| ||||||||||||||
|
|
|||||||||||||
Department of Anesthesiology, Julius-Maximilians-Universität, Würzburg, Germany
Address correspondence and reprint requests to Dr. C.-A. Greim, Klinik für Anaesthesiologie der Universität Würzburg, Josef-Schneider-Str. 2, 97080 Würzburg, Germany. Address e-mail to cgreim{at}anaesthesie.uni-wuerzburg.de
| Abstract |
|---|
|
|
|---|
Implications: Transnasal insertion of a miniaturized monoplane transesophageal echocardiography (TEE) probe was studied in endotracheally intubated patients. Nasal passage was well tolerated even by patients with only mild sedation. Imaging quality was similar to conventional transoral monoplane TEE with larger transducers, but technical restraints cause a deficit in complete cardiac diagnosing obtained with multiplane TEE.
| Introduction |
|---|
|
|
|---|
Recent publications suggest that transnasal TEE with a miniaturized monoplane TEE probe may provide advantages compared with conventional transoral TEE (6,7). In the current study, we assumed that transnasal insertion and examination with this probe would be easy in orally intubated patients independent of their level of sedation. The technical features of the miniaturized TEE probe may facilitate the single-plane approach to left ventricular (LV) function, which is useful to control LV loading and contractility (8,9). However, such probes may also limit the quality of a complete echocardiographic examination. Therefore, we evaluated cardiac diagnostic ability and monitoring of LV function with a transnasal TEE technique using a miniaturized esophageal probe.
| Methods |
|---|
|
|
|---|
Sedation was continuously provided with IV fentanyl and midazolam. Patients were categorized as deeply sedated when their ventilation was controlled ventilation and there was no response to verbal commands, transnasal pharyngeal suction of saliva, or stimulation of gagging reflex. Spontaneously breathing patients with respiratory support who were aroused at verbal commands or had discomfort on transnasal suction or stimulation of gagging reflex were categorized as mildly sedated.
Transnasal TEE was goal-oriented and focused on the assessment of LV function (n = 20), right ventricular function (n = 8), valvular function (n = 2), suspected endocarditis (n = 4), exclusion of intracardiac thrombi or mass (n = 6), and suspected aortic pathology (n = 2). All patients had a complete TEE examination of standard transgastric and transesophageal cardiac cross-sections (10). For comparison, all patients with deep sedation were additionally investigated with transoral monoplane and multiplane TEE.
The prototype of a recently developed miniature TEE probe in combination with a commercially available ultrasound unit (Sonos 2500; Hewlett-Packard, Andover, MA) was used for transnasal TEE. The probe is similar in diameter to a nasogastric 18-gauge tube and allows ante- and retroflexion of the transducer over 135°. Ultrasound is emitted and received in the monoplane fashion by 32 elements. For conventional TEE, a standard multiplane 5-MHz transducer probe with 64 elements that allows for only 110° anteflexion and 70° retroflexion was used with the same unit.
The nasal passages of the patients were prepared by the administration of 0.4 mL of 1% xylometazolin solution for vasoconstriction, followed by topical anesthesia with lidocaine jelly 2%. Nasal intubation with the prototype TEE probe has been described elsewhere (6) and was attempted starting with the right ostium and continuing with the left if the first attempt failed. Sedation was deepened if the patient expressed discomfort by activation of protective reflexes, e.g., by moving of the head, during nasal probe insertion. Nasal intubation was abandoned after three unsuccessful attempts, and the oral route was then chosen. All TEE studies were performed by one of two experienced investigators who had participated in a pilot study (7).
Echocardiograms of specific tomographic planes obtained from transnasal and transoral TEE were recorded on video tape and later analyzed by two independent observers. Each imaged cross-section was categorized as complete versus incomplete when it allowed for plane-related interpretation of left and right ventricular myocardial wall motion, mitral and tricuspid valvular function, evaluation of the left atrial appendage, main and right pulmonary arteries, or imaging of the aortic arch and ascending aorta. The numbers of complete versus incomplete imaging planes from both observers were averaged, and the average values from the two methods were then analyzed for a significant difference by using the
2 test. To assess the diagnostic capability of transnasal TEE, findings with this technique were compared with those of conventional monoplane and multiplane TEE. To compare the quantification by transnasal and conventional monoplane TEE, end-diastolic and end-systolic cross-sectional dimensions of left and right ventricular cavities, and Doppler flow velocities at the mitral valve, pulmonary vein, and main pulmonary artery were determined in patients who were examined by using both methods. Measurements were obtained from echocardiograms that were recorded when the patient was hemodynamically stable, and heart rate and mean arterial pressure during the two investigations did not vary by more than 15%. Endocardial borders and Doppler curves were manually traced and averaged from three cardiac cycles based on recommendations of the American Society of Echocardiography (12). The mean error of both methods was calculated as (100 x [mTNT - mCONV])/(0.5 x [mTNT + mCONV]), where mTNT = transnasal TEE measurement and mCONV = conventional TEE measurement. Interobserver and intraobserver variabilities were calculated as previously described (9). To compare end-diastolic and end-systolic LV area measurements obtained from transnasal and conventional monoplane TEE, a linear regression model and the Bland-Altman analysis of agreement were used.
| Results |
|---|
|
|
|---|
|
Data from 26 patients were available for comparison of transnasal and conventional transoral TEE. Specific imaging properties of both techniques are demonstrated in Table 2. Complete transgastric LV short-axis and transesophageal four-chamber long-axis views were obtained by all techniques without significant difference. Contrast and brightness of the transnasal TEE images were of worse quality than conventional TEE images; however, overall visual information on anatomical structure and myocardial motion was unaffected. Maximal power of the miniaturized TEE transducer was required in two thirds of patients to focus the transgastric short-axis view for evaluation of LV filling and contraction. No additional information on cardiac function was obtained by conventional transoral TEE in the monoplane approach. In the multiplane mode, however, conventional TEE revealed information for five patients (19%) that was not available from monoplane TEE. The diagnoses missed by monoplane TEE were left atrial thrombi (n = 2) and LV wall motion abnormalities (n = 3).
|
In 22 of 26 patients examined with transnasal and conventional monoplane TEE, tomographic cardiac views and hemodynamic conditions allowed for direct comparison of both methods. Analysis for agreement in quantification revealed that both methods diverged by less than 7% and 9% from the target values for area and diameter, calculated as the average value of the two techniques. This target value was chosen because, in the clinical setting, there is no reference for both echocardiographic and Doppler-echo measurements. Method-related differences in end-diastolic and end-systolic LV areas were within a 95% confidence range of 7.6 cm2 and 4.8 cm2, respectively (Figs. 1 and 2). Discrepancies in blood flow velocity measurements and the velocity curve integral were extremely large (Table 3), but the relative error in the ratio of early-diastolic to late-diastolic or systolic peak velocities was <4%. Intra- and interobserver variabilities for area, dimension, and velocity measurements were less than 5%, 9%, and 13%, respectively.
|
|
|
| Discussion |
|---|
|
|
|---|
The transducer and shaft diameters of the miniaturized TEE probe are small enough for transnasal rather than transoral probe insertion to the esophagus. This provides a number of advantages for TEE in endotracheally intubated patients. With conventional TEE, the use of a bite guard is common to protect the patient, as well as the probe, from dental trauma or damage. Intraoral guidance to introduce the probe or even laryngoscopy is often required in the anesthetized patient when blind insertion of the TEE probe proves difficult, and may cause pharyngeal injury or perforation (5,12). Possibly because of the anatomic guidance along the nasopharynx, esophageal insertion of the miniaturized TEE probe in the current study required no further instrumentation, whereas laryngoscopy was required in four patients undergoing transoral TEE. However, all the attempts of transoral TEE were finally successful, whereas the transnasal approach failed in 10% of patients. Further studies should focus on the causes for such limitations and the value of supportive equipment (such as nasal trumpets) to facilitate transnasal TEE.
A major advantage of transnasal TEE is its applicability in mildly sedated patients with an endotracheal tube. Our study shows that transnasal passage of the miniaturized TEE probe is well tolerated even by patients who are not fully cooperative. Topical anesthesia with lidocaine jelly 2% did prevent physical stress in most patients, but deepening of sedation was necessary in 18% of patients. In those patients, a more profound topical anesthesia, e.g., by applying lidocaine aerosol 10% or swabs soaked with cocaine 4%, could have allowed for transnasal TEE without additive analgesia or sedation. This would be advantageous in patients in whom additional sedation is undesirable, e.g., during weaning from prolonged artificial respiration.
When two diagnostic methods compete, studies are required to determine which technique is superior. The current study in 42 endotracheally intubated patients support the conclusions of an investigation in 139 nonintubated patients that valvular and ventricular function can be assessed by transnasal TEE with an imaging quality close to conventional monoplane TEE standards (6). This study additionally shows that there are no major quality differences between transnasal and conventional monoplane TEE with regard to LV short-axis and four-chamber long-axis imaging. Both views are useful in detecting acute changes in LV preload (8,9) or myocardial contraction (13) and to diagnose particulate or gaseous embolisms (14,15) that occur during surgical operations. Our study, however, shows the diagnostic inferiority of both transnasal and conventional monoplane TEE compared with multiplane TEE. Compared with multiplane TEE, monoplane TEE missed the diagnosis in approximately 20% of patients. Because the miniaturized probe lacks the superior technology of the multiplane examination mode, transnasal TEE is less suited for a comprehensive echocardiographic examination. Goal-oriented transnasal TEE in our study nevertheless led to a direct change in patient management in 47% of patients. The new diagnosis or certain exclusion of suspected cardiac pathology did not immediately affect treatment in all patients but could have had an impact on long-term patient prognosis and management. Outcome studies are required in this context but should clearly define the expected benefit from transnasal monoplane TEE in contrast to conventional multiplane TEE.
An important issue in monitoring rather than diagnosing cardiac function is the quantification of echocardiographic parameters. The intraobserver and interobserver variability of one echocardiographic method may be as high as 15% in both adults and children (16,17). In our study, the discrepancies in area and diameter measurements between both methods were of similar magnitude and could be the result of observer variabilities only. However, because the bias was small, the confidence limits in the LV area measurements obtained by transnasal TEE were large. Therefore, the reproducibility of conventional monoplane TEE measurements by transnasal TEE was only moderate. With Doppler-echocardiographic measurements, the relative error of transnasal TEE was >40% and, thus, far beyond tolerability. Calculated values for systolic to diastolic ratios of valvular flow, however, were prone to a relative error <5%. Because such ratios serve for certain interpretation of cardiac filling rather than cycle-related values (18,19), this information obtained by transnasal Doppler-TEE measurements may be the same as those obtained by the transoral method. A shortcoming of the study in this regard is that both techniques were not performed simultaneously, and differences in measurement may, in part, be explained by slightly different probe locations during echocardiographic recording. In addition, despite all precautions, minor changes in cardiac loading in the time interval between both investigations cannot be excluded and may have affected intracardiac volume, dimensions, and blood flow velocity.
Apart from endotracheal malpositioning and nasal bleeding that required treatment in only one patient, no complications related to transnasal TEE were observed. Another potential source for patient traumatization is the maximal ultrasound energy that, in most patients in this study, was intermittently required and bears a risk for thermal injury (20). Because of the enhanced probe angulation and smaller size, the pressure of the probe on the nasopharyngeal, esophageal and gastric mucosal walls could also be higher than previously determined for conventional TEE probes (20). The occurrence of bacteremia after transnasal TEE also requires investigation. In a study in 139 patients undergoing conventional transoral TEE, the incidence of bacteremia was 1.4% with a 95% probability that the true rate was <5.1% (21). Because the frequency of positive blood cultures was independent of prophylactic antibiotic treatment, routine antimicrobial prophylaxis before TEE seemed not to be justified; however, the occurrence of actual infection after TEE was not assessed (21). In the current study, the high incidence of minor nasal bleeding during probe insertion indicates that transnasal TEE may carry an increased risk for bacterial translocation because a highly vulnerable tissue surface is exposed to the probe.
In summary, this study demonstrates a major benefit of transnasal TEE for cardiac examination in ventilated patients. The technique can be performed even in mildly sedated patients with an endotracheal tube who are easily aroused but who are not fully oriented or cooperative. With regard to cardiac diagnostic capability, transnasal TEE is as good as transoral monoplane TEE, but it lacks the superior diagnostic quality of a conventional multiplane TEE examination, e.g., of the left atrial appendage. Four-chamber view imaging with the technique is suitable to assess common causes for hemodynamic instability, such as hypovolemia or mitral valve dysfunction, but LV short-axis imaging may not be satisfactory to examine myocardial wall motion.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. A. Aronson Transnasal Placement of Biplane Transesophageal Echocardiography Probe Intraoperatively in an Adolescent with Congenital Heart Disease Anesth. Analg., December 1, 2003; 97(6): 1617 - 1619. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Zimmermann, C. Greim, H. Trautner, U. Sagmeister, K. Kraemer, and N. Roewer Echocardiographic Monitoring During Induction of General Anesthesia with a Miniaturized Esophageal Probe Anesth. Analg., January 1, 2003; 96(1): 21 - 27. [Abstract] [Full Text] [PDF] |
||||
![]() |
C.-A. Greim, H. Trautner, K. Kramer, P. Zimmermann, C. C. Apfel, and N. Roewer The Detection of Interatrial Flow Patency in Awake and Anesthetized Patients: A Comparative Study Using Transnasal Transesophageal Echocardiography Anesth. Analg., May 1, 2001; 92(5): 1111 - 1116. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|