A Multivariable Predictive Model Utilizing Heart Rate Variability to Predict Future Deterioration in Emergency Department Patients with Sepsis
CCCF ePoster library. Barnaby D. Nov 2, 2016; 150977; 96
Dr. Doug Barnaby
Dr. Doug Barnaby
Login now to access Regular content available to all registered users.

You may also access this content "anytime, anywhere" with the Free MULTILEARNING App for iOS and Android
Abstract
Rate & Comment (0)
#96

Topic: Retrospective or Prospective Cohort Study

A Multivariable Predictive Model Utilizing Heart Rate Variability to Predict Future Deterioration in Emergency Department Patients with Sepsis


Barnaby, Douglas P., MD, MS1; Fernando, Shannon M., MD, MSc2; Herry, Christophe L., PhD3; Scales, Nathan PhD3; Bijur, Polly E., PhD1; Seely, Andrew J. E., MD, PhD3,4,5, Gallagher, E. John, MD1
1
Department of Emergency Medicine, Albert Einstein College of Medicine, Bronx, NY, USA; 2Department of Emergency Medicine, University of Ottawa, Ottawa, ON, Canada; 3Ottawa Hospital Research Institute, The Ottawa Hospital, Ottawa, ON, Canada; 4Division of Thoracic Surgery, Department of Surgery, University of Ottawa, Ottawa, ON, Canada; 5Division of Critical Care, Department of Medicine, University of Ottawa, Ottawa, ON, Canada

Conflicts of Interest: AJES holds patents related to multiorgan variability analysis, and has shares in Therapeutic Monitoring Systems Inc, a company whose mission is to help deliver variability-directed clinical decision support products to the bedside to improve care. CLH is a patent holder related to waveform quality assessment necessary for variability analysis.



Abstract:

Background: Sepsis remains an enormously costly and leading cause of mortality worldwide. A majority of patients with sepsis are initially assessed in the Emergency Department (ED), where early identification of patients at increased risk for future deterioration is essential to optimize individual patient care, manage resources efficiently, and potentially reduce incidence of deterioration. Heart Rate Variability (HRV) has been shown to predict clinical deterioration. However, its added value to clinical and laboratory measures in sepsis has not been studied.
 
Objectives: In a prospective observational cohort study of ED patients diagnosed with sepsis, as defined by the 1992 SCCM/ACCP clinical criteria, we aim to evaluate and compare the predictive capacity of laboratory values, clinical parameters and HRV metrics (see Table 1) to predict clinical deterioration within 72 hours, as determined by specific, pre-defined end-points.
 
Methods: Patients meeting inclusion criteria underwent bedside ECG monitoring for 15 minutes (Lifewatch Services Inc., Rosemont, IL) within 2 hours of ED arrival. ECG recordings were analyzed off-line to obtain indices of HRV using Continuous Individualized Multiorgan Variability Analysis (CIMVA™) on 1000 beat R-peak-to-R-peak time-series. Patients were followed for 72 hours to identify those meeting one or more of five defined study end-points indicative of clinical deterioration: 1) Intubation; 2) Non-invasive ventilation (>1hr); 3) Vasopressor/inotrope support (>1hr); 4) ICU admission (LOS > 24 hrs); or 5) Death. Patients meeting any of these end-points within 1 hour of arrival were excluded. Predictive modeling was performed using an ensemble of logistic regression models using training, validation and test sets, to derive the predictive performance of HRV, clinical, and laboratory indices, individually and in combinations.
 
Results: 1174 patients presenting to either of two urban high-volume EDs within a single healthcare system were enrolled, of whom 89 (7.6%) met an endpoint within 72 hours (mean 12.8 ± 10.8(SD) hours after presentation). Results of predictive modeling are depicted in Table 1. Overall, the highest receiver operating characteristic (ROC) area was derived from a combination of HRV, clinical and lab values (AUC = 0.84, 95%CI: 0.83-0.85). The values included clinical (age, blood pressure, heart rate, respiratory rate, temperature), laboratory (lactate, creatinine, INR) and 5 HRV metrics. Using this combined predictive model, the population was divided into sextiles on the basis of their outcome score (Figure 1). The sextile with the highest outcome score had a 3.7-fold increased likelihood of meeting one of the study end-points. The overall performance of the combined HRV+Clinical+Lab predictive model was greater than the combination Clinical+Lab values (AUC = 0.80, 95% CI: 0.79-0.81; Table 1), which was equivalent to the ROC area for HRV alone (AUC = 0.80, 95% CI: 0.79-0.81).
 
Conclusions: Our analysis produced a HRV+Clinical+Lab predictive model that may help identify risk of future deterioration in ED patients presenting with a diagnosis of sepsis, with HRV demonstrating added benefit. Further evaluation of this predictive performance in multicenter studies is required along with studying methods of clinical integration.


References:

No references.



Document: File 1 File 2
    This eLearning portal is powered by:
    This eLearning portal is powered by MULTIEPORTAL
Anonymous User Privacy Preferences

Strictly Necessary Cookies (Always Active)

MULTILEARNING platforms and tools hereinafter referred as “MLG SOFTWARE” are provided to you as pure educational platforms/services requiring cookies to operate. In the case of the MLG SOFTWARE, cookies are essential for the Platform to function properly for the provision of education. If these cookies are disabled, a large subset of the functionality provided by the Platform will either be unavailable or cease to work as expected. The MLG SOFTWARE do not capture non-essential activities such as menu items and listings you click on or pages viewed.


Performance Cookies

Performance cookies are used to analyse how visitors use a website in order to provide a better user experience.


Save Settings