- Email: [email protected]
Echocardiographic assessment of tricuspid regurgitation during ventricular demand pacing
EchocardiographicAssessmentof Tricuspid Regurgitation During Ventricular Demand Pacing DENNIS E. MORGAN, MD, ROSEMARY NORMAN, RN, ROXROY O. WEST, MD, and GARY BURGGRAF, MD
Twenty patients from our pacemaker clinic population were assessed clinically and by saline contrast echocardiography (subxiphoid view) to determine the prevalence of tricuspid regurgitation (TR) and, if TR was present, its mechanism. The patients had no known TR before lead placement, a single transvenous right ventricular pacing lead present more than 6 months (mean 52, range 7 to 138), ventricular demand pacing alternating with sinus rhythm and rate programmability. Each patient was studied in sinus rhythm and during ventricular pacing. Using the criterion of inferior vena cava (IVC) contrast reflux during ventricular systole to diagnose TR, no patient had evidence of TR in sinus rhythm, consistent with clinical examination. During ventricular demand pacing, jugular venous pulse cannon A waves developed in 10 patients, and 18 patients (including these 10) had IVC contrast reflux during ventricular
systole. Analysis of the timing of IVC reflux revealed its close temporal relation to the timing of atrial systole rather than a fixed timing during ventricular systole. This reflux occurred with loss of normal atrioventricular (AV) synchrony and the underlying mechanism in all cases was shown to be right atrial contraction against a closed tricuspid valve. Two patients who did not have such a pattern with pacing maintained normal AV synchrony. These obserrations indicate that: (1) TR is an uncommon accompaniment of ventricular demand pacing; (2) the jugular venous pulse and IVC echocardiographic contrast patterns during ventricular demand pacing simulate TR when AV synchrony occurs; and (3) the IVC contrast pattern of pacing-induced AV synchrony is best termed the cannon A wave synchronous pattern. (Am J Cardiol 1986;58:1025-1029)
R i g h t ventricular pacing will result in hemodynamic compromise in some patients--the so-called pacemaker syndrome. This has been attributed to atrioventricular (AV) valve regurgitation, among other mechanisms. 1,2Animal studies show that mitral regurgitation may occur with ventricular pacing and that its severity varies with the electrode site. 3 Hypotension with mitral regurgitation has been described in a patient during right ventricular pacing. 4 Tricuspid valve malfunction due to a right ventricular pacemaker lead has been noted. 5 Little is known, however, about the prevalence or mechanism of AV valve regurgitation in humans with ventricular pacemakers. Because the electrode crosses the tricuspid valve, patients with ventricular pacemakers are at risk of tri-
cuspid regurgitation (TR). Causes for TR include direct interference with tricuspid valve closure, inappropriately timed atrial systole due to asynchronous pacing, and the disordered sequence of right ventricular and papillary muscle activation during pacing. We assessed the prevalence of TR by M-mode contrast echocardiography in patients with a ventricular demand pacemaker and determined the mechanism of TR, if present.
From the Division of Cardiology, Queen's University, Kingston, Ontario, Canada. Manuscript received March 3, 1986; revised manuscript received July 1, 1986, accepted ]uly 2, 1986. Address for reprints: Gary W. Burggraf, MD, Hotel Dieu Hospital, Division of Cardiology, Kingston, Ontario, Canada K7L 5G2.
1025
Methods Study population: We selected 20 patients (10 men, 10 women) from our pacemaker population who fulfilled the following criteria: no known TR before pacemaker lead placement, a single transvenous right ventricular pacing lead present at least 6 months, ventricular demand pacing alternating with sinus rhythm, and rate programmability. The age range was 51 to 77 years (mean 69). Average duration of lead placement was 52 months (range 7 to 138). The indication for pacing was sick sinus syndrome in 11 patients and AV block in 9 patients. Eight patients had known ath-
1020
TRICUSPIDREGURGITATION DURING VENTRICULAR PACING
FIGURE 1. Subxiphold 2-dimensional echogram and corresponding M-mode recording with simultaneous electrocardiogram used to time occurrence of inferior vena cava (IVC) contrast reflux. RA =
right atrium.
erosclerotic coronary artery disease and 1 patient had pulmonary hypertension secondary to recurrent emboli. Methods: After the patient gave informed consent, an electrocardiogram was recorded to determine if the patient was in sinus rhythm or ventricular pacing. The patient then underwent a physical examination to look for clinical signs of TR, which included jugular venous pulse abnormality, left parasternal systolic murmur and hepatic pulsation. Echocardiography was then performed with a Hewlett-Packard phased-array instrument {model 77020A} using a 3.5-MHz transducer. Strip-chart re-
cordings were obtained with a Honeywell recorder at paper speeds of 50 mm/s for the M-mode contrast echocardiograms and at 50 or 100 mm/s for the jugular venous pulse and phonocardiogram tracings. Saline contrast echocardiograms were recorded in the subxiphoid view as described by Lieppe et al. 6 In this view 2-dimensional visualization outlined the right atrium, hepatic veins and the inferior vena cava {IVC} and a simultaneous M-mode echocardiogram was recorded from the IVC just proximal to its junction with the right atrium {Fig. 1). Boluses of agitated saline solution {10 ml} were then administered through a right antecubital vein while an M-mode recording of the IVC was obtained. Using this method TR may be diagnosed when contrast appears in the IVC during ventricular systole. e-8 Results were obtained at least in duplicate, with each patient undergoing simultaneous M-mode and 2dimensional visualization to ensure adquate bolus delivery to the right heart. Jugular venous and phonocardiographic tracings were recorded from the 10 patients in whom a clinical abnormality of the jugular waveform was apparent. To record studies in both paced and sinus rhythm, the heart rate was reprogrammed up or down, depending on the initial rhythm. This usually required a rate change of 5 to 15 beats/min and was successful in all patients. Subsequently, the clinical examination, contrast echocardiogram, jugular venous pulse recordings and phonocardiograms were repeated. Blood pressure was determined in both rhythms, and when the jugular venous waveform was clinically abnormal it was recorded in both rhythms for comparison. All studies were performed by the same examiners.
Results In sinus rhythm none of the 20 patients displayed clinical evidence of TR, and their contrast M-mode
FIGURE 2. Top, M-mode contrast inferior vena cava (IVC) echogram during sinus rhythm. Vertical a r r o w Indicates contrast reflux with atrial depolarIzatlon. This Is the pattern of "a-wave synchronous r e f l u x . " Bottom, normal Jugular venous pulse (JVP) tracing, electrocardiogram (ECG) and phonocardiogram. Timing of the a wave Is the same as IVC reflux. LLSB = left lower sternal border phonocardiogram; $1 and $2 = first and second head sounds.
November 1, 1986 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 58
echocardiographic findings were normal. The echocardiograms showed either no IVC reflux after bolus injection (10 patients) or reflux that occurred only with atrial contraction and before ventricular systole (10 patients). Figure 2 shows this reflux pattern, the socalled a wave synchronous pattern, which is normal. 8 Ten of the 20 patients also had jugular venous recordings done in sinus rhythm, and all findings were normal (Fig. 2). Thus, in sinus rhythm no patient had clinical or echocardiographic evidence of TR. In ventricular paced rhythm at rates slightly above the patient's sinus rate, 10 of 20 patients displayed a clinically detectable and recordable jugular venous pulse abnormality in the form of cannon A waves. No patient had a new systolic parasternal murmur or detectable hepatic pulsations. One patient had a drop in blood pressure from 170/90 to 130/80 mm Fig with pacing, but remained asymptomatic. In the other patients systolic blood pressure decreased less than 10 mm Fig. Eighteen patients had systolic IVC reflux with pacing, which occurred only when there was loss of AV synchrony (Fig. 3). The 2 patients who failed to have IVC reflux during pacing fortuitously maintained normal AV synchrony. Figure 4 shows the pattern of "variable systolic reflux" that occurred in 13 subjects. The upper panel displays a simultaneous electrocardiogram and the Mmode contrast study and the lower panel includes a jugular venous pulse and phonocardiographic recording. The electrocardiographic P waves are labeled and the onset of contrast appearance is indicated by arrows. Contrast appears during ventricular systole, but the timing of systolic reflux is variable. In the first cycle, contrast appears in the IVC 360 ms after the pacemaker spike, compared with 180 ms in the last cycle. However, the P waves consistently closely pre-
1027
FIGURE 3. M-mode contrast inferior vena cava (IVC) echogram showing appearance of reflux when normal atrloventrlcular synchrony Is lost during pacing. Note the lack of reflux when the P wave of the electrocardiogram precedes the pacemaker spike In cycle 4. Abbreviations as In Figures 1 and 2.
cede the contrast appearance. Seven of these 13 patients had jugular venous pulse recordings in both rhythms, and all recordings revealed that the waveform was normal in sinus rhythm or when AV synchrony was maintained despite pacing (cycles I and 2 of Fig. 4, lower). However, as AV synchrony was lost, cannon A waves developed that bore the same relation to the electrocardiographic P wave as did the contrast appearance (Fig. 4, upper) and showed the same variable timing in systole. This variable systolic reflux pattern can thus be attributed to right atrial contraction against a closed tricuspid valve. Figure 5 illustrates the pattern of "fixed systolic reflux" that occurred in 5 subjects. In contrast to Figure 4, the IVC reflux occurs at a constant time during ventricular systole. Once again, the contrast reflux occurs soon after the electrocardiographic P wave, which in this case maintains a constant time relation to the QRS complex. Also, the timing of systolic reflux is later
i ~
FIGURE 4. Top, M-mode contrast inferior vena cava (IVC) echogram during asynchronous pacing shows that reflux bears a constant temporal relation to the P wave of the electrocardiogram (ECG) and occurs st variable times in ventricular systole. This is the variable reflux pattern. Bottom, corresponding jugular venous pulse (JVP) tracing. Cannon A waves occur according to the timing of the electrocardiographic P wave. HV = hepatic vein; other abbreviations as in Figures 1 and 2.
1020
TRICUSPIDREGURGITATION DURING VENTRICULAR PACING
in onset than expected for TR. The jugular venous recording in the lower panel illustrates the pattern in 3 patients who had tracings available for both rhythms. Again, cannon A waves developed during paced rhythm and the large systolic waves occur later than one might expect for a tricuspid regurgitant venous wave. This fixed systolic reflux pattern can thus be attributed to right atrial contraction against a closed tricuspid valve.
Discussion The prevalence or mechanism of pacemaker-induced TR or mitral valvular regurgitation has not been defined in patients with or without the pacemaker syndrome. We thus attempted to determine the prevalence of TR in a general pacemaker population and elucidate the mechanism of TR if it was present. Our patient selection allowed us to determine whether TR was due to a direct effect of the electrode, in which case it would be present in sinus rhythm, or whether it was due to an indirect effect {loss of normal AV synchrony or abnormal ventricular depolarization}, where it would occur only during pacing. We believe that our patients were representative of the'general pacemaker population because although we applied select entry criteria, none of these would have excluded patients at risk for TR or the pacemaker syndrome. Although TR has been previously reported in patients with pacemakers, we did not see clinical or echocardiographic evidence of TR in the 20 patients during sinus rhythm. 5 If one relied solely on clinical examination, one might underestimate the prevalence of TR because clinical recognition of mild to moderate TR is often difficulty However, our results were conf i r m e d b y subxiphoid contrast M-mode echocardiog-
raphy, which has a sensitivity of 88 to 100% and a specificity of 100% when IVC systolic reflux is used to diagnose TR. a-l° Also, the jugular venous waveform was normal in the 10 patients in whom it was recorded during sinus rhythm. The absence of TR in these patients during sinus rhythm indicates that a right ventricular pacemaker lead does not usually interfere with normal function of the tricuspid valve. The patterns of systolic IVC reflux seen in this study are noteworthy because they may be mistakenly taken as evidence of TR. During ventricular demand pacing, clinical examination and jugular venous pulse recording revealed that 10 of the 20 patients displayed an abnormality due to cannon A waves. Eighteen patients, including these 10, had IVC contrast reflux during ventricular systole, a pattern that has been considered specific for the diagnosis of TR. However, it became obvious with successive studies that systolic IVC reflux occurred only when normal AV synchrony was lost and, more precisely, when atrial contraction overlapped ventricular systole. One may argue that we were actually seeing TR that manifested because of loss of AV synchrony. This phenomenon has been shown in relation to the mitral valve using angiographic methods11; but in every case in which IVC reflux occurred during ventricular systole, it followed atrial contraction and its timing was variable in systole according to the time of onset of the P wave. This observation and the occurrence of cannon A waves during pacing led us to conclude that the mechanism of contrast reflux during ventricular systole was atrial contraction against a closed tricuspid valve rather than TR. This is similar to the conclusion of Kay et al, 12who studied right-sided cardiac hemodynamics and performed right atrial angiography in a pacemaker patient
FIGURE 5. Top, M-mode contrast inferior vena cava (IVC) echogram during pacing shows a fixed timing of systolic reflux due to a fixed temporal relation between the pacemaker spike and P wave, This Is the fixed systolic reflux pattern. Bob tom, corresponding Jugular venous pulse (JVP) recording during pacing. There are late-occurring cannon A waves with constant timing. The mltral valve M-mode echogram is also seen. MVO = mitral valve opening; other abbreviations as in Figures 1 and 2.
November 1, 1986
THE AMERICAN JOURNAL OF CARDIOLOGY
Volume 58
1029
FIGURE 6. Right atrial (RA) and ventricular (RV) M-mode contrast echogram derived from a 2-dlmensional parasternal RV long-axis view. The tricuspid valve (TV) echogram can be seen within the contrast echoes. When P waves occur during ventricular systole (cycles 3 and 4), contrast can be seen "bouncing" off the tricuspid valve (arrows). When P waves precede ventrlcular systole (cycles 5 and 6), contrast flows toward the tricuspid valve in systole (arrow). RV = right ventricle; "13/ = tricuspid valve echogram; other abbrevlations as in Figures 1 and 2.
with clinically suspected TR.Their data indicated that this patient had pacemaker-induced pseudo-TR due to cannon A waves simulating TR. Would Doppler echocardiography have been more sensitive in detecting TR in these patients? Curtius et al 1~compared the sensitivity and specificity of contrast M-m0de and Doppler echocardiography for diagnosis of TR. Using contrast ventricu!ography as the standard in 68 patients, sensitivities for contrast echocardiography and Doppler techniques were, respectively, 82% and 91%, and specificities were 100% and 86%. This was in keeping with reported sensitivities of 88 to 100% for Contrast echocardiography and 74 to 100% for Doppler. 8-10,14-2° In these studies specificities of contrast echocardiography and Doppler were reported as 100% and 85 to 100%, respectively. Thus, Doppler may have lowered the potential for false-negative diagnosis at the risk of increased false-positive diagnosis. Furthermore, although it was not our aim to compare other methods Of detecting TR, we did look at right atrial and right ventricular linear M-mode contrast echocardiography in some patients. Figure 6 shows a pattern that may have resulted in a false-positive diagnosis of TR by the Doppler technique71 This pattern illustrates contrast bouncing off the tricuspid valve, traveling retrogradely in the right atrium during ventricular systole when AV asynchrony occurs. As noted, the appearance of IVC contrast during ventricular systole is considered highly specific for TR. This has been validated in patients who are in sinus rhythm or atrial fibrillation. This is clearly not the case when ventricular pacing induces AV asynchrony, Therefore, one must be aware of the potential for overdiagnosis, which can be prevented by observing for loss of normal AV synchrony during the contrast study. Acknowledgment: We thank Brenda Arniel and Cheryl Morgan for help in preParing the manuscript.
References 1. Morse D. What's wrong with pacing? PACE 1982:5:455-456. 2. Asubel K. Furman S. The pacemaker syndrome. Ann Intern Mad
1985;103:420-429. 3. Maurer G, Torres MAR, Corday E, Haendchen RV, Meerbaum S. Twodimensional echocardiographic contrast assessment of pacing induced mitral
regurgitation: relation to altered regional left ven!ricular functio n. JACG 1984;3:986-991. 4. Haas ]M, Strait GB. Pacemaker induced cardiovascular failure. Am l Cardiol 1974;33:295-299. 5. Gibson TC, Davidson RC, DeSilvey DL. Presumptive tricuspid valve malfunction induced by a pacemaker lead: a case report and review Of the literature. PACE 1980;3:89-95. 6. Lieppe W, Behar VS, Scallion R, Kisslo JA. Detection of tricuspid regurgitation with 2-dimensional echocardiagraphy and peripheral vein injections. Circulation 1978;57:12-132. , 7. DePace NL. Ross ], Iskandrian AS, Nestico PF, Kotler MN, Mintz GS, Segal BL, Hakki A, Morganroth ]. Tricuspid regurgitation: noninvasive techniques for determining causes and severity. ]ACC 1984;3:i540-1550. 8. Meltzer RS, van Hoogenhuyze D, Serruys PW, Haalebos MP, Hugenholtz PC, Roelandt J. Diagnosis of tricuspid regurgitation by contrast echocardiography. Circulation 1981;63:1093-1099. 9. Amano K, Sakamoto T, Hada y: Yamaguchi T, Ishimitsu KT, Takenaka K. Detection of tricuspid regurgitation by contrast echocardiography, lpn Cite l
1982;46:395-4o!. 10. Lambertz H, Schweizer P, Erbel R, Meyer ], Effert S. Value of the use of contrast echocardiagraphy in th e diagnosis of tricuspid insufficiency. Z Kar-
diol 1982;71:771-776. 11. Tsakiris AG, Gordon DA, Mathieu Y, Padiyar R, Labrosse C. Sudden
interuption of leafle! opening by ventriculur contractions: a mechanism of mitre! regurgitation. ] Appl Physiol 1976;40:132-137. 12. Kay RH, Ambrose ]A, Schek L, Blake ], Rubin D, Herman MV. Pacemaker-induced pseudotricuspid regurgitation. Chest 1983i83:282-284. 13. Curtius ]M, Thyssen M, Breuer HM, Loogen F. Doppler versus contrast echocardiography for diagnosis of tricuspid regurgitation. Am [ Cardiol 1985;56:333-336. 14. B]anchard D, Diebold B, Guermonprez ]L, Peronneau P, Forman J, Mau-
rice P. Non-invasive quantification of tricuspid regurgitation by Doppler echocardiography (abstr}. Circulation 1981;64:suppl IV:IV-256. 15. Diebold B, Touati R, Blanchard D, Colonna G, Guermonprez JL, Peronneau P, Forman ], Maurice P. Quantitative assessement of tricuspid regurgitation using pulsed Doppler echocardiography. Sr Heart [ 1993;50:
443-449. 16. Garcia-Dorado D, Fatzgraf F, Almazan A, Delcan ]L, Lopez-Bescos L, Menarguez L. Diagnosis of functional tricuspi d insufficiency by pulsed-wave
Doppler ultrasound. Circulation 1982;66:1315-1321.
•
17. Miyatake K, Okamoto M, Kinoshita N, Ohta M, Kozuka T, Sakakil~ara H, Nimura Y. Evaluation of tricuspid regurgitation by PUlsed Doppler and 2-
dimensional echocardiagraphy. Circulation 1982;66:777-784. 18. Stevenson G, Kawabori I, Guntheroth W. Validation of Doppler diagnosis of tricuspid regurgitation {abstr). Circulation 1981;64:suppl IV:IV-255. 19. Veyrat C, Kalmanson D, Farjon M, Manin ]P, Arb{tbol G. Non-invasive diagnosis and assessment of tricuspid regurgitation and stenosls using 1- and 2dimensional echopulsed Doppler. Br He0rt I 1982;47:596-605. 20. Waggoner AD, Quin'ones MA, Young JB, Brandon TA, Shah AA, Verani MS, M!ller RR. Pulsed Doppler echocurdiographic detection of right-sided valve regurgitation. Am [ Cardiol 1981;47:279-266. 21. Tei C, Shah PM, Ormiston ]A. Assessment of tricuspid regurgitation by directional analysis of right atrial systolic linear reflux echoes with contrast M-mode echocardiagraphy Am Heart [ 1982;193:1925-1030.
Twenty patients from our pacemaker clinic population were assessed clinically and by saline contrast echocardiography (subxiphoid view) to determine the prevalence of tricuspid regurgitation (TR) and, if TR was present, its mechanism. The patients had no known TR before lead placement, a single transvenous right ventricular pacing lead present more than 6 months (mean 52, range 7 to 138), ventricular demand pacing alternating with sinus rhythm and rate programmability. Each patient was studied in sinus rhythm and during ventricular pacing. Using the criterion of inferior vena cava (IVC) contrast reflux during ventricular systole to diagnose TR, no patient had evidence of TR in sinus rhythm, consistent with clinical examination. During ventricular demand pacing, jugular venous pulse cannon A waves developed in 10 patients, and 18 patients (including these 10) had IVC contrast reflux during ventricular
systole. Analysis of the timing of IVC reflux revealed its close temporal relation to the timing of atrial systole rather than a fixed timing during ventricular systole. This reflux occurred with loss of normal atrioventricular (AV) synchrony and the underlying mechanism in all cases was shown to be right atrial contraction against a closed tricuspid valve. Two patients who did not have such a pattern with pacing maintained normal AV synchrony. These obserrations indicate that: (1) TR is an uncommon accompaniment of ventricular demand pacing; (2) the jugular venous pulse and IVC echocardiographic contrast patterns during ventricular demand pacing simulate TR when AV synchrony occurs; and (3) the IVC contrast pattern of pacing-induced AV synchrony is best termed the cannon A wave synchronous pattern. (Am J Cardiol 1986;58:1025-1029)
R i g h t ventricular pacing will result in hemodynamic compromise in some patients--the so-called pacemaker syndrome. This has been attributed to atrioventricular (AV) valve regurgitation, among other mechanisms. 1,2Animal studies show that mitral regurgitation may occur with ventricular pacing and that its severity varies with the electrode site. 3 Hypotension with mitral regurgitation has been described in a patient during right ventricular pacing. 4 Tricuspid valve malfunction due to a right ventricular pacemaker lead has been noted. 5 Little is known, however, about the prevalence or mechanism of AV valve regurgitation in humans with ventricular pacemakers. Because the electrode crosses the tricuspid valve, patients with ventricular pacemakers are at risk of tri-
cuspid regurgitation (TR). Causes for TR include direct interference with tricuspid valve closure, inappropriately timed atrial systole due to asynchronous pacing, and the disordered sequence of right ventricular and papillary muscle activation during pacing. We assessed the prevalence of TR by M-mode contrast echocardiography in patients with a ventricular demand pacemaker and determined the mechanism of TR, if present.
From the Division of Cardiology, Queen's University, Kingston, Ontario, Canada. Manuscript received March 3, 1986; revised manuscript received July 1, 1986, accepted ]uly 2, 1986. Address for reprints: Gary W. Burggraf, MD, Hotel Dieu Hospital, Division of Cardiology, Kingston, Ontario, Canada K7L 5G2.
1025
Methods Study population: We selected 20 patients (10 men, 10 women) from our pacemaker population who fulfilled the following criteria: no known TR before pacemaker lead placement, a single transvenous right ventricular pacing lead present at least 6 months, ventricular demand pacing alternating with sinus rhythm, and rate programmability. The age range was 51 to 77 years (mean 69). Average duration of lead placement was 52 months (range 7 to 138). The indication for pacing was sick sinus syndrome in 11 patients and AV block in 9 patients. Eight patients had known ath-
1020
TRICUSPIDREGURGITATION DURING VENTRICULAR PACING
FIGURE 1. Subxiphold 2-dimensional echogram and corresponding M-mode recording with simultaneous electrocardiogram used to time occurrence of inferior vena cava (IVC) contrast reflux. RA =
right atrium.
erosclerotic coronary artery disease and 1 patient had pulmonary hypertension secondary to recurrent emboli. Methods: After the patient gave informed consent, an electrocardiogram was recorded to determine if the patient was in sinus rhythm or ventricular pacing. The patient then underwent a physical examination to look for clinical signs of TR, which included jugular venous pulse abnormality, left parasternal systolic murmur and hepatic pulsation. Echocardiography was then performed with a Hewlett-Packard phased-array instrument {model 77020A} using a 3.5-MHz transducer. Strip-chart re-
cordings were obtained with a Honeywell recorder at paper speeds of 50 mm/s for the M-mode contrast echocardiograms and at 50 or 100 mm/s for the jugular venous pulse and phonocardiogram tracings. Saline contrast echocardiograms were recorded in the subxiphoid view as described by Lieppe et al. 6 In this view 2-dimensional visualization outlined the right atrium, hepatic veins and the inferior vena cava {IVC} and a simultaneous M-mode echocardiogram was recorded from the IVC just proximal to its junction with the right atrium {Fig. 1). Boluses of agitated saline solution {10 ml} were then administered through a right antecubital vein while an M-mode recording of the IVC was obtained. Using this method TR may be diagnosed when contrast appears in the IVC during ventricular systole. e-8 Results were obtained at least in duplicate, with each patient undergoing simultaneous M-mode and 2dimensional visualization to ensure adquate bolus delivery to the right heart. Jugular venous and phonocardiographic tracings were recorded from the 10 patients in whom a clinical abnormality of the jugular waveform was apparent. To record studies in both paced and sinus rhythm, the heart rate was reprogrammed up or down, depending on the initial rhythm. This usually required a rate change of 5 to 15 beats/min and was successful in all patients. Subsequently, the clinical examination, contrast echocardiogram, jugular venous pulse recordings and phonocardiograms were repeated. Blood pressure was determined in both rhythms, and when the jugular venous waveform was clinically abnormal it was recorded in both rhythms for comparison. All studies were performed by the same examiners.
Results In sinus rhythm none of the 20 patients displayed clinical evidence of TR, and their contrast M-mode
FIGURE 2. Top, M-mode contrast inferior vena cava (IVC) echogram during sinus rhythm. Vertical a r r o w Indicates contrast reflux with atrial depolarIzatlon. This Is the pattern of "a-wave synchronous r e f l u x . " Bottom, normal Jugular venous pulse (JVP) tracing, electrocardiogram (ECG) and phonocardiogram. Timing of the a wave Is the same as IVC reflux. LLSB = left lower sternal border phonocardiogram; $1 and $2 = first and second head sounds.
November 1, 1986 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 58
echocardiographic findings were normal. The echocardiograms showed either no IVC reflux after bolus injection (10 patients) or reflux that occurred only with atrial contraction and before ventricular systole (10 patients). Figure 2 shows this reflux pattern, the socalled a wave synchronous pattern, which is normal. 8 Ten of the 20 patients also had jugular venous recordings done in sinus rhythm, and all findings were normal (Fig. 2). Thus, in sinus rhythm no patient had clinical or echocardiographic evidence of TR. In ventricular paced rhythm at rates slightly above the patient's sinus rate, 10 of 20 patients displayed a clinically detectable and recordable jugular venous pulse abnormality in the form of cannon A waves. No patient had a new systolic parasternal murmur or detectable hepatic pulsations. One patient had a drop in blood pressure from 170/90 to 130/80 mm Fig with pacing, but remained asymptomatic. In the other patients systolic blood pressure decreased less than 10 mm Fig. Eighteen patients had systolic IVC reflux with pacing, which occurred only when there was loss of AV synchrony (Fig. 3). The 2 patients who failed to have IVC reflux during pacing fortuitously maintained normal AV synchrony. Figure 4 shows the pattern of "variable systolic reflux" that occurred in 13 subjects. The upper panel displays a simultaneous electrocardiogram and the Mmode contrast study and the lower panel includes a jugular venous pulse and phonocardiographic recording. The electrocardiographic P waves are labeled and the onset of contrast appearance is indicated by arrows. Contrast appears during ventricular systole, but the timing of systolic reflux is variable. In the first cycle, contrast appears in the IVC 360 ms after the pacemaker spike, compared with 180 ms in the last cycle. However, the P waves consistently closely pre-
1027
FIGURE 3. M-mode contrast inferior vena cava (IVC) echogram showing appearance of reflux when normal atrloventrlcular synchrony Is lost during pacing. Note the lack of reflux when the P wave of the electrocardiogram precedes the pacemaker spike In cycle 4. Abbreviations as In Figures 1 and 2.
cede the contrast appearance. Seven of these 13 patients had jugular venous pulse recordings in both rhythms, and all recordings revealed that the waveform was normal in sinus rhythm or when AV synchrony was maintained despite pacing (cycles I and 2 of Fig. 4, lower). However, as AV synchrony was lost, cannon A waves developed that bore the same relation to the electrocardiographic P wave as did the contrast appearance (Fig. 4, upper) and showed the same variable timing in systole. This variable systolic reflux pattern can thus be attributed to right atrial contraction against a closed tricuspid valve. Figure 5 illustrates the pattern of "fixed systolic reflux" that occurred in 5 subjects. In contrast to Figure 4, the IVC reflux occurs at a constant time during ventricular systole. Once again, the contrast reflux occurs soon after the electrocardiographic P wave, which in this case maintains a constant time relation to the QRS complex. Also, the timing of systolic reflux is later
i ~
FIGURE 4. Top, M-mode contrast inferior vena cava (IVC) echogram during asynchronous pacing shows that reflux bears a constant temporal relation to the P wave of the electrocardiogram (ECG) and occurs st variable times in ventricular systole. This is the variable reflux pattern. Bottom, corresponding jugular venous pulse (JVP) tracing. Cannon A waves occur according to the timing of the electrocardiographic P wave. HV = hepatic vein; other abbreviations as in Figures 1 and 2.
1020
TRICUSPIDREGURGITATION DURING VENTRICULAR PACING
in onset than expected for TR. The jugular venous recording in the lower panel illustrates the pattern in 3 patients who had tracings available for both rhythms. Again, cannon A waves developed during paced rhythm and the large systolic waves occur later than one might expect for a tricuspid regurgitant venous wave. This fixed systolic reflux pattern can thus be attributed to right atrial contraction against a closed tricuspid valve.
Discussion The prevalence or mechanism of pacemaker-induced TR or mitral valvular regurgitation has not been defined in patients with or without the pacemaker syndrome. We thus attempted to determine the prevalence of TR in a general pacemaker population and elucidate the mechanism of TR if it was present. Our patient selection allowed us to determine whether TR was due to a direct effect of the electrode, in which case it would be present in sinus rhythm, or whether it was due to an indirect effect {loss of normal AV synchrony or abnormal ventricular depolarization}, where it would occur only during pacing. We believe that our patients were representative of the'general pacemaker population because although we applied select entry criteria, none of these would have excluded patients at risk for TR or the pacemaker syndrome. Although TR has been previously reported in patients with pacemakers, we did not see clinical or echocardiographic evidence of TR in the 20 patients during sinus rhythm. 5 If one relied solely on clinical examination, one might underestimate the prevalence of TR because clinical recognition of mild to moderate TR is often difficulty However, our results were conf i r m e d b y subxiphoid contrast M-mode echocardiog-
raphy, which has a sensitivity of 88 to 100% and a specificity of 100% when IVC systolic reflux is used to diagnose TR. a-l° Also, the jugular venous waveform was normal in the 10 patients in whom it was recorded during sinus rhythm. The absence of TR in these patients during sinus rhythm indicates that a right ventricular pacemaker lead does not usually interfere with normal function of the tricuspid valve. The patterns of systolic IVC reflux seen in this study are noteworthy because they may be mistakenly taken as evidence of TR. During ventricular demand pacing, clinical examination and jugular venous pulse recording revealed that 10 of the 20 patients displayed an abnormality due to cannon A waves. Eighteen patients, including these 10, had IVC contrast reflux during ventricular systole, a pattern that has been considered specific for the diagnosis of TR. However, it became obvious with successive studies that systolic IVC reflux occurred only when normal AV synchrony was lost and, more precisely, when atrial contraction overlapped ventricular systole. One may argue that we were actually seeing TR that manifested because of loss of AV synchrony. This phenomenon has been shown in relation to the mitral valve using angiographic methods11; but in every case in which IVC reflux occurred during ventricular systole, it followed atrial contraction and its timing was variable in systole according to the time of onset of the P wave. This observation and the occurrence of cannon A waves during pacing led us to conclude that the mechanism of contrast reflux during ventricular systole was atrial contraction against a closed tricuspid valve rather than TR. This is similar to the conclusion of Kay et al, 12who studied right-sided cardiac hemodynamics and performed right atrial angiography in a pacemaker patient
FIGURE 5. Top, M-mode contrast inferior vena cava (IVC) echogram during pacing shows a fixed timing of systolic reflux due to a fixed temporal relation between the pacemaker spike and P wave, This Is the fixed systolic reflux pattern. Bob tom, corresponding Jugular venous pulse (JVP) recording during pacing. There are late-occurring cannon A waves with constant timing. The mltral valve M-mode echogram is also seen. MVO = mitral valve opening; other abbreviations as in Figures 1 and 2.
November 1, 1986
THE AMERICAN JOURNAL OF CARDIOLOGY
Volume 58
1029
FIGURE 6. Right atrial (RA) and ventricular (RV) M-mode contrast echogram derived from a 2-dlmensional parasternal RV long-axis view. The tricuspid valve (TV) echogram can be seen within the contrast echoes. When P waves occur during ventricular systole (cycles 3 and 4), contrast can be seen "bouncing" off the tricuspid valve (arrows). When P waves precede ventrlcular systole (cycles 5 and 6), contrast flows toward the tricuspid valve in systole (arrow). RV = right ventricle; "13/ = tricuspid valve echogram; other abbrevlations as in Figures 1 and 2.
with clinically suspected TR.Their data indicated that this patient had pacemaker-induced pseudo-TR due to cannon A waves simulating TR. Would Doppler echocardiography have been more sensitive in detecting TR in these patients? Curtius et al 1~compared the sensitivity and specificity of contrast M-m0de and Doppler echocardiography for diagnosis of TR. Using contrast ventricu!ography as the standard in 68 patients, sensitivities for contrast echocardiography and Doppler techniques were, respectively, 82% and 91%, and specificities were 100% and 86%. This was in keeping with reported sensitivities of 88 to 100% for Contrast echocardiography and 74 to 100% for Doppler. 8-10,14-2° In these studies specificities of contrast echocardiography and Doppler were reported as 100% and 85 to 100%, respectively. Thus, Doppler may have lowered the potential for false-negative diagnosis at the risk of increased false-positive diagnosis. Furthermore, although it was not our aim to compare other methods Of detecting TR, we did look at right atrial and right ventricular linear M-mode contrast echocardiography in some patients. Figure 6 shows a pattern that may have resulted in a false-positive diagnosis of TR by the Doppler technique71 This pattern illustrates contrast bouncing off the tricuspid valve, traveling retrogradely in the right atrium during ventricular systole when AV asynchrony occurs. As noted, the appearance of IVC contrast during ventricular systole is considered highly specific for TR. This has been validated in patients who are in sinus rhythm or atrial fibrillation. This is clearly not the case when ventricular pacing induces AV asynchrony, Therefore, one must be aware of the potential for overdiagnosis, which can be prevented by observing for loss of normal AV synchrony during the contrast study. Acknowledgment: We thank Brenda Arniel and Cheryl Morgan for help in preParing the manuscript.
References 1. Morse D. What's wrong with pacing? PACE 1982:5:455-456. 2. Asubel K. Furman S. The pacemaker syndrome. Ann Intern Mad
1985;103:420-429. 3. Maurer G, Torres MAR, Corday E, Haendchen RV, Meerbaum S. Twodimensional echocardiographic contrast assessment of pacing induced mitral
regurgitation: relation to altered regional left ven!ricular functio n. JACG 1984;3:986-991. 4. Haas ]M, Strait GB. Pacemaker induced cardiovascular failure. Am l Cardiol 1974;33:295-299. 5. Gibson TC, Davidson RC, DeSilvey DL. Presumptive tricuspid valve malfunction induced by a pacemaker lead: a case report and review Of the literature. PACE 1980;3:89-95. 6. Lieppe W, Behar VS, Scallion R, Kisslo JA. Detection of tricuspid regurgitation with 2-dimensional echocardiagraphy and peripheral vein injections. Circulation 1978;57:12-132. , 7. DePace NL. Ross ], Iskandrian AS, Nestico PF, Kotler MN, Mintz GS, Segal BL, Hakki A, Morganroth ]. Tricuspid regurgitation: noninvasive techniques for determining causes and severity. ]ACC 1984;3:i540-1550. 8. Meltzer RS, van Hoogenhuyze D, Serruys PW, Haalebos MP, Hugenholtz PC, Roelandt J. Diagnosis of tricuspid regurgitation by contrast echocardiography. Circulation 1981;63:1093-1099. 9. Amano K, Sakamoto T, Hada y: Yamaguchi T, Ishimitsu KT, Takenaka K. Detection of tricuspid regurgitation by contrast echocardiography, lpn Cite l
1982;46:395-4o!. 10. Lambertz H, Schweizer P, Erbel R, Meyer ], Effert S. Value of the use of contrast echocardiagraphy in th e diagnosis of tricuspid insufficiency. Z Kar-
diol 1982;71:771-776. 11. Tsakiris AG, Gordon DA, Mathieu Y, Padiyar R, Labrosse C. Sudden
interuption of leafle! opening by ventriculur contractions: a mechanism of mitre! regurgitation. ] Appl Physiol 1976;40:132-137. 12. Kay RH, Ambrose ]A, Schek L, Blake ], Rubin D, Herman MV. Pacemaker-induced pseudotricuspid regurgitation. Chest 1983i83:282-284. 13. Curtius ]M, Thyssen M, Breuer HM, Loogen F. Doppler versus contrast echocardiography for diagnosis of tricuspid regurgitation. Am [ Cardiol 1985;56:333-336. 14. B]anchard D, Diebold B, Guermonprez ]L, Peronneau P, Forman J, Mau-
rice P. Non-invasive quantification of tricuspid regurgitation by Doppler echocardiography (abstr}. Circulation 1981;64:suppl IV:IV-256. 15. Diebold B, Touati R, Blanchard D, Colonna G, Guermonprez JL, Peronneau P, Forman ], Maurice P. Quantitative assessement of tricuspid regurgitation using pulsed Doppler echocardiography. Sr Heart [ 1993;50:
443-449. 16. Garcia-Dorado D, Fatzgraf F, Almazan A, Delcan ]L, Lopez-Bescos L, Menarguez L. Diagnosis of functional tricuspi d insufficiency by pulsed-wave
Doppler ultrasound. Circulation 1982;66:1315-1321.
•
17. Miyatake K, Okamoto M, Kinoshita N, Ohta M, Kozuka T, Sakakil~ara H, Nimura Y. Evaluation of tricuspid regurgitation by PUlsed Doppler and 2-
dimensional echocardiagraphy. Circulation 1982;66:777-784. 18. Stevenson G, Kawabori I, Guntheroth W. Validation of Doppler diagnosis of tricuspid regurgitation {abstr). Circulation 1981;64:suppl IV:IV-255. 19. Veyrat C, Kalmanson D, Farjon M, Manin ]P, Arb{tbol G. Non-invasive diagnosis and assessment of tricuspid regurgitation and stenosls using 1- and 2dimensional echopulsed Doppler. Br He0rt I 1982;47:596-605. 20. Waggoner AD, Quin'ones MA, Young JB, Brandon TA, Shah AA, Verani MS, M!ller RR. Pulsed Doppler echocurdiographic detection of right-sided valve regurgitation. Am [ Cardiol 1981;47:279-266. 21. Tei C, Shah PM, Ormiston ]A. Assessment of tricuspid regurgitation by directional analysis of right atrial systolic linear reflux echoes with contrast M-mode echocardiagraphy Am Heart [ 1982;193:1925-1030.