1. Compare measurements of cardiac output derived from electrical measurements from electrodes on an endotracheal tube (ECOM or Endotracheal Cardiac Output Monitor) to those made from a pulmonary artery thermodilution catheter.
2. Establish the safety and efficacy of the ECOM system.

Introduction: Cardiac output is a key physiological parameter. Unfortunately, it is often difficult to measure without using invasive techniques with associated risk. The pulmonary artery catheter has, since its introduction, been considered the gold standard for the measurement of cardiac output in man. Unfortunately, pulmonary artery catheters have been associated with serious complications.1,2 Thoracic electrical bio-impedance (TEB) has been suggested as a possible non-invasive technique for the measurement of cardiac output. The accuracy and reliability of TEB has been evaluated multiple times with some studies demonstrating good correlation with thermodilution3-7 and others with poor correlation.6,8-15. TEB is not reliable in patients after cardiopulmonary bypass8,9, kidney transplants11, congestive heart failure16, pulmonary edema14, sepsis12, pregnancy15, abdominal surgery11,17, or critical illness13,14.

One of the limitations of TEB is the signal to noise ratio. Commonly TEB systems use an alternating current (1 - 4 ma at 20-100 Khz) applied to the skin and have little control over the percentage of current passing through the vascular structures in the chest. Changes in the electrical impedance of the lungs with respiration change the percentage of the total current passing through blood containing structures. Finally, treating all of the blood containing structures in the chest as a single impedance to be measured, does not allow separation of the signal into its various components. A technique which stabilizes the percentage of current delivered to the target structure and records the electrical impedance signals directly from that target structure improves the accuracy and reliability of impedance based measurement of cardiac output.

The Endotracheal Cardiac Output Monitor (ECOM) is a system which records the voltages produced by a current (2 ma 100 Khz) delivered to the tracheal mucosa by electrodes on an endotracheal tube. The proximity of the ascending aorta and trachea allows the design of a device that can optimize the current delivery and signal recording from impedance changes in the ascending aorta. This study will test the accuracy and efficacy of the ECOM system in anesthetized patients.

Significance: A simple, inexpensive, continuous, non-invasive, accurate, reproducible method of measuring cardiac output in patients would reduce the need for pulmonary artery catheter measurements of cardiac output. Such a technique would allow the rapid diagnosis and treatment of hemodynamic instability in the critically ill patient. The maintenance of cardiac output is essential for the prevention of end organ injury which results in morbidity and mortality. A technique that accurately measured cardiac output would greatly assist the management of the critically ill patient.

Methods:

We propose to compare measurement of cardiac output made with the Conmed ECOM 6-3D endotracheal tube to those made with a pulmonary artery catheter using. We will do correlation studies of cardiac output against those performed using the standard thermodilution technique in patients, who in the normal course of their clinical care, are having cardiac outputs measured by the thermodilution technique. We will compare cardiac output measured by impedance cardiography to transit time measurements.

These correlation studies will be done in the O.R. and intensive care units on patients scheduled for cardiac surgery who routinely have cardiac output measurements using the standard thermodilution method.

Many different algorithms have been developed for the conversion of thoracic impedance signals into a calculation of cardiac output including Kubicek18 , Bernstein-Sramek19, Shmulewitz20, adaptive algorithms.20 Each of these algorithms requires some sort of empirically derived factor to convert from measurements of resistance into measurements of volume. The ECOM system uses an adaptive multi-parameter algorithm which allows for the reduction of between subject variability. We have derived the empiric fit for the porcine model but have not done so for patients. This study will be therefore divided into two components. The first data collection of 15 patients will provide the data for the empiric fit between the ECOM system and true cardiac output. The second set of 300 patients will be the test group for the algorithm.

1. General Study Design

   Two Component Study: Two component study. The first section is a 15 patient study which will provide the data for the empiric fit between the ECOM signal and true cardiac output. The second section is a 300 patient study which will provide the test data for validation of the algorithm.

   Cardiac Output Measurements: Cardiac outputs measured by the impedance device will be compared with measurements from the thermodilution catheter. The primary comparison will be between ECOM measurements and the thermodilution measurement.

   Safety Data: No complications have been identified with ECOM system use in patients studied to date. Despite this safety profile, we would like to identify any potential problems with ECOM tube use. These problems could potentially include tracheal stenosis, laryngeal injury, vocal cord injury, and changes in the voice. We have looked for these complications with clinical history and follow up. We have also included a more extensive post operative interview to look for any airway injury.

2. Methods of Data Analysis

   Primary outcome comparison will be the correlation between thermodilution cardiac output and ECOM impedance measurements. Both Linear regression and Bland-Altman statistics of thermodilution cardiac output measurements versus impedance cardiography will be performed. Significance will be at p < 0.05.

   Power Calculations:

   A sample size of 15 patients gives us a 90% power to show a correlation R2 of 0.75. In the ECOM porcine study we have an R2 of 0.77 to 0.84. The 300 patient second section will give us sufficient patient variability (large, small, male, female, young, old, aortic atherosclerosis, aortic regurgitation) to assess the technique across a wide spectrum of patients.

3. Subject Selection

   1. Who and Why:

      Patients scheduled for cardiac surgery using extracorporeal circulation and median sternotomy will be consented. As part of their routine care for cardiac surgery, these patients will be intubated and pulmonary artery catheters placed. The measurements proposed in this study will therefore add little additional risk. Study patients will be intubated with an endotracheal tube which will allow endotracheal electrode measurements (ECOM 6-3D). Thermodilution cardiac output is routinely measured which will allow correlation between ECOM and thermodilution cardiac output measurements at minimal additional risk.

   2. Total Number/Number per Group Three hundred (300) patients undergoing cardiac surgery with median sternotomy will be studied.

4. Subject Recruitment 1) Source: Patients scheduled for cardiac surgery at the San Francisco Veterans Affairs Medical Center (VAMC) hospital.

   2) Initial Contact Method: Prospective subjects will be approached by a study physician and consent obtained in the anesthesia pre-operative (preop) clinic prior to admission to the hospital, or in the hospital, if the patient is already an in-patient. Subjects will be approached by a study physician who will obtain consent.

5. Consent Process and Documentation All consent will be obtained by study physicians who will explain the risks to the patient. The VA population will be used. No minors or non-English speakers will be included. No patients who are mentally incompetent to give consent will be included. Only signed, written, witnessed, informed consent will be used.

6. Procedures 1) Study Procedures The patient will undergo routine monitor placement for surgery. Only patients who would have a pulmonary artery catheter placed for their surgery will be included. The following exceptions in standard anesthetic preparation, induction, maintenance, and emergence will be made.

   1. Extra ECG electrode patches will be placed.

   2. The endotracheal tube will have 7 silver doped plastic electrodes (Conmed ECOM Tube).

   3. Pulse oximetry will be measured and recorded using finger, nose, and forehead pulse oximetry measurements continuously. Pulse oximetry readings will be compared between the three sites and compared to cardiac output measurements. The percentage of time that a signal is obtained from the three sites will be calculated. The accuracy of pulse oximetry measurements will be compared to routinely drawn blood saturation measurements. Pulse oximetry is required during surgery. This addition will record the values obtained during routine care. If there are differences between brands or sensor types, those results may be reported.

   4. Cardiac output will be measured by iced-thermodilution and impedance cardiography . All personnel involved in the study understand that management of the patient will be done using the thermodilution derived cardiac output measurements only. The ECOM endotracheal cardiac output measurements are for research purposes only and will not be used in the management of the patient.

   5. ECOM Impedance cardiography will be measured in the ICU when routine thermodilution cardiac output measurements are made. Endotracheal impedance measurements (ECOM) will be continued until tracheal extubation. Correlation with thermodilution measurements will be stopped when either the endotracheal tube or the thermodilution catheter is removed (post op day 0 routinely).

   6. The patient will be interviewed and the medical records will be reviewed for any adverse outcomes from the study. Interviews will occur prior to hospital discharge, at one month post discharge, and yearly thereafter.

   7. The study will be conducted and data collected by the research team.

   2) Time: The study will be performed during surgery and the ICU stay. It should not delay surgery more than 5 minutes. Placement of the electrode patches is rapid. The endotracheal electrodes are attached to the endotracheal tube and placement of the tube is routine. Recording of cardiac outputs and impedance data will not delay surgery. Placement and removal of the aortic flow probe is rapid but may delay surgery for a few minutes. Data collected in the ICU will only be correlations between thermodilution and impedance data, since no transit time data can be collected outside the operating room. ICU data will not delay ICU or hospital discharge.

   3) Study Site: All studies will be performed at the VAMC either in the operating rooms or ICU.

7. Risks/Discomforts

   1. The extra ECG electrode patches have minimal risk. All circuits will have electrical isolation and will be approved by biomedical engineering to avoid ground loops and leakage currents. There is always the chance of electrical burns or shocks but these are rare.
   2. The endotracheal tube will have seven silver doped plastic electrodes. In addition to the electrical risks there is the chance of tracheal mucosal injury from the electrodes. The electrodes are biocompatible, medically approved, silver doped plastic. They are smooth and should not pose a risk to the tracheal mucosa. All endotracheal tubes will be sterilized after manufacture.
   3. Pulse oximetry measurements will be recorded. Pulse oximetry measurements are required during surgery. They add no additional risk.
   4. Cardiac output will be measured by iced-thermodilution and impedance cardiography every ten minutes during the operation. Samples will be obtained in with a ten cc volume of iced saline. A five hour operation would have a total volume of 800 cc of fluid. We routinely give a 1000 cc of saline to patients for cardiac cases so we will just reduce routine fluid administration to account for this extra volume from cardiac output measurements. Iced-injectate has no effect on patient temperature since active warming of the patient continues throughout the study.
   5. Impedance cardiography measurements will be high frequency (100 Khz), low current (4 mA), with input impedance of 1 mOhms. All circuits will be electrically isolated and approved by bio-medical engineering. There is always the risk of burns or shocks but these are rare.

   There is no interference with other electrical equipment on the patient (e.g., ECG, pacemaker, etc.) nor perception of the low current output by the patient. Impedance cardiography devices have been in use for over 30 years. They are fully approved by the FDA and some are commercially available for clinical use.

   6. The patient interview and review of the medical records will increase patient safety by identifying problems caused by the study.

8. Treatment and Compensation for Injury: There is minimal risks to the patient from this study. If there were any injury from the study, treatment will be available. If the patient is eligible for veteran's benefits, the costs of such treatment will be covered by the Department of Veterans Affairs. If not, the costs of such treatment may be covered by the Department of Veterans Affairs or the University of California, depending on a number of factors. The Department of Veterans Affairs and the University do not normally provide any other form of compensation for injury. For further information about this, the patient may call the V.A. District counsel at (415) 750-2288 or the office of the UCSF Committee on Human Research at (415) 476-1814.

9. Alternatives: Consent or refusal to participate in the study will have no effect on the patient's care, type of care, or access to care.

Informed consent: Informed, written consent is to be obtained from each study subject before any study-related procedures are begun. A full disclosure of the nature of the study is to be made in accordance with the guidelines described in the Code of Federal Regulations (21CFR 50.20).

Withdrawal: A patient may choose to or be withdrawn from the study for any reason, including:

a. Refusal of the patient to participate further; b. Refusal of the physician to allow patient participation. j. Costs to the subject: There will be no costs to the subject. The cost of all study related testing will be covered by Conmed.

k. Reimbursement of Subjects: None will be provided.

l. Confidentiality of records: All records will be confidential. All data will be analyzed by code number only. Code numbers will be kept at the VAMC under the control of the principal investigator. No reports will include patient identifiers.

5. Qualifications of Investigators: Dr. Arthur Wallace, M.D., Ph.D. has done medical research since 1978 including large animal, human clinical trials, and human physiological measurement. He has a Ph.D. in biomedical engineering and completed a medical internship, anesthesia residency, and a fellowship in cardiac anesthesiology. He has been a faculty member of anesthesiology at UCSF since 1992 during this time he has been the PI on several clinical trials in high risk patients including the Afterload, Warm Cardioplegia, Atenolol, Clonidine, EDRF Intraop, L-ARG, AVD Study, PCRRT, and Aneurysm Trials. Dr. Wallace has worked with impedance devices since 1991 and has been the chief scientific advisor on the ECOM projects since their inception in 1996. He has been responsible for the tube design, algorithm development, electronic design, in-vivo experiments, as well as the animal and human trials.

Inclusion Criteria:

1. Patients able to give informed consent.

2. Patients undergoing cardiac surgery which would routinely require:

   • an endotracheal tube
   • pulmonary artery catheter

Exclusion Criteria:

1. Patients who do not speak English.
2. Patients not competent to give informed consent.

Study analyses included the final n=101 who had the approved clinical electronics and clinical tube; no changes in tube, algorithm or electronic design were allowed during this part of the study. Experimental electronics and version of the tube were used to test and finalize the design and algorithm during enrollment of the earlier patient group.