBT with an AUC of 0.834 ± 0.006 and an F-measure of 42.96 per cent ± 1.26 per cent. Logistic regression performed with an AUC of 0.827 ± 0.004 and an F-measure of 28.29 per cent ± 1.01 per cent. Deep learning performed with an AUC of 0.817 ± 0.005 and an F-measure of 42.89 per cent ± 0.91 per cent. The most important variable for GBT was the slope of the minimum creatinine (30.32%). This study identifies the best performing machine learning algorithms for predicting AKI using trends in laboratory values in ICU patients. Early identification of these patients using readily available data indicates that incorporating machine learning predictive models into electronic medical record systems is an inevitable requisite for improving patient outcomes.

scholarly journals Acute kidney injury requiring dialysis in obstetric patients: management and clinical outcome in case series of 30 patients in a tertiary care centre

2017 ◽  
Vol 6 (7) ◽  
pp. 2997
Author(s):  
Molina U. Patel ◽  
Yuvraj Jadeja ◽  
Niket Patel ◽  
Nayana Patel ◽  
Smruti Vaishnav ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Tertiary Care ◽  
Case Series ◽  
Kidney Injury ◽  
Therapeutic Interventions ◽  
Risk Category ◽  
Care Hospital ◽  
Tertiary Care Centre ◽  
Significant Difference ◽  
The Impact

Background: Acute Kidney Injury is a common medical problem affecting approximately 5% of all hospitalized and 30% of critically ill patients. The incidence in obstetric patients ranges from 1 in 2000 to 1 in 25000 pregnancies. In India till date, the impact of AKI on fetomaternal outcome and pertaining therapeutic interventions is only sparsely studied.Methods: It is a retrospective cross-sectional study. All obstetric patients with AKI on dialysis, admitted to Shree Krishna Hospital, a tertiary care hospital in Karamsad village in Gujarat from January 2013 to August 2015. Multivariate statistical analysis of clinical and laboratory parameters was performed using SPSS program to obtain the results.Results: The incidence of dialysis was 1.6%. HELLP syndrome and pre-eclampsia (80%) was found to be the most common etiology of AKI followed by Congestive cardiac failure (34.5%), hemorrhage and sepsis in 30% resp. All patients were admitted to ICU care. No significant difference was found between SAP II and SOFA monitoring system. Mechanical ventilation was done to support 53.3% and inotropic support was needed by 56.7% patients. According to the RIFLE criteria, majority of the patients fall under risk category followed by injury. 18% of the patients developed End Stage Renal Disease.Conclusions: In view of the multifaceted etiologies and complexity of management of AKI, a multi-disciplinary approach involving nephrologist, intensivists, obstetricians and neonatologists is extremely important.

scholarly journals Utilization of Deep Learning for Subphenotype Identification in Sepsis-Associated Acute Kidney Injury

2020 ◽  
Vol 15 (11) ◽  
pp. 1557-1565 ◽  
Author(s):  
Kumardeep Chaudhary ◽  
Akhil Vaid ◽  
Áine Duffy ◽  
Ishan Paranjpe ◽  
Suraj Jaladanki ◽  
...  
Keyword(s):  
Intensive Care ◽  
Deep Learning ◽  
Medical Information ◽  
Tertiary Care ◽  
Vital Signs ◽  
Kidney Injury ◽  
The United States ◽  
Care Hospital ◽  
Electronic Health Record Data ◽  
Electronic Health

Background and objectivesSepsis-associated AKI is a heterogeneous clinical entity. We aimed to agnostically identify sepsis-associated AKI subphenotypes using deep learning on routinely collected data in electronic health records.Design, setting, participants, & measurementsWe used the Medical Information Mart for Intensive Care III database, which consists of electronic health record data from intensive care units in a tertiary care hospital in the United States. We included patients ≥18 years with sepsis who developed AKI within 48 hours of intensive care unit admission. We then used deep learning to utilize all available vital signs, laboratory measurements, and comorbidities to identify subphenotypes. Outcomes were mortality 28 days after AKI and dialysis requirement.ResultsWe identified 4001 patients with sepsis-associated AKI. We utilized 2546 combined features for K-means clustering, identifying three subphenotypes. Subphenotype 1 had 1443 patients, and subphenotype 2 had 1898 patients, whereas subphenotype 3 had 660 patients. Subphenotype 1 had the lowest proportion of liver disease and lowest Simplified Acute Physiology Score II scores compared with subphenotypes 2 and 3. The proportions of patients with CKD were similar between subphenotypes 1 and 3 (15%) but highest in subphenotype 2 (21%). Subphenotype 1 had lower median bilirubin levels, aspartate aminotransferase, and alanine aminotransferase compared with subphenotypes 2 and 3. Patients in subphenotype 1 also had lower median lactate, lactate dehydrogenase, and white blood cell count than patients in subphenotypes 2 and 3. Subphenotype 1 also had lower creatinine and BUN than subphenotypes 2 and 3. Dialysis requirement was lowest in subphenotype 1 (4% versus 7% [subphenotype 2] versus 26% [subphenotype 3]). The mortality 28 days after AKI was lowest in subphenotype 1 (23% versus 35% [subphenotype 2] versus 49% [subphenotype 3]). After adjustment, the adjusted odds ratio for mortality for subphenotype 3, with subphenotype 1 as a reference, was 1.9 (95% confidence interval, 1.5 to 2.4).ConclusionsUtilizing routinely collected laboratory variables, vital signs, and comorbidities, we were able to identify three distinct subphenotypes of sepsis-associated AKI with differing outcomes.

Acute Kidney Injury in Neonatal Sepsis in a Tertiary Care Hospital

2020 ◽  
Vol 7 (34) ◽  
pp. 1730-1734
Author(s):  
Sreelekha Palle ◽  
Kavitha Shanigaram ◽  
Raghava Polanki
Keyword(s):  
Acute Kidney Injury ◽  
Neonatal Sepsis ◽  
Tertiary Care Hospital ◽  
Tertiary Care ◽  
Kidney Injury ◽  
Care Hospital

scholarly journals Haemodynamically Unstable Acute Kidney Injury Patients Treated with Sustained Low Efficiency Dialysis: Outcome in an ICU of Bangladesh

2017 ◽  
Vol 42 (1) ◽  
pp. 14-20
Author(s):  
Kaniz Fatema ◽  
Mohammad Omar Faruq
Keyword(s):  
Acute Kidney Injury ◽  
Renal Replacement Therapy ◽  
Replacement Therapy ◽  
De Novo ◽  
Tertiary Care ◽  
Kidney Injury ◽  
Cost Effective ◽  
Apache Ii Score ◽  
Care Hospital ◽  
Low Efficiency

Acute kidney injury (AKI) is a risk factor for increased mortality in critically ill patients. Sustained low efficiency dialysis (SLED) is a new approach in renal replacement therapy (RRT) and it combines the advantages of continuous renal replacement therapy (CRRT) and intermittent haemodialysis (HD). The study was aimed to evaluate the outcome of the hae-modynamically unstable patients with AKI in Bangladesh who were treated with SLED. So far this is the first reported study on SLED in intensive care unit (ICU) in Bangladesh. This quasi-experimental study was conducted in a 10-bed adult ICU of a tertiary care hospital in Bangladesh from June 2012 to May 2013. A total of 153 sessions of SLED were performed on 43 AKI patients. Mean age of the patients was 60.12 ± 15.57 years with male preponder-ance (67.4% were male). Mean APACHE II score was 26.88 ± 6.25. Fourteen patients (32.55%) had de novo AKI. Twenty nine patients (67.4%) had chronic kidney disease (CKD) with baseline mean serum creatinine 2.56 mg/dl, but did not require any RRT before admis-sion in ICU. After giving SLED, AKI of the study patients were completely resolved in 27.9%. Some forty two percent patients became dialysis dependant and 30.23% patients died. Patients who had AKI on CKD became dialysis dependant more often than the patients with de novo AKI (p <0.01). Mortality rate was significantly higher in patients who were on inotrope support (p= 0.017). Otherwise, there was no relation of 28 day mortality with age, prior renal function and mechanical ventilator requirement (p>0.05). Thus, SLED is an excellent renal replacement therapy for the haemodynamically unstable AKI patients of ICU. It is also cost-effective compared to CRRT.

Sign in / Sign up

Export Citation Format

Share Document