Author + information
- Received December 27, 1999
- Revision received April 24, 2000
- Accepted June 21, 2000
- Published online November 1, 2000.
- Andrew J Rainbird, MBBS, FRACP∗,
- Patricia A Pellikka, MD, FACC∗,*,
- Vicky L Stussy, RN∗,
- Douglas M Mahoney, MS† and
- James B Seward, MD, FACC∗
- ↵*Reprint requests and correspondence to: Dr. Patricia A. Pellikka, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905
We compared a new two-stage transesophageal atrial pacing stress echocardiography (TAPSE) protocol with a standard dobutamine stress echocardiography (DSE) protocol.
Transesophageal atrial pacing stress echocardiography has been proposed as an efficient alternative to DSE.
Two-stage TAPSE (85% and 100% of age-predicted maximum heart rate) and DSE (5 to 40 μg/kg/min at 3-min stages with or without atropine) were both performed, in random sequence, in each patient of a study group of 36 patients. Regional wall-motion analysis, patient acceptance (1 = low, 5 = high), hemodynamics and duration for performing and interpreting tests were compared.
Transesophageal atrial pacing stress echocardiography was successful in 35 of the 36 patients (feasibility 97%). More TAPSE than DSE studies were called “ischemic” (37% vs. 14%; p = 0.005). Peak heart rate was higher with TAPSE (144 ± 18 vs. 129 ± 15 beats/min, p = 0.0001). Peak cardiac index (4.6 ± 2.1 vs. 5.1 ± 1.9 liters/min/m2, p = 0.14), patient acceptance score (4.2 ± 0.7 vs. 3.8 ± 1.3, p = 0.17) and study duration (14.2 ± 9.3 vs. 13.3 ± 3.3 min, p = 0.59) were similar. Recovery time (7.1 ± 7.6 vs. 16.2 ± 15.9 min, p = 0.0003) and interpretation time (9.1 ± 2.8 vs. 13.5 ± 4.4 min, p = 0.0001) were shorter for TAPSE than for DSE.
Two-stage TAPSE permits rapid evaluation of cardiac patients. Peak cardiac index and patient acceptance scores were similar for TAPSE and DSE. Ischemia was detected more often with TAPSE; this result was attributed to the higher peak heart rate obtained with this protocol.
Transesophageal atrial pacing stress echocardiography (TAPSE) has been proposed as an efficient, inexpensive alternative (1–4) to dobutamine stress echocardiography (DSE) for the detection of coronary artery disease (CAD). The heart rate can be rapidly increased, with the result that myocardial ischemia is produced in regions subtended by stenosed coronary arteries. By contrast to DSE, termination of pacing results in nearly instantaneous restoration of the patient’s intrinsic heart rate.
We reported on 100 patients who underwent DSE and TAPSE with the use of an 18F flexible pacing catheter (5). Transesophageal atrial pacing was started 10 beats above the resting heart rate and was increased 20 beats every 2 min until the target heart rate of 85% of age-predicted maximum heart rate or another end point was reached. Feasibility of TAPSE was 96%. Transesophageal atrial pacing studies were significantly shorter than those with DSE. There was excellent concordance between both tests for inducing specific regional wall-motion abnormalities.
In this study, we evaluated a new rapid two-stage protocol in which transesophageal atrial pacing was performed at 85% and 100% of age-predicted maximum heart rate, and we compared the results with those obtained with our standard dobutamine protocol. Studies were performed with the use of a new 10F pacing catheter, which was inserted orally. The target rate of 100% was used for transesophageal atrial pacing because, in our experience, the ischemic threshold was 10 beats/min higher with TAPSE than was observed with DSE (5). In addition, the stroke volume and cardiac output responses to the two forms of stress testing were compared by using Doppler methods (6).
All patients were outpatients more than 40 years of age who were scheduled for clinically indicated DSE for the evaluation of known or suspected CAD. No patient had atrial fibrillation or flutter or a permanent pacemaker or automatic implantable cardiac defibrillator, and none had moderate or severe valvular heart disease, gastroesophageal varices, unstable angina pectoris, recent myocardial infarction (within two weeks), untreated glaucoma, uncontrolled systemic hypertension (systolic blood pressure (BP) >190 mm Hg, diastolic BP ≥100 mm Hg) or inadequate transthoracic images. No patient was pregnant. Thirty-six patients met the study criteria and agreed to participate. Institutional approval had been obtained.
After fasting for 3 h, each patient underwent both two-stage TAPSE and DSE in a randomly determined sequence. The second test was started after a 30-min rest period, which began with the resolution of any stress-induced evidence of myocardial ischemia and after the heart rate had returned to within 10 beats of the baseline heart rate.
During each protocol, symptoms, heart rate, BP and adverse effects were recorded. Stroke volume and cardiac output were measured by the Doppler technique (6) at rest and at peak stress for both tests. The duration of each study (start of pacing catheter insertion to termination of pacing or start of dobutamine infusion to termination of dobutamine infusion) was recorded. Also recorded were duration of recovery period (termination of pacing or dobutamine infusion to termination of monitoring), time for interpretation of each study and videotape time. At any stage, mild sedation with midazolam, 1 mg to 2 mg, was given if necessary to alleviate anxiety, discomfort or gagging. At the completion of both studies, each patient completed a questionnaire, scoring the tests for comfort and acceptance on a scale of 1 (intolerable) to 5 (very satisfactory).
Transthoracic two-dimensional echocardiography was performed with standard cardiac ultrasound equipment. Atrial pacing was performed with bipolar esophageal cardiac pacing and a recording catheter housed in a 10F sheath connected to a Model 7A pulse generator and Model 3 preamplifier (TAPSCOPE-210-S and TAPSTRESS systems, CardioCommand, Tampa, Florida). All studies were recorded on 3/4-in. (1.9 cm) videotape. Images at each level were digitally stored on a Nova Microsonics system (Allendale, New Jersey). During TAPSE, the start–delay time interval to digital image acquisition was increased in accordance with the duration between the pacing spike and the R wave because the high-amplitude pacing spike triggered digital image acquisition.
Protocol for two-stage TAPSE
The TAPSE was performed by a cardiologist who had no prior experience with this technique (A.J.R.), under the direct supervision of a cardiologist who was experienced with TAPSE (P.A.P.). The oropharynx was anesthetized with 10% lidocaine spray, and the pacing catheter was introduced orally by instructing the patient to swallow while lying in the left lateral decubitus position. Catheter position was optimized by maximizing the size of the esophageal P wave on the electrocardiogram (ECG). Pacing was initiated at 85% of the age-predicted maximum heart rate and at the lowest current that provided stable atrial capture. Images were then recorded. If there were no new regional wall-motion abnormalities and if the patient was tolerating the study, the pacing rate was increased to 100% of the age-predicted maximum heart rate. After image acquisition, the pacing catheter was removed.
Blood pressure was recorded at each stage of the protocol. The test was terminated if there were intolerable symptoms, significant arrhythmia, ECG evidence of significant ischemia (ST depression ≥2 mm, at 80 ms after the J point), hypertension (BP >220 mm Hg systolic, ≥110 mm Hg diastolic) or hypotension (≤90 mm Hg systolic). Atropine (0.5 mg at 1-min intervals to a maximum dose of 2.0 mg) was administered if Wenckebach heart block occurred.
Protocol for DSE
The DSE was performed according to our usual protocol (7) with 3-min stages and a maximum dose of dobutamine of 40 μg/kg/min. Atropine, starting at 0.5 mg intravenously, repeated every minute to a maximum total dose of 2.0 mg, was administered at the start of the 40-μg/kg/min stage, as needed, to augment the heart rate. Standard end points, including achievement of a heart rate 85% of age-predicted maximum, were used.
Image analysis and interpretation
Dobutamine and pacing studies were scored separately by an experienced echocardiographer blinded to the clinical information and the results of the other study. Interpretation was based on review of both videotape and digitized images. Regional wall motion was scored at rest and with stress in each of 16 segments by the standard 1 through 5 scoring system of the American Society of Echocardiography (8). A normal response to stress was characterized by normal or hyperdynamic systolic function of all left ventricular segments. Infarction was characterized by a resting wall-motion abnormality that did not change with stress. The development of new wall-motion abnormalities in one or more segments or worsening of a resting wall-motion abnormality, including deterioration after initial improvement, was considered as an ischemic response (9). Ejection fraction at baseline and at peak stress was determined by the method of Quinones (10) or by visual estimation (11).
Follow-up was obtained by review of medical records and scripted telephone interviews. Cardiac death, myocardial infarction, coronary artery bypass surgery and percutaneous transluminal coronary angioplasty were recorded.
Statistical analysis and estimation of agreement of results
Wilcoxon matched-pairs signed rank and McNemar tests were used for paired continuous data and paired proportions, respectively. A p value of <0.05 was considered significant. The degree of agreement between results of DSE and those of TAPSE was estimated by the kappa coefficient (1 = perfect agreement, >0.5 = good agreement, and >0.8 = excellent agreement) (12). The degree of agreement for wall segment scoring was estimated as the average kappa for all 16 segments, with the standard error estimated by means of resampling techniques (13).
Transesophageal atrial pacing could not be performed in 1 of the 36 patients enrolled in this study (feasibility 97%). That patient had a large hiatus hernia. Results in the remaining 35 patients, who completed both forms of testing, are included here.
The demographics for the patients are described in Table 1. Stress testing was performed for pre-operative evaluation of noncardiac surgery in 20 patients (57%), for evaluation of known CAD in 14 (40%) or for evaluation of symptoms of dyspnea or chest pain in 3 (9%) patients. Thirty of the patients were unable to exercise, because of orthopedic limitations in 16 (46%), peripheral vascular disease in 9 (26%), lung disease in 3 (9%) and debility in 2 (6%).
The average pacing current was 18 ± 4 milliamps. The mean peak dose of dobutamine was 33 ± 10 μg/kg/min. Four patients (11%) required midazolam during TAPSE (dose 1.5 ± 0.6 mg) but none during DSE. Atropine was used more often with TAPSE (63% vs. 37%, p = 0.05), but doses were similar (0.95 ± 0.67 vs. 0.62 ± 0.74 mg, p = 0.4). Esmolol was used frequently after DSE (49% vs. 0%, dose 39 ± 18 mg) to alleviate symptoms.
Rest and peak systolic and diastolic BPs, rest heart rate, double product, rest stroke volume and rest ejection fraction were similar for the two forms of stress. A higher peak heart rate was achieved with TAPSE compared with DSE (144 ± 18 vs. 129 ± 14 beats/min, p = 0.0001). Target heart rate was achieved in 86% with TAPSE (100% of age-predicted maximum heart rate) and 89% with DSE (85% of age-predicted maximum heart rate), p = 0.74. Stroke volume decreased from rest to peak stress with TAPSE (82 ± 18 vs. 62 ± 24 ml, p = 0.0001); there was no change in stroke volume from rest to peak stress for DSE (81 ± 17 vs. 77 ± 22 ml, p = 0.24). The peak stroke volume was greater for DSE than for TAPSE (77 ± 22 vs. 62 ± 24 ml, p = 0.0001). However, because of the greater heart rate achieved with TAPSE, the peak cardiac index was similar (4.6 ± 2.1 vs. 5.1 ± 1.9 liter/min/m2, p = 0.14). The ejection fraction increased at peak stress for DSE (62 ± 12 vs. 69 ± 16%, p = 0.0001) but did not change from rest to peak stress with TAPSE. Ejection fraction was greater at peak stress for DSE compared with TAPSE (69 ± 16 vs. 61 ± 16%, p = 0.0001) (Tables 2 and 3). ⇓⇓
Wall-motion score index at rest was similar for the two techniques. At peak stress, the peak wall-motion score index was greater for TAPSE compared with DSE (1.5 ± 0.6 vs. 1.4 ± 0.5, p = 0.001). More ischemic studies were reported for TAPSE than for DSE (37 vs. 14%, p = 0.005) (Table 2). Kappa coefficient values for abnormal individual wall segments between TAPSE and DSE at rest and peak stress were 0.82 ± 0.10 and 0.59 ± 0.08, respectively.
Times needed to perform TAPSE and DSE were similar (14.2 ± 9.3 vs. 13.3 ± 3.3 min, p = 0.59). However, the last 10 TAPSE studies could be performed in less time than the first 10 TAPSE studies (8.5 ± 3.3 vs. 21.5 ± 11.8 min, p = 0.01) and in less time than the last 10 DSE studies (8.5 ± 3.3 vs. 13.2 ± 2.4 min, p = 0.01). No difference was seen in a similar comparison between the last and first 10 DSE studies (14.6 ± 3.7 vs. 13.2 ± 2.4 min, p = 0.14). Times for recovery, interpretation and videotape recording were shorter for TAPSE (Table 4).
Overall, patient acceptance scores were similar for TAPSE and DSE (4.2 ± 0.8 vs. 3.8 ± 1.3, p = 0.17); however, more patients rated TAPSE 4 or 5 compared with DSE (89% vs. 64%, p = 0.02). Side effects occurred less often with TAPSE than with DSE (34% vs. 83%, p = 0.0001). No major side effects occurred with TAPSE. Nonsustained ventricular tachycardia occurred in one patient with DSE. Minor side effects with DSE included palpitations 60%, chest pain 20%, nausea 11%, headache 9%, dyspnea 6% and tremor 3%; minor side effects for TAPSE included gagging 20% and chest pain 6%.
Follow-up was complete in 34 patients (97%); the mean duration was 9.0 ± 2.8 months. Five patients underwent coronary angiography, and at least one coronary stenosis of more than 70% of the luminal diameter was found in each of these patients. Both TAPSE and DSE were positive for the detection of ischemia in each of these five patients. Myocardial infarction and cardiac death occurred in one patient, myocardial infarction and coronary angioplasty in one patient, coronary artery bypass operation in one patient and coronary angioplasty in another patient. Both TAPSE and DSE were positive for the detection of ischemia in each of these four patients.
The results of this study, in which a specially designed 10F atrial pacing catheter was used, confirm our previous finding that TAPSE is a safe, feasible and effective method for the assessment of patients with suspected CAD. This rapid two-stage technique, using 85% and 100% of age-predicted maximal heart rate, can be performed and reported in a shorter period of time than DSE. Transesophageal atrial pacing was performed by an operator who had no previous experience with this technique. Experience was rapidly achieved and resulted in shorter study times, as seen with the comparison between the first and last 10 TAPSE studies. Omitting the acquisition of Doppler measurements would have reduced the time needed to perform the study. The recovery period was significantly shorter for the TAPSE protocol, despite the frequent use of esmolol in the recovery period of the DSE protocol.
Cardiac stroke volume and cardiac index, obtained by Doppler techniques, have not been previously reported for transesophageal atrial pacing. Although stroke volume fell from rest to peak stress, we were able, by obtaining a higher peak heart rate with TAPSE, to achieve an equivalent cardiac index with the two techniques. The hemodynamic changes seen with TAPSE were similar to results obtained from previous studies using transvenous atrial pacing (14–19). Peak stroke volumes and ejection fractions were higher with DSE because of the positive inotropic action of dobutamine. No differences in systolic or diastolic BPs were noted.
Agreement for individual wall segments of the two tests was excellent at rest (kappa = 0.82) and good at peak stress (kappa = 0.59). Myocardial ischemia was detected more frequently with transesophageal atrial pacing, owing, we believe, to the higher peak heart rate achieved with TAPSE. Additional reasons include hyperdynamic response and left ventricular cavity obliteration seen with DSE, which may obscure hypokinetic regions, and the lack of gradual warmup or ischemic conditioning of the coronary bed with the rapid-pacing protocol. However, because coronary angiography was not routinely performed, false-positive results with TAPSE are also a possibility.
The transesophageal atrial pacing catheter used in this study, TAPSCOPE-210-S, was a new, smaller (10F) catheter. It was highly effective, adequate atrial capture being obtained in all but one patient. In this patient, a previously known large hiatal hernia may have interfered with our obtaining correct catheter position and esophageal wall contact. Positioning of the pacing catheter was not difficult and was very well tolerated by the patients. There were fewer symptoms, and less midazolam was required for sedation with this smaller atrial pacing catheter compared with our previous experiences with the larger (18F) catheter (5). The smaller catheter could also have been placed nasally; this might have reduced the frequency of gagging.
The number of patients studied was small. Coronary angiographic data were insufficient for an estimation of the relative diagnostic accuracy of TAPSE and DSE. The use of midazolam in four patients during TAPSE may have affected their preference of forms of testing.
Transesophageal atrial pacing appears to be an acceptable technique for the detection of CAD. Potential roles include its use as an alternative to DSE—especially in patients who experience intolerable symptoms with dobutamine—and in patients who have chronotropic incompetence, including those taking beta-blocking medications, in whom peak heart rate may be difficult to achieve with dobutamine. This technique may also be useful in the emergency room assessment of chest pain syndromes, as results can be obtained quickly. The rapid return of the heart rate to pretest level may confer additional safety on this test. The shorter recovery, interpretation and videotape times offer potential cost savings for this technique as a substitute for DSE.
Finally, as no major adverse side effects occurred with two-stage TAPSE, this rapid technique could be simplified further to one stage using 100% of the age-predicted maximum heart rate, which may add other benefits to those noted above.
☆ Dr. Rainbird was supported by a grant from the Queensland Heart Clinic, Brisbane, Australia, and Mallinckrodt Inc., St. Louis, Missouri.
Presented as a poster at the American Society of Echocardiography Annual Meeting, July 1999, Washington, DC. Additional support was provided by a grant from CardioCommand Inc., Tampa, Florida. Nothing in this publication implies that the Mayo Foundation endorses the products of CardioCommand Inc.
- blood pressure
- coronary artery disease
- dobutamine stress echocardiography
- transesophageal atrial pacing stress echocardiography
- Received December 27, 1999.
- Revision received April 24, 2000.
- Accepted June 21, 2000.
- American College of Cardiology
- Don Michael T.A.,
- Rao G.,
- Balasingam S.
- Iliceto S.,
- Sorino M.,
- D’Ambrosio G.,
- et al.
- Kamp O.,
- DeCock C.C.,
- Visser C.A.
- Chapman P.D.,
- Doyle T.P.,
- Troup P.J.,
- Gross C.M.,
- Wann L.S.
- Lee C.Y.,
- Pellikka P.A.,
- McCully R.B.,
- Mahoney D.W.,
- Seward J.B.
- Schiller N.B.,
- Shah P.M.,
- Crawford M.,
- et al.
- Roger V.L.,
- Pellikka P.A.,
- Oh J.K.,
- Bailey K.R.,
- Tajik A.J.
- Quinones M.A.,
- Waggoner A.D.,
- Reduto L.A.,
- et al.
- Efron B.,
- Tibshirani R.
- Linhart J.W.,
- Hildner F.J.,
- Barold S.S.,
- Lister J.W.,
- Samet P.
- Dwyer E.M. Jr.
- Slutsky R.,
- Watkins J.,
- Peterson K.,
- Karliner J.
- Ross J. Jr.,
- Linhart J.W.,
- Brauwald E.