Non-invasive biomarkers for detecting progression toward hypovolemic cardiovascular instability in a lower body negative pressure model.

Occult hemorrhages after trauma can be present insidiously, and if not detected early enough can result in patient death. This study evaluated a hemorrhage model on 18 human subjects, comparing the performance of traditional vital signs to multiple off-the-shelf non-invasive biomarkers. A validated lower body negative pressure (LBNP) model was used to induce progression towards hypovolemic cardiovascular instability. Traditional vital signs included mean arterial pressure (MAP), electrocardiography (ECG), plethysmography (Pleth), and the test systems utilized electrical impedance via commercial electrical impedance tomography (EIT) and multifrequency electrical impedance spectroscopy (EIS) devices. Absolute and relative metrics were used to evaluate the performance in addition to machine learning-based modeling. Relative EIT-based metrics measured on the thorax outperformed vital sign metrics (MAP, ECG, and Pleth) achieving an area-under-the-curve (AUC) of 0.99 (CI 0.95-1.00, 100% sensitivity, 87.5% specificity) at the smallest LBNP change (0-15 mmHg). The best vital sign metric (MAP) at this LBNP change yielded an AUC of 0.6 (CI 0.38-0.79, 100% sensitivity, 25% specificity). Out-of-sample predictive performance from machine learning models were strong, especially when combining signals from multiple technologies simultaneously. EIT, alone or in machine learning-based combination, appears promising as a technology for early detection of progression toward hemodynamic instability.

Prediction of Occult Hemorrhage in the Lower Body Negative Pressure Model: Initial Validation of Machine Learning Approaches.

INTRODUCTION: Detection of occult hemorrhage (OH) before progression to clinically apparent changes in vital signs remains an important clinical problem in managing trauma patients. The resource-intensiveness associated with continuous clinical patient monitoring and rescue from frank shock makes accurate early detection and prediction with noninvasive measurement technology a desirable innovation. Despite significant efforts directed toward the development of innovative noninvasive diagnostics, the implementation and performance of the newest bedside technologies remain inadequate. This poor performance may reflect the limitations of univariate systems based on one sensor in one anatomic location. It is possible that when signals are measured with multiple modalities in multiple locations, the resulting multivariate anatomic and temporal patterns of measured signals may provide additional discriminative power over single technology univariate measurements. We evaluated the potential superiority of multivariate methods over univariate methods. Additionally, we utilized machine learning-based models to compare the performance of noninvasive-only to noninvasive-plus-invasive measurements in predicting the onset of OH. MATERIALS AND METHODS: We applied machine learning methods to preexisting datasets derived using the lower body negative pressure human model of simulated hemorrhage. Employing multivariate measured physiological signals, we investigated the extent to which machine learning methods can effectively predict the onset of OH. In particular, we applied 2 ensemble learning methods, namely, random forest and gradient boosting. RESULTS: Analysis of precision, recall, and area under the receiver operating characteristic curve showed a superior performance of multivariate approach to that of the univariate ones. In addition, when using both invasive and noninvasive features, random forest classifier had a recall 95% confidence interval (CI) of 0.81 to 0.86 with a precision 95% CI of 0.65 to 0.72. Interestingly, when only noninvasive features were employed, the results worsened only slightly to a recall 95% CI of 0.80 to 0.85 and a precision 95% CI of 0.61 to 0.73. CONCLUSIONS: Multivariate ensemble machine learning-based approaches for the prediction of hemodynamic instability appear to hold promise for the development of effective solutions. In the lower body negative pressure multivariate hemorrhage model, predictions based only on noninvasive measurements performed comparably to those using both invasive and noninvasive measurements.

Detection of subclinical hemorrhage using electrical impedance: a porcine study.

Objective.Analyze the performance of electrical impedance tomography (EIT) in an innovative porcine model of subclinical hemorrhage and investigate associations between EIT and hemodynamic trends.Approach. Twenty-five swine were bled at slow rates to create an extended period of subclinical hemorrhage during which the animal's heart rate (HR) and blood pressure (BP) remained stable from before hemodynamic deterioration, where stable was defined as <15% decrease in BP and <20% increase in HR-i.e.hemorrhages were hidden from standard vital signs of HR and BP. Continuous vital signs, photo-plethysmography, and continuous non-invasive EIT data were recorded and analyzed with the objective of developing an improved means of detecting subclinical hemorrhage-ideally as early as possible.Main results. Best area-under-the-curve (AUC) values from comparing bleed to no-bleed epochs were 0.96 at a 80 ml bleed (∼15.4 min) using an EIT-data-based metric and 0.79 at a 120 ml bleed (∼23.1 min) from invasively measured BP-i.e.the EIT-data-based metric achieved higher AUCs at earlier points compared to standard clinical metrics without requiring image reconstructions.Significance.In this clinically relevant porcine model of subclinical hemorrhage, EIT appears to be superior to standard clinical metrics in early detection of hemorrhage.

Another head's up on head's up.

A commentary on D. K. Lee et al and "head's-up" CPR.

Epinephrine plus chest compressions is superior to epinephrine alone in a hypoxia-induced porcine model of pseudo-pulseless electrical activity.

AIM: Pseudo-pulseless electrical activity (pseudo-PEA) is a global hypotensive ischemic state with retained coordinated myocardial contractile activity and an organized ECG with no clinically detectable pulses. The role of standard external chest compressions (CPR) and its associated intrinsic hemodynamics remains unclear in the setting of pseudo-PEA. We undertook an experimental trial to compare epinephrine alone versus epinephrine with CPR in the treatment of pseudo-PEA. METHODS: Using a porcine model of hypoxic pseudo-PEA, we randomized 12 Yorkshire male swine to resuscitation with epinephrine only (control) (0.0015 mg/kg) versus epinephrine plus standard CPR (intervention). Animals who achieved return of spontaneous circulation (ROSC) were stabilized, fully recovered to hemodynamic and respiratory baseline, and rearrested up to 6 times. Primary outcome was ROSC defined as a sustained systolic blood pressure (SBP) of 60 mmHg for 2 min. Secondary outcomes included time to ROSC, coronary perfusion pressure (CoPP), and end-tidal carbon dioxide (ETCO2). RESULTS: Among 47 events of pseudo-PEA in 12 animals, we observed significantly higher proportion of ROSC when treatment included CPR (14/21 - 67%) compared to epinephrine alone (4/26 - 15%) (p = 0.0007). CoPP, aortic pressures and ETCO2 were significantly higher, and right atrial pressures were lower in the intervention group. CONCLUSIONS: In a swine model of hypoxia-induced pseudo-PEA, epinephrine plus CPR was associated with improved intra-arrest hemodynamics and higher probability of ROSC. Thus, epinephrine plus CPR may be superior to epinephrine alone in the treatment of patients with pseudo-PEA.

Quantitative characterization of left ventricular function during pulseless electrical activity using echocardiography during out-of-hospital cardiac arrest.

BACKGROUND: Several prospective studies have demonstrated that the echocardiographic detection of any myocardial activity during PEA is strongly associated with higher rates of return of spontaneous circulation (ROSC). We hypothesized that PEA represents a spectrum of disease in which not only the presence of myocardial activity, but more specifically that the degree of left ventricular (LV) function would be a predictor of outcomes. The purpose of this study was to retrospectively assess the association between LV function and outcomes in patients with OHCA. MATERIALS AND METHODS: Using prospectively obtained data from an observational cohort of patients receiving focused echocardiography during cardiopulmonary resuscitation (CPR) in the Emergency Department (ED) setting, we analyzed 312 consecutive subjects with available echocardiography images with initial rhythm of PEA. We used left ventricular systolic fractional shortening (LVFS), a unidimensional echocardiographic parameter to perform the quantification of LV function during PEA. Regression analyses were performed independently to evaluate for relationships between LVFS and a primary outcome of ROSC and secondary outcome of survival to hospital admission. We analyzed LVFS both as a continuous variable and as a categorial variable using the quartiles and the median to perform multiple different comparisons and to illustrate the relationship of LVFS and outcomes of interest. We performed survival analysis using Cox proportional hazards model to evaluate the hazard corresponding to length of resuscitation. RESULTS: We found a positive association between LVFS and the primary outcome of ROSC (OR 1.04, 95%CI 1.01-1.08), but not with the secondary outcome of survival to hospital admission (OR 1.02, 95%CI 0.96-1.08). Given that the relationship was not linear and that we observed a threshold effect in the relationship between LVFS and outcomes, we performed an analysis using quartiles of LVFS. The predicted probability of ROSC was 75% for LVFS between 23.4-96% (fourth quartile) compared to 47% for LVFS between 0-4.7% (first quartile). The hazard of not achieving ROSC was significantly greater for subjects with LVFS below the median (13.1%) compared to the subgroup with LVFS greater than 13.1% (p < 0.05), with the separation of the survival curves occurring at approximately 40 min of resuscitation duration. CONCLUSIONS: Left ventricular function measured by LVFS is positively correlated with higher probability of ROSC and may be associated with higher chances of survival in patients with PEA arrest.

Early Detection of Hypotension Using a Multivariate Machine Learning Approach.

INTRODUCTION: The ability to accurately detect hypotension in trauma patients at the earliest possible time is important in improving trauma outcomes. The earlier an accurate detection can be made, the more time is available to take corrective action. Currently, there is limited research on combining multiple physiological signals for an early detection of hemorrhagic shock. We studied the viability of early detection of hypotension based on multiple physiologic signals and machine learning methods. We explored proof of concept with a small (5 minutes) prediction window for application of machine learning tools and multiple physiologic signals to detecting hypotension. MATERIALS AND METHODS: Multivariate physiological signals from a preexisting dataset generated by an experimental hemorrhage model were employed. These experiments were conducted previously by another research group and the data made available publicly through a web portal. This dataset is among the few publicly available which incorporate measurement of multiple physiological signals from large animals during experimental hemorrhage. The data included two hemorrhage studies involving eight sheep. Supervised machine learning experiments were conducted in order to develop deep learning (viz., long short-term memory or LSTM), ensemble learning (viz., random forest), and classical learning (viz., support vector machine or SVM) models for the identification of physiological signals that can detect whether or not overall blood loss exceeds a predefined threshold 5 minutes ahead of time. To evaluate the performance of the machine learning technologies, 3-fold cross-validation was conducted and precision (also called positive predictive value) and recall (also called sensitivity) values were compared. As a first step in this development process, 5 minutes prediction windows were utilized. RESULTS: The results showed that SVM and random forest outperform LSTM neural networks, likely because LSTM tends to overfit the data on small sized datasets. Random forest has the highest recall (84%) with 56% precision while SVM has 62% recall with 82% precision. Upon analyzing the feature importance, it was observed that electrocardiogram has the highest significance while arterial blood pressure has the least importance among all other signals. CONCLUSION: In this research, we explored the viability of early detection of hypotension based on multiple signals in a preexisting animal hemorrhage dataset. The results show that a multivariate approach might be more effective than univariate approaches for this detection task.

Predicting Future Occurrence of Acute Hypotensive Episodes Using Noninvasive and Invasive Features.

INTRODUCTION: Early prediction of the acute hypotensive episode (AHE) in critically ill patients has the potential to improve outcomes. In this study, we apply different machine learning algorithms to the MIMIC III Physionet dataset, containing more than 60,000 real-world intensive care unit records, to test commonly used machine learning technologies and compare their performances. MATERIALS AND METHODS: Five classification methods including K-nearest neighbor, logistic regression, support vector machine, random forest, and a deep learning method called long short-term memory are applied to predict an AHE 30 minutes in advance. An analysis comparing model performance when including versus excluding invasive features was conducted. To further study the pattern of the underlying mean arterial pressure (MAP), we apply a regression method to predict the continuous MAP values using linear regression over the next 60 minutes. RESULTS: Support vector machine yields the best performance in terms of recall (84%). Including the invasive features in the classification improves the performance significantly with both recall and precision increasing by more than 20 percentage points. We were able to predict the MAP with a root mean square error (a frequently used measure of the differences between the predicted values and the observed values) of 10 mmHg 60 minutes in the future. After converting continuous MAP predictions into AHE binary predictions, we achieve a 91% recall and 68% precision. In addition to predicting AHE, the MAP predictions provide clinically useful information regarding the timing and severity of the AHE occurrence. CONCLUSION: We were able to predict AHE with precision and recall above 80% 30 minutes in advance with the large real-world dataset. The prediction of regression model can provide a more fine-grained, interpretable signal to practitioners. Model performance is improved by the inclusion of invasive features in predicting AHE, when compared to predicting the AHE based on only the available, restricted set of noninvasive technologies. This demonstrates the importance of exploring more noninvasive technologies for AHE prediction.

Intravenous calcium as a pressor in a swine model of hypoxic pseudo-pulseless electrical mechanical activity-a preliminary report.

BACKGROUND: Pseudo-pulseless electrical activity (pseudo-PEA) is a lifeless form of profound cardiac shock characterized by measurable cardiac mechanical activity without clinically detectable pulses. Pseudo-PEA may constitute up to 40% of reported cases of cardiac arrest. Resuscitation from pseudo-PEA is often associated with hypotension refractory to catecholamine pressors. We hypothesized that this post-resuscitation state may be associated with hypocalcemic hypotension responsive to intravenous calcium. METHODS: Using pre-existing data from our hypoxic swine pseudo-PEA model, we measured blood pressure, hemodynamics, and electrolytes. Physiological data were analyzed on a heartbeat by heartbeat basis. The midpoint of the calcium response was defined using change of curvature feature detection. Hemodynamic parameters were shifted such that the value at the midpoint was equal to zero. RESULTS: In 9 animals with refractory hypotension, we administered 37 boluses of intravenous calcium in the dosage range of 5-20 mg. Comparisons were made between the average values in the time period 40-37 s before the midpoint and 35-40 s after the midpoint. Of the 37 administered boluses, 34 manifested a change in the blood pressure, with mean aortic pressure, systolic and diastolic pressures all increasing post bolus administration. CONCLUSIONS: Administration of intravenous calcium may be associated with a pressor-like response in refractory hypotension after resuscitation from pseudo-PEA. Relative ionized hypocalcemia may cause hypotension after resuscitation from pseudo-PEA. Therapy with intravenous calcium should be further investigated in this setting.

Addition of a biomarker panel to a clinical score to identify patients at low risk for appendicitis.

BACKGROUND: The diagnosis of pediatric acute appendicitis can be difficult. Although scoring systems such as the Pediatric Appendicitis Score (PAS) are helpful, they lack adequate sensitivity and specificity as standalone diagnostics. When used for risk stratification, they often result in large percentages of moderate-risk patients requiring further diagnostic evaluation. METHODS: We applied a biomarker panel (the APPY1 Test) that has high sensitivity and negative predictive value (NPV) to patients with PAS in the moderate-risk range (3-7) and reclassified those patients with a negative result to the low-risk group. We compared the specificity, sensitivity, and NPV of the original and reclassified low-risk groups at several different PAS low-risk cutoffs. RESULTS: The application of a negative biomarker panel to a group of patients with a moderate risk for appendicitis (PAS, 3-7) resulted in 4 times more patients (586 vs 145) being safely classified as low risk. Reclassification increased the overall specificity or the proportion of patients without appendicitis who were correctly identified as low risk, from 10.3% to 42.0%. The high NPV (97.2%) in the original group was preserved (97.6%) in the reclassified low-risk group, as was the sensitivity (original 99.1% vs reclassified 96.9%). CONCLUSION: The addition of negative biomarker test results to patients with a moderate risk of appendicitis based on the PAS can safely reclassify many to a low-risk group. This may allow clinicians to provide more conservative management in children with suspected appendicitis and decrease unnecessary resource utilization.

Instrument to detect syncope and the onset of shock.

Currently the diagnosis of hemorrhagic shock is essentially clinical, relying on the expertise of nurses and doctors. One of the first measurable physiological changes that marks the onset of hemorrhagic shock is a decrease in capillary blood flow. Diffuse correlation spectroscopy (DCS) quantifies this decrease. DCS collects and analyzes multiply scattered, coherent, near infrared light to assess relative blood flow. This work presents a preliminary study using a DCS instrument with human subjects undergoing a lower body negative pressure (LBNP) protocol. This work builds on previous successful DCS instrumentation development and we believe it represents progress toward understanding how DCS can be used in a clinical setting.

A disposable, flexible skin patch for clinical optical perfusion monitoring at multiple depths.

Stable, relative localization of source and detection fibers is necessary for clinical implementation of quantitative optical perfusion monitoring methods such as diffuse correlation spectroscopy (DCS) and diffuse reflectance spectroscopy (DRS). A flexible and compact device design is presented as a platform for simultaneous monitoring of perfusion at a range of depths, enabled by precise location of optical fibers in a robust and secure adhesive patch. We will discuss preliminary data collected on human subjects in a lower body negative pressure model for hypovolemic shock. These data indicate that this method facilitates simple and stable simultaneous monitoring of perfusion at multiple depths and within multiple physiological compartments.

Coronary perfusion pressure during external chest compression in pseudo-EMD, comparison of systolic versus diastolic synchronization.

BACKGROUND: The fraction of cardiac arrest patients presenting with pulseless electrical activity is increasing, and it is likely that many of these patients have pseudo-electromechanical dissociation (P-EMD), a state in which there is residual cardiac contraction without a palpable pulse. The efficacy of cardiopulmonary resuscitation (CPR) with external chest compression synchronized with the P-EMD cardiac systole and diastole has not been fully evaluated. HYPOTHESIS: During external chest compression in P-EMD, the coronary perfusion pressure (CPP) will be greater with systolic synchronization compared with diastolic phase synchronization. METHODS: A porcine model of P-EMD induced by progressive hypoxia with peak aortic pressures targeted to 50 mmHg was used. CPR chest compressions were performed by either load distributing band or vest devices. Paired 10s intervals of systolic and diastolic synchronization were performed randomly during P-EMD, and aortic, right atrial and CPP were compared. RESULTS: Stable P-EMD was achieved in 8 animals, with 2.6±0.5 matched synchronization pairs per animal. Systolic synchronization was association with increases in relaxation phase aortic pressure (41.7±8.9 mmHg vs. 36.9±8.2 mmHg), and coronary perfusion pressure (37.6±11.7 mmHg vs. 30.2±9.6 mmHg). Diastolic synchronization was associated with an increased right atrial pressure (6.7±4.1 mmHg vs. 4.1±5.7 mmHg). CONCLUSION: During P-EMD, synchronization of external chest compression with residual cardiac systole was associated with higher CPP compared to synchronization with diastole.

Emergency department initiation of cardiopulmonary bypass: a case report and review of the literature.

BACKGROUND: Out-of-hospital cardiac arrest carries a dismal prognosis. Percutaneous extracorporeal membrane oxygenation (ECMO) has been used with success for in-hospital arrests, and some literature suggests improvement in long-term survival for out-of-hospital arrests as well. OBJECTIVES: This case highlights the use of ECMO in the emergency department. CASE REPORT: We report a case in which emergency physician-initiated ECMO was used as a bridge to definitive care in an out-of- hospital cardiac arrest in the United States. CONCLUSIONS: ECMO is a novel adjunct for patients in cardiac arrest in whom the usual advanced life support techniques have failed.

Inhomogeneity and temporal effects in AutoPulse Assisted Prehospital International Resuscitation--an exception from consent trial terminated early.

BACKGROUND: The ASPIRE trial (AutoPulse Assisted Prehospital International Resuscitation) was multicenter exception from consent clinical trial that compared mechanical cardiopulmonary resuscitation (CPR) with a device (AutoPulse-CPR) to traditional manual CPR (manual-CPR) in out-of-hospital cardiac arrest. Enrollment was suspended early due to safety concerns. One site (site C) made a potentially important protocol change midtrial, and enrollment at that site was noted to be independently associated with outcome. METHODS: The study used a post hoc reanalysis of source data and documentation using standard statistical approaches evaluating for possible secular, temporal, and trial design, factors that may have related to the trial's outcome. RESULTS: The protocol change at site C also appears to have resulted in a delay in application of AutoPulse-CPR. Before and after the protocol change survival in patients receiving AutoPulse-CPR decreased from 19.6% to 4% (P = .024). Logistic regression analysis showed site C was significantly different (P = .008) from the remaining sites with respect to survival. Unlike site C, the other sites actually showed an increase over time in the primary end point of 4-hour survival (P = .008) favorable to AutoPulse-CPR. There did not appear to be significant safety (P = .42) nor efficacy concerns (P = .17) at these sites. CONCLUSIONS: The difference in survival that caused early suspension of ASPIRE appears to have been limited to one site after its protocols change. At the time the trial was suspended, the outcomes of patients at the other sites appear to have been trending in favor of the intervention.

Outcomes from low versus high-flow cardiopulmonary resuscitation in a swine model of cardiac arrest.

BACKGROUND: Return of spontaneous circulation (ROSC) is improved by greater vital organ blood flow during cardiopulmonary resuscitation (CPR). We tested the hypothesis that myocardial flow above the threshold needed for ROSC may be associated with greater vital organ injury and worse outcome. METHODS: Aortic and right atrial pressures were measured with micromanometers in 27 swine. After 10 minutes of untreated ventricular fibrillation, chest compression was performed with an automatic, load-distributing band. Animals were randomly assigned to receive flows just sufficient for ROSC (low flow: target coronary perfusion pressure = 12 mm Hg) or well above the minimally effective level (high flow: coronary perfusion pressure = 30 mm Hg). Myocardial flow was measured with microspheres, defibrillation was performed after 3.5 minutes of CPR, and ejection fraction was measured with echocardiography. RESULTS: Return of spontaneous circulation was achieved by 9 of 9 animals in the high-flow group and 15 of 18 in the low-flow group. All animals in the high-flow group defibrillated initially into a perfusing rhythm, whereas 12 of 15 animals achieving ROSC in the low-flow group defibrillated initially into pulseless electrical activity (P < .05, Fisher exact test). Compared with animals in the low-flow group, animals in the high-flow group had shorter resuscitation times, higher mean aortic pressures at ROSC, and higher ejection fractions at 2 hours post-ROSC (all P < .05). CONCLUSION: High-flow CPR significantly improved arrest hemodynamics, rates of ROSC, and post-ROSC indicators of myocardial status, all indicating less injury with higher flows. No evidence of organ injury from vital organ blood flow substantially above the threshold for ROSC was found.