Closed loop stimulation in prevention of vasovagal syncope. Inotropy controlled pacing in vasovagal syncope (INVASY): a multicentre randomized, single blind, controlled study
Closed loop stimulation in prevention of vasovagal syncope. Inotropy controlled pacing in vasovagal syncope (INVASY): a multicentre * randomized, single blind, controlled study
Eraldo Occhetta 1 2
Miriam Bortnik 1 2
Roberto Audoglio 0 2
Corrado Vassanelli 1 2
for the INVASY Study Investigators 2
0 SRA , Linarolo, Pavia , Italy
1 Division of Cardiology, Facolta` di Medicina e Chirurgia di Novara, Universita` degli Studi del Piemonte Orientale , Novara , Italy
2 Ospedale Civile, Acqui Terme: Pierluigi Roncarolo; Ospedale S.M. Annunziata, Bagno a Ripoli: Leandro Chiodi; Ospedale degli Infermi, Biella: Marco Marcolongo, Davide Torta; Ospedale S.S. Trinita`, Borgomanero: Marco Zanetta; Policlinico S. Orsola, Bologna: Giuseppe Boriani; Ospedale S. Giovanni, Cagliari: Raimondo Pirisi; Ospedale Civico, Chivasso: Antonio Mazza; Ospedale Careggi, Firenze: Paolo Marconi; Ospedale Civile, Legnago: Diran Igidbashian; Ospedale Policlinico, Milano: Salvatore Romano, Ruggero Manfredini; Centro Cardiologico Monzino, Milano: Claudio Tondo; Policlinico, Monza: Massimo Arlotti, Marcello Chimienti; Ospedale Monaldi, Napoli: Luigi Zamparelli, Luciano Cioffi; Azienda Ospedaliera, Parma: Angelo Carboni; Ospedale di Cisanello, Pisa: Maria Grazia Bongiorni; Ospedale S. Maria degli Ungheresi , Polistena: Rocco Polimeni
Objectives To determine whether dual-chamber rate-adaptive Closed Loop Stimulation (CLS) could prevent recurrence of Vasovagal Syncope (VVS). Background During VVS, an increase in myocardial contractility associated with a reduction of ventricular filling produces an increase in baroreceptor afferent flow and a consequent decrease in the heart rate. The CLS algorithm is a form of rateadaptive pacing, which responds to myocardial contraction dynamics, by measuring variations in right ventricular intracardiac impedance: during an incipient VVS it could increase paced heart rate and avoid bradycardia, arterial hypotension and syncope. Methods Fifty patients (27 males, mean age 59 G 18 year) with severe and recurrent vasovagal syncope and positive Head Up Tilt Test (HUTT) with cardioinhibition, received a CLS pacemaker (INOS2, Biotronik GmbH Co., Germany). The primary end point was recurrence of two VVSs during a minimum of 1 year of follow-up. Randomization between DDD-CLS and DDI mode (40 bpm) pacing was performed only during the first stage of the study (first year): 9/26 randomized to DDI mode (control group) and 17/26 in DDD-CLS mode. All the 24 patients recruited in the second stage of the study (second year) were programmed in DDD-CLS mode.
eol>vasovagal syncope; closed loop stimulation; DDD pacing
Results Of the nine patients randomized to the DDI mode, seven had recurrences
of syncope during the first year. At the end of the first year the nine patients were
reprogrammed to the CLS mode and no syncope occurred after reprogramming. The
41 patients programmed to CLS had a mean follow-up of 19 G 4 months: none
reported VVS, only four (10%) reported occasional presyncope and their quality of
life greatly improved. Positive HUTT at the end of the first year failed to predict the
clinical response to CLS pacing.
Conclusions The study demonstrates the effectiveness of CLS pacing in
preventing cardioinhibitory VVS. A possible placebo effect of pacemaker
implantation occurred in 22% of patients.
ª 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights
In patients with recurrent, severe, cardioinhibitory
vasovagal syncope (VVS), significant bradycardia or
prolonged asystole (up to more than 60 s) and
concomitant hypotension, can produce serious
physical injuries and psychological impairment,
including a substantial limitation of social and
working life [
When VVS is refractory to conventional
measures and/or to pharmacological treatment, the
implantation of a pacemaker may induce clinical
Closed loop stimulation (CLS) and rationale for its use in vasovagal syncope
During VVS, the diminished venous return
stimulates a sympathetic compensatory tone that leads
to a positive inotropic effect [
ventricular filling is reduced, the left ventricular systolic
pressure may increase participating in a
baroreceptor induction of bradycardia [
creating a paradoxical situation: increased inotropic
effect associated with decreased chronotropic
state. This anomalous situation inhibits
sympathetic activity and promotes a vagal effect that
causes VVS by increasing peripheral vasodilatation
with associated reflex bradycardia [
Loop Stimulation (CLS) function performed by the
model INOS2 CLS dual-chamber pacemaker (by
Biotronik GmbH & Co., Germany) tracks the
variations of intracardiac impedance during the systolic
phase of the cardiac cycle on a beat-to-beat
basis . Changes in intracardiac impedance are
closely correlated with both the right and left
ventricular dP/dtmax, making this pacing system
suitable for the detection of changes in
contractility in the early phase of VVS [
]. The CLS
detection of the increased contractility in the first
stage of vasovagal syncope (Fig. 1) could activate
atrioventricular (AV) sequential pacing [
that may anticipate withdrawal of sympathetic
tone and counterbalance the increase in vagal
tone, thus preventing arterial hypotension,
bradycardia and possibly syncope (Fig. 2) [
The promising results of the CLS Pre-INVASY
] justified a larger prospective
randomized controlled study. Thus, the INVASY study was
designed to investigate whether CLS pacing may be
more effective in preventing syncope than a
placebo pacemaker implantation programmed in DDI
INVASY was a prospective, controlled, randomized,
single blind, multicentre study: DDD-CLS versus
DDI pacing mode, with crossover after the second
recurrence of syncope. The tested hypothesis was
that the implantation of a DDD-CLS pacemaker
would reduce the recurrence of syncope by at
least 50% compared with the placebo implantation
of a similar pacemaker programmed in DDI mode at
The ethics committee of the institution of the
principal investigator approved the study protocol.
Because of the clinical characteristics of the
enroled patients (several syncopal recurrences
and cardioinhibitory response to head up tilt test),
the ethics committee did not allow programming
of ODO mode.
All enroled patients gave their written,
Patients had to meet the following criteria for
inclusion in the study:
- more than five syncopal episodes and/or O2 in
the last year before enrolment;
- positive type 2A or 2B (in accordance with the
VASIS classification) cardioinhibitory response
to Head Up Tilt Test (HUTT) [
- age O 18 years;
- proven refractoriness to conventional drug
therapy and tilt training (when performed).
The diagnosis of VVS was based on a positive tilt
test, after exclusion of other possible causes of
syncope by complete systematic cardiac and
neurological evaluation [
]. Exclusion criteria
included: previous myocardial infarction; congestive
heart failure and concomitant severe chronic
diseases (e.g., neurological disorders, metabolic
diseases, neoplasm and life expectance ! 1 year).
Tilt test protocol
The end point of the HUTT was syncope.
First the Westminster protocol [
] (tilting to
60( for 45 min) was used. The blood pressure was
measured every minute. If the response was
negative at the end of 45 min, 300 mg sublingual
nitroglycerine was administered  with continued
Fourteen patients (11 in the CLS arm and three
in the placebo arm) were enroled following a
positive response to nitroglycerine. Only patients with
a cardioinhibitory positive response, i.e. the type 2
of the VASIS classification [
] were eligible for
Patients were assigned to two study arms
according to a central computer-generated 2:1 (DDD-CLS
to DDI ratio) randomization list. This kind of
randomization was chosen after the positive
results observed in the previous CLS Pre-INVASY
All patients received an INOS2 CLS pacemaker.
Patients were not informed of the pacemaker’s
programme, in accordance with a single blind
In the CLS arm the device was programmed in
DDD-CLS mode. Since the response of the CLS
algorithm is self adjusting on intracardiac
impedance variations, the parameters programmed by
the physician were:
(1) the maximum sensor driven pacing rate at
a value of at least 100 ppm (programmed range
100e120 bpm), in accordance with the
experience achieved during the registry and with
(2) the maximum sinus tracking rate (programmed
range 140e160 bpm) to avoid interference with
sinus rate during exercise and emotion;
Time since tilt-up (minutes)
(3) the optimal dynamic AV interval at rest
(programmed range 120e160 ms) to allow
persistent ventricular pacing and
intraventricular impedance detection.
In the control group the devices were
programmed in DDI mode, (lower rate 40 bpm and
AV interval of 150 ms).
Unfortunately, the diagnostic memory of the
device used had no capability to store data about
syncopal episodes requiring pacing.
Patients in both groups received no medication
for syncope. Other drug treatments in progress
were continued in both groups without dosage
Patients were followed every 6 months with the
same pre-defined protocol. A HUTT was performed
after each recurrence of syncope or at the end of
the first year.
The primary end point of the study was the
recurrence of the second syncopal spell.
Secondary end points included:
- the predictive value of HUTT for VVS
- the effectiveness of CLS pacing during HUTT
- the quality of life [
] at 1 year: the quality
of life questionnaire included a total of 35
areas, covering health, psychological, physical
and lifestyle feelings of the patient. The
evaluation of the answers was based on
attributing low and high scores to responses
reflecting negative/worsening and positive/
improving status, respectively (score range
35e105). All patients were asked to fill the
questionnaire at the time of recruitment and at
the end of the study.
The original assumption to calculate the power of
the study was that the CLS paced arm would have
a cumulative risk of recurrence of syncope 50%
lower than the control group, and that only 75% of
the patients included in this latter group would
have a recurrence of syncope during the first year
We anticipated that a total of 100 patients would
have to be followed for 1 year to yield a statistically
significant reduction in the risk of recurrence of
syncope in the CLS paced arm (P % 0.05).
For the primary analysis, all outcomes were
analyzed on the intention-to-treat principle. The
time to first syncopal recurrence in the two
treatment groups was analyzed by means of
Kaplan-Meier event-free curves.
The secondary end points were analyzed by
means of the chi-square test (HUTT predictive
value) and t test for combined estimation of
variance (time-to-syncope and quality of life score).
Data are reported as mean G standard
deviation, unless otherwise specified.
Fifty patients were enroled between May 2000 and
June 2002 in 19 Italian hospitals and all were
followed for a minimum of 12 months and a
maximum of 36 months (18.9 G 4.2 months).
Twenty-six patients were enroled between May
2000 and June 2001. Seventeen in this group were
included in the CLS group and nine in the control
group. During this period no patients randomized to
CLS pacing had syncope, but seven patients in the
DDI group experienced at least one syncopal spell.
For this reason the steering committee decided
to discontinue randomization in July 2001.
From August 2001 to June 2002 all 24 patients
were enroled only in the CLS arm.
The baseline characteristics of patients in the
three groups are summarized in Table 1.
The CLS programmed patients had slightly, but
insignificantly, fewer syncopal episodes in the last
year before enrolment, than the control group
patients. A cardioinhibitory response HUTT with
asystole of O3 s (VASIS 2B) was present in 37 of 41
patients in the CLS group (90%) and in seven of nine
patients (78%) in the control group.
As the clinical characteristics of the three groups
of patients did not significantly differ, the 41
patients who received CLS therapy were combined in
the analysis, versus nine patients in the control
group (pacemaker programmed in DDI) (Table 2).
Primary end point
During the follow-up period (mean 18.9 G 4.2
month/patient; range 12e36 months), none of
the 41 patients in the CLS arm experienced
syncope (0%); four out of 41 patients (9.7%)
reported rare mild presyncopal symptoms and
two (4.9%) reported rare brief dizziness, without
any impairment of daily file. In three patients the
pacemaker upper rate was reprogrammed to a
lower value because of palpitations. No other
clinical complications occurred.
In the control arm, four of nine patients (44%)
had two syncopal spells before the end of the first
year (mean time to first recurrence 1.5 months;
range 0.5e2 months) and they were crossed over
to the CLS pacing group after the second spell;
three patients (33%) had one syncopal recurrence
in the first year (mean time to recurrence 7.3
month; range 4e11 months) and were crossed over
to the CLS pacing group at the end of the study
period; two patients (22%) experienced only
The Table 2 summarizes the primary end point
data and Fig. 3 shows the Kaplan-Meier actuarial
estimation of first recurrence of syncope for the
The seven patients crossed over to CLS pacing
were followed until June 2003 and none of them
reported syncope after device reprogramming.
The two remaining patients in the control group
were reprogrammed to CLS pacing: one
immediately after the end of the study and the second at
the 18-month follow-up because of frequent
episodes of presyncope.
Secondary end points
Predictive value of HUTT
At the end of the first year, or just before pacing
mode crossover, a HUTT was performed in 39/50
patients (30 in CLS and nine in control group)
without reprogramming the pacemaker. Eleven
patients (all CLS) refused testing.
The protocol was identical to that used at
Group I: 15 HUTTs (38%) produced a response
concordant with the clinical picture during
followup e seven were positive (all in the placebo arm
and before crossing over) and eight negative (all
Group II: the remaining 24 HUTTs (62%) e 22 CLS
arm and two syncope-free in the DDI arm had
a positive response with hypotensive presyncopal
symptoms, but without asystole because of pacing
For the 22 patients in the CLS arm the early
increase in pacing rate was insufficient to avoid
a fall in systolic blood pressure of at least 30%
relative to the basal value. None of these patients
had reported syncopal spells during follow-up. The
3 m 6 m 9 m 1 y
in CLS arm
in Placebo arm 9 5
chi-square test performed on these data confirm
that HUTT is an unreliable predictor of clinical
recurrence of VVS in CLS treated patients
(P ! 0.0001).
Effect of CLS pacing during HUTT at 1 year of follow-up
The time from tilt-up to the development of
presyncope was measured in 22 patients in the
CLS arm in Group II who exhibited a positive
response to HUTT at the end of the first year. This
time was compared with that recorded during
HUTT at recruitment.
The mean time-to-syncope/presyncope was
19.7 G 3.8 min (range 8e31 min) at 1 year,
compared with 13.9 G 4.2 min (range 5e22 min) at
recruitment, a median increase of 42% (P ! 0.001).
Further attempts to persuade these patients to
undergo additional HUTT with different pacemaker
Quality of life
At recruitment, the QoL score for all patients
showed a median of 49 (range 35e57).
After 1 year of CLS therapy, the median QoL
score for the 41 treated patients was 85 (range
72e101), with an improvement from 46% to 81% in
a normalized scale of QoL.
QoL scores of the two syncope-free patients in
the control arm were 44 versus 66 and 51 versus
74, at recruitment and follow-up, respectively.
This study shows that dual-chamber CLS pacing
prevents recurrence of vasovagal syncope in
patients with a cardioinhibitory response to HUTT.
Zero recurrence of syncope during CLS treatment
is in accordance with the data achieved in the
PreINVASY registry [
]. This benefit is maintained for
up to 2 years, in the present INVASY study, but it
may extend to over 5 years as the results of the
Pre-INVASY registry suggest.
Despite this relatively long follow-up, it cannot
be excluded that patients implanted with INOS2
CLS pacemakers could experience some
recurrence of syncope during their lives. It is generally
accepted that the natural history of VVS is variable
with long periods free of recurrences: follow-up
for decades would be required to confirm
definitively the clinical benefit.
Syncopal history before enrolment and selection
criteria were basically similar in all the previous
studies (Table 1).
The mean age of treated patients in the INVASY
study (59 years) (Table 1) was higher than of those
enroled in VPS-1 (43 years) and VPS-2 (50.8 years)
] , similar to those of SYDIT (58.1 years)  , and
lower than those in the VASIS study (64 years) [
The follow-up periods (Table 3) are quite
variable (range 6e44 months) and may have influenced
the outcome of the treated groups.
It appears that the better results of the INVASY
study compared with VPS-2 [
] may be related to
a different pacing algorithm rather than
differences in the length of follow-up or patient age.
In cardioinhibitory vasovagal syncope (VASIS
Type 2A and 2B during HUTT) bradycardia/asystole
and the associated hypotension develop almost
simultaneously, but neuroendocrine mechanisms
in the two events are different and somewhat
independent. When a conventional dual-chamber
pacemaker, which intervenes on the basis of
falling heart rate (as in all studies except INVASY),
prevention of syncope may not be optimal. Since
the pacemaker must detect the onset and stability
of bradycardia (DDD with Rate Drop Response
(RDR) algorithm in VPS, VPS II and SYDIT studies)
] or wait until it reaches the hysteresis
rate (DDI with Rate Hysteresis in the VASIS study)
, pacing may be inadequate to prevent
syncope as delay allows hypotension to become
In contrast, CLS detects variations in myocardial
contractility at the beginning of vasovagal syncope
and reacts quickly with a rate increase after a mean
of 4 min from the beginning of the HUTT [
while the mean time for syncope to occur is about
14 min. This timely intervention suppresses the
resultant bradycardia and counterbalances the
hypotension of decreased venous return by
maintaining cardiac output. In patients with dominant
vasodepression, syncopal symptoms at the peak of
the reflex may occur despite CLS pacing.
These positive results were also documented in
other smaller studies in which INOS2 CLS
pacemaker was used to prevent VVS [
Our patients in the control group had a percentage
of VVS recurrences (78%) higher than that
Even if the number of these patients is small,
their baseline clinical characteristics were similar
to those randomized to CLS stimulation; and
clinical recurrence of syncope during follow-up in
this group was in accordance with the statistical
The most probable reason for this should be
sought in the study protocol. In order to avoid
inconsistencies in the data and to assess the
efficacy of CLS, or the real influence of the
placebo effect of surgery, any pharmacological
therapy to suppress or attenuate vasovagal
symptoms was not allowed in both groups.
While in other studies, b-blockers were used in
] in the control group and in VASIS [
] and VPS-2 [
] medication was
permitted to continue during these studies (Table 3).
The placebo effect of pacemaker implantation is
a crucial question raised by the VPS-2 study [
Because VVS is neurally-mediated, its occurrence
in some patients may be influenced by
psychological factors. The VPS-2 study showed that in
patients with a DDD-RDR device the placebo
effect may be important, because the cumulative
risk of syncope at 6 months was 31% in the paced
group versus 40% in the placebo group (relative
risk reduction 30% e not statistically significant).
Remarkably, the INVASY study showed that the
cumulative risk of syncope at 12 months in
patients with the pacemaker programmed in DDI
(without medication) was 78%, while the risk
dropped to zero in the treated CLS group.
Head up tilt
The value of HUTT in vasovagal syncope remains
controversial [21e23,34]. The INVASY and other
studies have demonstrated the inability of tilt
testing to predict adverse events in syncopal
patients treated with pacing [
]. A positive
HUTT, but no spontaneous recurrence of syncope
during follow-up, was the most frequent result in
our study (62%).
During vasovagal syncope induced by HUTT, CLS
showed some possible effectiveness in increasing
the time-to-syncope by 42% compared with that at
enrolment. The two tests were performed 1 year
apart so that the influence of a tilt training effect
is unlikely. The effect of CLS pacing in the
prevention of vasovagal syncope induced by HUTT was
recently confirmed by Griesbach et al. in a group
of 22 syncopal patients using a particularly
aggressive three step tilt protocol [
Quality of life
The improvement of the quality of life in the CLS
paced patients was great and in accordance with
the clinical outcome. The possible placebo effect
of surgery significantly and positively influenced
the quality of life of the two patients in the control
arm that did not experienced syncopal recurrence
The decision of the steering committee to
prematurely to discontinue the randomization,
forced by the relevant number of syncopal
recurrences in the control group, caused the low
number of control subjects, which represents the
main limitation of the study.
The total population enroled in the study was
relatively low because a substantial number of
eligible patients were included in the Pre-INVASY
Moreover, screening logs were not maintained
throughout the study recruitment period to
calculate the total of potentially eligible patients.
Recurrences of presyncope and dizziness were not
collected at enrolment, because they are difficult
to evaluate and because patients are more likely to
remember syncope rather than dizziness.
Finally, as the two syncope-free patients in the
control arm were crossed over early to CLS pacing,
the follow-up of patients in DDI mode is limited to
The INVASY study, supported by the Pre-INVASY
registry, has yielded favourable long-term results,
using Closed Loop Stimulation, whose rate
response is driven by variations in myocardial
contractility, in prevention of vasovagal syncope.
Patients paced with Closed Loop Stimulation did
not exhibit any recurrence of syncope and had an
improved quality of life. Further follow-up is
required to confirm lifetime benefit.
Pre-INVASY registry and INVASY study
Participating Centers and Investigators
Ospedale degli Infermi, Rimini: Sergio Sermasi;
Ospedale S. Giovanni-Addolorata, Roma: Giovanni
Ospedale Generale Provinciale, Saronno: Adriano
Ospedale S.S. Annunziata, Sassari: Giovanni
Battista Tola, Pierfranco Terrosu;
Ospedale San Paolo, Savona: Paolo Bellotti,
Ospedale Maggiore, Torino: Luigi Libero, Marcella
Ospedale Giovanni Bosco, Torino: Mauro Bensoni;
Ospedale Galmarini, Tradate: Daniela Barbieri,
Ospedale Panico, Tricase: Antonio Galati;
Ospedale di Circolo, Varese: Jorge Salerno,
Stabilimento Ospedaliero, Verbania: Enzo Maria
Bianchi, Renato Glenzer;
Ospedale Civile, Vigevano: Francesco Zolezzi,
Ospedale Belcolle, Viterbo: Massimo Sassara,
Fernando De Luca.
Organizing Committee (Steering
E. Occhetta (Chairman), R. Audoglio, M.G. Bongiorni,
L. Chiodi, S. Favale, S. Romano.
External Monitoring and Safety
A. Malliani and R. Furlan.
 Maloney JD , Jaeger FJ , Fouad-Tarazi FM , Morris HH . Malignant vasovagal syncope prolonged asystolie provoked by head-up tilt . Cleveland Clin Med J 1988 ; 55 : 542e8 .
 Sutton R. Vasovagal syndrome . Could it be malignant? Eur J Cardiac Pacing Electrophysiol 1992 ; 2 : 89 .
 Ellenbogen KA , Morillo CA , Wood MA , et al. Neural monitoring of vasovagal syncope . Pacing Clin Electrophysiol 1997 ; 20 : 788e94 .
 Sra JS , Jazayeri MR , Avitall B , et al. Comparison of cardiac pacing with drug therapy in the treatment of neurocardiogenic (vasovagal) syncope with bradycardia or asystole . N Engl J Med 1993 ; 328 : 1085e90 .
 Petersen ME , Chamberlain-Webber R , Fitzpatrick AP , et al. Permanent pacing for cardioinhibitory malignant vasovagal syncope . Br Heart J 1994 ; 71 : 274e81 .
 Thomson HL , Atherton JJ , Khafagi FA , Frenneaux MP . Failure of reflex venoconstriction during exercise in patients with vasovagal syncope . Circulation 1996 ; 93 : 953e9 .
 Lee TM , Chen MF , Su SF , et al. Excessive myocardial contraction in vasovagal syncope demonstrated by echocardiography during head-up tilt . Clin Cardiol 1996 ; 19 : 137e40 .
 Quan KJ , Carlson MD , Thames MD . Mechanisms of heart rate and arterial blood pressure control: implications for the pathophysiology of neurocardiogenic syncope . Pacing Clin Electrophysiol 1997 ; 20 : 764e74 .
 Sra JS , Murthy V , Natale A , et al. Circulatory and catecholamines changes during head up tilt testing in neurocardiogenic (vasovagal) syncope . Am J Cardiol 1994 ; 73 : 33e7 .
 Petersen MEV , Williams TR , Erickson M , Sutton R . Right ventricular pressure , dP/dt, and preejection interval during tilt induced vasovagal syncope . Pacing Clin Electrophysiol 1997 ; 20 : 806e9 .
 Mosqueda-Garcia R , Furlan R , Violante RF , et al. Sympathetic and baroreceptor reflex function in neurally mediated syncope evoked by tilt . J Clin Invest 1997 ; 99 : 2736e44 .
 Schaldach M , Hutten H . Intracardiac impedance to determine sympathetic activity in rate responsive pacing . Pacing Clin Electrophysiol 1992 ; 15 : 1778e86 .
 Osswald S , Cron T , Gra¨del C, et al. Closed-loop stimulation using intracardiac impedance as a sensor principle: correlation of right ventricular dP/dtmax and intracardiac impedance during dobutamine stress test . Pacing Clin Electrophysiol 2000 ; 23 : 1502e8 .
 Ravazzi AP , Carosio G , Diotallevi P , et al. Clinical assessment of the correlation between right ventricular impedance and left ventricular contractility . Prog Biomed Res 2000 ; 5 : 478e81 .
 Da Costa A , Ostermeier M , Schaldach M , Isaaz K. Closed loop pacing in a young patient with vasovagal syncope during tilt test . Arch Mal Coeur Vaiss 1998 ; 91 ( Suppl III ): 48 [abstract].
 Occhetta E , Bortnik M , Paffoni P , et al. Closed loop stimulation in neuromediated syncope . Meditteranean Soc Pacing Electrophysiol J 1999 ; 1 : 143e8 .
 Occhetta E , Bortnik M , Paffoni P , et al. Neurohumoral effects on closed loop stimulation . In: Santini M, editor. Seventh International Symposium on Progress in Clinical Pacing . CEPI ed. Rome; 1998 . p. 39e43 .
 Occhetta E , Bortnik M , Vassanelli C , et al. The DDDR closed loop stimulation for the prevention of vasovagal syncope: results from the INVASY prospective feasibility registry . Europace 2003 ; 5 : 153e62 .
 Sutton R , Brignole M , Menozzi C , et al. Dual-chamber pacing in the treatment of neurally mediated tilt-positive cardioinhibitory syncope . Circulation 2000 ; 102 : 294e303 .
 Occhetta E , Bongiorni MG , Chiodi L , et al. Studio INVASY . Protocollo di studio clinico per la valutazione della stimolazione CLS nella prevenzione delle sincopi neuromediate di tipo cardioinibitorio e misto (sincope vasovagale) . G Ital Aritmol Cardiostim 2001 ; 1 : 28e36 .
 Fitzpatrick A , Sutton R . Tilting towards a diagnosis in recurrent unexplained syncope . Lancet 1989 ; 1 : 658e60 .
 Fitzpatrick A , Theodorakis G , Vardas P , et al. Methodology of head-up tilt testing in unexplained syncope . J Am Coll Cardiol 1991 ; 17 : 125e30 .
 Raviele A , Menozzi C , Brignole M , et al. Value of head-up tilt testing potentiated with sublingual nitroglycerin to assess the origin of unexplained syncope . Am J Cardiol 1995 ; 76 : 267e72 .
 Shumacher M , Olschewski M , Schulgen G . Assessment of quality of life in clinical trials . Stat Med 1991 ; 10 : 1915e30 .
 Testa MA , Simonson DC . Assessment of quality of life outcomes . N Engl J Med 1996 ; 334 : 835e40 .
 Connolly SJ , Sheldon R , Roberts RS , et al. The North American vasovagal pacemaker study (VPS). A randomized trial of permanent cardiac pacing for the prevention of vasovagal syncope . J Am Coll Cardiol 1999 ; 33 ( 1 ): 16e20 .
 Connolly SJ , Sheldon R , Thorpe KE , et al. Pacemaker therapy for prevention of syncope in patients with recurrent severe vasovagal syncope. Second vasovagal pacemaker study (VPS II): a randomized trial . JAMA 2003 ; 289 ( 17 ): 2224e9 .
 Ammirati F , Colivicchi F , Santini M. Permanent cardiac pacing versus medical treatment for the prevention of recurrent vasovagal syncope . Circulation 2001 ; 104 : 52e63 .
 Sheldon R , Koshman ML , Wilson W , et al. Effect of dualchamber pacing with automatic rate-drop sensing on recurrent neurally mediated syncope . Am J Cardiol 1998 ; 81 : 158e62 .
 Benditt D , Sutton R , Gammage M , et al. Clinical experience with Thera DR rate-drop response pacing algorithm in carotid sinus syndrome and vasovagal syncope . Pacing Clin Electrophysiol 1997 ; 20 : 832e9 .
 Occhetta E , Bortnik M , Paffoni P , et al. Closed loop stimulation in vasovagal syncope e one year follow-up in selected patients . Prog Biomed Res 1999 ; 4 : 176e80 .
 Griesbach L , Huber T , Knote B , et al. Closed loop stimulation: therapy for malignant neurocardiogenic syncope . Prog Biomed Res 2002 ; 7 : 242e7 .
 Guyomar Y , Graux P , Nicolas E , et al. INOS2 DR and neurocardiogenic syncope: a first experience about four patients . In: Adornato E, editor. Rhythm control from cardiac evaluation to treatment . Proceedings Of the Sixth Southern Symposium on Cardiac Pacing (Taormina 1998 ). L. Pozzi ed., vol. 2 . Roma; 1998 . p. 170e5 .
 Kurbaan AS , Franze´n AC , Bowker TJ , et al. Usefulness of tilt test-induced patterns of heart rate and blood pressure using a two-stage protocol with glyceryl trinitrate provocation in patients with syncope of unknown origin . Am J Cardiol 1999 ; 84 : 665e70 .