Background and aims:Multidrug-resistant organisms (MDROs) are a growing threat in critical care settings, causing prolonged hospitalization, increased costs, and mortality. Irrational antibiotic use, invasive procedures, and limited diagnostic times contribute to the prevalence of MDROs. Machine learning (ML) offers a promising approach for early risk identification. This study aimed to develop and pilot validate an ML-based tool to predict MDRO infection risk in Critical Care Unit (CCU) patients using baseline clinical data. Patients and methods:Retrospective data from 323 patients admitted to the CCU of a tertiary care hospital in Kerala, India, were used to develop and evaluate five ML models using Python (version 3.11). Predictor variables included demographics, comorbidities, and initial medical device usage. A pilot prospective validation was subsequently conducted on 49 new CCU admissions to assess real-world predictive performance within 48 hours of admission. Results:Among 323 patients, 76 developed MDRO infections (23.5%). Major risk factors identified included prolonged hospitalization and invasive device use. The random forest model demonstrated a training accuracy of 96.0% and a test-set accuracy of 75.3%. In the pilot prospective validation, the tool yielded a sensitivity of 81.8%, a specificity of 100%, and an overall accuracy of 95.9%. Conclusions:This study demonstrates that the ML-based tool can effectively stratify MDRO risk in a CCU setting. While the performance discrepancy between training and testing indicates overfitting, the high specificity and sensitivity observed in the prospective pilot phase support its potential as a bedside aid for early antimicrobial stewardship (AMS). Clinical significance:The ML tool enables early risk stratification, allowing for timely infection control and efficient resource utilization in constrained clinical settings.
Background and aims:We studied changes in nutritional status from baseline, as measured by serial skinfold thickness measurements and their association with patient factors, disease factors, and biochemical factors, as well as time to the start of enteral feeds. Patients and methods:We enrolled 50 consecutive children aged 2-5 years from the pediatric intensive care unit (PICU) of a public sector hospital in India between 14 February, 2024 and 31 August, 2024. Baseline demographic and clinical data were recorded. Serial anthropometric parameters were measured for all children at the time of admission (+4 hours), day 3, day 7, and day 14 or at hospital discharge, if earlier. We used the World Health Organization (WHO) growth charts to interpret the anthropometric data. Results:Of the 50 children enrolled (mean age 3.17 ± 1.22 years; 46% females), 58% were underweight, 52% had wasting, and 52% had short stature at admission. Only 10% children (n = 5) had triceps skinfold thickness for age <-2 standard deviation (SD) at baseline. 32 (64%) of the children demonstrated a reduction in triceps skinfold thickness, 72% (n = 36) experienced weight loss, and 58% (n = 29) showed a decrease in mid-upper arm circumference (MUAC). However, we did not find a statistically significant association of change in triceps skinfold thickness with age, severity of illness, outcome, duration of stay, and early start of enteral feeds. Conclusion:This study shows the feasibility of skinfold calipers as a serial monitoring tool in the PICU. Nutritional status worsens during the initial 2 weeks of PICU stay, despite early enteral feeding policies.
Aims and background:Paraquat (PQ) poisoning has a mortality rate of 50-90% with no specific antidote. Hemoperfusion (HP) is used for PQ removal, but evidence for its efficacy is limited, and optimal strategies for guiding HP remain poorly defined. The study aims to describe a urine sodium dithionite test (UDT)-guided HP protocol and evaluate clinical outcomes in patients with acute PQ poisoning. Patients and methods:A retrospective observational study of 27 consecutive patients with acute PQ poisoning admitted to a tertiary intensive care unit (ICU) in South India was conducted. Serial UDT was used to guide HP initiation, continuation, and cessation. Clinical data, treatment details, and outcomes were analyzed. Survivors and non-survivors were compared. Results:Mean age was 30.5 ± 11.2 years (74.1% male). Fourteen patients (51.9%) survived. Ingestion-to-door (I-to-D) time ≤4 h was present in 71.4% of survivors vs 20% of non-survivors. Urine sodium dithionite test negative conversion within 10 h occurred in 77.8% of survivors vs 16.7% of non-survivors. Survivors received more HP sessions (mean: 3.3 vs 2.6). Acute kidney injury (AKI) occurred in 80% of non-survivors vs 14.3% of survivors. Conclusion:Early presentation combined with a UDT-guided HP protocol was associated with favorable outcomes. Serial UDT provides a practical, low-cost tool for guiding extracorporeal treatment (ECTR) in resource-limited settings. Clinical significance:The UDT-guided HP protocol offers intensivists a bedside decision-making tool that replaces costly plasma PQ assays, enabling evidence-based ECTR decisions in Indian ICU settings where quantitative assays are unavailable.
How to cite this article: Vijayasimha M, Srikanth M, Rao K, Mishra P, Shweta, Juneja A. From Tele-ICU "Alerts" to Tele-ICU "Assurance": Making Hemodynamic Surveillance Decision-grade and Globally Transferable. Indian J Crit Care Med 2026;30(4):344.
