1

Paradoxical Change in Atrial Fibrillation Dominant Frequencieswith Baroreflex-Mediated Parasympathetic Stimulation with

Phenylephrine InfusionJASON NG, Ph.D., ROD S. PASSMAN, M.D., RISHI ARORA, M.D., ALAN H. KADISH, M.D.,

and JEFFREY J. GOLDBERGER, M.D.

From the Division of Cardiology and Bluhm Cardiovascular Institute, Feinberg School of Medicine,Northwestern University, Chicago, Illinois, USA

Baroreflex Response and AF Dominant Frequency. Introduction: Parasympathetic stimulationis known to promote atrial fibrillation (AF) through shortening of atrial refractory periods. We hypothesizedthat baroreflex-mediated parasympathetic stimulation via phenylephrine (PE) infusion would increase AFrate as measured by dominant frequency (DF).

Methods and Results: The protocol was performed in 27 patients (24 M, 59 ± 1 years old) prior toAF ablation. For 10 patients in AF, PE was infused until systolic blood pressure increased ≥30 mmHg.Electrograms were recorded in the left atrium before and after PE. DFs of each recording were calculatedoffline. Atrial effective refractory periods (ERPs) were measured before and after PE in 11 patients whowere in sinus rhythm during the procedure. DFs were also measured in 6 patients in AF before and aftercomplete parasympathetic blockade with atropine (0.04 mg/kg).

PE resulted in increased RR intervals during sinus rhythm (1,170 ± 77 to 1,282 ± 85 ms, P = 0.03)and AF (743 ± 32 to 826 ± 30 ms, P = 0.03), consistent with parasympathetic effect on the sinus and AVnodes, respectively. DFs were decreased by PE in the left atrium (6.2 ± 0.2 to 6.0 ± 0.2 Hz, P = 0.004).Correspondingly, atrial ERPs significantly increased from 218 ± 13 to 232 ± 11 ms (P = 0.04). Atropineresulted in a decreasing trend in DF in the left atrium (5.9 ± 0.1 to 5.8 ± 0.1 Hz, P = 0.07).

Conclusions: Despite baroreflex-mediated parasympathetic effect, PE produced a slowing of AF alongwith lengthening of ERP, while parasympathetic blockade also slowed DF. It is therefore likely that the directand indirect adrenergic effects of PE on atrial electrophysiology are more prominent than its parasympa-thetic effects. (J Cardiovasc Electrophysiol, Vol. pp. 1-6)

alpha-adrenergic receptors, atrial fibrillation, autonomic nervous system, baroreflex, dominant frequency,electrograms

Introduction

The autonomic nervous system has been shown to facili-tate the induction and maintenance of atrial fibrillation (AF).Experimental studies have shown significant shortening ofatrial refractory periods, increasing AF activation rates, andspontaneous AF from electrical vagal stimulation,1,2 stimu-lation of the ganglionated plexi,3,4 and acetylchloline appli-cation or infusion.5,6 Despite recent studies demonstratingdecreased AF in response to denervation of the atria,7 theextent of the role of the autonomic nervous system in AFis still not well understood. Furthermore, methods to eval-uate the autonomic influences on AF substrate are not well

Drs. Goldberger and Kadish received compensation from NCME for par-ticipation on a speaker’s bureau relevant to this topic. Dr. Kadish reportsownership of a patent on gene therapy for AFib.

Other authors: No disclosures.

Address for correspondence: Jeffrey J. Goldberger, M.D., FeinbergSchool of Medicine, Northwestern University, 251 East Huron,Feinberg 8-503, Chicago, IL 60611, USA. Fax: 312-926-2707; E-mail: j-goldberger@northwestern.edu

Manuscript received 17 September 2011; Revised manuscript received9 March 2012; Accepted for publication 23 March 2012.

doi: 10.1111/j.1540-8167.2012.02362.x

established, particularly when using pharmacologic or othernoninvasive methods.

Acute increases in arterial blood pressure are knownto elicit baroreflex-mediated parasympathetic stimulation.Baroreflex responses are typically characterized by brady-cardia, decreased cardiac contractility, peripheral vascularresistance, and venous return with a resultant reflective de-cline in blood pressure towards baseline.8 Stimulation of thearterial alpha-1 adrenergic receptors via phenylephrine (PE)has been a common means to induce baroreflex-mediatedparasympathetic stimulation. Baroreflex testing with PE, forexample, has been proposed as a means to evaluate auto-nomic function and predict risk for ventricular arrhythmiapostmyocardial infarction based on its ability to stimulateparasympathetic mediated slowing of the sinus rate.9-11

The aim of this study was to examine the effect of PEon AF activation rates in patients undergoing AF ablationusing dominant frequency (DF) analysis. We hypothesizedthat baroreflex-mediated parasympathetic stimulation wouldresult in increased atrial DFs and decreased atrial effectiverefractory periods (ERPs).

Methods

Subjects

Twenty-seven patients undergoing AF ablation at North-western Memorial Hospital were recruited. Patients withprior left atrial ablation or surgery in the atria were not

2 Journal of Cardiovascular Electrophysiology Vol. No.

eligible for this study. All patients provided written informedconsent for participation. The study was approved by theInstitutional Review Board of Northwestern University.

Experimental Protocol

The experimental protocol was performed after position-ing all catheters (coronary sinus [CS] catheter, Lasso catheter,and ablation catheters in the left atrium via transeptal access)and prior to any ablation lesions being delivered. For 10 pa-tients who were in AF at the time of the procedure, baselinebipolar AF electrogram recordings were obtained with theLasso catheter sequentially positioned at the ostium of theleft superior pulmonary vein (LSPV), ostium of the right su-perior pulmonary vein (RSPV), and the left atrial appendage(LAA) with the recordings at each site lasting at least 30 sec-onds. The position of the Lasso catheter was marked using anelectroanatomical mapping system (Ensite NavX, St. Jude,St. Paul, MN, USA). PE was then administered via intra-venous infusion12-16 with an initial rate of 0.6 μg/kg/min. Toprovoke a baroreflex response, the infusion rate was increasedby 0.3 μg/kg/min every 5 minutes until systolic blood pres-sure increased by at least 30 mmHg above baseline or whena maximum infusion rate of 1.5 μg/kg/min was achieved.17

Once the target systolic blood pressure was achieved, a sec-ond set of AF recordings with the Lasso catheter was madein the same locations as for the baseline recordings. Simul-taneous surface ECG (to determine RR intervals) and CSrecordings were made during each Lasso recording.

Preablation ERPs were measured in a separate set of11 patients who were in sinus rhythm at the time of the pro-cedure to determine whether phenyleprhine-induced changesin AF electrograms were attributed to changes in refractoryperiod. The Lasso catheter was positioned at either the LSPV,RSPV, or LAA for the duration of this evaluation. Four tofive ERP measurements were performed at each site from themultielectrode Lasso catheter. A drive train (S1) of 8 beatsand cycle length of 400 ms was used and the extrastimulusbeat (S2) was incremented or decremented by 10 ms to deter-mine the ERP. The ERP measurements were repeated at thesame site after the PE infusion protocol as described above.

In a third set of 6 patients with AF at the time of theprocedure, atropine was given to determine the effects ofcomplete parasympathetic blockade on AF electrograms andto compare the results with the baroreflex-mediated parasym-pathetic stimulation via PE. Again, AF electrogram record-ings were sequentially made with the Lasso catheter from theLSPV, RSPV, and LAA and with the CS catheter before andafter atropine. Atropine was given intravenously at a dose of0.04 mg/kg over 2 minutes to achieve complete parasympa-thetic blockade.18 The electrogram recordings were obtainedstarting 2 minutes after completion of atropine administra-tion.

Signal Processing

DF analysis was performed on all signals recorded duringAF, as previously described.19 Preprocessing of the electro-grams (40 to 250 Hz bandpass filtering, rectification, and20 Hz low pass filtering) was performed.20 The processedsignals, which were between 30 seconds and 1 minute induration, were divided into contiguous 4 second segments(7 to 15). The power spectrum of each segment was thencalculated using the Fast Fourier Transform. The DF was

then selected as the frequency with the highest power in thepower spectrum. The DFs of all segments were then averagedto account for temporal variability due to changes in AF cy-cle length and due to the DF technique itself, as previouslydescribed.19

Data Analysis

DF data from noisy signals were visually excluded fromanalysis. The median DF of all the signals from the Lassocatheter was chosen as the representative value of the record-ing site. The mean ERP for each set of Lasso measurementswas chosen as the representative ERP.

Stable CS electrograms recorded over the course of the ex-perimental protocol were used to assess the temporal stabilityof the DF measurements before and after PE. The DFs andcorresponding heart rates were calculated at the followingtime points: (1) earliest available stable CS recording dur-ing baseline, (2) baseline immediately before PE infusion,(3) post-PE immediately after the systolic blood pressure re-sponse (>30 mmHg) was achieved, and (4) latest availablepost-PE recording before the PE infusion was stopped.

Repeated-measures ANOVA was used to compare the DFsbefore and after PE or atropine for the multiple sites. Pairedt-tests were used to compare the ERPs and RR intervalsbefore and after PE. P-values less than or equal to 0.05 wereconsidered statistically significant. All data are presented asmean ± standard error.

Results

Patient Clinical Characteristics

There were 10 patients with persistent AF (9 male, 58 ±2 years old) receiving PE while in AF. Eight patients wereunder conscious sedation and 2 under general anesthesia.Four patients had a history of hypertension, 3 had hyper-lipidemia, 1 had diabetes mellitus, and 3 had a history ofsleep apnea. One patient each had a bicuspid aortic valve, apermanent pacemaker, and atrial septal defect repair. Sevenpatients were treated with beta blockers, 4 with diltiazem, 5with digoxin, and 1 with sotalol. The mean left ventricularejection fraction of this group was 57 ± 2%.

Eleven patients with paroxysmal AF (10 male, 58 ± 2years old) received PE during sinus rhythm. All patients inthis group were under conscious sedation. Seven patients hada history of hypertension, 5 had hyperlipidemia, and 2 haddiabetes mellitus. There was 1 patient each with coronaryartery disease—status postpercutaneous coronary interven-tion, tachycardia related cardiomyopathy, and a permanentpacemaker. Nine patients were treated with beta blockers,2 with digoxin, and 4 with antiarrhythmic drugs (dofetilide,sotalol, propafenone, amiodarone). The mean left ventricularejection fraction of this group was 54 ± 2%.

Six patients with persistent AF (5 male, 62 ± 1 yearsold) received atropine during AF. Three patients were underconscious sedation and 3 under general anesthesia. Threepatients had a history of hypertension and 1 patient eachhad coronary artery disease, hyperlipidemia, nonischemiccardiomyopathy, sleep apnea, and tachycardia related car-diomyopathy. Five patients were treated with beta blockers,1 with diltiazem, 2 with digoxin, and 1 with dronedarone.The mean left ventricular ejection fraction for this group was49±3%.

Ng et al. Baroreflex response and AF dominant frequency 3

Figure 1. Results for dominant frequency (DF) at multiple atrial sites beforeand after phenylephrine (PE) infusion. LSPV = left superior pulmonary veinostium, RSPV = right superior pulmonary vein ostium, LAA = left atrialappendage, CS = coronary sinus.

Figure 2. Examples of bipolar electrograms recording from the left supe-rior pulmonary vein (LSPV) ostium and the corresponding power spectrumbefore (A) and after (B) phenylephrine infusion.

Phenylephrine Effects During AF

PE in the AF group resulted in increases in systolic bloodpressure from 121 ± 3 to 151 ± 4 mmHg (P < 0.0001). MeanRR intervals showed an increasing trend from 743 ± 32 to826 ± 30 ms (P = 0.067). A significant decrease in DF afterPE was found with repeated-measures ANOVA (6.2 ± 0.2to 6.0 ± 0.2 Hz, P = 0.004). Figure 1 shows the DF resultsfrom the multiple left atrial sites before and after PE. Figure 2shows an example of electrograms recorded from the LSPVbefore and after PE. The corresponding power spectra showa lower DF after PE.

Figure 3. Results for ERPs at different atrial sites and overall before andafter phenylephrine (PE) infusion. LSPV = left superior pulmonary veinostium, RSPV = right superior pulmonary vein ostium, LAA = left atrialappendage.

Phenylephrine Effects during Sinus Rhythm

PE in the sinus rhythm group resulted in increases insystolic blood pressure from 117 ± 5 to 150 ± 6 mmHg(P < 0.0001). PE significantly increased average RR intervalsfrom 1,150 ± 70 to 1,248 ± 77 ms (P = 0.03). The ERPs weremeasured in the LSPV in 4 patients, the RSPV in 4 patients,and the LAA in 3 patients. As a group, ERPs significantlyincreased from 217 ± 13 to 231 ± 11 ms (P = 0.03). ERPresults at each of the sites are shown in Figure 3.

Temporal Analysis of AF Electrograms and Heart Ratewith Phenylephrine

The temporal results for the CS DFs are shown inFigure 4A. The DFs at the two baseline time points (withan average 23 ± 6 min interval between the two measure-ments) were not significantly different (P = 0.37). The DFsimmediately after a >30 mmHg blood pressure increasewas achieved with PE (22 ± 3 min after the start of PE)were significantly lower than the pre-PE DFs (P < 0.0006).The significant decrease from the pre-PE DF was main-tained at the last post-PE recording (P = 0.02), 10 ± 3min after the first post-PE recording. Heart rates obtainedat the same times points during AF (Figure 4B) showedno significant difference between the two pre-PE timepoints (P = 0.43) and a significant drop immediately af-ter the >30 mmHg blood pressure increase was achievedwith PE (P = 0.027). In contrast to the DFs, the heartrates at the last post-PE recordings were no longer signifi-cantly different from the pre-PE heart rates (P = 0.55). Heartrates analyzed during sinus rhythm showed similar trends(Figure 4C).

Atropine Effects During AF

Complete parasympathetic blockade with atropine de-creased average RR intervals from 728 ± 46 to 573 ± 45 ms(P = 0.002) as expected. A decreasing trend in DF after PEwas found with repeated-measures ANOVA (5.9 ± 0.1 to 5.8± 0.1 Hz, P = 0.07). Figure 5 shows the DF results for allsites before and after atropine.

Discussion

The role of parasympathetic tone on AF characteristicsin humans is largely unknown. In this study, PE producedslowing of heart rate during both AF and sinus rhythm,consistent with the expected effect of baroreflex-activatedparasympathetic stimulation and sympathetic attenuation on

4 Journal of Cardiovascular Electrophysiology Vol. No.

Figure 4. Temporal analysis of (A) dominant frequencies (DF) from thecoronary sinus (CS), (B) heart rate (HR) during atrial fibrillation (AF), and(C) HR during sinus rhythm. Four time points were analyzed. Baseline 1: thefirst baseline measurement. Baseline 2: baseline measurement immediatelybefore phenylephrine (PE) infusion. BP Achieved: post-PE measurementimmediately after systolic blood pressure response (>30 mmHg). PE Final:latest available recording before PE is stopped.

sinus activation rate and atrioventricular conduction. PE alsodecreased DF, opposite to the expected increase in AF acti-vation rate with parasympathetic stimulation, suggesting thatalpha and/or beta-adrenergic influences have stronger effectson AF activation. Testing ERPs during sinus rhythm withPE also showed an increase in ERP, consistent with the DFdecrease observed with PE. While parasympathetic block-ade with atropine produced expected increases in heart rate,it also caused the expected (albeit modest) decrease in DF,confirming that parasympathetic effects serve to increase ac-tivation rate and DF in AF. Our recent study in canines alsoshowed increases in ERP with atropine.21 Taken together,these responses allow us to conclude that the observed ef-

Figure 5. Results for dominant frequency (DF) at multiple atrial sites beforeand after atropine administration. LSPV = left superior pulmonary veinostium; RSPV = right superior pulmonary vein ostium; LAA = left atrialappendage; CS = coronary sinus.

fect due to PE infusion is either a direct or indirect adren-ergic response. Further investigation is required to explorethis heretofore unrecognized effect of parasympathetic andadrenergic signaling on human AF characteristics.

PE stimulates alpha-1 receptors in the arteries causingconstriction of the arteries and results in the rise in ar-terial blood pressure.22,23 The change in blood pressurethen is sensed by carotid and aortic baroceptors which thensignal the cardiovascular control centers to shift the auto-nomic balance to increase parasympathetic tone relative tosympathetic tone. Parasympathetic stimulation and sympa-thetic attenuation work in tandem to reduce heart rate whichsubsequently reduces blood pressure. Whereas baroreflex-meditated parasympathetic stimulation and sympatheticattenuation concordantly decrease sinus node rate and atri-oventricular node conduction properties, they cause discor-dant effects on atrial myocardium. Acetylcholine, which isa neural transmitter released with parasympathetic stimula-tion, shortens atrial action potential duration though activa-tion of IKAch, the G-protein coupled, acetylcholine-activatedinward rectifier K+ current. Beta-adrenergic stimulation isknown to decrease atrial and ventricular refractory periods24;thus, sympathetic/adrenergic withdrawal would be expectedto result in refractory period prolongation. Thus, in the set-ting of baroreflex stimulation, in which both parasympatheticstimulation and sympathetic withdrawal occur, the resultingeffect on atrial electrical properties is not a priori predictable.Furthermore, the sympathovagal interaction when both theparasympathetic and sympathetic arms of the autonomic ner-vous system are manipulated may be a complex process. Thiscomplexity may in part be responsible for the nonparallel na-ture of the sinus/AV nodal response and the atrial responseto the baroreflex provocation as seen in this study.

In addition to its indirect effects mediated by the barore-flex, PE has been shown to have direct alpha-1 adrenergic ef-fects on cardiac myocytes.25,26 Bruckner et al. demonstratedthat a prolongation of bovine ventricular action potentialduration occurs with PE.25 Robinson et al. showed that pro-longation of canine ventricular (epicardial) action potentialduration with PE was due to reduction of IKS.27 Positive in-otropic effects of PE have been shown in several studies.In a canine in vivo study, Talajic et al. studied the effectsof PE on cardiac electrophysiology in the presence of auto-nomic blockade. 28 They found that PE did not have a signif-icant effect on atrial, AV nodal, or ventricular refractoriness.Another study by Morady et al. showed that epinephrine,a nonselective agonist of all adrenergic receptors, shortensatrial refractory periods in humans without beta blockade, but

Ng et al. Baroreflex response and AF dominant frequency 5

lengthens atrial refractory periods with beta blockade.26 Thissuggests that alpha adrenergic stimulation with epinephrinedoes lengthen refractory periods if the beta adrenergic short-ening effect is blocked. The reasons for these conflictingresults are not clear but could be in part due to the heteroge-neous response of the atrium to adrenergic stimulation.

Previous data on the effect of PE on AF are limited. Taiet al. reported that PE-induced baroreflex response sup-pressed focal activity (premature atria beats and AF bursts)from the pulmonary veins in patients with focal AF.29

Marrouche et al. examined the effects of PE in dissociatedpulmonary veins after AF ablation and found that PE resultedin the modest slowing of the PV firing rate.30 The reducedDFs with PE found in this study could therefore be due tocombined effects of the lengthening of atrial refractory peri-ods and slowing of pulmonary vein activity.

The current study is the first, to our knowledge, to show theeffects of baroreflex response with PE on atrial activity duringAF. The results reflect possible regional differences in themagnitude of the PE effects. With the known heterogeneityof the effect of vagal stimulation in the atrium,1 there maybe regions that are more sensitive to vagal stimulation, someareas that are more sensitive to adrenergic stimulation, andother areas where the vagal and adrenergic effects of PEcancel out each other. If this is the case, the DF response toPE will vary accordingly.

The reproducibility data demonstrated that the effects ofPE on DF persisted for the duration of the infusion. Interest-ingly, the effects on heart rate may have dissipated in bothpatients who were in AF and those in sinus rhythm. The sta-ble effect of PE infusion on DF suggests that the observedeffects may be due to the direct, persistent alpha-adrenergiceffects that would be present during a constant PE infusion.Complete beta-blockade would provide the definitive answerto the relative roles of the alpha- and beta-adrenergic compo-nents in the DF decrease, but the time course analysis appearsto support an alpha effect.

Limitations

Electrogram recordings were obtained at only selectedsites in the atrium and may not be representative of the atriumas a whole. The sample size is underpowered to detect dif-ferences among sites. The patient population consisted of amix of pathologies and patients were on a variety of medi-cations that may have an effect on the autonomic response.However, each patient in this protocol serves as his or herown control. As the directional changes with the autonomicinterventions should be similar despite these differences, thediversity of pathologies and medications should not affectthe qualitative findings. It is also possible that sedation mayhave an effect on the baseline parasympathetic tone. How-ever, the modest effect of parasympathetic blockade on atrialDFs suggests that vagal activity in the atrium was not ata level too high for further effect. Complete beta-blockadewas not performed during this study, which would allow thedefinitive determination of the relative roles of alpha versusbeta-adrenergic effects with PE and its baroreflex response.Nevertheless, the data clearly demonstrate that despite en-hanced parasympathetic effects with phenylephrine infusion,the expected parasympathetic effects on DF and ERP werenot observed.

Conclusions

With PE, a stronger than expected adrenergic effect on AFactivation rate and atrial refractory periods was observed. Abetter understanding of the combined roles and interactionsof cholinergic and adrenergic effects on the atrium in AFcould lead to the development of new anti-arrhythmic ap-proaches to the treatment of AF.

Acknowledgments: We thank Aleksey Borodyanskiy, M.D., DerrickFansler, M.D., Roger Villuendas, M.D., Jason Rubenstein, M.D., IndranilSen-Gupta, M.D., and Eric T. Chang, M.D. for their assistance in carryingout these studies.

References

1. Allessi R, Nusynowitz M, Abildskov JA, Moe GK: Nonuniform dis-tribution of vagal effects on the atrial refractory period. Am J Physiol1958;194:406-410.

2. Goldberger JJ, Kadish AH, Johnson D, Qi X: New technique for vagalnerve stimulation. J Neurosci Methods 1999;15:109-114.

3. Scherlag BJ, Yamanashi W, Patel U, Lazzara R, Jackman WM: Auto-nomically induced conversion of pulmonary vein focal firing into atrialfibrillation. J Am Coll Cardiol 2005;45:1878-1886.

4. Takahashi Y, Jaıs P, Hocini M, Sanders P, Rotter M, Rostock T, HsuLF, Sacher F, Clementy J, Haıssaguerre M: Shortening of fibrillatorycycle length in the pulmonary vein during vagal excitation. J Am CollCardiol 2006;47:774-780.

5. Burashnikov A, Antzelevitch C: Reinduction of atrial fibrillation im-mediately after termination of the arrhythmia is mediated by latephase 3 early afterdepolarization-induced triggered activity. Circula-tion 2003;107:2355-2360.

6. Sharifov OF, Fedorov VV, Beloshapko GG, Glukhov AV, YushmanovaAV, Rosenshtraukh LV: Roles of adrenergic and cholinergic stimu-lation in spontaneous atrial fibrillation in dogs. J Am Coll Cardiol2004;43:483-490.

7. Pappone C, Santinelli V, Manguso F, Vicedomini G, Gugliotta F,Augello G, Mazzone P, Tortoriello V, Landoni G, Zangrillo A, LangC, Tomita T, Mesas C, Mastella E, Alfieri O: Pulmonary vein den-ervation enhances long-term benefit after circumferential ablation forparoxysmal atrial fibrillation. Circulation 2004;109:327-334.

8. Kirchheim HR: Systemic arterial baroreceptor reflexes. Physiol Rev1976;56:100-177.

9. Schwartz PJ, Vanoli E, Stramba-Badiale M, De Ferrari GM, BillmanGE, Foreman RD: Autonomic mechanisms and sudden death. Newinsights from analysis of baroreceptor reflexes in conscious dogs withand without a myocardial infarction. Circulation 1988;78:969-979.

10. Exner DV, Kavanagh KM, Slawnych MP, Mitchell LB, Ramadan D,Aggarwal SG, Noullett C, Van Schaik A, Mitchell RT, Shibata MA,Gulamhussein S, McMeekin J, Tymchak W, Schnell GG, AM, SheldonRS, Fick GH, Duff HJ, REFINE Investigators: Noninvasive risk assess-ment early after a myocardial infarction the REFINE study. J Am CollCardiol 2007;50:2275-2284.

11. La Rovere MT, Bigger Jr JT, Marcus FI, Mortara A, Schwartz PJ:Baroreflex sensitivity and heart-rate variability in prediction of totalcardiac mortality after myocardial infarction. ATRAMI (AutonomicTone and Reflexes after Myocardial Infarction) Investigators. Lancet1998;351:478-484.

12. Page RL, Tang AS, Prystowsky EN: Effect of continuous enhancedvagal tone on atrioventricular nodal and sinoatrial nodal function inhumans. Circ Res 1991;68:1614-1620.

13. Page RL, Wharton JM, Prystowsky EN: Effect of continuous vagalenhancement on concealed conduction and refractoriness within theatrioventricular node. Am J Cardiol 1996;77:260-265.

14. Goldberger J, Ahmed M, Parker M, Kadish A: Dissociation of heart ratevariability from parasympathetic tone. Am J Physiol 1994;266:H2152-H2157.

15. Goldberger J, Kim Y, Ahmed M, Kadish A: Effect of graded increasesin parasympathetic tone on heart rate variability. J Cardiovasc Electro-physiol 1996;7:594-602.

16. Goldberger JJ, Challapalli S, Tung R, Parker MA, Kadish AH: Rela-tionship of heart rate variability to parasympathetic effect. Circulation2001;103:1977-1983.

6 Journal of Cardiovascular Electrophysiology Vol. No.

17. Smyth HS, Sleight P: Reflex regulation of aterial pressure during sleepin man. Circ Res 1969;24:109-121.

18. Jose AD, Collison D: The normal range and determinants of the intrinsicheart rate in man. Cardiovasc Res 1970;4:160-167.

19. Ng J, Kadish AH, Goldberger JJ: Effect of electrogram characteristicson the relationship of dominant frequency to atrial activation rate inatrial fibrillation. Heart Rhythm 2006;3:1295-1305.

20. Botteron GW, Smith JM: A technique for measurement of the extent ofspatial organization of atrial activation during atrial fibrillation in theintact human heart. IEEE Trans Biomed Eng 1995;42:579-586.

21. Ng J, Villuendas R, Cokic I, Schliamser JE, Gordon D, Koduri H,Benefield B, Simon J, Murthy SN, Lomasney JW, Wasserstrom JA,Goldberger JJ, Aistrup GL, Arora R: Autonomic remodeling in theleft atrium and pulmonary veins in heart failure: Creation of a dy-namic substrate for atrial fibrillation. Circ Arrhythm Electrophysiol2011;4:388-396.

22. Eckstein JW, Abboud FM: Circulatory effects of sympathometicamines. Am Heart J 1962;63:119-135.

23. Thames MD, Abboud FM: Interaction of somatic and cardiopul-monary receptor in control of renal circulation. Am Heart J 1979;237:H560-H5.

24. Farges JP, Ollagnier M, Faucon G: Influence of acetylcholine, isopro-terenol, quinidine and ouabain on effective refractory periods of atrial

and ventricular myocardium in the dog. Arch Int Pharmacodyn Ther1977;227:206-219.

25. Bruckner R, Scholz H: Effects of a-adrenoceptor stimulation withphenylephrine in the presence of propranolol on force of contrac-tion, slow inward current and cyclic AMP content in the bovine heart.Br J Pharmac 1984;82:223-232.

26. Morady F, Nelson SD, Kou WH, Pratley R, Schmaltz S, De Buitleir M,Halter JB: Electrophysiologic effects of epinephrine in humans. J AmColl Cardiol 1988;11:1235-1244.

27. Robinson RB, Liu Q-Y, Rosen MR: Ionic basis for action potential pro-longation by phenylephrine in canine epicardial myocytes. J CardiovascElectrophysiol 2000;11:70-76.

28. Talajic M, Villemaire C, Nattel S: Electrophysiological effects of alpha-adrenergic stimulation. Pacing Clin Electrophysiol 1990;13:578-582.

29. Tai CT, Chiou CW, Wen ZC, Hsieh MH, Tsai CF, Lin WS, Chen CC,Lin YK, Yu WC, Ding YA, Chang MS, Chen SA: Effect of phenyle-phrine on focal fibrillation originating in the pulmonary veins and su-perior vena cava. J Am Coll Cardiol 2000;36:788-793.

30. Marrouche N, Wazni OM, Martin DO, Rossillo A, Saliba W, ErciyesD, Schweikert R, Khaykin Y, Burkhardt D, Bhargava M, Verma A,Abdul-Karim A, Natale A: Response to pharmacological challengeof dissociated pulmonary vein rhythm. J Cardiovasc Electrophysiol2005;16:122-126.