William C. Shoemaker
Howard Belzberg
Charles C.J. Wo
George Hatzakis
Demetrios Demetriades
Definition of the Problem
Research in acute life-threatening emergencies has clearly shown that early recognition and vigorous therapy of shock facilitated by monitoring in acute life-threatening emergencies are the keys to successful resuscitation, because delayed therapy increases mortality and morbidity (1,2,3,4,5,6,7,8). Acute critical illness and shock from trauma, hemorrhage, high-risk surgery, sepsis, burns, stroke, and cardiac emergencies are circulatory problems that can be described by hemodynamic monitoring. There were significantly reduced mortality and morbidity in prospective randomized trials of optimal hemodynamic goals when they were achieved early, that is, <24 hours after emergency department (ED) admission and before onset of organ failure (1,2,3,4). However, outcome was not improved when optimal goals were accomplished late, defined as >24 hours after ED admission or after the onset of organ failure (3,4). The central focus of hemodynamic monitoring used is to provide appropriate therapy as determined by the adequacy of tissue perfusion and oxygenation. Recent experiences have demonstrated that early intervention is associated with improved outcome. Identification of the adequacy of perfusion is a function of balancing oxygen delivery with oxygen demand. Invasive monitoring requires time and technology, often leading to delays that reduce the potential benefit of the information gathered. In addition to the need for rapid monitoring, the challenge of continuous data collection and analysis as opposed to intermittent static observations is of major concern in the selection of monitoring techniques.
Pulmonary artery catheters (PACs) or Swan-Ganz catheters provide the maximum circulatory data at the bedside to describe the hemodynamic status in a wide variety of clinical states, emergencies, and other acute conditions (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18). However, PACs require intensive care unit (ICU) environments, and by the time the patient gets an ICU bed, it is often too late for early therapy. Since circulatory diseases are the leading cause of acute death, it is appropriate to develop and test hemodynamic methods and concepts for early use in emergencies and acute illnesses.
The principal difference between noninvasive monitoring compared with invasive PAC monitoring with blood gases taken at intervals is that the former provides online, real-time visual displays of multiple hemodynamic patterns describing cardiac, pulmonary, and tissue perfusion functions and their interrelationships (5,6,7). By contrast, invasive monitoring is taken once or at intervals as a set of static values and is primarily used to describe departures from the normal range as well as thresholds for circulatory disorders. For example, cardiac index (CI) <2.0 L/minute/m2, systolic blood pressure (SBP) <90 mm Hg, or mean arterial pressure (MAP) <70 mm Hg is used to define shock. However, noninvasive monitoring uniquely displays early hemodynamic patterns of acute illnesses that may underlie circulatory mechanisms (8,9,10). These patterns are not like static signposts, but are patterns of curves that may resemble biphasic or U waves. Stress responses may initially elevate CI, heart rate (HR), and MAP values, but initially high values are not sustained in the presence of hypovolemia, anemia, or impaired cardiac function and subsequently they rapidly fall to low levels along with arterial saturation of oxygen (SpO2) and transcutaneous partial pressure of oxygen/fraction of inspired oxygen (PtcO2/FiO2) values (see below). These sequential hemodynamic patterns provide new descriptive dimensions to understand the process of acute circulatory problems and resuscitation. Moreover, noninvasive patterns with information systems can predict outcome and support therapeutic decision making (16,17,18).
Noninvasive hemodynamic monitoring was found to be safe for patients and staff, convenient, inexpensive, reasonably accurate, and available anywhere in the hospital, doctors' offices, or prehospital areas. Early recognition of abnormal hemodynamic or incipient shock patterns allows therapy to be given earlier, more vigorously, and with greater effectiveness (6,7,8,9,10). Timing of treatment is as essential as the therapy itself in achieving good outcome.
There are a variety of modalities that have been developed to provide for hemodynamic monitoring by invasive, minimally invasive, and noninvasive methods. Many of these methods are presented in detail elsewhere in this textbook, but are mentioned here to emphasize their interrelationships. These methods can be grouped by their basic technologies, including Doppler ultrasound techniques, transcutaneous/transmucosal sensors of gases, monitors of hemoglobin saturation, expired gases, and a variety of dilution techniques. Each of these methods has its own strengths and weaknesses; taken together they have the potential to create a robust multidimensional picture of patients' hemodynamic states and their responses to therapeutic interventions. When only one method is used, its limitations may obscure the real problem and limit clinical usefulness.
Noninvasive Hemodynamic Monitoring
Conventional Vital Signs
Conventional vital signs include heart rate and mean arterial pressure; heart rate may be automatically derived from the electrocardiogram (ECG), and the systolic, diastolic, and mean arterial pressure measured directly or automated by DynaMAP (Tampa, FL), or from an indwelling arterial catheter if one is already in place.
Arterial Hemoglobin Oxygen Saturation by Pulse Oximeter
The standard pulse oximeter (Nellcor, Pleasanton, CA) is routinely used for continuous hemoglobin saturation (SpO2) monitoring (16). Two different wavelengths of infrared light are reflected from the capillaries to identify and evaluate the oxygen saturation of the arterial component. Sudden changes in the arterial hemoglobin saturation are useful in assessing changes in pulmonary function. This has had a major impact on clinical practice both inside and outside the operating room. Unfortunately, arterial oxygen saturation by itself does not indicate the oxygen delivery (DO2) or the oxygen consumption/demand (VO2), and is not a reliable measure of tissue oxygenation/perfusion or the hemodynamic status.
Cardiac Output Monitoring by Thoracic Electric Bioimpedance
In thoracic electric bioimpedance (TEB) methods, pairs of noninvasive disposable prewired hydrogel electrodes consisting of an outer injecting electrode and an inner sensing electrode are positioned on the anterior thoracic skin surface at the levels of the lung apex and the base; ECG leads are also placed on the precordium and shoulders (6,7,11,19,20,21,22,23,24,25,26,27,28,29,30,31,32). The outer pairs of electrodes pass a small-amplitude (0.2–5.0 mA) alternating current at 40 to 100 kHz through the patient's thorax to produce an electrical field. The injected electrical signals travel predominantly down the aorta, which has lower electrical resistance than aerated lung. Each ventricular contraction propels the stroke volume down the aorta, increasing aortic blood volume and aortic flow and lowering impedance. The impedance is sensed by the inner recording electrodes that capture the baseline impedance (Zo), the first derivative of the impedance waveform (dZ/dt), and the ECG (6,7,19). Changes in aortic blood flow throughout the cardiac cycle are quantitatively related to changes in the electrical impedance.
We have routinely used two TEB devices, PhysioFlow (VasoCOM, Bristol, PA) and IQ 101 (Noninvasive Medical Technologies, Las Vegas, NV), which were applied to the patient as soon as possible after arrival into the ED (5,6,7,19). When clinically indicated, the PAC was inserted in the operating room (OR) or ICU and was used to validate the noninvasive TEB method. Both of these TEB instruments gave satisfactory results when compared with the cardiac output values measured by the thermodilution (COtd) method using PACs. In 907 simultaneous pairs of TEB and thermodilution measurements in 267 patients, r was 0.92, r2 was 0.84, and p was <0.001 (11). The IQ 101 recently has been taken off the market for revisions and re-engineering.
In paired simultaneous TEB and COtd or other hemodynamic measurements, studies have reported satisfactory agreement between the methods (6,7,28,29,30,31,32,33,34,35,36,37,38,39,40,41). Raaijmakers et al. (25) found general agreement between several TEB systems reported over the past three decades, particularly in recent models using improved technology and algorithm. Simultaneous CI measurements by TEB and COtd during cardiac evaluation were satisfactorily correlated at rest (r = 0.83) and during exercise for cardiology workup (r = 0.86) (42,43). Reported errors between TEB systems and COtd were usually within twice the average difference in successive COtd estimations. We reported an average difference of 9.7% between TEB and COtd and an average difference of 9.5% between rapidly repeated COtd measurements in the same patient (6,7). Therefore, we routinely take four or five measurements and delete the most aberrant values. Some of the older systems such as the NCOM3 (BoMED, San Diego, CA) and its successor BioZ (CardioDynamics, San Diego, CA) had satisfactory correlations in many patients, but wide variations in about one fourth of patients, resulting in an overall average error of 25%.
Thoracic Electric Bioimpedance Applications
TEB cardiography was found to be helpful for diagnosis and therapy in the initial ED resuscitation since it is difficult clinically to detect early (occult) shock from BP and HR alone (13,23,24,33). In clinical trials, the diagnosis differed in 12 (13%) patients and the treatment changed in 35 (39%) patients after TEB results became available (33). Using TEB did not adversely affect defibrillation, resuscitation, or survival in patients during out-of-hospital cardiac arrest treated with biphasic waveform defibrillators (26). Ambulatory TEB provided useful values in both supine and tilt positions compared with Doppler cardiography (27). TEB has been used in the home, OR, ED, and ICU for thoracic fluid status monitoring after resuscitation from acute circulatory problems and for pacemaker optimization (6,7,28,29,30,31,32,33,34,35,36,37,38,39,40,41).
TEB provides a reasonably accurate, cost-effective, and noninvasive system that can replace PAC monitoring for various types of acute ED patients as well as for coronary care unit patients, those with pulmonary hypertension, intra- and postoperative coronary artery bypass, pediatric cardiac conditions (44,45), and diagnostic evaluation (46,47) and therapeutic responses (31,32,33,34,35,36,37,38,39). This technology may be used for healthy outpatient settings as well (46). For example, the effect of 1 unit of blood donation in 197 healthy volunteers demonstrated that stroke volume index values decreased from a baseline of 47.0 ± 6.9 (standard deviation [SD]) mL/m2 to 43.9 ± 7.3 mL/m2 after blood donation (32). Several groups reported that TEB was a feasible and accurate method for noninvasive monitoring of stroke index and cardiac index (44,45,46,47). Pianosi (44,45) measured exercise cardiac output by thoracic impedance in healthy children and with cystic fibrosis.
Use of Thoracic Electric Bioimpedance in Various Cardiac Conditions
The annual heart failure costs are about $56 billion, with two thirds of the cost consumed in the management of acutely decompensated patients. Systolic blood pressure alone did not reliably reflect the cardiac output status in 245 outpatient visits for heart failure (39,40,41,48). TEB may provide an outpatient tool to evaluate cardiomyopathies, heart failure, and pulmonary edema in the early and late stages (49,50) and also guide vasodilation therapy (49).
TEB has been used to assess cardiovascular status in hypertensive heart disease and to provide more complete physiologic characterization for effective targeted drug management in the ED (51). Physicians' clinical assessments were compared with measured TEB values in 186 patients. Diagnosis changed in 51% with more incidence of changed diagnosis in patients with a low CI (52). In another study, TEB and COtd values were compared with physicians' clinical judgments, assessed as high, medium, or low, and clinical judgment was incorrect 58% of the time (53).
TEB was found to be reliable for chronic fluid management in 33 patients diagnosed as New York Heart Association class III or IV. Of the ten patients hospitalized for fluid overload on 25 occasions in 20 months, TEB cardiac output identified patients in pulmonary edema who were subsequently hospitalized with PAC monitoring for fluid overload and PAC measurements reflected TEB values (54). TEB measurements were reproducible in both intra- and interday values in stable coronary artery disease patients (55), and diagnosed chronic heart failure and severity of cardiac dysfunction in ED patients with dyspnea (56). Other cardiac functions presented in the bioimpedance technology include velocity index, thoracic fluid content, and the left ventricular ejection time (57,58), and these values correlated well with outcome in 52 hypertensive patients who eventually demonstrated a 33% mortality (59).
Thoracic Electric Bioimpedance Combined with Other Noninvasive Methods for Comprehensive Monitoring: Transcutaneous Oxygen and Carbon Dioxide for Evaluation of Tissue Perfusion
TEB combined with conventional vital signs (MAP and HR), pulse oximetry (SpO2), and transcutaneous O2 and CO2 tensions and organized at the bedside of emergency patients in the ED, OR, or ICU provides the maximum noninvasive physiologic information previously only available by the PAC (5,6,7,18). Noninvasive variables were chosen to reflect (a) cardiac function by CI, MAP, and HR; (b) changes in respiratory function by SpO2; and (c) tissue perfusion reflected by transcutaneous pressure of CO2 (mm Hg) (PtcCO2) and transcutaneous pressure of O2 (mm Hg) indexed to the fractional inspired O2concentration (PtcO2/FiO2) (6,7). These noninvasive hemodynamic data were continuously displayed online in real time beginning shortly after ED admission. The values were continuously monitored, recorded by an interfaced personal computer, and filed directly into a database to describe sequential hemodynamic patterns related to survival or death (6,7,11,18). Martin et al. (20) used TEB with SpO2, and tissue perfusion by PtcO2/FiO2 values in pediatric trauma patients to predict outcome and track their hospital course.
PtcO2 reflects delivery of oxygen to the local area of skin; it also parallels the mixed venous oxygen tension. It uses the same Clark polarographic oxygen electrode routinely used in standard blood gas analyses. Oxygen tensions are determined in a representative area of the skin surface heated to 44°C to increase emissivity of oxygen across the stratum corneum, and to prevent vasoconstriction in the local area being measured. Transcutaneous CO2 tensions are continuously monitored by a standard Stowe-Severinghaus electrode in the O2 sensor unit.
Whole Body Electrical Bioimpedance
Whole body electrical bioimpedance (WBEB) uses proprietary electrodes arranged in a wrist-to-ankle configuration (43,60,61,62). An alternating current of 30 kHz, 1.4 mA is delivered through the two electrodes, and the bioimpedance fluctuations are measured. The ECG is used to measure HR. CI by WBEB compared satisfactorily with the COtd in 28 patients with bypass grafts (60).
Echocardiography: Transesophageal and Transtracheal Cardiography
Two-dimensional (2-D) Doppler echocardiography maps the anatomic relations of the heart by ultrasonic wave pulses sent out from transducers placed on the chest wall, esophageal tube, or endotracheal tube. The sound waves are bounced off the walls and valves of the heart and are captured by sensors as echoes that are electronically plotted to produce images corresponding to anatomic features of the heart, including the right and left atrium, the atrial septum, valves, pulmonary vessels, and the thoracic duct.
Transesophageal echocardiography (TEE) is useful in evaluating heart valve abnormalities, tumors, blood clots, dissecting aortic aneurysms, and congenital defects. Doppler techniques are for diagnostic imaging rather than a continuous hemodynamic assessment tool; the images may be distorted by obesity, chronic obstructive pulmonary disease (COPD), hyperinflation of the lungs, chest trauma, and surgical dressings. Doppler echocardiology favorably compared with TEB in 16 normal subjects; the mean differences in CI between both methods were relatively small (r = 0.87) (43).
Doppler techniques are able to provide some qualitative information about flow through vessels, but are limited by the need for expert interpretation of the images and the need for very sensitive transducer placement. These factors provide for a useful diagnostic tool, but limit the reproducibility of the measurements and their use for continuous monitoring.
Lithium Dilution Cardiac Output Measurements
Lithium chloride, 0.3 mmol, is used as the indicator for calculation of cardiac output by the standard indicator dilution technique (63). The system requires an intravenous catheter for rapid infusion of the lithium indicator and a disposable lithium-selective electrode in the arterial catheter tubing with an attached battery-operated roller pump to withdraw blood through the lithium sensor.
Pulse Contour Cardiac Output System
The pulse contour method analyzed the systolic part of the arterial pressure waveform based on a complex analysis of waveform harmonics (64,65). The disadvantage of this approach is that it requires calibration by invasive methods, and the calibration lasts only as long as the patient's arterial vasoactive status remains constant. This is a major limitation for unstable postoperative patients.
Noninvasive Partial CO2 Rebreathing Cardiac Output Measurement
Noninvasive partial CO2 rebreathing cardiac output (NICO) measurement is a cardiac output monitoring system based on CO2 elimination measured by the direct Fick principle. The system measures CO2 on a breath-by-breath basis to provide a continuous estimation of metabolic demand, with precise tidal volume and flow analyses (flow/volume). The cardiac output is proportional to the changes in CO2elimination divided by the change in end-tidal CO2 measured by a CO2 sensor, which periodically adds a rebreathing volume into the ventilatory circuit (NICO, Novametrics Medical Systems, Wallingford, CT). The system also provides compliance assessments that facilitate adjustment of ventilatory parameters to optimize compliance with positive end-expiratory pressure (PEEP) and tidal volume settings. Three NICO values averaged over 1 minute were compared with 418 paired thermodilution measurements in 122 patients during cardiac catheterization prior to cardiac surgery. The overall correlation was r = 0.89 with a small bias (66,67,68).
The NICO system was prospectively compared with thermodilution in 122 patients (a) during cardiac catheterization; (b) before, during, and after coronary bypass surgery; and (c) during treatment for acute congestive heart failure (CHF). The authors concluded that NICO provided an accurate noninvasive measurement of CI (67,68). The system is minimally invasive but it requires the patient to have tracheal intubation, a closed ventilatory circuit, and mechanically assisted ventilation.
These techniques share the shortcomings of thermodilution with PACs in that they give only episodic sets of measurements, require critical care conditions, and need frequent recalibration.
|
Table 17.1 Hemodynamic values in trauma patients by invasive and noninvasive monitoring |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastric Transmucosal Oxygen and Carbon Dioxide Tensions
By positioning a balloon against the stomach wall and allowing for gas to equilibrate between the gastric mucosa and the lumen of the balloon, O2 and CO2 tensions in tissue adjacent to the probe may be extrapolated to assess the degree of anaerobic metabolism. Some investigators have found this technique useful in titrating resuscitation in shock. The difficulty with this is the limited tissue bed assessed and the assumption that perfusion of the monitored tissue reflects the other tissues. Details of this technology are described in other chapters.
Near-infrared Spectroscopy
Near-infrared spectroscopy (NIRS) is used to measure tissue myoglobin oxygen saturation (StO2) of the hypothenar space as an alternative to transcutaneous O2 and CO2 to assess regional tissue perfusion (69,70,71). Regional perfusion markers have been extrapolated to estimate the overall tissue perfusion status. These measurements were reported to be reliable for resuscitation and monitoring, but their role remains to be defined throughout in the clinical environment. NIRS feasibility studies were undertaken in 46 ED patients, six of whom were hospitalized for heart failure with one death (69).
|
|
|
Figure 17.1. Data from a 66-year-old man who was involved in a motor vehicle accident and sustained lacerations of the spleen and descending aorta, hemothorax, and multiple rib fractures. The injury severity score was 59, the Glasgow coma scale was 9, and he sustained an estimated initial blood loss of 4,000 mL. He was operated on between 1 and 5 hours after emergency department (ED) admission (splenectomy and repair of thoracic aortic laceration). He was given 7 liters of Ringer lactate, 8 units of packed red cells, and 1,500 mL starch in the operating room (OR) followed by 5 units of fresh frozen plasma in the intensive care unit (ICU) and 2 more units of packed red cells and 2 units of fresh frozen plasma (lowest section of the graph). Upper row: Cardiac index (CI). Second row: Mean arterial pressure (MAP). Third row: Pulse oximetry (SpO2). Fourth row: PtcO2/FiO2. Lowest row: Survival probability. Time, in hours from ED admission, is noted below the bottom horizontal line. Therapies are outlined in boxes: FFP, fresh frozen plasma; HES, hydroxyethyl starch; RBC, packed red blood cell transfusion; DOP, dopamine; LR, lactated Ringer solution, which was given at the rate 150 mL/hr postoperatively. Time and place of resuscitation (ER, OR, ICU) indicated at lowest line. Note: The CI fell to around 2 L/min/m2 throughout most of the intraoperative and immediate postoperative period, the MAP fell to around 60 to 70 mm Hg, the PtcO2/FiO2 was less than 100 torr throughout most of the course, and the SP was less than 50% throughout most of the course. Despite this evidence of hypovolemia, the operating surgeon was reluctant to give more fluids for fear of disrupting the repaired aortic laceration. The patient developed acute respiratory distress syndrome (ARDS), sepsis, cardiac failure, and renal failure. He died 31 days after admission. |
|
|
|
Figure 17.2. Survivors' (solid line) and nonsurvivors' (dashed line) temporal patterns are shown for the first 48 hours after their emergency department (ED) admission. Mean values ± standard error of means (SEM) are shown for cardiac index (CI) in mL/min/m2, heart rate (HR), mean arterial pressure (MAP), pulse oximetry (SpO2), transcutaneous oxygen tension indexed to the fractional inspired oxygen concentration (PtcO2/FiO2), and survival probability (SP). All values are keyed to at the time of admission to the ED. Note that the survivors' cardiac index, MAP, SpO2, PtcO2/FiO2, and SP values were generally higher than those of the nonsurvivors. The mean survivors' SP values were significantly higher than the mean nonsurvivors' SP values in this observation period. |
Summary of Survivors' and Nonsurvivors' Data by Invasive and Noninvasive Monitoring
Table 17.1 summarizes the monitored data of 661 surviving and nonsurviving trauma patients. Survivors had greater CI, MAP, SpO2, PtcO2/FiO2, DO2, VO2, and survival probability (SP) values. Figure 17.1illustrates data of a 66-year-old man who sustained an auto accident with lacerations of the spleen, thoracic aorta, and multiple rib fractures with an estimated initial blood loss of over 4,000 mL. Despite the data and other evidence of hypovolemia, the operating surgeon was reluctant to give more fluids for fear of disrupting the repaired aortic laceration. The patient died of multiple organ failure after 31 days. Figure 17.2 illustrates data of 185 consecutive trauma patients obtained during the first 2 hours in the ED for CI, MAP, HR, SpO2, PtcO2/FiO2, and SP. Table 17.2 lists the outcome predictions by SP at the beginning of monitoring compared with the actual outcome at the time of hospital discharge. The misclassifications were 14.2% compared with greater errors by conventional analyses (Table 17.3 and Fig. 17.3).
|
Table 17.2 Summary of classifications of emergency trauma patients exclusive of severe head injury and brain death (N = 661) |
|||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||
Quantitative Assessment of Noninvasive Variables as Net Cumulative Excess or Deficit
While continuous noninvasively monitored values are directly observed, they may have considerable variability that obscures the underlying patterns. To overcome this problem, we calculated the net cumulative excess or deficit of physiologic variables by integrating the areas between the curves of monitored data and normal values to provide a quantitative measure of the overall net cumulative deficit in cardiac, pulmonary, and tissue perfusion functions.
|
Table 17.3 Misclassifications in prediction by various outcome predictors |
||||||||||||||||||||||||
|
||||||||||||||||||||||||
Information Systems to Predict Outcome and to Support Therapy
Continuous noninvasive monitoring needs information systems to make sense of the vast amount of data generated. Information systems are designed to translate massive continuous data streams from complex sources into useful knowledge and then to fashion these arrays of knowledge into intelligent clinical decisions. Bayard et al. (11,16,17,18) developed and tested an information system based on a stochastic (probability) analysis and control program. It uses a large database of emergency patients to identify similar patients with identical diagnoses, covariates, and very close hemodynamic patterns. These similar patients, defined as “nearest neighbors,” are used as surrogates to compute in real time the SP of newly admitted study patients. A patient's SP for a given state is calculated by first extracting from the database 40 or more “nearest neighbors” whose clinical and hemodynamic patterns most closely resemble those of the study patient. Then the SP is calculated as the fraction of nearest neighbors that survived. The SP predicts outcome and is also a digitized measure of severity of illness; low SP values are quantitatively related to the likelihood of death and reflect illness severity (11,17,18).
|
|
|
Figure 17.3. Receiver operating characteristic (ROC) curves calculated for data collected in trauma patients over the first 24-hour period after emergency department (ED) admission. The area under the curves represents the sensitivity and specificity for each variable: 1.00 represents 100% correct, 0.50 represents no correlation. These areas under the ROC curve were 0.85 for survival probability (SP), 0.81 for transcutaneous oxygen tension indexed to the fractional inspired oxygen concentration (PtcO2/FiO2), 0.64 for pulse oximetry (SpO2), 0.61 for mean arterial pressure (MAP), and 0.59 for cardiac index (CI). |
The therapeutic decision support program uses the study patient's nearest neighbors' responses to various therapies that were given to them and recorded in the database. The likely responses of the study patient are suggested by the nearest neighbors' responses known from the database (11,17). This approach includes the following characteristics: (a) the patient's course is described by a three-dimensional trajectory; (b) the first derivative of the initial vector projects the patient's course if there are no inherent changes or external influences; (c) the second derivative tracks changes in the patient's course from either internal compensations, further deterioration, spontaneous improvement, or external influences such as changes in therapy; and (d) the integral sums up the total influences. Methods of machine learning (37,38), and dynamic programming for stochastic control (39,40) motivated the program.
Minimally Invasive Monitoring
Noninvasive Hemodynamic Monitoring with Central Venous Pressure Catheters and Oximeters
Rivers et al. (10) used noninvasive hemodynamic monitoring and central venous pressure (CVP) catheters with attached oximeters at their tips to guide resuscitation of septic and septic shock patients in the ED. They showed significantly improved survival in septic patients with early goal-directed therapy started in the ED based on minimally invasive monitoring. They identified their septic patients in the first hour after ED admission; this was followed by 6 hours of goal-directed therapy in the ED in protocol patients, and compared for a similar period with standard therapy in the control group. Then, the patients were sent to the appropriate in-hospital department without the knowledge of whether the patient had been randomized to the protocol or the control group (10). The treatment administered in the first 6 hours impacted outcomes. This demonstrates that the following are important: early treatment, and resuscitation to a flow-related goal rather than the traditional blood pressure goal.
Summary
Noninvasive systems are able to continuously display variables online, in real time throughout the patient's hospital course, and with information systems to predict outcome and support therapeutic decision making. Because they are simple, safe, easy to use, inexpensive, reasonably accurate, and available anywhere in the hospital, some form of noninvasive hemodynamic monitoring with an appropriate information system will replace invasive monitoring for acutely ill patients.
Pearls
1. One-dimensional monitoring with one type of monitor may obscure the real problem. Therefore, it is essential to monitor cardiac, pulmonary, and tissue perfusion/oxygenation functions and their interrelationships to show the whole picture.
2. Comprehensive noninvasive hemodynamic monitoring with an information system will predict outcome and guide therapy.
3. Continuous, online, real-time visual displays of hemodynamic patterns are safer for both patient and staff, are inexpensive, are user friendly, approximate PAC values, and may be used throughout the hospital and prehospital areas.
4. There is no single magic number for optimal blood flow, because the microcirculation needed for tissue perfusion/oxygenation must also be adequate, just as rush hour traffic may greatly affect flow in main arteries, or side streets, or both. It is necessary to optimize both total blood flow and tissue perfusion/oxygenation.
References
1. Shoemaker WC, Appel PL, Kram HB. Prospective trial of supranormal values of survivors as therapeutic goals in high risk surgical patients. Chest. 1988;94:1176.
2. Boyd O, Grounds M, Bennett D. Preoperative increase of oxygen delivery reduces mortality in high risk surgical patients. JAMA. 1993;270:2699.
3. Boyd O, Hayes M. The oxygen trail: the goal. Brit Med Bull. 1999;55:125–139.
4. Kern JW, Shoemaker WC. Meta-analysis of hemodynamic optimization in high-risk patients. Crit Care Med. 2002;30:1686–1692.
5. Shoemaker WC, Appel PL, Kram HB, et al. Multicomponent noninvasive physiologic monitoring of circulatory function. Crit Care Med. 1988;16:482–490.
6. Shoemaker WC, Belzberg H, Wo CCJ, et al. Multicenter study of noninvasive monitoring systems as an alternative to invasive monitoring of acutely ill emergency patients. Chest. 1998;114:1643–1652.
7. Shoemaker WC, Wo CCJ, Chan L, et al. Outcome prediction of emergency patients by noninvasive hemodynamic monitoring. Chest. 2001;120:528–537.
8. Bishop MW, Shoemaker WC, Appel PL, et al. Relationship between supranormal values, time delays and outcome in severely traumatized patients. Crit Care Med. 1993;21:56.
9. Bishop MH, Shoemaker WC, Kram HB, et al. Prospective randomized trial of survivor values of cardiac index, oxygen delivery, and oxygen consumption as resuscitation endpoints in severe trauma. J Trauma. 1995;38:780–787.
10. Rivers E, Nguyen B, Havsted S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2002;345:1368–1377.
11. Shoemaker WC, Wo CCJ, Cheien L-C, et al. Evaluation of invasive and noninvasive hemodynamic monitoring in trauma patients. J Trauma. 2006;60:82–90.
12. Moore FA, Haemel JB, Moore EE, et al. Incommensurate oxygen consumption in response to maximal oxygen availability predicts postinjury oxygen failure. J Trauma. 1992;33:58.
13. Yu M, Levy MM, Smith P, et al. Effect of maximizing oxygen delivery on mortality and morbidity rates in critically ill patients: a prospective randomized control study. Crit Care Med. 1993;21:830–838.
14. Creamer JE, Edwards JD, Nightingale P. Hemodynamic and oxygen transport variables in cardiogenic shock secondary to acute myocardial infarction. Am J Cardiol. 1990;65:1287.
15. Rady MY, Edwards JD, Rivers EP, et al. Measurement of oxygen consumption after uncomplicated acute myocardial infarction. Chest. 1993;103:886–895.
16. Bayard DS, Botnen A, Shoemaker WC, et al. Stochastic analysis of therapeutic modalities using a database of patient responses. IEEE Symposium on Computer Based Medical Systems (CBMS). 2001;11:439–444.
17. Shoemaker WC, Wo CCJ, Botnan A, et al. Development of a hemodynamic database in severe trauma patients to define optimal goals and predict outcome. IEEE Symposium on Computer Based Medical Systems (CBMS). 2001;11:445–452.
18. Shoemaker WC, Bayard DS, Botnen A, et al. Mathematical program for outcome prediction and therapeutic support for trauma beginning within 1 hr of admission: a preliminary report. Crit Care Med. 2005;33:1499–1506.
19. Charloux A, Lonsdorfer-Wolf E, Richard R, et al. A new impedance cardiograph device for the noninvasive evaluation of cardiac output at rest and during exercise: comparison with the “direct” Fick method. Eur J Appl Physiol. 2000;82:313–320.
20. Martin M, Brown C, Bayard DS, et al. Continuous noninvasive monitoring of cardiac performance and tissue perfusion in pediatric trauma. J Pediatr Surg. 2005;40:1957–1963.
21. Shoemaker WC, Appel PL, Kram HB. Role of oxygen debt in the development of organ failure, sepsis and death in high-risk surgical patients. Chest. 1992;102:208–215.
22. Shoemaker WC, Wo CCJ, Bishop MH, et al. Multicenter trial of a new thoracic electrical bioimpedance device for cardiac output estimations. Crit Care Med. 1994;22:1907–1912.
23. Summers RL, Vogel J, Emerman CE. Impact of impedance cardiography on diagnosis and therapy of emergent dyspnea: the ED-IMPACT trial. Acad Emerg Med. 2006;13:365–371.
24. Summers RL, Shoemaker WC, Peacock WF, et al. Bench to bedside: electrophysiologic and clinical principles of noninvasive hemodynamic monitoring using impedance cardiography. Acad Emerg Med. 2003;10:669–680.
25. Raaijmakers E, Faes TJ, Scholten RJ, et al. A meta-analysis of three decades of validating thoracic impedance cardiography. Crit Care Med. 1999;27:1203–1213.
26. White RD, Blackwell TH, Russell JK, et al. Transthoracic impedance does not affect defibrillation, resuscitation or survival in patients with out-of-hospital cardiac arrest treated with a non-escalating biphasic waveform defibrillator. Resuscitation. 2005;64:63–69.
27. Cybulski G, Michalak E, Kozluk E, et al. Stroke volume and systolic time intervals: beat-to-beat comparison between echocardiography and ambulatory impedance cardiography in supine and tilted positions. Med Biol Eng Comp. 2004;42:707–711.
28. Critchley LA. Impedance cardiography: the impact of new technology. Anaesthesiology. 1998;53:677–684.
29. Haryadi DG, Westenskow DR, Critchley LA, et al. Evaluation of a new advanced thoracic bioimpedance device for estimation of cardiac output. J Clin Monit Comp. 1999;15:131–138.
30. Bucklar GB, Kaplan V, Bloch KE. Signal processing technique for non-invasive real-time estimation of cardiac output by inductance cardiography (thoracocardiography). Med Biol Eng Comput. 2003;41:302–309.
31. Van De Water JM, Miller TW, Vogel RL, et al. Impedance cardiography: the next vital sign technology? Chest. 2003;123:2028–2033.
32. Wilson M, Davis DP, Coimbra R. Diagnosis and monitoring of hemorrhagic shock during the initial resuscitation of multiple trauma patients: a review. J Emerg Med. 2003;24:413–422.
33. Neath SX, Lazio L, Guss DA. Utility of impedance cardiography to improve physician estimation of hemodynamic parameters in the emergency department. Cong Heart Fail. 2005;11:17–20.
34. Drazner MH, Thompson B, Rosenberg PB, et al. Comparison of impedance cardiography with invasive hemodynamic measurements in patients with heart failure secondary to ischemic or nonischemic cardiomyopathy. Am J Cardiol. 2002;89:993–995.
35. Kosowsky JM, Han JH, Collins SP, et al. Assessment of stroke index using impedance cardiography: comparison with traditional vital signs for detection of moderate acute blood loss in healthy volunteers. Acad Emerg Med. 2002;9:775–780.
36. Sageman WS, Riffenburgh RH, Spiess BD. Equivalence of bioimpedance and thermodilution in measuring cardiac index after cardiac surgery. J Cardiothor Vasc Anesth. 2002;16:8–14.
37. Richard R, Lonsdorfer-Wolf E, et al. Non-invasive cardiac output evaluation during a maximal progressive exercise test, using a new impedance cardiograph device. Euro J Appl Physiol. 2001;85:202–207.
38. Engoren M, Barbee D. Comparison of cardiac output determined by bioimpedance, thermodilution, and the Fick method. Am J Crit Care. 2005;14:40–45.
39. Albert NM, Hail MD, Li J, et al. Equivalence of the bioimpedance and thermodilution methods in measuring cardiac output in hospitalized patients with advanced, decompensated chronic heart failure. Am J Crit Care. 2004;13:469–479.
40. Imhoff M, Lehner JH, Lohlein D. Noninvasive whole-body electrical bioimpedance cardiac output and invasive thermodilution cardiac output in risk surgical patients. Crit Care Med. 2000;28:2812–2818.
41. Barin E, Haryadi DG, Schookin SI, et al. Evaluation of a thoracic bioimpedance cardiac output monitor during cardiac catheterization. Crit Care Med. 2000;28:698–702.
42. Smit HJ, Vonk Noordegraaf A, Marcus JT, et al. Determinants of pulmonary perfusion measured by electrical impedance tomography. Euro J Appl Physiol. 2004;92:45–49.
43. Cotter G, Moshkovitz Y, Kaluski E, et al. Accurate, noninvasive continuous monitoring of cardiac output by whole-body electrical bioimpedance. Chest. 2004;125(4):1431–1440.
44. Pianosi PT. Measurement of exercise cardiac output by thoracic impedance in healthy children. Euro J Appl Physiol. 2004;92:425–430.
45. Pianosi PT. Impedance cardiography accurately measures cardiac output during exercise in children with cystic fibrosis. Chest. 1997;111:333–337.
46. Scherhag A, Kaden JJ, Kentschke E, et al. Comparison of impedance cardiography and thermodilution-derived measurements of stroke volume and cardiac output at rest and during exercise testing. Cardiovasc Drug Ther. 2005;19:141–147.
47. Weiss SJ, Ernst AA, Godorov G, et al. Bioimpedance-derived differences in cardiac physiology during exercise stress testing in low-risk chest pain patients. South Med J. 2003;96:1121–1127.
48. Parrott CW, Quale C, Lewis DL, et al. Systolic blood pressure does not reliably identify vasoactive status in chronic heart failure. Am J Hypertens. 2005;18(Pt 2):82S–86S.
49. Bhalla V, Isakson S, Bhalla MA, et al. Diagnostic ability of B-type natriuretic peptide and impedance cardiography: testing to identify left ventricular dysfunction in hypertensive patients. Am J Hypertens. 2005;18(Pt 2):73S–81S.
50. Yancy C, Abraham WT. Noninvasive hemodynamic monitoring in heart failure: utilization of impedance cardiography. Congest Heart Fail. 2003;9:241–250.
51. Ventura HO, Taler SJ, Strobeck JE. Hypertension as a hemodynamic disease: the role of impedance cardiography in diagnostic, prognostic, and therapeutic decision making. Am J Hypertens. 2005;18(Pt 2):26S–43S.
52. Van De Water JM, Dalton ML, Parish DC, et al. Cardiopulmonary assessment: is improvement needed? World J Surg. 2005;29(Suppl 1):S95–98.
53. Veale WN Jr, Morgan JH, Beatty JS, et al. Hemodynamic and pulmonary fluid status in the trauma patient: are we slipping? Am Surgeon. 2005;71:621–626.
54. Yu CM, Wang L, Chau E, et al. Intrathoracic impedance monitoring in patients with heart failure: correlation with fluid status and feasibility of early warning preceding hospitalization. Circulation. 2005;112:841–848.
55. Treister N, Wagner K, Jansen PR. Reproducibility of impedance cardiography parameters in outpatients with clinically stable coronary artery disease. Am J Hypertens. 2005;18(Pt 2):44S–50S.
56. Barcarse E, Kazanegra R, Chen A, et al. Combination of B-type natriuretic peptide levels and non-invasive hemodynamic parameters in diagnosing congestive heart failure in the emergency department. Cong Heart Fail. 2004;10:171–176.
57. Packer M, Abraham WT, Mehra MR, et al. Utility of impedance cardiography for the identification of short-term risk of clinical decompensation in stable patients with chronic heart failure. J Am Coll Cardiol. 2006;47:2245–2252.
58. Sanford T, Treister N, Peters C. Use of noninvasive hemodynamics in hypertension management. Am J Hypertens. 2005;8(Pt 2):87S–91S.
59. Ramirez MF, Tibayan RT, Marinas CE, et al. Prognostic value of hemodynamic findings from impedance cardiography in hypertensive stroke. Am J Hypertens. 2005;18(Pt 2):65S–72S.
60. Kaukinen S, Koobi T, Bi Y, et al. Cardiac output measurement after coronary artery bypass grafting using bolus thermodilution, continuous thermodilution, and whole-body impedance cardiography. J Cardiothor Vasc Anesth. 2003;17:199–203.
61. Kauppinen PK, Koobi T, Hyttinen J, et al. Segmental composition of whole-body impedance cardiogram estimated by computer simulations and clinical experiments. Clin Physiol. 2000;20:106–113.
62. Koobi T, Kaukinen S, Turjanmaa VM, et al. Whole-body impedance cardiography in the measurement of cardiac output. Crit Care Med. 1997;25:779–785.
63. Linton RA, Band DM, Haire KM. A new method for measuring cardiac output using lithium dilution. Br J Anaesth. 1993;71:262–266.
64. Nichols WW, O'Rourke MF. MacDonald's Blood Flow in Arteries. 5th ed. London: Hodder Arnold Publication; 2005.
65. Linton NWP, Linton RAF. Estimation of changes in cardiac output from the arterial blood pressure waveform in the upper limb. Br J Anaesth. 2001;86:486–494.
66. Capek JM, Roy RJ. IEEE Trans BME. 1988:653–661.
67. Jaffe MB:. Partial CO2 rebreathing cardiac output: operating principles of the NICO system. J Clin Monit. 1999;15:387–401.
68. Tordi N, Mourot L, Matusheski B, et al. Measurements of cardiac output during constant exercises: comparison of two non-invasive techniques. Int J Sports Med. 2004;25:145–149.
69. Beilman GJ, Groehler KE, Lazaron V, et al. Near infrared spectroscopy measurement of regional tissue oxyhemoglobin saturation during hemorrhagic shock. Shock. 1999;12:196–200.
70. Cohn SM, Crookes BA, Proctor KG. Near-infrared spectroscopy in resuscitation. J Trauma. 2003;54(5):S199–S202.
71. Myers DE, Anderson LD, Seifert RP, et al. Noninvasive method for measuring local hemoglobin oxygen saturation in tissue using wide gap second derivative near-infrared spectroscopy. J Biomed Optics. 2005;10(3):1–18.
