RESUMO
Right ventricular (RV) dysfunction is a major cause of morbidity and mortality in intensive care and cardiac surgery. Early detection of RV dysfunction may be facilitated by continuous monitoring of RV waveform obtained from a pulmonary artery catheter. The objective is to evaluate the extent to which RV pressure monitoring can detect changes in RV systolic performance assess by RV end-systolic elastance (Ees) following the development of an acute RV ischemic in a porcine model. HYPOTHESIS: RV pressure monitoring can detect changes in RV systolic performance assess by RV Ees following the development of an acute RV ischemic model. METHODS AND MODELS: Acute ischemic RV dysfunction was induced by progressive embolization of microsphere in the right coronary artery to mimic RV dysfunction clinically experienced during cardiopulmonary bypass separation caused by air microemboli. RV hemodynamic performance was assessed using RV pressure waveform-derived parameters and RV Ees obtained using a conductance catheter during inferior vena cava occlusions. RESULTS: Acute ischemia resulted in a significant reduction in RV Ees from 0.26 mm Hg/mL (interquartile range, 0.16-0.32 mm Hg/mL) to 0.14 mm Hg/mL (0.11-0.19 mm Hg/mL; p < 0.010), cardiac output from 6.3 L/min (5.7-7 L/min) to 4.5 (3.9-5.2 L/min; p = 0.007), mean systemic arterial pressure from 72 mm Hg (66-74 mm Hg) to 51 mm Hg (46-56 mm Hg; p < 0.001), and mixed venous oxygen saturation from 65% (57-72%) to 41% (35-45%; p < 0.001). Linear mixed-effect model analysis was used to assess the relationship between Ees and RV pressure-derived parameters. The reduction in RV Ees best correlated with a reduction in RV maximum first derivative of pressure during isovolumetric contraction (dP/dtmax) and single-beat RV Ees. Adjusting RV dP/dtmax for heart rate resulted in an improved surrogate of RV Ees. INTERPRETATION AND CONCLUSIONS: Stepwise decreases in RV Ees during acute ischemic RV dysfunction were accurately tracked by RV dP/dtmax derived from the RV pressure waveform.
RESUMO
BACKGROUND: Early deterioration indicators have the potential to alert hospital care staff in advance of adverse events, such as patients requiring an increased level of care, or the need for rapid response teams to be called. Our work focuses on the problem of predicting the transfer of pediatric patients from the general ward of a hospital to the pediatric intensive care unit. OBJECTIVES: The development of a data-driven pediatric early deterioration indicator for use by clinicians with the purpose of predicting encounters where transfer from the general ward to the PICU is likely. METHODS: Using data collected over 5.5 years from the electronic health records of two medical facilities, we develop machine learning classifiers based on adaptive boosting and gradient tree boosting. We further combine these learned classifiers into an ensemble model and compare its performance to a modified pediatric early warning score (PEWS) baseline that relies on expert defined guidelines. To gauge model generalizability, we perform an inter-facility evaluation where we train our algorithm on data from one facility and perform evaluation on a hidden test dataset from a separate facility. RESULTS: We show that improvements are witnessed over the modified PEWS baseline in accuracy (0.77 vs. 0.69), sensitivity (0.80 vs. 0.68), specificity (0.74 vs. 0.70) and AUROC (0.85 vs. 0.73). CONCLUSIONS: Data-driven, machine learning algorithms can improve PICU transfer prediction accuracy compared to expertly defined systems, such as a modified PEWS, but care must be taken in the training of such approaches to avoid inadvertently introducing bias into the outcomes of these systems.