Forecasting MAP Patterns to Anticipate Hypotension in ICUs

Overview

This pilot study evaluates a clinical decision support system (CDSS) designed to predict hypotension in ICU patients undergoing continuous vasopressor therapy. The CDSS demonstrated a 100% detection rate for hypotension clusters and predicted 91.7% of vasopressor dose increases.

Background

Hypotension in ICU patients is a significant concern, often leading to adverse outcomes. Previous analyses have shown that patients frequently experience cumulative hours of hypotension, typically defined as a mean arterial pressure (MAP) below 65 mmHg. Early detection and intervention are crucial to mitigate the risks associated with prolonged hypotension.

Data Highlights

Key Findings

• The CDSS detected every hypotension-cluster (100%) during early flickering hypotension.
• It predicted 91.7% of vasopressor dose increases.
• Before hypotension began, the CDSS predicted 44% of hypotension-clusters.
• The median cumulative hypotension duration associated with positive predictions was 26.4 min.
• Hypotension occurred for a median of 119.7 min per day in the study population.

Clinical Implications

The findings indicate that the CDSS could enhance early detection of hypotension in ICU settings.

Conclusion

The pilot evaluation of the CDSS indicates its effectiveness in predicting hypotension and vasopressor needs.

Related Resources & Content

1. Faulkner et al., ClinicalTrials.gov, 2020 -- Forecasting MAP Patterns to Anticipate Hypotension
2. Frontiers in Medicine — Machine learning model for predicting hypotension following continuous renal replacement therapy initiation in end-stage kidney disease patients: A SHAP-interpretable approach
3. Intensive Care Medicine — The Impact of ICU Hypotension on Mortality and Morbidity Among Septic Patients During Hospitalization
4. Frontiers in Anesthesiology — Impact of hypotension prediction index on mean arterial pressure and stroke volume index during pancreatic surgery: a retrospective pre-post observational pilot study
5. Frontiers in Digital Health — Intra-Session recurrence in intradialytic hypotension prediction: evaluation implications and recurrence-aware modeling
6. Machine learning model for predicting hypotension following continuous renal replacement therapy initiation in end-stage kidney disease patients
7. The Impact of ICU Hypotension on Mortality and Morbidity Among Septic Patients During Hospitalization
8. Impact of hypotension prediction index on mean arterial pressure and stroke volume index during pancreatic surgery
9. Surviving Sepsis Campaign Adult Guidelines
10. VA ICME-D-25-00526V1.2
11. High versus Low Blood-Pressure Target in Patients with Septic Shock | New England Journal of Medicine
12. Effect of Reduced Exposure to Vasopressors on 90-Day Mortality in Older Critically Ill Patients With Vasodilatory Hypotension: A Randomized Clinical Trial - PMC
13. Efficacy of targeting high mean arterial pressure for older patients with septic shock (OPTPRESS): a multicentre, pragmatic, open-label, randomised controlled trial - PMC
14. Blood Pressure Targets for Adults with Vasodilatory Shock — An Individual Patient Data Meta-Analysis | NEJM Evidence
15. Mortality effect of higher versus lower blood pressure targets in vasodilatory shock: an updated systematic review and meta-analysis | Critical Care | Springer Nature Link
16. The relationship between ICU hypotension and in-hospital mortality and morbidity in septic patients - PMC
17. Hypotension During Vasopressor Infusion Occurs in Predictable Clusters: A Multicenter Analysis.
18. Early prediction of vasopressor initiation in ICU sepsis patients using an interpretable EHR-based ML model | BMC Medical Informatics and Decision Making | Springer Nature Link
19. Effect of a Machine Learning-Derived Early Warning Tool With Treatment Protocol on Hypotension During Cardiac Surgery and ICU Stay: The Hypotension Prediction 2 (HYPE-2) Randomized Clinical Trial.
20. Prediction of hypotension events with physiologic vital sign signatures in the intensive care unit - PMC