Postablation-acquired short atrioventricular Mahaim-type fibers: Observations on their clinical, electrocardiographic, and electrophysiologic profile

a

Postablation-acquired short atrioventricular Mahaim-type fibers:Observations on their clinical, electrocardiographic, andelectrophysiologic profileEduardo Back Sternick, MD, PhD, FHRS,*† Frederico Soares Correa, MD,† Silvia Rego, MD,†

Daniela Moreira Santos, MD,† Fernando Damascena, MD,† Ricardo Scarpelli, MD,†

Luiz Márcio Gerken, MD,† Hein J.J. Wellens, MD, PhD, FACC‡

From the *Post Graduate Institute, Faculty of Medical Sciences of Minas Gerais, Belo Horizonte, Brazil, †Arrhythmiand Electrophysiology Unit, Biocor Instituto, Nova Lima, Brazil, and ‡Cardiovascular Research Institute, Maastricht, The

Netherlands.

BACKGROUND The electrophysiologic characteristics of decre-mentally conducting accessory pathways (APs) are well described;however, little is known about decrementally conducting APscaused by the radiofrequency ablation of a rapidly conducting AP.

OBJECTIVE To report the clinical, electrocardiographic, and elec-trophysiologic characteristics of 6 patients who developed a dec-remental AP after an attempt at ablation.

METHODS We compared the clinical and electrophysiologic char-acteristics of 295 consecutive patients with the Wolff-Parkinson-White syndrome who underwent radiofrequency ablation of 311manifest APs (group A) with those of 6 patients with the Wolff-Parkinson-White syndrome in whom a decrementally conductingAP was detected after an attempt at ablation.

RESULTS The AP ablation site in group B patients was at thecoronary sinus ostium region in 3 patients, middle cardiac vein in2 patients, and left posteroseptal region in 1 patient. Sixty-twobypass tracts in group A patients and all 6 in group B patients wereablated at these locations, while 249 bypass tracts in group Apatients and none in group B patients were ablated elsewhere

wtctt

[email protected].

1547-5271/$ -see front matter © 2012 Heart Rhythm Society. All rights reserved

physiology had an AP located in the venous system. The odds fordeveloping an acquired decremental antegrade atrioventricular APwhen it was located inside the venous system were 1 in 6. Allgroup B decremental APs were sensitive to adenosine, but none in85 group A patients (P �.0001).

CONCLUSIONS The risk for developing decremental conductionafter the ablation of a rapidly conducting AP is greater for APsinside the coronary venous system. Acquired decremental ante-grade atrioventricular APs are electrophysiologically similar to denovo ones. They are capable of being part of an arrhythmia circuitand, therefore, should be targeted for ablation.

KEYWORDS Decrementally conducting accessory pathways; Mahaimfibers; acquired Mahaim fibers; Radiofrequency catheter ablation;Accessory pathways; Wolff-Parkinson-White syndrome; Adenosine

ABBREVIATIONS AP � accessory pathway; AV � atrioventricular;ECG � electrocardiogram; RF � radiofrequency; VA � ventricu-loatrial

(Heart Rhythm 2012;9:850–858) © 2012 Heart Rhythm Society.
(P � .0001). Five of the 6 patients (83%) with acquired Mahaim All rights reserved.
IntroductionThe electrocardiogram (ECG) during sinus rhythm in patientshaving both atrioventricular (AV) conduction over a manifestAV bypass tract and a decrementally conducting AP showsventricular depolarization through the rapidly conducting AP.The presence of a Mahaim fiber is usually diagnosed aftercatheter ablation of the rapidly conducting AP, either becausea minimal preexcitation pattern appears during sinus rhythm1

or because it is brought out during atrial pacing at increasingrates. However, one should be suspicious of an iatrogenic slowand decrementally conducting AP when the preexcited QRScomplex resembles the one recorded before ablation particu-

Address for reprint requests and correspondence: Dr Eduardo BackSternick, MD, PhD, FHRS, Alameda do Morro 85, ap. 1900, bairro Vila daSerra, Nova Lima, Minas Gerais 34.000-000, Brazil. E-mail address:

larly when it is associated with a prolonged P-delta interval.Radiofrequency (RF) slow and decremental conduction in 1patient with a previous rapidly conducting AP was first re-ported by Haissaguerre et al,2 but there has been no systematicstudy about acquired decremental antegrade AV APs causedby the injury of a rapidly conducting AP.3 The aim of our study

as to describe the clinical and electrophysiologic charac-erisctics of a cohort of 6 patients having a decrementallyonducting AP following RF catheter ablation attempt in pa-ients with AV conduction over a rapidly conducting AP, ando discuss the likely mechanisms for its occurrence.

MethodsStudy populationA cohort of 301 consecutive patients with the Wolff-Par-kinson-White syndrome who were referred for RF catheter
ablation between January 2005 and December 2011 was
. doi:10.1016/j.hrthm.2012.02.011

mailto:[email protected]

w

nMAA

M

A

C

A

851Sternick et al Postablation-Acquired Mahaim Fibers

retrospectively analyzed. Informed and written consent wasobtained from study subjects. The study was approved bythe Ethics Committee of Biocor Institute.

Group A had 295 patients with 311 rapidly antegrade con-ducting AV accessory pathways (APs) who underwent a suc-cessful RF ablation procedure. The group had 172 men (53%),with a mean age of 30 � 15 years (range 8–81 years) (Table 1).

Group B had 6 patients with a single fast bidirectionalconducting AV AP (except case 1 who had unidirectionalanterograde conduction through the AP) developing slowAV conduction with decremental properties after an attemptat ablation (Figure 1 and Table 2). Three of the 6 patients

ere men (50%), with a mean age of 26 � 10 years.

Exclusion criteriaSixteen patients with variants of preexcitation—7 atriofascicu-lar pathways, 6 fasciculoventricular pathways, and 3 short decre-mentally conducting AV APs—were excluded because decre-mental conduction was already present before ablation (6 of 7atriofascicular fibers and 2 of 3 short AV fibers) or because theywere diagnosed after the ablation of a rapidly conducting AP froma different site. The fasciculoventricular pathways were excludedbecause in them catheter ablation is not indicated.

Definition of terms

Decremental conductionDecremental conduction is defined as cycle length–depen-dent prolongation of the impulse conduction time � 30 ms

Table 1 Comparative characterization of group A and group Bpatients

Group A Group B P

295 6ale gender 172 (58%) 3 (50%) nsge (y) 30 � 15 26 � 10 nsPsManifest 311 6Concealed 8 0

ultiple APs2 APs 18 03 APs 3 0

P site (manifest AP)Left free wall 95 0Anteroseptal 34 0Right free wall 36 0Right posteroseptal 69 0Coronary sinus ostium 15 3 �.0001Middle cardiac vein 10 2Left posteroseptal 36 1Midseptal 16 0

onduction block withintravenous adenosine

0 of 85 6 of 6 �.0001

dditional heart diseaseEbstein 3 0Reversible left ventricle

dysfunction2 0

Hypertrophic cardiomyopathy 1 0

AP � accessory pathway; ns � not significant.

through the AP.

Acquired decremental antegrade AV APFollowing the ablation attempt, each one of the 6 group Bpatients showed electrophysiologic criteria for a decremen-tally conducting bypass tract during atrial pacing, with pro-gressive AH and AV interval prolongation coupled with adecreasing HV interval leading to a greater degree of ven-tricular preexcitation. These electrophysiologic propertiesare similar to decrementally conducting de novo AV AP.

Electrophysiologic studyProgrammed electrical stimulation and recordings of the12-lead surface ECG and intracardiac electrograms weremade by using the EP Tracer (CardioTek BV, Maastricht,The Netherlands). Programmed electrical stimulation in-cluded atrial and ventricular pacing at increasing rates andextrastimuli during sinus rhythm and during atrial and ven-tricular pacing. The ventricular stimulation protocol with upto 3 extra stimuli was repeated during isoproterenol infu-sion. Group B patients underwent electrophysiologic assess-ment including pharmacologic testing with adenosine andisoproterenol when it became clear that the electrophysi-ologic properties of the AP changed after the ablation at-tempt. Programmed electrical stimulation was also repeatedin all group A patients after the ablation of the AP. Allgroup A patients were challenged with adenosine after ab-lation, but only 85 patients were challenged before ablation.Patients challenged with intravenous adenosine received abolus of 6 �g followed by 12 �g, and up to 18 �g, until asignificant effect on the AV nodal conduction was seen.

Statistical analysisValues are given as mean � standard deviation. The sig-nificance of differences (P �.05) in clinical, electrocardio-graphic, or electrophysiologic parameters between groupswas assessed by using the Student t test or the Fisher exacttest, using Stata 11 software.

ResultsRF catheter ablation

Group AWe used a regular 4-mm-tip ablation catheter (DAIG, SJM,Minnetonka, Minnesota) in 285 patients. A 3.5-mm cool-tip(Biosense-Webster, Diamond Bar, California) ablation cath-eter (open-irrigated tip) was used in 8 of the 36 patients witha right free wall bypass tract and in 2 of the 10 patients withAPs located in the middle cardiac vein.

Group BWe used an open-irrigated tip catheter in 3 of the 6 patients(case numbers 1, 2, and 6) in whom high impedance oc-curred, with a regular 4-mm-tip ablation catheter being usedwhile ablating in the middle cardiac vein, the coronarysinus, and the origin of a posterior vein, respectively (20–25W, 17 mL/min). The mean number of RF pulses was 2.5 �1.6 as compared with 5.6 � 3.4 in group A (P � ns).Successful ablation was accomplished in 5 of the 6 patients
using the same procedure.

.psat(

CaAasf m

ipob

CG pre

852 Heart Rhythm, Vol 9, No 6, June 2012

The relationship between the location of the distalend of the AP with the occurrence of an acquireddecremental antegrade AV APThe initial ablation site in group B patients was at thecoronary sinus ostium region in 3 patients, in the middlecardiac vein in 2 patients, and at the left posteroseptal regionin 1 patient (transseptal access). The decrementally con-ducting AP in case number 6 was ablated at the origin of aposterior vein (Table 1). Sixty-two bypass tracts in group Apatients and in all 6 patients from group B were ablated atthese locations, while 249 bypass tracts in group A patientsand none in group B patients were ablated elsewhere (P �0001). Five of the 6 patients (83%) with acquired Mahaimhysiology had the AP located in the coronary venousystem. The odds for developing an acquired decrementalntegrade AV AP fiber in patients with an AP located insidehe coronary venous system were 1 in 6 (5 of 30 patients)Table 1).

omparison of the 12-lead ECG pre- and postablationttempt in group B patientsll group B patients had overt preexcitation before RF

blation. Following the first ablation, 3 of the 6 patients weretill preexcited. The PR interval in these patients increasedrom a mean of 0.11 � 0.01 to 0.21 � 0.04 ms (P � .01).

Patients 2, 4, and 6 did not have preexcitation during sinusrhythm; however, preexcitation was unmasked during atrialpacing at increasing rates as well as by adenosine infusion.Careful assessment of the QRS during maximal preexcita-

Figure 1 Twelve-lead ECG showing pre- and postablation pattern of opreexcited QRS complexes. Atrial pacing was used in patients 1 and 4 (E

tion (Figure 1) showed that there were slight changes from o

the preablation QRS complex configuration in all patients,albeit the delta wave axis did not change in any patient northe QRS transition in the precordial leads. Three types ofchanges of the QRS complex were seen during maximalpreexcitation (Figure 1): different amplitude of the R waveor the S wave, appearance of a new small terminal r wave,and a more narrow QRS complex.

The electrophysiologic profile of patients with anacquired decremental antegrade AV AP

Adenosine sensitivity. All patients with acquired “Mahaimphysiology” were sensitive to intravenous adenosine. Therewere 2 types of response to adenosine: a progressive pro-longation of the P-delta interval and of the QRS width(Figure 2A) in 1 case or the same response ending in blockof the decremental fiber (5 patients) (Figure 2B).

Decremental conduction. After ablation of the fast conduct-ing AP, all patients had a Wenckebach block at the decre-mental AP during atrial pacing with a mean cycle length of476 � 160 ms. Maximal decrement during atrial pacing atincreasing rates ranged from 30 to120 ms (mean of 76 � 41

s). Figure 3 shows prolongation of the stimulus-deltanterval (case 6) after ablation of the fast conducting AP. Inatient number 3, preexcitation disappeared within 3 sec-nds of ablation, and RF was discontinued 17 seconds laterecause of junctional automaticity followed by prolongation

eexcitation for each of the group B patients. The aim was to assess fully) and patients 4 and 6 (ECG post). ECG � electrocardiogram.

vert pr

f the PR interval. Conduction through the AP resumed 50

lle2

Clin

ical

,el

ectr

ocar

diog

raph

ic,

and

elec

trop

hysi

olog

icfe

atur

esof

grou

pB

pati

ents

Age

(y)

Sex

Clin

ical

arrh

ythm

iaCS ve

nogr

aphy

Abla

tion

site

1RF pu

lses

ECG

post

-RF1

Pri

(s)

basa

lPr

i(s

)po

st-R

F12:

1AP

pre-

RFW

APpo

st-R

F1M

axim

umde

crem

ent

Aden

osin

epr

e-RF

1Ad

enos

ine

sens

itiv

eAn

tidr

omic

indu

cibl

eAb

lati

onsi

te2

Cool

-tip

cath

eter

28M

AFNo

tdo

neM

CV4

Delt

a0.

130.

2532

055

040

NoYe

sNo

MCV

Yes

33F

AVRT

Norm

alCS

os5

Node

lta

340

730

120

NoYe

sYe

sCS

osYe

s20

FAV

RTNo

tdo

neCS

os1

Delt

a0.

110.

2133

035

010

0No

Yes

No38

FAV

RTNo

rmal

CSos

1No

delt

a30

034

012

0Ye

sYe

sNo

CSos

10M

AVRT

Norm

alLP

S2

Delt

a0.

110.

1731

055

030

Yes

Yes

NoLP

S27

MAV

RTNo

rmal

MCV

2No

delt

a25

034

050

Yes

Yes

NoPV

Yes

AF�

atri

alfib

rilla

tion

;AP

�ac

cess

ory

path

way

;AVR

T�

atri

oven

tric

ular

reen

tran

tta

chyc

ardi

a;CS

�co

rona

rysi

nus;

CSos

�co

rona

rysi

nus

osti

um;D

elta

�de

lta

wav

e;EC

G�

elec

troc

ardi

ogra

m;F

�fe

mal

e;LP

S�

left

post

eros

epta

l;m

ale;

MCV

�m

iddl

eca

rdia

cve

in;

pre-

RF�

befo

reab

lati

on;

PRi

�PR

inte

rval

;PV

�po

ster

ior

vein

;RF

�ra

diof

requ

ency

;si

te1

�si

tew

here

first

abla

tion

atte

mpt

was

mad

e;si

te2

�si

teof

abla

tion

ofth

ede

crem

enta

lfibe

r;W

�ck

ebac

hcy

cle

leng

th;

2:1

�2:

1bl

ock

cycl

ele

ngth

.


seconds later. The 12-lead ECG (Figure 4) at this point showssinus rhythm with spontaneous second-degree Wenckebachblock in the AP and first-degree AV block at the AV node.Conduction gradually improved and during programmedatrial stimulation a nonsustained antidromic tachycardiawas induced. After discussing the risk of AV block withfamily members, the procedure was discontinued. One daylater, the patient developed spontaneous symptomatic anti-dromic tachycardia. No other group B patient had inducibleantidromic tachycardia.

Heat-induced automaticity. None of the 5 group B patientshad automaticity arising in the decremental AP during ab-lation, but case 2 had AV nodal automaticity (narrow QRScomplex). Patient 3 also had AV nodal automaticity in thefirst ablation procedure.

Ventriculoatrial conduction. Five of the 6 group B patientshad preablation ventriculoatrial (VA) conduction throughthe rapidly conducting APs. Following the first attempt atablation, no patient had VA conduction through the AP.Cases 3, 4, and 6 had VA conduction through the AV nodeHis-Purkinje axis (Figure 5), and cases 2 and 5 had VAdissociation during ventricular pacing.

Follow-up. We do not have long-term data of these acquireddecrementally conducting APs, because 5 of the 6 patientswere eventually ablated during the same procedure. How-ever, case 3 remained for almost 2 years with the decre-mental AP under medical treatment. One-to-one AV con-duction through the decremental AP is still present duringdaytime. At night, with increased vagal tonus, there is in-termittent second-degree Wenckebach block in the AP. AVnodal conduction still shows first-degree block.

DiscussionMain findingsIn the present study, we compared the clinical and electro-physiologic characteristics of 2 groups of patients with theWolff-Parkinson-White syndrome treated with RF ablation:Group A patients were successfully ablated, and group Bpatients developed slow conduction and decremental prop-erties after an attempt at ablation of the rapidly conductingAPs. We found that 5 of the 6 group B patients had an APlocated in the coronary sinus venous system. Another im-portant finding was acquired adenosine sensitivity in groupB patients (Table 2).

Slow conduction and decremental properties may occurin a variety of circumstances. Atriofascicular pathways orshort AV Mahaim fibers have decremental properties due tothe presence of AV nodal-like tissue.4 APs may also havedecremental properties due to a tortuous course.5 Fascicu-oventricular pathways are dependent on AV nodal decre-

mental properties.Tab

Case

1 2 3 4 5 6 M�

Wen

ccicl

tprcp

a


By which mechanism is RF ablation of an AP able totransform fast into slow conduction?Electrical impulse conduction is dependent on cell size andshape, microscopical tissue structure, expression of ionchannels, and cell-to-cell coupling.6 In linear models ofonducting cardiac tissue (appropriate for an AP), a de-rease in cell-to-cell coupling is associated with a decreasen propagation velocity. RF injury to the AP fiber body isonsistent with disruption in cell-to-cell coupling. Sleeve-

Figure 2 A: Case 6: Left panel: 12-lead ECG before ablation. Right pancomplexes). Following adenosine infusion and transient AV nodal block, aAV conduction time through the AP (P-delta interval prolongs). B: Case 1:in the decremental AP. Note that conduction blocks in the AV node befoTransient third-degree AV nodal block probably occurs before the fifthtrioventricular; ECG � electrocardiogram.

ike extensions of the coronary sinus myocardial coat cover

he terminal portion of the middle cardiac vein and theosterior coronary vein in 3% and 2% of normal hearts,espectively.7 These myocardial sleeves could serve as aonnection between the coronary sinus and the ventricle,robably with arborization at the distal end.8 This complex

anatomic configuration could explain a number of unusualelectrophysiologic findings such as a shift in the site ofearliest ventricular activation mimicking multiple pathwaysfollowing RF ablation at the ventricular insertion. RF abla-

ead ECG postablation does not show ventricular preexcitation (first 3 QRSr degree of preexcitation appears (arrow), and there is prolongation of theost prevalent response to adenosine, showing a Wenckebach type of blockking in the decremental fiber (PR interval prolongs in the second beat).

complex (which is fully preexcited). AP � accessory pathway; AV �

el: 12-lsmalleThe mre blocQRS

tion–mediated injury of part of the muscular coating fibers

3efct

WmpdWeooe

tosp(otcoma

RiAswAcip


might set the stage for slow conduction. The conductionvelocity of an impulse passing abruptly from a fiber of smalldiameter to a fiber of larger diameter will transiently slowbecause of impedance mismatch. The smaller amount ofconducting fibers remaining after the ablation attempt willconduct a smaller amount of excitation downstream.9 Case

(Figure 5) is an example consistent with the branchingxplanation. The dispersion of local current can produceurther propagation slowing and eventually block, as seen inase number 3, who had spontaneous Wenckebach block inhe postablation decrementally conducting AP (Figure 4).

hy do some patients have an almost identical QRSorphology while in others subtle differences areresent in the QRS complex during ventricularepolarization by the decrementally conducting fiber?e believe that the subtle QRS complex changes can be

xplained by a greater contribution to ventricular activationver the AV nodal axis because of the delay in conductionver the AP. However, AP distal arborization cannot bexcluded.

Haissaguerre et al2 reported a sudden change in theventricular preexcitation pattern after RF ablation at thedistal end of a right-sided decrementally conducting AVAP, strongly suggesting distal arborization. We likewisereported a slight change in the QRS configuration during

Figure 3 Case 6: Left and right panels show the signals recorded fromconducting AP. Note the prolongation of the stimulus-delta interval, withradiofrequency.

antidromic tachycardia after RF application at the distal end

of a long decremental fiber.10 A similar phenomenon seemso have happened in our 6 group B patients. The fact that 5f the 6 APs (83%) were located in the coronary venousystem add support to the likelihood that arborizationlayed a role. Case number 6 is an illustrative exampleFigures 3, 5, and 6). After the first RF pulse close to thestium of the middle cardiac vein (Figure 6, middle panel),he AP became decremental with a slight change in QRSomplex morphology (Figure 1). Eventually, another pulsef RF delivered at the ostium of a posterior vein, located 6m from the first application spot (Figure 6, right panel),

bolished AP conduction.

esponse to adenosine and its potential implicationsn the mechanism of acquired slow conductionll group B patients with slowly conducting APs were

ensitive to intravenous adenosine. Three group B patientsere challenged with adenosine before ablation (Table 2).denosine did not cause conduction block in the rapidly

onducting AP in any patient. Adenosine has been shown tomprove conduction in dormant pulmonary veins by hyper-olarizing the resting membrane potential.11 Adenosine

shortens action potential duration on atrial cells. One couldspeculate that adenosine would be able to improve conduc-tion in some of the patients if the mechanism of slowconduction would be related to injury in the AP body. The

ation catheter at the ablation site, before and after RF ablation of the fastjor changes in the QRS configuration. AP � accessory pathway; RF �

the ablout ma

lack of improvement with adenosine is consistent with AP

wa

TddCp

ory path


distal branching hypothesis. Adenosine blocks conductionon the AV node by abolishing the N cells action poten-tials.12 As long as no patient from our cohort had an AP

ith AV node-like properties, the mechanism of acquireddenosine sensitivity is unknown.

Figure 4 Case 3: 12-lead ECG after ablation of the fast conducting Adecremental AP and prolonged conduction in the AV node. AP � access

Figure 5 Case 6: Left (A) and middle (B) panels show the pattern of ectachycardia and right ventricle pacing before ablation, respectively. Right
VA � ventriculoatrial.
he electrophysiologic profile of acquiredecrementally conducting AV pathways is similar to aecremental antegrade AV APase 3 had inducible antidromic tachycardia during electro-hysiologic study after the ablation attempt and spontaneous

coronary sinus ostium. There is spontaneous Wenckebach block in theway; AV � atrioventricular; ECG � electrocardiogram.

atrial depolarization (bracketing in CS 4) during atrioventricular reentrantC) shows VA conduction pattern after ablation, midline and decremental.

P at the

centricpanel (

Tsabppttcc

AcdOchrbbaacdtoa

NcC


tachycardia on follow-up. All 6 group B cases had evidence ofdecremental conduction during atrial pacing at increasing rates.No patient had VA conduction through the decremental AP. Itis worth mentioning that none of the 5 patients who underwentRF ablation had thermo-induced automaticity. We previouslyreported13 8 patients with a decremental antegrade AV AP.

here are many similarities with the current cohort of acquiredlowly conducting AP. Only 1 of the 8 patients had induciblentidromic tachycardia; adenosine caused transient conductionlock in most of the patients tested (3 of the 5 patients). Noatient had minimal preexcitation (7 of the 8 patients had overtreexcitation). The major difference was the occurrence ofhermo-induced automaticity during RF ablation in 50% (4 ofhe 8 patients) of the patients from the previous study, which isonsistent with the fact that these RF-induced decrementallyonducting APs probably had no AV nodal-like tissue.

bsence of VA conduction through the decrementallyonducting APs: A common feature of nearly allecremental fibersne common feature of all different varieties of decrementally

onducting APs is the unidirectional anterograde conduction;owever, the mechanism for the absence of VA conductionemains speculative. All 6 group B patients had an AP withidirectional conduction, but following RF ablation the APsecame slowly conducting APs with decremental propertiesnd at the same time incapable of VA conduction. One couldrgue that the most likely explanation for the absence of VAonduction is linked to the presence of slow conduction andecremental properties, in other words, simultaneous change inhe electrophysiologic properties of A-V and V-A conductionver the AP dependent on anatomic changes induced by RFblation delivered at the distal end of the AP.

ever say never! An AP with bidirectional longonduction times and decremental propertiesritelli et al14 reported a patient with permanent junctional

Figure 6 Case 6: Fluoro images: Left panel shows a balloon veno-occluthe middle and right panels. The first RF application was done in the miposterior vein (right arrow). The distances from site 1 and site 2 were 0.5–radiofrequency.

reentrant tachycardia who after AV node ablation showed

anterograde AV conduction with long conduction times anddecremental properties, through the same posteroseptal AP.The AV conduction time was very long with spontaneousWenckebach during sinus rhythm. Theoretically, given thebalance between refractoriness and conduction velocity, arare patient could manifest bidirectional AP slow conduc-tion, but that will probably be an exception.

Is it possible that the same phenomenon could happenwhen ablating the atrial insertion?We are not aware of a previously published case of anacquired retrograde decrementally conducting AP followingan ablation attempt. We know of a single well-documentedcase from Maastricht with an atrioventricular reentranttachycardia using a concealed posteroseptal bypass tractretrogradely, who after a single RF pulse developed a pro-longed VA interval (120–260 ms) with decremental prop-erties. That pathway was permanently interrupted by thenext ablation impulse.

Clinical implicationsIt is critical to repeat electrophysiologic and pharmacologicmaneuvers after ablating a fast conducting bypass tract, notonly ventricular stimulation. The finding of decrementalmidline VA conduction during ventricular pacing is notsufficient. The control protocol after catheter ablation of anAP should include programmed right and left atrial stimu-lation and adenosine infusion, with the aim to unveil thepresence of an acquired decremental antegrade conduction,because ventricular preexcitation may be absent or difficultto assess during sinus rhythm after the ablation of therapidly conducting AP. Recurrent atrioventricular reentranttachycardia may result if the electrophysiologist fails torecognize such acquired decremental conduction.

ConclusionsThe risk for developing decremental conduction after the

ronary sinus angiogram. The arrows point to the ablation sites, as seen inrdiac vein. The decremental AP was ablated at the origin of an adjacent(ablation catheter tip measured 3.5 mm). AP � accessory pathway; RF �

sive coddle ca0.6 cm

ablation of a fast conducting AP is greater for an AP inside

1

1

1

1

1


the coronary venous system. Acquired “Mahaim-like” fiberscaused by RF ablation of a fast conducting AP are electro-physiologically similar to a short AV decrementally con-ducting AP. They are capable of being part of an arrhythmiacircuit and, therefore, should be targeted for ablation.

References1. Sternick EB, Timmermans C, Sosa E, et al. The electrocardiogram during sinus

rhythm and tachycardia in patients with anterograde conduction over Mahaimfibers: the importance of an “rS” pattern in lead III. J Am Coll Cardiol 2004;44:1626–1635.

2. Haissaguerre M, Cauchemez B, Marcus F, et al. Characteristics of the ventric-ular insertion sites of accessory pathways with anterograde decremental con-duction properties. Circulation 1995;91:1077–1085.

3. Sternick EB, Wellens HJJ. Variants of Ventricular Preexcitation: Recognitionand Treatment. Malden: Blackwell; 2006.

4. Klein GJ, Guiraudon GM, Kerr CR, et al. “Nodoventricular” accessory pathway:evidence for a distinct accessory atrioventricular pathway with atrioventricularnode-like properties. J Am Coll Cardiol 1988;11:1035–1040.

5. Critelli G, Gallagher JJ, Thiene G, et al. Electrophysiologic and histopathologic
correlations in a case of permanent form of reciprocating tachycardia. EurHeart J 1985;6:130–137.
6. Kléber AG. Gap junctions and conduction of cardiac excitation. Heart Rhythm2011;8:1981–1984.

7. Von Ludinghausen M, Schott C. Microanatomy of the human coronary sinus andits major tributary. In: Meerbaum S, ed. Myocardial Perfusion, Reperfusion,Coronary Venous Retroperfusion. Darmstadt: Steinkopff Verlag; 1990:93–122.

8. Sun Y, Arruda M, Otomo K, et al. Coronary sinus-ventricular accessory con-nections producing postero-septal and left posterior accessory pathways. Circu-lation 2002;106:1362–1367.

9. Sternick EB, Gerken LM, Good God E. Concealed accessory pathway with longconduction times and incremental properties: a case report. J Cardiovasc Elec-trophysiol 2001;12:103–107.

0. Sternick EB, Timmermans C, Rodriguez LM, Wellens HJJ. Mahaim fiber: anatriofascicular or a long atrioventricular pathway? Heart Rhythm 2004;1:724–727.

1. Datino T, Macle L, Qi XY, et al. Mechanisms by which adenosine restoresconduction in dormant canine pulmonary veins. Circulation 2010;121:963–972.

2. Belardinelli L, Shryock JC, Song Y, et al. Ionic basis of the electrophysiologicalactions of adenosine on cardiomyocytes. FASEB J 1995;9:359–365.

3. Sternick EB, Fagundes M, Cruz Filho FE, et al. Short atrioventricular Mahaimfiber: observations on their clinical, eletrocardiographic and electrophysiologicprofile. J Cardiovasc Electrophysiol 2005;16:127–134.

4. Critelli G, Perticone F, Coltorti F, et al. Anterograde slow bypass conductionafter closed-chest ablation of the His bundle in permanent junctional recipro-
cating tachycardia. Circulation 1983;67:687–692.

Documents

Postablation-acquired short atrioventricular Mahaim-type fibers: Observations on their clinical, electrocardiographic, and electrophysiologic profile