- Email: [email protected]
Follow-up of asystolic episodes in patients with cardioinhibitory, neurally mediated syncope and VVI pacemaker
Follow-Up of Asystolic Episodes in Patients with Cardioinhibitory, Neurally Mediated Syncope and WI Pai=emaker Carlo Menozzi, MD, Michele Brignole, MD, Gino Lolli, MD, Nicola Bottoni, MD, Daniele Oddone, MD, Lorella Gianfranchi, MD, and German0 Gaggioli, MD The occurrence rate of spontaneous asystolic episodes during long-term follow-up in patients with abnormal asystolic responses induced by means of vasovagal maneuvers was evaluated. The heart rate of 23 patients (mean age 64 + 12 years; 6 women and 17 men) affected by neurally mediated syncope (mean 4.3 +- 4.9 episodes) was continuously monitored by a specially designed implanted pacemaker able to detect and store in its memory all asystolic episodes lasting 3 to 6 or ~6 seconds. Asystolic, neurally mediated syncope was diagnosed when a reflex asystole of ~3 seconds was induced during carotid sinus massage (n q 22), eyeball compression test (n q 3) or headup tilt test (n = 2). During a total of 367 months (mean 15 5~ 7) of monitoring, asystolic episodes occurred in 17 patients (74%): 1,765 episodes of 8 to 6-second (median 3) duration occurred in 14 patients, and 47 episodes of M-second (median 2) duration occurred in 11. The actuarial estimates of occurrence of asystolic episodes of *3 and ~6 seconds were 62 and 53%, respectively, after 2 years of followup. Only 12 episodes of 3 to 6 seconds (0.7%), and 20 episodes of ~6 sea onds (43%) resulted in presyncopal or syncopal symptoms. Thus, an asystolic response to vavagal maneuvers predids the occurrence of spontaneous asystolic episodes during follow-up. With few exceptions, spontaneous episodes are asymptomatic and their incidence is low. (Am J Cardiol1993;72:1152-1155)
From the Laboratory of Electrophysiology and Pacing, First Service of Cardiology, Ospedale S. Maria Nuova, Reggio Emilia, and the Laboratory of Electrophysiology and Pacing, Service of Cardiology, Ospedali Riuniti, Lavagna, Italy. Manuscript received February 10,1993; revised manuscript received June 17, 1993, and accepted June 23. Address for reprints: Michele Brignole, MD, Via A. Grilli 164, I16041 Borzonasca (GE), Italy.
1152
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
I
n patients with syncope of unknown origin, a positive responseto carotid sinus massage,eyeball compression test or head-up tilt test suggestsa diagnosis of neurally mediated syncope, and when the cardioinhibitory reflex is dominant, pacemaker therapy is usually considered.‘” However, the spontaneousoccurrence of the induced asystolic reflex has only seldom been recorded by the conventional diagnostic methods owing to the rarity of the episodes.4Thus, whether the induction of an asystolic responseduring vasovagal maneuvers predicts its spontaneousoccurrence is largely speculative. This study evaluatesthe occurrence rate of spontaneousasystolic episodes,and the positive predictive value of vasovagal maneuversby long-term monitoring of the heart rate obtained with an implanted VVI pacemakerwith Holter function. METHODS
A prospective follow-up study was performed in 23 consecutive patients affected by cardioinhibitory, neurally mediated syncope who received a VVI pacemaker with diagnostic functions. Selection of patients: Patients were recruited from July 1990 to June 1992 from the emergency room, inpatient service or ambulatory program. For inclusion in the study,patients had to fulfill the following criteria: (1) recurrent or severeepisodesof syncope and presyncope that had caused major trauma or risk of death because of sudden onset or the activity of the patient, or both, or becauseof their frequency and intensity, had caused the patient discomfort and interfered with daily activity; (2) a positive asystolic response of >3 seconds during carotid sinus massageor eyeball compression,with reproduction of spontaneoussymptoms. Patients with a positive responseto the head-up tilt test, either asystolic or nonasystolic, were included only if they also had an associatedasystolic responseto carotid sinus massageor the eyeball compressiontest. It is generally acceptedthat VVI pacing is contraindicated in patients with a positive head-up tilt test.3Therefore, the responseto the head-up tilt test did not constitute an inclusion criterion, but the test was used for stratification of the results during follow-up. We also excluded patients with a positive responseto carotid sinus massagewho were consideredat risk of VVI pacemakersyndrome; on the basis of a previous study,5a VVI pacemakersyndrome was suspected in the cardioinhibitory type of carotid sinus syndrome with VVI pacemaker effect and in the mixed type of carotid sinus syndrome with VVI pacemakereffect,ventriculoatrial conduction or orthostatic hypotension; and (3) no other identiliable cause of symptoms, despite a careful history, thorough physical examination, baseline NOVEMBER15, 1993
TABLE I Patient
Data Follow-Up
Patient
Age (yr) & Sex
Underlying Heart Disease
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
51M 66M 68M 48M 86F 84F 78M 56M 72M 69M 78M 58M 61M 62F 54M 62M 51M 64M 71M 74F 65M 40F 65F
DCM Previous AMI HCD 0 CHD HCD CHD DCM Ml CHD HCD 0 0 HCD 0 Previous AMI 0 HCD CHD 0 0 0 0
Diagnostic Tests* CSM Cl Cl Cl Cl M Cl Cl M M M M Cl Cl Cl Cl Cl Cl Cl M M Cl
(8.1) (9.6) (8.5) (6.1) (10.2) (8.6) (8.9) (8.8) (7.7) (9.0) (7.1) (7.0) (6.5) (9.3) (7.3) (8.0) (7.4) (6.6) (9.2) (9.5) (9.0) 0 0
Pauses (number)
EBC
HUT
Duration (mos)
0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cl (10.2) 0 0 0 0 0 0 Cl (5.5) Cl (11.0)
0 0 0 0 0 0 0 0 0 0 0 0 0 0 M M M M (7.4) M (6.1) M VD 0 VD
25 25 24 22 22 16 16 14 12 10 7 7 6 3 24 24 16 14 13 10 6 24 17
Symptoms (number)
> 6 Seconds
3 to 6 Seconds
Syncope
Presyncope
1 0 2 1 0 1 0 0 2 0 0 2 0 1 0 30 0 0 3 0 1 3 0
1 0 14 3 6 1,325 0 200 1 0 1 8 0 0 0 173 1 i 30 0 0 0 1
0 0 1 1 0 2 0 0 0 0 0 2 0 0 2 1 0 0 1 0 0 0 0
0 0 1 1 0 3 0 0 2 0 0 3 0 0 0 6 1 0 4 2 1 14 2
-
*Numbers in parentheses are duration of maximal asystolic response (expressed in seconds) in cases with an asystolic response of > 3 seconds. AMI = acute myocardial infarction: CHD = congestive heart disease,; Cl = cardioinhibitory response; CSM = carotid sinus massage; DCM = dilated cardiomyopathy (nonischemic); EBC = eyeball compression: HCD = systemic hypertension with cardiac involvement; HUT = head-up tilting; M = mixed response; MI = mitral incompetence; VD = vasodepressor K?SPO”SC
laboratory testing, neurologic evaluation, 1Zlead electrocardiogram, prolonged electrocardiographic monitoring (ambulatory or telemetric) of >24-hour duration, chest X-ray, echocardiographic evaluation, intracavitaty electrophysiologic study, and definite evaluation of any clinical or historical tinding suggesting another causeof the symptoms. Exclusion criteria were described previously.’ ,6 The 23 enrolled patients (17 men and 6 women) (Table I) had a mean age of 64 + 12 years. Patients had complained of a mean of 4.3 f 4.9 syncopal episodes, 1.9 f 1.2 of which had occurred during the previous year. History of presyncopal episodes was also present in 15 patients. An associated mild heart disease was present in 14 patients, but none had signs of heart failure or were in class >I of the classification of the New York Heart Association. Induction tests: Patients underwent carotid sinus massage,eyeball compression and head-up tilt testing. The tests were performed in an isolated room (while patients were in the nonfasting state) between 8:OOA.M. and 1:00 P.M. All cardioactive and vasoactivedrugs were withdrawn. The methods of execution of the tests were described and validated previously.1,5,7 Carotid sinus massagewas performed in both right and left carotid sinusesin the supine and erect positions for 10 seconds.Vasodepressorreflex was evaluated after suppressionof the cardioinhibitory reflex by an-opineinfusion (0.02 mg/kg) when indicated.ls5The eyeball compression test was performed in the supine and erect positions. The maximal tolerated thumb pressure was applied to the closed eyelids of both orbits for 10 seconds.’
The head-up tilt test was performed on a tilt table at 60” for 60 minutes according to the Westminsterprotocol.1,7 Definitions: A positive test was defined as the development of syncope in association with bradycardia (or asystole) or hypotension, or both. An asystolic response was deiined as the development of ventricular asystolle of 23-second duration. Vasodepressorresponsewas delined as a decreasein systolic blood pressureof 2.50mm Hg (after atropine.infusion in the case of carotid sinus massageor eyeball compression). Mixed response was defined as the association of asystole of 23 seconds(or heart rate 145 beats/min in the case of the head-up tilt test) with a reduction in systolic blood pressureof 250 mm Hg. Pacemaker Hotter desaiption: After informed consent had been obtained, all patients received a speci,al pacemakermodified from a commercial device (Theorema 90, Sorin Biomedica S.p.A, Saluggia, Italy). This multiprogrammable, telemetric, bipolar VVI pacemaker has a very long hysteresis (maximal standby pacing cycle length of 6 seconds), and is equipped with a microprocessor circuit that can record and store in its virtually unlimited memory all spontaneousbradycardisc episodes. In each patient, the pacemakerwas initially programmed to record asystolic episodesof 3- to 6- and >6-second duration. If a spontaneousasystole of >6 secondsoccurred, the pacer switched to a basic pal;ing rate of 60 beats/min, and the asystolic event was memorized by the pacemaker; if a <6-second asystole occurred, the pacemakerdid not emit any pulse, and the sensedelectrical events were recorded. If during followup, a spontaneoussyncope occurred that was thought to LONG-TERMMONITORINGOFASYSTOLICREFLEXES 1153
be related to an asystolic episode, the pacemaker was rapidly reprogrammedto a hysteresis of 13 seconds. Followup: After implantation of the pacemaker,patients were trained to record their symptoms daily in a diary and were examined in our pacemakerclinics every 3 months or as soon as possible after a syncopal episode. During each visit, Holter data stored in the memory of the pacemakerwere analyzed, and the pacemaker was reset. Syncope was deIined as sudden and transient loss of consciousness,with inability to maintain postural tone. F’resyncopewas delined as the complex and premonitory signs and symptoms of imminent syncope (dizziness, suddenweakness,near fainting, blurring of the vision, and so forth), with difficulty in maintaining postural tone. RESULTS
During a total of 357 months of monitoring (mean 15.5 + 7.1 months/patient), 1,812 asystolic episodes of >3-second duration occurred in 17 patients (74%): 1,765 episodesof 3- to B-second(median 3) duration occurred in 14 patients (61%), and 47 episodesof >6-second (median 2) duration occurred in 11 (48%). Asystolic episodeswere found in 47 of 166 pacemakerexaminations (28%). The actuarial rates of occurrence of asystolic episodesare reported in Figure 1. The overall predicted occurrence rate of asystolic episodes (i.e., durations >6 and 3 to 6 seconds)was 64% at 12 months and 82% at 24 months of follow-up; the predicted actuarial rate of asystolic episodes of >6 secondsalone was 42% at 12 months and 53% at 24 months of follow-up. Most asystolic eventsoccurred in 3 patients (nos. 6, 8 and 16) who had a total of 1,729 events (mean 384/year), whereasthe remaining 14 had only 83 events (mean 4.7lyear). During follow-up, 10 episodesof syncope and 40 of presyncopeoccurred in 13 patients (57%). In all, 32 episodes (8 syncope and 25 presyncope) occurred in periods during which the subsequentpacemakerexamination revealed the presenceof an asystole of 3- to 6-second (n = 12) or >6-second (n = 20) duration. Therefore, only 12 of 1,765 asystolic episodes of 3 to 6 seconds (On%), and 20 of 47 of 6 seconds (43%) could have caused some symptoms. Nonasystolic symptoms occurred in 5 patients (nos. 15, 17,20, 22 and 23); all except 1 had had a positive nonasystolic responseduring ll-23 %
OCCURRENCE OF ASYSTOLE
ll=16
“=8
“=6
100
>3 set
MONTHS I
FIGURE 1. Actuarial rates of occunence of asystolic events of ~3 and >&second duration (estimated with KaplaMeier actuarial IifHeble method).
1154
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
head-up tilt test, and 3 had also had some asystolic episodesrecorded during other periods of examination. No injury or related illness occurred due to recurrence of syncope. After the syncopal recurrences,the hysteresis value was reprogrammed to 53 seconds in 5 patients; during the subsequent10 f 3 months of follow-up, no further episode of syncope occurred. DISCUSSION
The main result of this study is that patients affected by neurally mediated syncope with an asystolic responseof >3 secondsduring carotid sinus massageand eyeball compression have a high probability of occurrence of similar spontaneousasystolic events during the subsequent2 years of follow-up. Thus, a positive asystolic responseto 1 of these tests can predict the occurrence of similar spontaneousevents; in this study, the positive predictive value was 74%. We do not know the predictive value in patients with nonsevere,neurally mediated syncope. The natural history of asystolic neural reflexes appearsto be unpredictable; a few patients had’ a relatively high number of asystolic episodes,whereas others had no recurrence for a long period. The specificity of a pause of >3 seconds detected during long-term electrocardiographic monitoring using an implanted device is obviously unknown, becausenormal control subjects cannot be paced. However, analysis of many patients undergoing 24-hour, dynamic, electrocardiographic monitoring has revealed that pauses of >3 secondsare very unusual in healthy subjects, even during sleep, especially if they are old.8-11 Therefore, the results suggestthat in patients with an asystolic response to vasovagal maneuvers, most episodes of syncope are probably due to a spontaneous asystolic reflex, thus supporting the conviction that these patients are suitable candidates for pacemakertherapy (i.e., the most efficacioustherapy for asystole). The study was not designed to evaluate the recurrence rate of syncopeand presyncope.It is probable that the implanted pacemakercould prevent several syncopal recurrences,despite being programmedwith a long hysteresis; for example, in this study, syncope and presyncope recurred in 30 and 52% of patients, respectively, whereas the natural history of those affectedby carotid sinus syndrome is characterizedby 45 and 86% recurrence rates, respectively, over a similar follow-up period.12In some cases,symptoms recurred in the absence of asystolic pausesof >3 seconds;thus in these cases, severe bradycardia, or a dominant vasodepressorreflex or another etiology, was responsible for the symptoms. Moreover, some patients may have had nonasystolic symptoms during the sameperiod in which an asystolic event occurred. The results are in accordancewith the constant tinding that syncope or presyncoperecurs in a minority of permanently paced patients affectedby neurally mediated syncope.3,5J2 A limitation of this study is that only patients suitable for VVI pacing could be included. In patients with a severevasodepressorreflex associatedwith cardioinhibition (i.e., mixed forms), the test may not have had such a good predictive value. BecauseVVI pacing is generally supposed to be contraindicated in patients with a NOVEMBER15, 1993
positive head-up tilt test,3 we included only those who tions, spontaneousepisodes are asymptomatic and their had an associated dominant cardioinhibitory response incidence is low. during carotid sinus massage.This associationhas been reported to be present in approximately 50% of patients Brignole M, Menozd C, Gianfranchi L, Oddone D, Lolli G, Bexhdla A. Carotid with positive responsesto each test.l During the follow- 1.sinus massage, eyeball compression, and head-up tilt test in patients with syncope up period, asystolic events occurred in similar propor- of uncertain origin and in control subjects. Am Heart J 1991;122:1~1651. 2. Brignole M, Menozzi C, Gianfranchi L, Bottoni N, Lolli G. The clinical and tions among patients with both positive and negative prognostic significance of the asystolic response during the head-up tilt test. Eur J head-up tilt tests (6 of 8 and 11of 15, respectively), and Cardiac Pacing & Electrophysiol 19922: lG%113. asystolic events also occurred in 4 of 6 patients who had 3. Fitzpatrick A, Sutton R. Tilting towards a diagnosis in recurrent unexplained Lmlcet 1989;1:858-860. had a nonasystolic responseduring head-up tilting; how- 4.syncope. Oddone D, Brignole M, Menozzi C, Gianfranchi L, Lolli G. Spontaneous ccever, nonasystolic symptoms occurred more frequently currence of the induced cadioinhibitory vasovagal reflex. PACE 19??1;14:415-419. in those with a positive responseto tilting (4 of 8 vs 1 5. Brignole M, Menozzi C, Lolli G, Oddone D, Gianfranchi L. Validation of a new method for the choice of the pacing mode in carotid sinus syndrome with or withof 15; p = 0.03). Therefore, it appears that when con- out sinus bradycardia. PACE 1991;14:19&203. trasting results of vasovagalmaneuversare presentin the 6. Kapoor WN, Karpf M, Wieand S, Peterson JR, Levey GS. A prospective evaluation and follow-up of patients with syncope N En@ .I Med 1993;309:197-204. samepatients, both asystolic and nonasystolic symptoms 7. Kenny RA, Ingram A, Bayliss J, Sutton R. Head-up tilt: a useful test for incan occur during follow-up. vestigating unexplained syncope. Lancet 1986;2:1352-1354, 6. Viitasalo M, Halonen L, Pwtinen M, Livanainen M, Kala R, Eisalo A. Sleep The pacemakerused in this study could detect much cardiac rhythm in healthy men. Ann Med 1991;23:135-139. longer pausesthan those in previous studies13-16; there- 9.andMolgaard H, Sorensen E, Bjerregaard P. Minimal heart rates and longest pansfore, this device provides a unique opportunity to study es in healthy adult subjects on hvo occasions eight years apart. Eur Heart J 198#9; the symptomatology of prolonged spontaneousvasova- 10:758-764. Kant&p JP, Sage E, Duchene-Mamllaz P. Findings in ambulatory clecbogal pauses.We found that most asystolic pausesof up 10. cardiographic monitoring in subjects older than 80 years. Am J Cardiol 1986587: to 6-second duration do not cause symptoms; only 0.7% 398-401. 11. Wajngarten M, Gmpi C, Bellotti GM, Da Luz P, Azul L, Pileggi F. Frequenof asystolic episodes of 3- to 6-second duration and cy and significance of cardiac rhythm disturbances in healthy elderly individuals .I 43% of 6-second duration were symptomatic. This re- Electmcardiol 1990;23:171-176. sult suggestscaution in the use of pacemaker therapy 12. Brignole M, Menozzi C, Lolli G, Bottoni N, Gaggioli G. Long-term outcome paced and non-paced patients with severe carotid sinus syndrome. Am J Card~‘ol based only on the finding of asymptomatic prolonged of1992;69:1039-1043. pauseson electrocardiographic monitoring. This finding 13. Murdock CJ, Klein GJ, Yee R, Leitch JW, Tea WS, Norris C. Feasibility of is in contrast with that of Ector et all7 who found dur- long-tam electrccaxiiogmphic monitoring with an implanted device for syncope diPACE 1990;13:1374-1378. ing dynamic electrocardiographic monitoring, an as- agnosis. 14. Menozzi C, Brignole M, Pagani P, Lolli G, Casali G. Assessment of VVI dsociation of symptoms with pausesof >3 secondsin 45 agnostic mode in patients with caxlioinhibitory carotid sinus syndrome. PACE 1988; 11:164-1646. of 53 patients. However, our finding is in agreementwith 16. Rosenquist M, Edhag KO, V&in HO. Clinical experience with a bradycardia the study of Hilgard et all8 in which only 5 of 52 pa- indicating pacemaker. PACE 1983;6:515-524. tients with pausesof >3 secondshad related symptoms. 16. Shaw DB, Kekwick CA, Vale D, Whistance TWT. Unexplained syncope-a diagnostic pacemaker? PACE 1983;6:72&725. This discrepancy is probably due to a difference in study 17. Ector H, Relies L, De Geest H. Dynamic electrocardiography and ventticulla population. pauses of 3 seconds and more: etiology and therapeutic implications. PACE 1983; In conclusion, an asystolic responseto vasovagalma- 6548-551. Hilgad J, Ezri MD, Denes P. Significance of ventricular pauses of three. se%neuvers predicts the occurrence of similar spontaneous 16. ends or more detected on twenty-four-hour Halter recordings. Am J Cardiol 1985; asystolic episodes during follow-up. With few excep- 55:1005-1008.
LONG-TERMMONITORINGOF ASYSTOLICREFLEXES 1155
From the Laboratory of Electrophysiology and Pacing, First Service of Cardiology, Ospedale S. Maria Nuova, Reggio Emilia, and the Laboratory of Electrophysiology and Pacing, Service of Cardiology, Ospedali Riuniti, Lavagna, Italy. Manuscript received February 10,1993; revised manuscript received June 17, 1993, and accepted June 23. Address for reprints: Michele Brignole, MD, Via A. Grilli 164, I16041 Borzonasca (GE), Italy.
1152
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
I
n patients with syncope of unknown origin, a positive responseto carotid sinus massage,eyeball compression test or head-up tilt test suggestsa diagnosis of neurally mediated syncope, and when the cardioinhibitory reflex is dominant, pacemaker therapy is usually considered.‘” However, the spontaneousoccurrence of the induced asystolic reflex has only seldom been recorded by the conventional diagnostic methods owing to the rarity of the episodes.4Thus, whether the induction of an asystolic responseduring vasovagal maneuvers predicts its spontaneousoccurrence is largely speculative. This study evaluatesthe occurrence rate of spontaneousasystolic episodes,and the positive predictive value of vasovagal maneuversby long-term monitoring of the heart rate obtained with an implanted VVI pacemakerwith Holter function. METHODS
A prospective follow-up study was performed in 23 consecutive patients affected by cardioinhibitory, neurally mediated syncope who received a VVI pacemaker with diagnostic functions. Selection of patients: Patients were recruited from July 1990 to June 1992 from the emergency room, inpatient service or ambulatory program. For inclusion in the study,patients had to fulfill the following criteria: (1) recurrent or severeepisodesof syncope and presyncope that had caused major trauma or risk of death because of sudden onset or the activity of the patient, or both, or becauseof their frequency and intensity, had caused the patient discomfort and interfered with daily activity; (2) a positive asystolic response of >3 seconds during carotid sinus massageor eyeball compression,with reproduction of spontaneoussymptoms. Patients with a positive responseto the head-up tilt test, either asystolic or nonasystolic, were included only if they also had an associatedasystolic responseto carotid sinus massageor the eyeball compressiontest. It is generally acceptedthat VVI pacing is contraindicated in patients with a positive head-up tilt test.3Therefore, the responseto the head-up tilt test did not constitute an inclusion criterion, but the test was used for stratification of the results during follow-up. We also excluded patients with a positive responseto carotid sinus massagewho were consideredat risk of VVI pacemakersyndrome; on the basis of a previous study,5a VVI pacemakersyndrome was suspected in the cardioinhibitory type of carotid sinus syndrome with VVI pacemaker effect and in the mixed type of carotid sinus syndrome with VVI pacemakereffect,ventriculoatrial conduction or orthostatic hypotension; and (3) no other identiliable cause of symptoms, despite a careful history, thorough physical examination, baseline NOVEMBER15, 1993
TABLE I Patient
Data Follow-Up
Patient
Age (yr) & Sex
Underlying Heart Disease
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
51M 66M 68M 48M 86F 84F 78M 56M 72M 69M 78M 58M 61M 62F 54M 62M 51M 64M 71M 74F 65M 40F 65F
DCM Previous AMI HCD 0 CHD HCD CHD DCM Ml CHD HCD 0 0 HCD 0 Previous AMI 0 HCD CHD 0 0 0 0
Diagnostic Tests* CSM Cl Cl Cl Cl M Cl Cl M M M M Cl Cl Cl Cl Cl Cl Cl M M Cl
(8.1) (9.6) (8.5) (6.1) (10.2) (8.6) (8.9) (8.8) (7.7) (9.0) (7.1) (7.0) (6.5) (9.3) (7.3) (8.0) (7.4) (6.6) (9.2) (9.5) (9.0) 0 0
Pauses (number)
EBC
HUT
Duration (mos)
0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cl (10.2) 0 0 0 0 0 0 Cl (5.5) Cl (11.0)
0 0 0 0 0 0 0 0 0 0 0 0 0 0 M M M M (7.4) M (6.1) M VD 0 VD
25 25 24 22 22 16 16 14 12 10 7 7 6 3 24 24 16 14 13 10 6 24 17
Symptoms (number)
> 6 Seconds
3 to 6 Seconds
Syncope
Presyncope
1 0 2 1 0 1 0 0 2 0 0 2 0 1 0 30 0 0 3 0 1 3 0
1 0 14 3 6 1,325 0 200 1 0 1 8 0 0 0 173 1 i 30 0 0 0 1
0 0 1 1 0 2 0 0 0 0 0 2 0 0 2 1 0 0 1 0 0 0 0
0 0 1 1 0 3 0 0 2 0 0 3 0 0 0 6 1 0 4 2 1 14 2
-
*Numbers in parentheses are duration of maximal asystolic response (expressed in seconds) in cases with an asystolic response of > 3 seconds. AMI = acute myocardial infarction: CHD = congestive heart disease,; Cl = cardioinhibitory response; CSM = carotid sinus massage; DCM = dilated cardiomyopathy (nonischemic); EBC = eyeball compression: HCD = systemic hypertension with cardiac involvement; HUT = head-up tilting; M = mixed response; MI = mitral incompetence; VD = vasodepressor K?SPO”SC
laboratory testing, neurologic evaluation, 1Zlead electrocardiogram, prolonged electrocardiographic monitoring (ambulatory or telemetric) of >24-hour duration, chest X-ray, echocardiographic evaluation, intracavitaty electrophysiologic study, and definite evaluation of any clinical or historical tinding suggesting another causeof the symptoms. Exclusion criteria were described previously.’ ,6 The 23 enrolled patients (17 men and 6 women) (Table I) had a mean age of 64 + 12 years. Patients had complained of a mean of 4.3 f 4.9 syncopal episodes, 1.9 f 1.2 of which had occurred during the previous year. History of presyncopal episodes was also present in 15 patients. An associated mild heart disease was present in 14 patients, but none had signs of heart failure or were in class >I of the classification of the New York Heart Association. Induction tests: Patients underwent carotid sinus massage,eyeball compression and head-up tilt testing. The tests were performed in an isolated room (while patients were in the nonfasting state) between 8:OOA.M. and 1:00 P.M. All cardioactive and vasoactivedrugs were withdrawn. The methods of execution of the tests were described and validated previously.1,5,7 Carotid sinus massagewas performed in both right and left carotid sinusesin the supine and erect positions for 10 seconds.Vasodepressorreflex was evaluated after suppressionof the cardioinhibitory reflex by an-opineinfusion (0.02 mg/kg) when indicated.ls5The eyeball compression test was performed in the supine and erect positions. The maximal tolerated thumb pressure was applied to the closed eyelids of both orbits for 10 seconds.’
The head-up tilt test was performed on a tilt table at 60” for 60 minutes according to the Westminsterprotocol.1,7 Definitions: A positive test was defined as the development of syncope in association with bradycardia (or asystole) or hypotension, or both. An asystolic response was deiined as the development of ventricular asystolle of 23-second duration. Vasodepressorresponsewas delined as a decreasein systolic blood pressureof 2.50mm Hg (after atropine.infusion in the case of carotid sinus massageor eyeball compression). Mixed response was defined as the association of asystole of 23 seconds(or heart rate 145 beats/min in the case of the head-up tilt test) with a reduction in systolic blood pressureof 250 mm Hg. Pacemaker Hotter desaiption: After informed consent had been obtained, all patients received a speci,al pacemakermodified from a commercial device (Theorema 90, Sorin Biomedica S.p.A, Saluggia, Italy). This multiprogrammable, telemetric, bipolar VVI pacemaker has a very long hysteresis (maximal standby pacing cycle length of 6 seconds), and is equipped with a microprocessor circuit that can record and store in its virtually unlimited memory all spontaneousbradycardisc episodes. In each patient, the pacemakerwas initially programmed to record asystolic episodesof 3- to 6- and >6-second duration. If a spontaneousasystole of >6 secondsoccurred, the pacer switched to a basic pal;ing rate of 60 beats/min, and the asystolic event was memorized by the pacemaker; if a <6-second asystole occurred, the pacemakerdid not emit any pulse, and the sensedelectrical events were recorded. If during followup, a spontaneoussyncope occurred that was thought to LONG-TERMMONITORINGOFASYSTOLICREFLEXES 1153
be related to an asystolic episode, the pacemaker was rapidly reprogrammedto a hysteresis of 13 seconds. Followup: After implantation of the pacemaker,patients were trained to record their symptoms daily in a diary and were examined in our pacemakerclinics every 3 months or as soon as possible after a syncopal episode. During each visit, Holter data stored in the memory of the pacemakerwere analyzed, and the pacemaker was reset. Syncope was deIined as sudden and transient loss of consciousness,with inability to maintain postural tone. F’resyncopewas delined as the complex and premonitory signs and symptoms of imminent syncope (dizziness, suddenweakness,near fainting, blurring of the vision, and so forth), with difficulty in maintaining postural tone. RESULTS
During a total of 357 months of monitoring (mean 15.5 + 7.1 months/patient), 1,812 asystolic episodes of >3-second duration occurred in 17 patients (74%): 1,765 episodesof 3- to B-second(median 3) duration occurred in 14 patients (61%), and 47 episodesof >6-second (median 2) duration occurred in 11 (48%). Asystolic episodeswere found in 47 of 166 pacemakerexaminations (28%). The actuarial rates of occurrence of asystolic episodesare reported in Figure 1. The overall predicted occurrence rate of asystolic episodes (i.e., durations >6 and 3 to 6 seconds)was 64% at 12 months and 82% at 24 months of follow-up; the predicted actuarial rate of asystolic episodes of >6 secondsalone was 42% at 12 months and 53% at 24 months of follow-up. Most asystolic eventsoccurred in 3 patients (nos. 6, 8 and 16) who had a total of 1,729 events (mean 384/year), whereasthe remaining 14 had only 83 events (mean 4.7lyear). During follow-up, 10 episodesof syncope and 40 of presyncopeoccurred in 13 patients (57%). In all, 32 episodes (8 syncope and 25 presyncope) occurred in periods during which the subsequentpacemakerexamination revealed the presenceof an asystole of 3- to 6-second (n = 12) or >6-second (n = 20) duration. Therefore, only 12 of 1,765 asystolic episodes of 3 to 6 seconds (On%), and 20 of 47 of 6 seconds (43%) could have caused some symptoms. Nonasystolic symptoms occurred in 5 patients (nos. 15, 17,20, 22 and 23); all except 1 had had a positive nonasystolic responseduring ll-23 %
OCCURRENCE OF ASYSTOLE
ll=16
“=8
“=6
100
>3 set
MONTHS I
FIGURE 1. Actuarial rates of occunence of asystolic events of ~3 and >&second duration (estimated with KaplaMeier actuarial IifHeble method).
1154
THE AMERICANJOURNALOF CARDIOLOGY VOLUME72
head-up tilt test, and 3 had also had some asystolic episodesrecorded during other periods of examination. No injury or related illness occurred due to recurrence of syncope. After the syncopal recurrences,the hysteresis value was reprogrammed to 53 seconds in 5 patients; during the subsequent10 f 3 months of follow-up, no further episode of syncope occurred. DISCUSSION
The main result of this study is that patients affected by neurally mediated syncope with an asystolic responseof >3 secondsduring carotid sinus massageand eyeball compression have a high probability of occurrence of similar spontaneousasystolic events during the subsequent2 years of follow-up. Thus, a positive asystolic responseto 1 of these tests can predict the occurrence of similar spontaneousevents; in this study, the positive predictive value was 74%. We do not know the predictive value in patients with nonsevere,neurally mediated syncope. The natural history of asystolic neural reflexes appearsto be unpredictable; a few patients had’ a relatively high number of asystolic episodes,whereas others had no recurrence for a long period. The specificity of a pause of >3 seconds detected during long-term electrocardiographic monitoring using an implanted device is obviously unknown, becausenormal control subjects cannot be paced. However, analysis of many patients undergoing 24-hour, dynamic, electrocardiographic monitoring has revealed that pauses of >3 secondsare very unusual in healthy subjects, even during sleep, especially if they are old.8-11 Therefore, the results suggestthat in patients with an asystolic response to vasovagal maneuvers, most episodes of syncope are probably due to a spontaneous asystolic reflex, thus supporting the conviction that these patients are suitable candidates for pacemakertherapy (i.e., the most efficacioustherapy for asystole). The study was not designed to evaluate the recurrence rate of syncopeand presyncope.It is probable that the implanted pacemakercould prevent several syncopal recurrences,despite being programmedwith a long hysteresis; for example, in this study, syncope and presyncope recurred in 30 and 52% of patients, respectively, whereas the natural history of those affectedby carotid sinus syndrome is characterizedby 45 and 86% recurrence rates, respectively, over a similar follow-up period.12In some cases,symptoms recurred in the absence of asystolic pausesof >3 seconds;thus in these cases, severe bradycardia, or a dominant vasodepressorreflex or another etiology, was responsible for the symptoms. Moreover, some patients may have had nonasystolic symptoms during the sameperiod in which an asystolic event occurred. The results are in accordancewith the constant tinding that syncope or presyncoperecurs in a minority of permanently paced patients affectedby neurally mediated syncope.3,5J2 A limitation of this study is that only patients suitable for VVI pacing could be included. In patients with a severevasodepressorreflex associatedwith cardioinhibition (i.e., mixed forms), the test may not have had such a good predictive value. BecauseVVI pacing is generally supposed to be contraindicated in patients with a NOVEMBER15, 1993
positive head-up tilt test,3 we included only those who tions, spontaneousepisodes are asymptomatic and their had an associated dominant cardioinhibitory response incidence is low. during carotid sinus massage.This associationhas been reported to be present in approximately 50% of patients Brignole M, Menozd C, Gianfranchi L, Oddone D, Lolli G, Bexhdla A. Carotid with positive responsesto each test.l During the follow- 1.sinus massage, eyeball compression, and head-up tilt test in patients with syncope up period, asystolic events occurred in similar propor- of uncertain origin and in control subjects. Am Heart J 1991;122:1~1651. 2. Brignole M, Menozzi C, Gianfranchi L, Bottoni N, Lolli G. The clinical and tions among patients with both positive and negative prognostic significance of the asystolic response during the head-up tilt test. Eur J head-up tilt tests (6 of 8 and 11of 15, respectively), and Cardiac Pacing & Electrophysiol 19922: lG%113. asystolic events also occurred in 4 of 6 patients who had 3. Fitzpatrick A, Sutton R. Tilting towards a diagnosis in recurrent unexplained Lmlcet 1989;1:858-860. had a nonasystolic responseduring head-up tilting; how- 4.syncope. Oddone D, Brignole M, Menozzi C, Gianfranchi L, Lolli G. Spontaneous ccever, nonasystolic symptoms occurred more frequently currence of the induced cadioinhibitory vasovagal reflex. PACE 19??1;14:415-419. in those with a positive responseto tilting (4 of 8 vs 1 5. Brignole M, Menozzi C, Lolli G, Oddone D, Gianfranchi L. Validation of a new method for the choice of the pacing mode in carotid sinus syndrome with or withof 15; p = 0.03). Therefore, it appears that when con- out sinus bradycardia. PACE 1991;14:19&203. trasting results of vasovagalmaneuversare presentin the 6. Kapoor WN, Karpf M, Wieand S, Peterson JR, Levey GS. A prospective evaluation and follow-up of patients with syncope N En@ .I Med 1993;309:197-204. samepatients, both asystolic and nonasystolic symptoms 7. Kenny RA, Ingram A, Bayliss J, Sutton R. Head-up tilt: a useful test for incan occur during follow-up. vestigating unexplained syncope. Lancet 1986;2:1352-1354, 6. Viitasalo M, Halonen L, Pwtinen M, Livanainen M, Kala R, Eisalo A. Sleep The pacemakerused in this study could detect much cardiac rhythm in healthy men. Ann Med 1991;23:135-139. longer pausesthan those in previous studies13-16; there- 9.andMolgaard H, Sorensen E, Bjerregaard P. Minimal heart rates and longest pansfore, this device provides a unique opportunity to study es in healthy adult subjects on hvo occasions eight years apart. Eur Heart J 198#9; the symptomatology of prolonged spontaneousvasova- 10:758-764. Kant&p JP, Sage E, Duchene-Mamllaz P. Findings in ambulatory clecbogal pauses.We found that most asystolic pausesof up 10. cardiographic monitoring in subjects older than 80 years. Am J Cardiol 1986587: to 6-second duration do not cause symptoms; only 0.7% 398-401. 11. Wajngarten M, Gmpi C, Bellotti GM, Da Luz P, Azul L, Pileggi F. Frequenof asystolic episodes of 3- to 6-second duration and cy and significance of cardiac rhythm disturbances in healthy elderly individuals .I 43% of 6-second duration were symptomatic. This re- Electmcardiol 1990;23:171-176. sult suggestscaution in the use of pacemaker therapy 12. Brignole M, Menozzi C, Lolli G, Bottoni N, Gaggioli G. Long-term outcome paced and non-paced patients with severe carotid sinus syndrome. Am J Card~‘ol based only on the finding of asymptomatic prolonged of1992;69:1039-1043. pauseson electrocardiographic monitoring. This finding 13. Murdock CJ, Klein GJ, Yee R, Leitch JW, Tea WS, Norris C. Feasibility of is in contrast with that of Ector et all7 who found dur- long-tam electrccaxiiogmphic monitoring with an implanted device for syncope diPACE 1990;13:1374-1378. ing dynamic electrocardiographic monitoring, an as- agnosis. 14. Menozzi C, Brignole M, Pagani P, Lolli G, Casali G. Assessment of VVI dsociation of symptoms with pausesof >3 secondsin 45 agnostic mode in patients with caxlioinhibitory carotid sinus syndrome. PACE 1988; 11:164-1646. of 53 patients. However, our finding is in agreementwith 16. Rosenquist M, Edhag KO, V&in HO. Clinical experience with a bradycardia the study of Hilgard et all8 in which only 5 of 52 pa- indicating pacemaker. PACE 1983;6:515-524. tients with pausesof >3 secondshad related symptoms. 16. Shaw DB, Kekwick CA, Vale D, Whistance TWT. Unexplained syncope-a diagnostic pacemaker? PACE 1983;6:72&725. This discrepancy is probably due to a difference in study 17. Ector H, Relies L, De Geest H. Dynamic electrocardiography and ventticulla population. pauses of 3 seconds and more: etiology and therapeutic implications. PACE 1983; In conclusion, an asystolic responseto vasovagalma- 6548-551. Hilgad J, Ezri MD, Denes P. Significance of ventricular pauses of three. se%neuvers predicts the occurrence of similar spontaneous 16. ends or more detected on twenty-four-hour Halter recordings. Am J Cardiol 1985; asystolic episodes during follow-up. With few excep- 55:1005-1008.
LONG-TERMMONITORINGOF ASYSTOLICREFLEXES 1155