True Incidence of Pacemaker Syndrome DEBRA HELDMAN, DANIEL MULVIHILL, HAROLD NGUYEN, JOHN C. MESSENGER, ARLENE RYLAARSDAM, KARLA EVANS, and MARK J. CASTELLANET From the Long Beach Memorial Medical Center, Memorial Heart Institute, Long Beach, California
HELDMAN, D., ET AL.: True Incidence of Pacemaker Syndrome. Although the purported incidence of pacemaker syndrome according to the literature is only 5Y0-15%, this is based on a series of patients with VVI pacing. Increasing numbers of studies are being reported in which patients prefer the dual chamber mode despite little benefit being demonstrated on objective testing, suggesting that pacemaker syndrome may be more common than is generally reported. This study was designed to evaluate the reported symptoms in a series of patients programmed to both the VVI and one or more dual chamber modes. Forty unselected patients with dual chamber pacemakers were entered into a blind, randomized trial comparing the symptoms associated with VVI pacing to those associated with dual chamber pacing. Patients were randomized to either W I or dual chamber pacing. At the end of 1 week, questionnaires rating 16 different symptoms were completed. Blood pressure, LV function, presence of ventriculoatrial conduction, and ability to override the pacemaker were evaluated. The pacemaker was then programmed to the other mode. Overall, 12 of 16 symptoms were significantly worse in the VVI as compared to dual chamber mode. The most highly significant (P < 0.005) were shortness of breath, dizziness, fatigue, pulsations in the neck or abdomen, cough, and apprehension. Pacemaker syndrome was clinically recognized in 83% of patients paced in the W I mode with 65% of patients experiencing moderate to severe symptoms. There were no readily identified clinical, hemodynamic, or electrophysiological parameters that predicted which patients would develop pacemaker syndrome. Thus, when patients have an opportunity to experience both pacing modes in close proximity to one another, there is a high incidence of pacemaker syndrome in the VVI mode. (PACE, Vol. 13, December, Part 11 1990) pacemaker syndrome
Introduction One of the major advances in cardiac pacing is the development and increasing utilization of dual chamber, particularly DDD, pacemakers. The availability of these systems has been accompanied by an ongoing controversy concerning the appropriate utilization of these more.costly and complex systems with regard to the general pacemaker-patient population. At the present time, only 25% of permanent pacemakers implanted in the United States are dual chamber devices (based upon information from stock analysis reports and
Address for reprints: Mark J. Castellanet, M.D., 2888 Long Beach Boulevard, Suite #165, Long Beach, CA 90806.
1742
manufacturers’ corporate annual reports for 19881989). A question frequently raised by medicare and other funding sources concerns the clinical value of these systems over the less complicated and less costly single chamber devices. The complex of signs and symptoms that constitute the pacemaker syndrome has been well described in the 1 i t e r a t ~ r e . Historically, l~~ the incidence of pacemaker syndrome has been estimated at approximately 7%.’ This is based upon the development of dramatic symptoms in a relatively small number of patients who had VVI pacemakers. However, any syndrome represents a wide spectrum of symptoms. This may include: severe hemodynamic limitation with hypotension; syncope; and overt congestive heart failure
December 1990, Part I1
PACE, Vol. 13
PACEMAKER SYNDROME
to intermittent and relatively mild symptoms that may not even be appreciated as indicative of a problem by the patient if there is no basis for comparison. Thus, the true incidence of pacemaker syndrome is difficult to assess in a population of patients with a single chamber ventricular pacing system. Unless marked, lesser symptoms are often attributed to the patient's age or underlying disease process with the role of the VVI pacemaker being discounted. Recent studies have evaluated patients who served as their own control with respect to VVI and physiological pacing."-'' These studies suggest that incidence of pacemaker syndrome is much higher than that previously reported in the literat~re.~.~'." In addition, a majority of patients who spent time programmed to both dual chamber and single chamber (VVI) modes expressed a preference for dual chamber pacing even in the absence of an objectively measured difference suggesting that the key parameters were not being monitored or that there are features which defy quantification but are clinically significant. This study was designed to assess the incidence and severity of symptoms that have been reported in the literature as consistent with the pacemaker syndrome. An unselected consecutive series of patients all of whom had a dual chamber pacemaker were invited to participate. Each patient served as his own control enabling subtle symptoms present in one mode but not the other to be identified. This protocol did not allow for identification of subtle limitations in cardiac function induced by one mode but which were able to be compensated for by the patient's cardiac reserve mechanisms and thus protecting the patient from the development of symptoms.
Methods Patients The practice of the authors has been to implant dual chamber pacemakers in approximately 80% of their patients. Bedridden or markedly debilitated patients, and those with chronic atrial fibrillation who needed a pacemaker would be treated with VVI pacing. An unselected series of consecutive patients with a normally functioning dual chamber pacemaker were invited to partici-
PACE, Vol. 13
pate in this study. A total of 40 patients were recruited. They were successfully paced in either the DDD mode (30 patients) or DDI mode (10 patients) for more than 3 months (range 4-72 months) at the time of entry into the study. There were 23 men and 1 7 women. The group had a mean age of 68 2 10 years with a range of 47-86 years. The electrical indications for pacing included sinus node dysfunction in 21, AV block in 14, and 5 patients had combined sinus node dysfunction and AV block. The underlying cardiac disease included: coronary artery disease in nine patients; congestive heart failure in three; and hypertensive heart disease in two. One patient had AV block associated with aortic replacement and one had myotonic dystrophy. The remaining patients had primary conduction system disease with no other identified disease entity. Four of the patients had a history of pacemaker syndrome with a previous VVI pacemaker.
Study Protocol The study was approved by the hospital research council. After obtaining voluntary informed consent, each patient was blindly randomized to 1 week of VVI pacing and 1 week of dual chamber pacing. The dual chamber mode chosen was that present upon entry into the study (either DDD or DDI). The DDI mode had been intentionally programmed at either implant or subsequently to avoid tracking demonstrated supraventricular arrhythmias. Additionally, half of the patients whose baseline mode was DDD were randomized to an additional week of DVI pacing. At the beginning of the study and before the pacemaker was programmed to a randomly selected mode, the patient completed the first of a series of questionnaires. The same questionnaire was completed at the end of each week of the study. In addition, each patient's blood pressure was checked. After completing this weekly evaluation, the pacemaker was programmed to the next mode as determined by the randomization sequence for that patient. The questionnaire assessed the presence and relative severity of 16 different symptoms, which were reported in the literature as consistent with the pacemaker syndrome. Each symptom was rated on a scale from 0 (not present) to 10 (very
December 1990,Part I1
1743
HELDMAN, ET AL.
severe). The development or exacerbation of symptoms in one mode as compared to the other was arbitrarily considered to be mild if the increase in total symptom points was < 16 with no single symptom increase being > 5. It was considered moderate if the increase in total symptom score ranged from 1 7 to 32 points with no increase in a single symptom score being > 8. The symptoms were considered severe if the total symptom score exceeded 32 or a single symptom increased more than 8 from baseline or there was an early request to return to the previous mode. Each patient was evaluated for the presence or absence of ventriculoatrial conduction during VVI pacing. This was accomplished by a careful examination of the surface ECG recorded at 50 mm/sec or by telemetered intracardiac atrial electrograms when these were available. A symptom limited exercise treadmill test utilizing the Ellestad protocol was performed while the patient was in the VVI pacing mode. Six patients were unable to perform this test. This was due to joint disease in three, pulmonary disease in two, and patient refusal in one case. ECGs were recorded prior to exercise, at the end of each minute of exercise, and at the end of the protocol to document whether the patient’s intrinsic rhythm overrode the pacemaker. A two-dimensional echocardiogram to evaluate left ventricular systolic function at rest was obtained in the dual chamber pacing mode in each patient who had not had an assessment of ventricular function within the 6 months prior to participating in this protocol. Each patient’s baseline pacing rate was continued throughout the study. No pacemaker parameter was changed except for the mode. Baseline medications were maintained. During the course of the study, the patients were encouraged to notify the investigators if intolerable symptoms developed at any time. In these cases, the patients were seen the same day for blood pressure measurement and completion of the next questionnaire, after which they were programmed to the next mode. If this occurred during the last week of the study, they were programmed back to their baseline mode.
Data Analysis The data are expressed as mean i standard deviation. The paired Student’s t-test was used to
1744
compare means for continuous variables. The Yates corrected Chi-square test was used to compare group data.
Results Patients had significantly fewer symptoms in either the DDD or DDI mode as compared to the VVI mode. Table I lists the individual severity scores for each of the 16 symptoms listed on the questionnaire along with the mean r standard deviation and relevant P values. Twelve of the 16 symptoms were significantly less severe in the dual chamber mode as compared to the VVI mode. The most highly significant were shortness of breath, fatigue, dizziness, pulsations in the neck and/or abdomen, cough, and apprehension. In addition, the total symptom score was significantly lower in the dual chamber mode as compared to the VVI mode. Figure 1 shows the total symptom score for each patient in the VVI mode. These reflected either a new symptom or an increase in symptom as compared to the patient’s baseline questionnaire which was obtained when the patient’s pacemaker was programmed to the dual chamber mode. Of the ten patients whose baseline mode was DDI, eight had fewer symptoms in DDI mode than the VVI mode. One patient felt the same and one patient reported fewer symptoms in the VVI mode as compared to the DDI mode. In the 30 patients whose baseline mode was DDD, 25 had fewer symptoms in the DDD mode as compared to the VVI mode. Two patients felt the same and three expressed slight preference for VVI. One of the patients who reported a preference for VVI had a viral illness during the week that he randomized to the DDD mode. No other patient had an intercurrent illness. Figure 2 demonstrates the percentage of patients developing either no, mild, moderate, or severe additional symptoms when in the VVI mode as compared to the dual chamber modes. Sixtyfive percent of all patients had development or exacerbation of symptoms of a moderate or severe degree in the VVI mode as compared to their baseline dual chamber mode. An additional 18% had mild symptoms. Seventeen percent denied more severe symptoms with VVI pacing. Figure 3 shows the total symptom score in
December 1990, Part I1
PACE, Vol. 1 3
PACEMAKER SYNDROME
Table 1.
Severity of Symptoms in VVI Versus Dual-Chamber Pacing Symptom 1. 2. 3. 4. 5. 6. 7. 8. 9.
10. 11. 12. 13. 14. 15. 16.
Shortness of breath Fatigue Dizziness Apprehension Cough Pulsations in necwabdomen Orthopnea Headache Palpitations Chest pain Choking sensation Confusion Pedal edema Sensation of tachycardia Chest congestion Diaphoresis Total Symptoms
DDDlDDl +
VVI* 3.3 4.8 2.9 3.0 1.7 2.0 1.3 1.3 1.5 1.4 1.3 0.9 0.9
f 3.1 ?
*
f ?
f +?
f -+ f
*
f
1.1 +1.1 0.7 29.0
* ?
0.8 1.3 0.8 0.3 0.4 0.4 0.3 0.5 0.5 0.4 0.3 0.2 0.3 0.4 0.5
3.5 3.6 3.6 2.5 3.2 2.5 2.3 3.0 2.6 2.9 2.2 2.3 2.5
1.9 2.3 26.1
= Symptom score during VVI pacing expressed as mean
*
1.8
f 2.3 ?
1.3
2 0.8 f 1.6
1.1 ? 1.3 ? 0.9 +. 1.0 f 1.2 1.2 f 0.6 -+ 0.9 f 1.4 f 1.6 0.1 0.2 7.3 ? 12.4 %
*
*
p Value p p p p p p
< 0.001 < 0.001 < 0.001 < 0.001
0.001 0.002 p < 0.02 p < 0.02 p < 0.04 p < 0.04 p < 0.04 p < 0.05 NS NS NS NS p < 0.001 = =
*
standard deviation Symptom score during baseline dual-chamber pacing (either DDD or DDI) expressed as mean ’. standard deviation
+
=
each mode for those patients who were also programmed to the DVI mode. Eleven of these 16 patients had fewer symptoms in the DVI mode as compared to the VVI mode. Six of the 11 had a further improvement when programmed to the DDD mode. Of the five patients who did not demonstrate a preference for DVI over VVI, all showed a considerable reduction in symptoms when programmed to the DDD mode. Of note, our only patient who developed atrial fibrillation during this study did so when programmed to the DVI mode. This patient had no prior history of atrial arrhythmias. Tables 11,111, and IV compare individual clinical characteristics of patients with the likelihood of their developing symptoms during VVI pacing. None of the following clinical, hemodynamic, or electrophysiological parameters predicted which patients would develop the pacemaker syndrome: gender; body weight; past history of pacemaker syndrome; left ventricular systolic function; hypotension with ventricular pacing; type of conduction system disease; intact ventriculoatrial
PACE, Vol. 13
conduction; or ability of the patient’s endogenous rhythm to override the pacemaker during VVI pacing.
Discussion The signs, symptoms, and presumed mechanisms underlying the pacemaker syndrome have been well summarized by Ausubel and F ~ r m a n . ~ The most important mechanisms are believed to be the loss of atrioventricular synchrony, valvular regurgitation, asynchronous ventricular contraction, inappropriate or an alteration of circulatory reflexes, and ventriculoatrial conduction. Several studies have shown a 10Y0-35% increase in resting cardiac output with restoration of AV synchrony when compared to single chamber ventricular pacing.”-18 The loss of normal AV sequence in a ventricular pacing system contributes to AV valvular regurgitation; increased atrial distension may then affect vasoregulatory ref l e x e ~ . ’ ~ -Retrograde ~’ conduction is estimated to occur in roughly 50% of patients paced for all rea-
December 1990, Part I1
1745
HELDMAN, ET AL.
110
100
90 80
70
s!
8
cn
60
a 50
E
40 c
I-“ 30 20 10
0
I
I
I
I
I
I
I
I
I
I
I
-1 0
Individual Patients
Figure 1. Total symptom score in VVI mode above baseline symptoms (zero line) in dual chamber mode for each of 40 patients (30 DDD baseline, 10 DDI). Each vertical line represents an individual patient.
nea, dizziness, and palpitations were all more sons4 This leads to the atria contracting against a prominent in the VVI mode than dual chamber closed AV valve on a repeated basis leading to mode.’.’* A study by Kenny and associatesg pulmonary and systemic venous regurgitation and looked at ten patients with known coronary artery higher venous pressures that will induce congesdisease and clinically manifested angina. These tion. authors reported a higher incidence of angina durA multiplicity of previous studies, usually ing VVI pacing at a rate of 70 as compared to DDD looking at a small number of paced patients who pacing at a rate of 100 ppm. DDD pacing at higher had dual chamber pacemakers, have demonrates, however, was associated with an increase in strated that in these patients, dual chamber pacing angina. Rediker et al.7 entered 19 patients with was better tolerated than VVI pacing. This was DDD pacemakers into a blinded, randomized based upon exercise tests, echo-Doppler studies, VVI versus DDD pacing. Cardiac study comparing noninvasive and invasive hemodynamic studies, output, percent fractional shortening, and exerand AV 0, differences although no single study cise duration were all significantly greater during used all of these testing m ~ d a l i f i e s . ~ ” ~ ~ ~ ~ * ~ ~ DDD compared to VVI pacing. Sixty-three percent A number of studies have also evaluated the of the patients preferred DDD pacing and none patients’ level of symptoms in a double-blind ranpreferred VVI pacing. domized protocol with the pacing mode being Although there have been claims in the liteither dual chamber or VVI. Symptoms of dysp-
1746
December 1990, Part I1
PACE, Vol. 1 3
PACEMAKER SYNDROME
Figure 2. Severity of symptom development or exacerbation with Wl pacing compared to DDD or DDI mode in 40 patients. See Methods for definition of symptom severity.
erature that rate modulated VVI pacing (VVIR)was hemodynamically equivalent to DDD pacing and may even result in a greater exercise tolerance in those patients with chronotropic incompetence, those studies which evaluated other hemodynamic variables that just exercise duration and/or the patients' level of symptoms have favored dual chamber ~ a c i n g . ~ OAusubel * * ~ and a s s o ~ i a t e s 'ex~ amined 12 patients during moderate levels of exercise using radionuclide angiography to assess the hemodynamics associated with DDD pacing or the functional equivalent of VVIR. They found that DDD pacing resulted in greater ventricular filling and spared contractile reserve at comparable workloads. Dateling et a1.l' evaluated symptomatology in a series of 217 patients programmed to the VVI, VVIR, and DDD pacing modes over a 4-month block of time. Dual chamber pacing resulted in greater symptomatic relief than VVI and VVIR with respect to dizziness, dyspnea, and weakness. These authors found that the presence of ventricuoatrial conduction, percentage of time paced, and heart rate response to exercise were not predictive of the patient experiencing increased symptoms with VVI or VVIR compared to DDD pacing.
PACE, Vol. 13
WI
Figure 3. Total symptom scores for 16 individual patients placed in VVI, DVI, and DDD modes.
Our findings are in keeping with these earlier studies. Eighty-three percent of our patients who had a dual chamber pacemaker experienced either an exacerbation or development of symptoms in the VVI mode as compared to the dual chamber pacing mode. Sixty-five percent of this group experienced moderate-to-severe symptoms and 42% of patients were unable to tolerate VVI pacing for 1 week. Eighteen percent of our patients reported no discernible benefit from dual chamber pacing as compared to VVI pacing. Since our study did not utilize invasive hemodynamic monitoring or other noninvasive testing during each mode of pacing, the possibility that subtle limitations that were able to be corrected by the patient's endogenous reserve mechanisms cannot be excluded. Still, our results support the thesis that the majority of patients in whom atrial pacing or sensing is physiologically feasible are compromised by VVI pacing such that they develop symptoms that
December 1990, Part I1
1747
HELDMAN, ET AL.
Table II.
Relationship Between HistoryIPhysical Findings & Symptoms During VVI Pacing N01 Mild Symptoms
N
Moderat el Severe Symptoms
p Value
Sex
Female Male
17 23
5 (29%)
9 (39%)
14 (61%)
12 (71%)
NS
21 11
8 (38%) 3 (27%)
13 (62%) 8 (63%)
NS
4 36
0 (0%) 14 (39%)
4 (100%) 22 (61%)
NS
Body Weight
2140pounds 10 mmHg LV
1748
=
Left ventricular
December 1990, Part I1
PACE, Vol. 13
PACEMAKER SYNDROME
Table IV.
Relationship Between Conduction Disease and Symptoms During VVI Pacing No/ Mild Symptoms
Moderatel Severe Symptoms
13 (62%) 9 (64%) 4 (80%)
NS
5
8 (38%) 5 (36%) 1 (20%)
18 19
5 (28%) 8 (42%)
13 (72%) 11 (58%)
NS
27 7
9 (33%) 4 (57%)
18 (67%) 3 (43%)
NS
N
p Value
Conduction Disease
sss
AVB
SSS + AVB V-A Conduction
Present Absent
21 14
Ability to Override Pacer
Yes No AVB = Atrioventricular block SSS = Sick sinus syndrome V-A = Ventriculoatrial
A recent debate in both the literature and national meetings has centered on the importance of AV synchrony versus rate responsiveness in the prevention of the pacemaker syndrome. Thus, some patients in this study were also programmed to the DVI mode. DVI pacing provides AV synchrony but not rate responsiveness. Based upon symptom scores, DVI resulted in fewer symptoms than VVI pacing but more symptoms than DDD pacing. Thus, it appears that both AV synchrony (DVI)and rate modulation, in this case based upon the patients’ endogenous atrial activity (DDD),are important in reducing symptoms. Ten patients had their pacemakers programmed to the DDI mode as a baseline and in these cases, DDI was superior to VVI with regard to symptoms. These patients all had documented pathological atrial tachyarrhythmias and thus they. were not programmed to the DDD mode. These patients, however, had sinus node dysfunction and when their atrial rate increased, they had intact AV conduction. One needs to ask why our results and more recent studies in the literature vary to such a large degree from earlier articles with regard to the incidence of pacemaker syndrome. We have demonstrated that 83% of our patients experience in-
PACE, Vol. 13
creased symptoms when paced VVI; thus, if all our patients were to have received a VVI pacemaker, 83% would have the pacemaker syndrome. A smaller percentage of patients had moderate-tosevere symptoms and it is likely that only those patients with severe symptoms would have been labeled pacemaker syndrome had they received a VVI pacemaker. When patients have no basis for comparison and major symptoms such as syncope have been resolved with VVI pacing, both they and their physicians often attribute other nondiagnostic and minor symptoms to the aging process or progression of their underlying disease state. Thus, the true incidence of pacemaker syndrome is not fully appreciated.
Summary The majority of patients paced in the dual chamber mode derive an improved quality of life and decreased symptomatology as compared to single chamber ventricular pacing. Our results support the findings in recent reports when the different pacing modes can be compared in the same patient population. Since there are no clearcut clinical, hemodynamic, or electrophysiological parameters that predict which patients will
December 1990,Part I1
1749
HELDMAN, ET AL.
benefit from this more complex therapy, and more than 80% of an unselected series of patients demonstrate a lower incidence of symptoms with DDD as compared to VVI pacing, it is proposed that dual chamber pacing should be the mode of choice where the atrium is capable of being either sensed or stimulated. Currently, only 25% of paced patients in the United States receive a dual chamber unit and in many other countries, the use of this
modality is even lower. Clearly, a much larger percentage of patients would benefit from dual chamber pacing.
Acknowledgments: We greatly appreciate the assistance of Linda Lorenz in manuscript preparation and Randy Crump in statistical analysis. We also thank Paul Levine, M.D., F.A.C.C., and Adrian Shandling, M.D. for critical review of the manuscript.
References 1. Mitsui T, Hori M, Suma K, et al. The “pacemaking syndrome.” In JE Jacobs (ed.): Proceedings of the Eighth Annual International Conference on Medical and Biological Engineering. Chicago, Association for the Advancement of Medical Instrumentation. 1969, pp. 29-33. 2. Saigusa M. Artificial cardiac pacemaker: Six years experience with cardiac pacemaking for heart block. Jpn Circ J 1969; 33:1583-1586. 3. Mitsui T, Mizuno A, Hasegawa A, et al. Atrial rate as an indicator for optimal pacing rate and the pacemaking syndrome. Ann Cardiol Angiol 1971; 20:371-379. 4. Ausubel K, Furman S. The pacemaker syndrome. Ann Intern Med 1985; 103:420-429. 5. Cohen S, Frank H. Preservation of active atrial transport. Chest 1982: 81:51-54. 6. Kruse I, Arnman K, Conradson T, et al. A comparison of the acute and long-term hemodynamic effects of ventricular inhibited and atrial synchronous ventricular inhibited pacing. Circulation 1982; 65:846-855. 7. Rediker DE, Eagle KA, Herman S, et al. Clinical and hemodynamic comparison of VVI versus DDD pacing in patients with DDD pacemakers. Am J Cardiol 1988; 61:323-329. 8 . Kristensson B-E, Arnman K, Smedgard P, et al. Physiological versus single-rate ventricular pacing: A double-blind crossover study. PACE 1985; 8:73-84. 9. Kenny RA, Ingram A, Mitsuoban T, et al. Optimum pacing mode for patients with angina pectoris. Br Heart J 1986; 56:463-468. 10. Dateline F, Obel IWP. Clinical comparison of VVI, VVIR, and DDD pacemakers in symptomatic relief of bradyarrhythmias. (abstract) PACE 1989; 12:1278. 11. Perrins EJ, Morley CA, Chen SL, et al. Randomized
controlled trial of physiological and ventricular pacing. Br Heart J 1983; 50:112-117. 12. Benchimol A, Ellis JG, Dimond EG. Hemodynamic
1750
13.
14. 15.
16.
17.
18.
19.
20.
21.
22.
consequences of atrial and ventricular pacing in patients with normal and abnormal hearts. Am J Med 1965; 39:911-921. Morris JJ Jr, Entman M, North WC, et al. The changes in cardiac output with reversion of atrial fibrillation to sinus rhythm. Circulation 1965; 31:670-678. Sowton E. Artificial pacemaking and sinus rhythm. Br Heart J 1965; 27:311-318. Samet P, Bernstein WH, Levine S, et al. Hemodynamic effects of tachycardias produced by atrial and ventricular pacing. Am J Med 1965; 39:905910. Leinbach RC, Chamberlain DA, Kastor JA, et al. A comparison of the hemodynamic effects of ventricular and sequential A-V pacing in patients with heart block. Am Heart J 1969; 78:502-508. Samet P, Castillo C, Bernstein W. Hemodynamic consequences of atrial and ventricular pacing in subjects with normal hearts. Am J Cardiol 1966; 18:522-525. Stewart WJ, Dicola VC, Harthorne JW, et al. Doppler ultrasound measurement of cardiac output in patients with physiologic pacemakers. Am J Cardiol 1984; 54:308-312. Ausubel K, Steingart RM, Shimshi M, et al. Maintenance of exercise stroke volume during ventricular versus atrial synchronous pacing: Role of contractility. Circulation 1985; 72:1037-1043. Nishimura RA, Gersh BJ, Vlietstra RE, et al. Hemodynamics and symptomatic consequences of ventricular pacing. PACE 1982; 5:903-910. Alicandri C, Fouad FM, Tarazi RC, et al. Three cases of hypotension and syncope with ventricular pacing: Possible role of atrial reflexes. Am J Cardiol 1978; 42:137-142. Sutton R, Perrins EJ, Morley C, et al. Sustained improvement in exercise tolerance following physiological cardiac pacing. Eur Heart J 1983; 4~781-785.
December 1990, Part I1
PACE, Vol. 1 3
HELDMAN, D., ET AL.: True Incidence of Pacemaker Syndrome. Although the purported incidence of pacemaker syndrome according to the literature is only 5Y0-15%, this is based on a series of patients with VVI pacing. Increasing numbers of studies are being reported in which patients prefer the dual chamber mode despite little benefit being demonstrated on objective testing, suggesting that pacemaker syndrome may be more common than is generally reported. This study was designed to evaluate the reported symptoms in a series of patients programmed to both the VVI and one or more dual chamber modes. Forty unselected patients with dual chamber pacemakers were entered into a blind, randomized trial comparing the symptoms associated with VVI pacing to those associated with dual chamber pacing. Patients were randomized to either W I or dual chamber pacing. At the end of 1 week, questionnaires rating 16 different symptoms were completed. Blood pressure, LV function, presence of ventriculoatrial conduction, and ability to override the pacemaker were evaluated. The pacemaker was then programmed to the other mode. Overall, 12 of 16 symptoms were significantly worse in the VVI as compared to dual chamber mode. The most highly significant (P < 0.005) were shortness of breath, dizziness, fatigue, pulsations in the neck or abdomen, cough, and apprehension. Pacemaker syndrome was clinically recognized in 83% of patients paced in the W I mode with 65% of patients experiencing moderate to severe symptoms. There were no readily identified clinical, hemodynamic, or electrophysiological parameters that predicted which patients would develop pacemaker syndrome. Thus, when patients have an opportunity to experience both pacing modes in close proximity to one another, there is a high incidence of pacemaker syndrome in the VVI mode. (PACE, Vol. 13, December, Part 11 1990) pacemaker syndrome
Introduction One of the major advances in cardiac pacing is the development and increasing utilization of dual chamber, particularly DDD, pacemakers. The availability of these systems has been accompanied by an ongoing controversy concerning the appropriate utilization of these more.costly and complex systems with regard to the general pacemaker-patient population. At the present time, only 25% of permanent pacemakers implanted in the United States are dual chamber devices (based upon information from stock analysis reports and
Address for reprints: Mark J. Castellanet, M.D., 2888 Long Beach Boulevard, Suite #165, Long Beach, CA 90806.
1742
manufacturers’ corporate annual reports for 19881989). A question frequently raised by medicare and other funding sources concerns the clinical value of these systems over the less complicated and less costly single chamber devices. The complex of signs and symptoms that constitute the pacemaker syndrome has been well described in the 1 i t e r a t ~ r e . Historically, l~~ the incidence of pacemaker syndrome has been estimated at approximately 7%.’ This is based upon the development of dramatic symptoms in a relatively small number of patients who had VVI pacemakers. However, any syndrome represents a wide spectrum of symptoms. This may include: severe hemodynamic limitation with hypotension; syncope; and overt congestive heart failure
December 1990, Part I1
PACE, Vol. 13
PACEMAKER SYNDROME
to intermittent and relatively mild symptoms that may not even be appreciated as indicative of a problem by the patient if there is no basis for comparison. Thus, the true incidence of pacemaker syndrome is difficult to assess in a population of patients with a single chamber ventricular pacing system. Unless marked, lesser symptoms are often attributed to the patient's age or underlying disease process with the role of the VVI pacemaker being discounted. Recent studies have evaluated patients who served as their own control with respect to VVI and physiological pacing."-'' These studies suggest that incidence of pacemaker syndrome is much higher than that previously reported in the literat~re.~.~'." In addition, a majority of patients who spent time programmed to both dual chamber and single chamber (VVI) modes expressed a preference for dual chamber pacing even in the absence of an objectively measured difference suggesting that the key parameters were not being monitored or that there are features which defy quantification but are clinically significant. This study was designed to assess the incidence and severity of symptoms that have been reported in the literature as consistent with the pacemaker syndrome. An unselected consecutive series of patients all of whom had a dual chamber pacemaker were invited to participate. Each patient served as his own control enabling subtle symptoms present in one mode but not the other to be identified. This protocol did not allow for identification of subtle limitations in cardiac function induced by one mode but which were able to be compensated for by the patient's cardiac reserve mechanisms and thus protecting the patient from the development of symptoms.
Methods Patients The practice of the authors has been to implant dual chamber pacemakers in approximately 80% of their patients. Bedridden or markedly debilitated patients, and those with chronic atrial fibrillation who needed a pacemaker would be treated with VVI pacing. An unselected series of consecutive patients with a normally functioning dual chamber pacemaker were invited to partici-
PACE, Vol. 13
pate in this study. A total of 40 patients were recruited. They were successfully paced in either the DDD mode (30 patients) or DDI mode (10 patients) for more than 3 months (range 4-72 months) at the time of entry into the study. There were 23 men and 1 7 women. The group had a mean age of 68 2 10 years with a range of 47-86 years. The electrical indications for pacing included sinus node dysfunction in 21, AV block in 14, and 5 patients had combined sinus node dysfunction and AV block. The underlying cardiac disease included: coronary artery disease in nine patients; congestive heart failure in three; and hypertensive heart disease in two. One patient had AV block associated with aortic replacement and one had myotonic dystrophy. The remaining patients had primary conduction system disease with no other identified disease entity. Four of the patients had a history of pacemaker syndrome with a previous VVI pacemaker.
Study Protocol The study was approved by the hospital research council. After obtaining voluntary informed consent, each patient was blindly randomized to 1 week of VVI pacing and 1 week of dual chamber pacing. The dual chamber mode chosen was that present upon entry into the study (either DDD or DDI). The DDI mode had been intentionally programmed at either implant or subsequently to avoid tracking demonstrated supraventricular arrhythmias. Additionally, half of the patients whose baseline mode was DDD were randomized to an additional week of DVI pacing. At the beginning of the study and before the pacemaker was programmed to a randomly selected mode, the patient completed the first of a series of questionnaires. The same questionnaire was completed at the end of each week of the study. In addition, each patient's blood pressure was checked. After completing this weekly evaluation, the pacemaker was programmed to the next mode as determined by the randomization sequence for that patient. The questionnaire assessed the presence and relative severity of 16 different symptoms, which were reported in the literature as consistent with the pacemaker syndrome. Each symptom was rated on a scale from 0 (not present) to 10 (very
December 1990,Part I1
1743
HELDMAN, ET AL.
severe). The development or exacerbation of symptoms in one mode as compared to the other was arbitrarily considered to be mild if the increase in total symptom points was < 16 with no single symptom increase being > 5. It was considered moderate if the increase in total symptom score ranged from 1 7 to 32 points with no increase in a single symptom score being > 8. The symptoms were considered severe if the total symptom score exceeded 32 or a single symptom increased more than 8 from baseline or there was an early request to return to the previous mode. Each patient was evaluated for the presence or absence of ventriculoatrial conduction during VVI pacing. This was accomplished by a careful examination of the surface ECG recorded at 50 mm/sec or by telemetered intracardiac atrial electrograms when these were available. A symptom limited exercise treadmill test utilizing the Ellestad protocol was performed while the patient was in the VVI pacing mode. Six patients were unable to perform this test. This was due to joint disease in three, pulmonary disease in two, and patient refusal in one case. ECGs were recorded prior to exercise, at the end of each minute of exercise, and at the end of the protocol to document whether the patient’s intrinsic rhythm overrode the pacemaker. A two-dimensional echocardiogram to evaluate left ventricular systolic function at rest was obtained in the dual chamber pacing mode in each patient who had not had an assessment of ventricular function within the 6 months prior to participating in this protocol. Each patient’s baseline pacing rate was continued throughout the study. No pacemaker parameter was changed except for the mode. Baseline medications were maintained. During the course of the study, the patients were encouraged to notify the investigators if intolerable symptoms developed at any time. In these cases, the patients were seen the same day for blood pressure measurement and completion of the next questionnaire, after which they were programmed to the next mode. If this occurred during the last week of the study, they were programmed back to their baseline mode.
Data Analysis The data are expressed as mean i standard deviation. The paired Student’s t-test was used to
1744
compare means for continuous variables. The Yates corrected Chi-square test was used to compare group data.
Results Patients had significantly fewer symptoms in either the DDD or DDI mode as compared to the VVI mode. Table I lists the individual severity scores for each of the 16 symptoms listed on the questionnaire along with the mean r standard deviation and relevant P values. Twelve of the 16 symptoms were significantly less severe in the dual chamber mode as compared to the VVI mode. The most highly significant were shortness of breath, fatigue, dizziness, pulsations in the neck and/or abdomen, cough, and apprehension. In addition, the total symptom score was significantly lower in the dual chamber mode as compared to the VVI mode. Figure 1 shows the total symptom score for each patient in the VVI mode. These reflected either a new symptom or an increase in symptom as compared to the patient’s baseline questionnaire which was obtained when the patient’s pacemaker was programmed to the dual chamber mode. Of the ten patients whose baseline mode was DDI, eight had fewer symptoms in DDI mode than the VVI mode. One patient felt the same and one patient reported fewer symptoms in the VVI mode as compared to the DDI mode. In the 30 patients whose baseline mode was DDD, 25 had fewer symptoms in the DDD mode as compared to the VVI mode. Two patients felt the same and three expressed slight preference for VVI. One of the patients who reported a preference for VVI had a viral illness during the week that he randomized to the DDD mode. No other patient had an intercurrent illness. Figure 2 demonstrates the percentage of patients developing either no, mild, moderate, or severe additional symptoms when in the VVI mode as compared to the dual chamber modes. Sixtyfive percent of all patients had development or exacerbation of symptoms of a moderate or severe degree in the VVI mode as compared to their baseline dual chamber mode. An additional 18% had mild symptoms. Seventeen percent denied more severe symptoms with VVI pacing. Figure 3 shows the total symptom score in
December 1990, Part I1
PACE, Vol. 1 3
PACEMAKER SYNDROME
Table 1.
Severity of Symptoms in VVI Versus Dual-Chamber Pacing Symptom 1. 2. 3. 4. 5. 6. 7. 8. 9.
10. 11. 12. 13. 14. 15. 16.
Shortness of breath Fatigue Dizziness Apprehension Cough Pulsations in necwabdomen Orthopnea Headache Palpitations Chest pain Choking sensation Confusion Pedal edema Sensation of tachycardia Chest congestion Diaphoresis Total Symptoms
DDDlDDl +
VVI* 3.3 4.8 2.9 3.0 1.7 2.0 1.3 1.3 1.5 1.4 1.3 0.9 0.9
f 3.1 ?
*
f ?
f +?
f -+ f
*
f
1.1 +1.1 0.7 29.0
* ?
0.8 1.3 0.8 0.3 0.4 0.4 0.3 0.5 0.5 0.4 0.3 0.2 0.3 0.4 0.5
3.5 3.6 3.6 2.5 3.2 2.5 2.3 3.0 2.6 2.9 2.2 2.3 2.5
1.9 2.3 26.1
= Symptom score during VVI pacing expressed as mean
*
1.8
f 2.3 ?
1.3
2 0.8 f 1.6
1.1 ? 1.3 ? 0.9 +. 1.0 f 1.2 1.2 f 0.6 -+ 0.9 f 1.4 f 1.6 0.1 0.2 7.3 ? 12.4 %
*
*
p Value p p p p p p
< 0.001 < 0.001 < 0.001 < 0.001
0.001 0.002 p < 0.02 p < 0.02 p < 0.04 p < 0.04 p < 0.04 p < 0.05 NS NS NS NS p < 0.001 = =
*
standard deviation Symptom score during baseline dual-chamber pacing (either DDD or DDI) expressed as mean ’. standard deviation
+
=
each mode for those patients who were also programmed to the DVI mode. Eleven of these 16 patients had fewer symptoms in the DVI mode as compared to the VVI mode. Six of the 11 had a further improvement when programmed to the DDD mode. Of the five patients who did not demonstrate a preference for DVI over VVI, all showed a considerable reduction in symptoms when programmed to the DDD mode. Of note, our only patient who developed atrial fibrillation during this study did so when programmed to the DVI mode. This patient had no prior history of atrial arrhythmias. Tables 11,111, and IV compare individual clinical characteristics of patients with the likelihood of their developing symptoms during VVI pacing. None of the following clinical, hemodynamic, or electrophysiological parameters predicted which patients would develop the pacemaker syndrome: gender; body weight; past history of pacemaker syndrome; left ventricular systolic function; hypotension with ventricular pacing; type of conduction system disease; intact ventriculoatrial
PACE, Vol. 13
conduction; or ability of the patient’s endogenous rhythm to override the pacemaker during VVI pacing.
Discussion The signs, symptoms, and presumed mechanisms underlying the pacemaker syndrome have been well summarized by Ausubel and F ~ r m a n . ~ The most important mechanisms are believed to be the loss of atrioventricular synchrony, valvular regurgitation, asynchronous ventricular contraction, inappropriate or an alteration of circulatory reflexes, and ventriculoatrial conduction. Several studies have shown a 10Y0-35% increase in resting cardiac output with restoration of AV synchrony when compared to single chamber ventricular pacing.”-18 The loss of normal AV sequence in a ventricular pacing system contributes to AV valvular regurgitation; increased atrial distension may then affect vasoregulatory ref l e x e ~ . ’ ~ -Retrograde ~’ conduction is estimated to occur in roughly 50% of patients paced for all rea-
December 1990, Part I1
1745
HELDMAN, ET AL.
110
100
90 80
70
s!
8
cn
60
a 50
E
40 c
I-“ 30 20 10
0
I
I
I
I
I
I
I
I
I
I
I
-1 0
Individual Patients
Figure 1. Total symptom score in VVI mode above baseline symptoms (zero line) in dual chamber mode for each of 40 patients (30 DDD baseline, 10 DDI). Each vertical line represents an individual patient.
nea, dizziness, and palpitations were all more sons4 This leads to the atria contracting against a prominent in the VVI mode than dual chamber closed AV valve on a repeated basis leading to mode.’.’* A study by Kenny and associatesg pulmonary and systemic venous regurgitation and looked at ten patients with known coronary artery higher venous pressures that will induce congesdisease and clinically manifested angina. These tion. authors reported a higher incidence of angina durA multiplicity of previous studies, usually ing VVI pacing at a rate of 70 as compared to DDD looking at a small number of paced patients who pacing at a rate of 100 ppm. DDD pacing at higher had dual chamber pacemakers, have demonrates, however, was associated with an increase in strated that in these patients, dual chamber pacing angina. Rediker et al.7 entered 19 patients with was better tolerated than VVI pacing. This was DDD pacemakers into a blinded, randomized based upon exercise tests, echo-Doppler studies, VVI versus DDD pacing. Cardiac study comparing noninvasive and invasive hemodynamic studies, output, percent fractional shortening, and exerand AV 0, differences although no single study cise duration were all significantly greater during used all of these testing m ~ d a l i f i e s . ~ ” ~ ~ ~ ~ * ~ ~ DDD compared to VVI pacing. Sixty-three percent A number of studies have also evaluated the of the patients preferred DDD pacing and none patients’ level of symptoms in a double-blind ranpreferred VVI pacing. domized protocol with the pacing mode being Although there have been claims in the liteither dual chamber or VVI. Symptoms of dysp-
1746
December 1990, Part I1
PACE, Vol. 1 3
PACEMAKER SYNDROME
Figure 2. Severity of symptom development or exacerbation with Wl pacing compared to DDD or DDI mode in 40 patients. See Methods for definition of symptom severity.
erature that rate modulated VVI pacing (VVIR)was hemodynamically equivalent to DDD pacing and may even result in a greater exercise tolerance in those patients with chronotropic incompetence, those studies which evaluated other hemodynamic variables that just exercise duration and/or the patients' level of symptoms have favored dual chamber ~ a c i n g . ~ OAusubel * * ~ and a s s o ~ i a t e s 'ex~ amined 12 patients during moderate levels of exercise using radionuclide angiography to assess the hemodynamics associated with DDD pacing or the functional equivalent of VVIR. They found that DDD pacing resulted in greater ventricular filling and spared contractile reserve at comparable workloads. Dateling et a1.l' evaluated symptomatology in a series of 217 patients programmed to the VVI, VVIR, and DDD pacing modes over a 4-month block of time. Dual chamber pacing resulted in greater symptomatic relief than VVI and VVIR with respect to dizziness, dyspnea, and weakness. These authors found that the presence of ventricuoatrial conduction, percentage of time paced, and heart rate response to exercise were not predictive of the patient experiencing increased symptoms with VVI or VVIR compared to DDD pacing.
PACE, Vol. 13
WI
Figure 3. Total symptom scores for 16 individual patients placed in VVI, DVI, and DDD modes.
Our findings are in keeping with these earlier studies. Eighty-three percent of our patients who had a dual chamber pacemaker experienced either an exacerbation or development of symptoms in the VVI mode as compared to the dual chamber pacing mode. Sixty-five percent of this group experienced moderate-to-severe symptoms and 42% of patients were unable to tolerate VVI pacing for 1 week. Eighteen percent of our patients reported no discernible benefit from dual chamber pacing as compared to VVI pacing. Since our study did not utilize invasive hemodynamic monitoring or other noninvasive testing during each mode of pacing, the possibility that subtle limitations that were able to be corrected by the patient's endogenous reserve mechanisms cannot be excluded. Still, our results support the thesis that the majority of patients in whom atrial pacing or sensing is physiologically feasible are compromised by VVI pacing such that they develop symptoms that
December 1990, Part I1
1747
HELDMAN, ET AL.
Table II.
Relationship Between HistoryIPhysical Findings & Symptoms During VVI Pacing N01 Mild Symptoms
N
Moderat el Severe Symptoms
p Value
Sex
Female Male
17 23
5 (29%)
9 (39%)
14 (61%)
12 (71%)
NS
21 11
8 (38%) 3 (27%)
13 (62%) 8 (63%)
NS
4 36
0 (0%) 14 (39%)
4 (100%) 22 (61%)
NS
Body Weight
2140pounds 10 mmHg LV
1748
=
Left ventricular
December 1990, Part I1
PACE, Vol. 13
PACEMAKER SYNDROME
Table IV.
Relationship Between Conduction Disease and Symptoms During VVI Pacing No/ Mild Symptoms
Moderatel Severe Symptoms
13 (62%) 9 (64%) 4 (80%)
NS
5
8 (38%) 5 (36%) 1 (20%)
18 19
5 (28%) 8 (42%)
13 (72%) 11 (58%)
NS
27 7
9 (33%) 4 (57%)
18 (67%) 3 (43%)
NS
N
p Value
Conduction Disease
sss
AVB
SSS + AVB V-A Conduction
Present Absent
21 14
Ability to Override Pacer
Yes No AVB = Atrioventricular block SSS = Sick sinus syndrome V-A = Ventriculoatrial
A recent debate in both the literature and national meetings has centered on the importance of AV synchrony versus rate responsiveness in the prevention of the pacemaker syndrome. Thus, some patients in this study were also programmed to the DVI mode. DVI pacing provides AV synchrony but not rate responsiveness. Based upon symptom scores, DVI resulted in fewer symptoms than VVI pacing but more symptoms than DDD pacing. Thus, it appears that both AV synchrony (DVI)and rate modulation, in this case based upon the patients’ endogenous atrial activity (DDD),are important in reducing symptoms. Ten patients had their pacemakers programmed to the DDI mode as a baseline and in these cases, DDI was superior to VVI with regard to symptoms. These patients all had documented pathological atrial tachyarrhythmias and thus they. were not programmed to the DDD mode. These patients, however, had sinus node dysfunction and when their atrial rate increased, they had intact AV conduction. One needs to ask why our results and more recent studies in the literature vary to such a large degree from earlier articles with regard to the incidence of pacemaker syndrome. We have demonstrated that 83% of our patients experience in-
PACE, Vol. 13
creased symptoms when paced VVI; thus, if all our patients were to have received a VVI pacemaker, 83% would have the pacemaker syndrome. A smaller percentage of patients had moderate-tosevere symptoms and it is likely that only those patients with severe symptoms would have been labeled pacemaker syndrome had they received a VVI pacemaker. When patients have no basis for comparison and major symptoms such as syncope have been resolved with VVI pacing, both they and their physicians often attribute other nondiagnostic and minor symptoms to the aging process or progression of their underlying disease state. Thus, the true incidence of pacemaker syndrome is not fully appreciated.
Summary The majority of patients paced in the dual chamber mode derive an improved quality of life and decreased symptomatology as compared to single chamber ventricular pacing. Our results support the findings in recent reports when the different pacing modes can be compared in the same patient population. Since there are no clearcut clinical, hemodynamic, or electrophysiological parameters that predict which patients will
December 1990,Part I1
1749
HELDMAN, ET AL.
benefit from this more complex therapy, and more than 80% of an unselected series of patients demonstrate a lower incidence of symptoms with DDD as compared to VVI pacing, it is proposed that dual chamber pacing should be the mode of choice where the atrium is capable of being either sensed or stimulated. Currently, only 25% of paced patients in the United States receive a dual chamber unit and in many other countries, the use of this
modality is even lower. Clearly, a much larger percentage of patients would benefit from dual chamber pacing.
Acknowledgments: We greatly appreciate the assistance of Linda Lorenz in manuscript preparation and Randy Crump in statistical analysis. We also thank Paul Levine, M.D., F.A.C.C., and Adrian Shandling, M.D. for critical review of the manuscript.
References 1. Mitsui T, Hori M, Suma K, et al. The “pacemaking syndrome.” In JE Jacobs (ed.): Proceedings of the Eighth Annual International Conference on Medical and Biological Engineering. Chicago, Association for the Advancement of Medical Instrumentation. 1969, pp. 29-33. 2. Saigusa M. Artificial cardiac pacemaker: Six years experience with cardiac pacemaking for heart block. Jpn Circ J 1969; 33:1583-1586. 3. Mitsui T, Mizuno A, Hasegawa A, et al. Atrial rate as an indicator for optimal pacing rate and the pacemaking syndrome. Ann Cardiol Angiol 1971; 20:371-379. 4. Ausubel K, Furman S. The pacemaker syndrome. Ann Intern Med 1985; 103:420-429. 5. Cohen S, Frank H. Preservation of active atrial transport. Chest 1982: 81:51-54. 6. Kruse I, Arnman K, Conradson T, et al. A comparison of the acute and long-term hemodynamic effects of ventricular inhibited and atrial synchronous ventricular inhibited pacing. Circulation 1982; 65:846-855. 7. Rediker DE, Eagle KA, Herman S, et al. Clinical and hemodynamic comparison of VVI versus DDD pacing in patients with DDD pacemakers. Am J Cardiol 1988; 61:323-329. 8 . Kristensson B-E, Arnman K, Smedgard P, et al. Physiological versus single-rate ventricular pacing: A double-blind crossover study. PACE 1985; 8:73-84. 9. Kenny RA, Ingram A, Mitsuoban T, et al. Optimum pacing mode for patients with angina pectoris. Br Heart J 1986; 56:463-468. 10. Dateline F, Obel IWP. Clinical comparison of VVI, VVIR, and DDD pacemakers in symptomatic relief of bradyarrhythmias. (abstract) PACE 1989; 12:1278. 11. Perrins EJ, Morley CA, Chen SL, et al. Randomized
controlled trial of physiological and ventricular pacing. Br Heart J 1983; 50:112-117. 12. Benchimol A, Ellis JG, Dimond EG. Hemodynamic
1750
13.
14. 15.
16.
17.
18.
19.
20.
21.
22.
consequences of atrial and ventricular pacing in patients with normal and abnormal hearts. Am J Med 1965; 39:911-921. Morris JJ Jr, Entman M, North WC, et al. The changes in cardiac output with reversion of atrial fibrillation to sinus rhythm. Circulation 1965; 31:670-678. Sowton E. Artificial pacemaking and sinus rhythm. Br Heart J 1965; 27:311-318. Samet P, Bernstein WH, Levine S, et al. Hemodynamic effects of tachycardias produced by atrial and ventricular pacing. Am J Med 1965; 39:905910. Leinbach RC, Chamberlain DA, Kastor JA, et al. A comparison of the hemodynamic effects of ventricular and sequential A-V pacing in patients with heart block. Am Heart J 1969; 78:502-508. Samet P, Castillo C, Bernstein W. Hemodynamic consequences of atrial and ventricular pacing in subjects with normal hearts. Am J Cardiol 1966; 18:522-525. Stewart WJ, Dicola VC, Harthorne JW, et al. Doppler ultrasound measurement of cardiac output in patients with physiologic pacemakers. Am J Cardiol 1984; 54:308-312. Ausubel K, Steingart RM, Shimshi M, et al. Maintenance of exercise stroke volume during ventricular versus atrial synchronous pacing: Role of contractility. Circulation 1985; 72:1037-1043. Nishimura RA, Gersh BJ, Vlietstra RE, et al. Hemodynamics and symptomatic consequences of ventricular pacing. PACE 1982; 5:903-910. Alicandri C, Fouad FM, Tarazi RC, et al. Three cases of hypotension and syncope with ventricular pacing: Possible role of atrial reflexes. Am J Cardiol 1978; 42:137-142. Sutton R, Perrins EJ, Morley C, et al. Sustained improvement in exercise tolerance following physiological cardiac pacing. Eur Heart J 1983; 4~781-785.
December 1990, Part I1
PACE, Vol. 1 3
