Poly mor phic Ven tric u lar Tachy car dia in a

Pa tient with Cor o nary Arteriovenous Fistulae

with Mild Cor o nary Ar tery Stenoses

Wei-Ru Chiou,1 Charles Jia-Yin Hou, Hung-I Yeh, Yo-Hsang Chou and Cheng-Ho Tsai

Tor sade de Pointes can be trig gered by sev eral mech a nisms. Car diac ischemia is one of the eti ol o gies of Tor sade de

Pointes. About 50% of pa tients are asymp tom atic in the pres ence of a cor o nary arteriovenous fis tula; oth ers may

have con ges tive heart fail ure, in fec tive endocarditis, myo car dial ischemia, or rup ture of an aneurysmal fis tula. They

of ten pres ent with an gina be cause of cor o nary ar tery steal phe nom e non. We re port a 79-year-old fe male with

re peated at tacks of Tor sade de Pointes even af ter cor rect ing pos si ble causes of Tor sade de Pointes. Cor o nary

angiography re vealed two cor o nary arteriovenous fistulae, one of them aris ing from the prox i mal left an te rior

de scend ing ar tery and drain ing into the pul mo nary ar tery and the other from the right cor o nary ar tery and also

drain ing into the pul mo nary ar tery. There were co ex ist ing atherosclerotic le sions at the mid dle and dis tal left

an te rior de scend ing ar tery and the ob tuse mar ginal branch of left cir cum flex ar tery. Af ter fistulectomy and coronary

artery bypass graft surgery, the patient remained asymptomatic for 2 years.

Key Words: Cor o nary arteriovenous fistulae · Poly mor phic ven tric u lar tachy car dia

IN TRO DUC TION

Cor o nary arteriovenous fis tula (CAVF) is an un com -

mon con gen i tal anom aly. If the fis tula is small, myo car -

dial blood flow will not be com pro mised, and the pa tient

re mains asymp tom atic. How ever, ischemic events can

still be in duced by “cor o nary ar tery steal phe nom e non”

if fistulae co ex ist with cor o nary ar tery dis ease. Long QT

syn drome with Tor sade de Pointes (TdP) can be due to

many eti ol o gies, in clud ing elec tro lyte im bal ance, med i -

ca tions, struc tural heart dis ease, HIV in fec tion, stroke,

brain in ju ries, and eat ing dis or ders. We re port an aged

fe male who had re peated at tacks of TdP as so ci ated with

stenoses of cor o nary ar ter ies, and two cor o nary fistulae,

one from the prox i mal left an te rior de scend ing ar tery

(LAD) and the other from the right cor o nary ar tery

(RCA). TdP and chest pain re solved af ter fistulectomy

and cor o nary ar tery by pass graft sur gery. The pa tient re -

mained un event ful dur ing sub se quent fol low-up.

CASE RE PORT

This pa tient was a 79-year-old fe male with a his tory

of hy per ten sion, type 2 di a be tes mellitus, and atrial fi -

bril la tion with fre quent long pause, and had had a VVIR

per ma nent pace maker im planted about 2 months prior to

this ep i sode. She com plained of dizzy spells and chest

tight ness one day be fore ad mis sion. Diz zi ness was pre -

ceded by chest tight ness and diaphoresis, which could be

ag gra vated by ex er tion and re lieved by rest ing. The pa -

tient’s con scious ness was clear on ar rival at the emer -

gency room, with sta ble vi tal signs. Elec tro car di og ra phy

(ECG) showed pace maker rhythm and chest X-ray re -

109 Acta Cardiol Sin 2009;25:109-14

CAVF with Cor o nary Ar tery Stenoses In duces Poly mor phic Ven tric u lar Tachy car diaCase Report Acta Cardiol Sin 2009;25:109-14

Re ceived: March 25, 2008 Ac cepted: May 22, 2008Di vi sion of Car di ol ogy, De part ment of In ter nal Med i cine, MackayMe mo rial Hos pi tal, Tai pei, 1Taitung, Taiwan.Ad dress cor re spon dence and re print re quests to: Dr. Charles Jia-YinHou, Di vi sion of Car di ol ogy, De part ment of In ter nal Med i cine, Mackay Me mo rial Hos pi tal, No. 92, Sec tion 2, Chungshan North Road,Tai pei 10449, Tai wan. Tel: 886-2-2543-3535, Fax: 886-2-2543-3642; E-mail: clementchiou@yahoo.com.tw

vealed cardiomegaly and pul mo nary con ges tion. Car diac

en zymes, bio chem i cal pro file, and hemogram were with -

in nor mal lim its ex cept her po tas sium was 2.9 mEq/L.

Her chest tight ness was re solved af ter giv ing ox y gen, as -

pi rin load ing, ni tro glyc er ine in fu sion and po tas sium sup -

ple ment; yet sub se quent sud den on set loss of con scious -

ness oc curred at the emer gency room. The ECG at that

time showed poly mor phic ven tric u lar tachy car dia with

chang ing QRS mor phol ogy im pli cat ing TdP. The pa tient

re gained her con scious ness and pace maker rhythm af ter

200 joules of defibrillation shock.

Af ter be ing ad mit ted to car diac care unit, her heart

rate was 81 bpm, and blood pres sure 129/77 mmHg and

phys i cal ex am i na tion re vealed a GrII/IV con tin u ous

mur mur at the left up per ster nal bor der. ECG showed

atrial fi bril la tion with mod er ate ven tric u lar re sponse

with T-wave in verted at leads II, III, aVF, V3-V6, and

with pro longed QT in ter val; the cor rected QT in ter val

(QTc) mea sured 642 msec (Fig ure 1A). TdP re peat edly

at tacked in spite of in tra ve nous lidocaine plus po tas sium

chlo ride and MgSO4 sup ple ment. There were no of fend -

ing med i ca tions that could at trib ute to her ven tric u lar

tachyarrhythmia, and her po tas sium level was 4.6

mEq/L, mag ne sium 2.6 mg/dL, cal cium 7.7 mg/dL, pho -

s phate 2.0 mg/dL af ter ap pro pri ate treat ment, but QTc

re mained at 666 msec. We in creased pac ing rate up to 80

bpm but in vain (Fig ure 1B). Se rial car diac en zymes

showed her CK was 171 U/L, CKMB 7.9 U/L, and

troponin-I 1.20 ng/mL. It was pos tu lated that her long

QT and TdP could be due to myo car dial ischemia. An

emer gent cor o nary angiography showed 40% ste no sis of

the mid dle LAD, 58% ste no sis of the dis tal LAD (Fig ure

2A) and a CAVF aris ing from the prox i mal LAD and

drain ing into the pul mo nary ar tery (Fig ure 2B); there

was also 48% ste no sis of the left cir cum flex ar tery

(LCX) ob tuse mar ginal branch (Fig ure 2B) and a fis tula

aris ing from the prox i mal RCA and drain ing into the

pul mo nary ar tery (Fig ure 2C). Shunt ing mea sure ment

could not be ob tained due to un sta ble hemodynamics

sec ond ary to fre quent TdP by echocardiography ini tially

but right-side car diac catheterization re vealed step-up of

ox y gen sat u ra tion and pul mo nary-to-sys temic flow ra -

tios (Qp/Qs) was 1.26. Percutaneous cor o nary in ter ven -

tion (PCI) was car ried out in an at tempt to smooth out

le sions of the LAD, with re sult of re sid ual stenoses of

LAD be ing around 30% with thrombolysis in myo car dial

in farc tion (TIMI) flow grade 3.

The pa tient re mained in un sta ble hemodynamics af ter

PCI, and sub se quent echocardiography re vealed peri -

cardial ef fu sion which may be due to wire per fo ra tion

dur ing PCI. A sur geon was called upon for an em er gent

operaion, which proved 2 cor o nary fistulae aris ing from

the LAD and RCA and drain ing into the main pul mo nary

ar tery. About 350 ml peri cardial ef fu sion was drained out,

with no ac tive bleeder found ex cept some ecchymoses

over the dis tal LAD. The two cor o nary arteriovenous

fistulae were suc cess fully li gated af ter cor o nary ar tery by -

pass graft sur gery. TdP re solved with QTc short ened to

459 msec about 1 month af ter the pa tient’s sur gery (Fig -

ure 1C). The pa tient re mained chest pain-free and had no

more at tacks of TdP in the fol low ing 2 years.

DIS CUS SION

CAVF is an un com mon, ab nor mal com mu ni ca tion

between the epicardial cor o nary ar tery and vena cava,

subpulmonary veins, pul mo nary ar tery, mediastinal ves sels,

cor o nary si nus or a car diac cham ber, oc cur ring as an in -

ci den tal find ing in 0.1% to 0.2% of rou tine car diac

angiographic stud ies. In ci dence of fistulae aris ing from the

RCA is around 50%, from the left cor o nary ar tery 42%,

from both ves sels 5% and the re main ing 3% not spec i fied.1

41% of fistulae drain into the right ven tri cle, 26% into the

right atrium, 17% into the pul mo nary ar tery, 7% into the

cor o nary si nus, 5% into the left atrium, 3% into the left

ven tri cle, and 1% into the su pe rior vena cava.1 Around

50% of pa tients are asymp tom atic when CAVF is found

dur ing eval u a tion of a car diac mur mur or on a cor o nary

angiography. In the oth ers, CAVF was found to be due to

symp toms like con ges tive heart fail ure, in fec tive endo -

carditis, myo car dial ischemia, or rup ture of an an eu -

rysmal fis tula. There is a ten dency to oc cur in early in -

fancy or at > 40 years of age. Phys i cal ex am i na tions may

re veal a con tin u ous mur mur. Se lec tive cor o nary arterio -

graphy is the gold stan dard for di ag no sis. Other di ag nos -

tic mo dal i ties such as transthoracic echocardiography,

transesophageal echocardiography, or multidetector

com puted to mog ra phy have also been re ported.2 If the

shunt is large enough, the step-up of ox y gen sat u ra tion

can be mea sured by right-sided car diac catheterization.

Among pa tients with atherosclerotic cor o nary ar tery

Acta Cardiol Sin 2009;25:109-14 110

Wei-Ru Chiou et al.

111 Acta Cardiol Sin 2009;25:109-14

CAVF with Cor o nary Ar tery Stenoses In duces Poly mor phic Ven tric u lar Tachy car dia

Fig ure 1. (A) ECG af ter 1st ven tric u lar tachy car dia ep i sode showed atrial fi bril la tion with mod er ate ven tric u lar re sponse, T-wave in verted at leads

II, III, aVF, V3-V6, and pro longed QT in ter val, with cor rected QT in ter val mea sured 642 msec. (B) Long QT in ter val with Tor sade de Pointes

re peat edly at tacked in spite of in tra ve nous lidocaine, po tas sium, MgSO4 sup ple ment and pac ing rate in creas ing to 80 bpm. (C) Tor sade de Pointes

re solved with cor rected QT in ter val short ened to 459 msec about 1 month af ter the pa tient’s sur gery.

A

B

C

dis ease, CAVF can lead to cer tain par tic u lar symp toms.

In this case, the CAVF arose from the prox i mal LAD, and

there were two dis crete le sions with around 40%-60% ste -

no sis. Our pro posed the ory is a “cor o nary steal phe nom e -

non”, with shunt ing from the high-pres sure cor o nary ar tery

into the lower-re sis tance pul mo nary sys tem. If the pa tient

only had CAVF or in sig nif i cant cor o nary ar tery dis eases

(CAD) with her myo car dial blood flow re main ing un -

compromised, she would be asymp tom atic. We pro pose

that the ef fect of cor o nary steal phe nom e non in con junc tion

with CAD caused myo car dial ischemia dis tal to the fis tula.

In our case, hypokalemia was found ini tially, but

long QT in ter val with TdP re peat edly at tacked de spite

sup ple ment of po tas sium and MgSO4. Ac cord ing to the

pa tient’s ischemic ECG change, pos i tive troponin-I level

and the find ings on cor o nary angiography, the ischemia-

in duced long QT syn drome and Tor sade de Pointes are

the most prob a ble causes, since her TdP re solved and

QT in ter val short ened af ter fistulectomy and cor o nary

ar tery by pass graft sur gery.

It is gen er ally ac knowl edged that early afterdepo -

lari zations (EADs) can in duce TdP.3 EADs are sin gle or

mul ti ple os cil la tions of the transmembrane volt age or

de po lar iza tion and fail ure of nor mal de po lar iza tion.3

Pro lon ga tion of the ac tion po ten tial du ra tion and the QT

in ter val and repolarization fail ure may oc cur by ac ti va -

tion of de layed so dium cur rent, an in creased in ward cal -

cium cur rent and de creased out ward po tas sium cur rent.

So dium in flux may pre cede po tas sium efflux dur ing

ischemia, and the ac ti va tion of pro long de po lar iza tion

due to in ward so dium cur rent causes net po tas sium loss.4

Hypoxia in creases the ac tiv ity and open prob a bil ity of

per sis tent so dium cur rent in rat ven tric u lar myocytes.5

Oth er wise, a prod uct of ischemic me tab o lite lysophos -

Acta Cardiol Sin 2009;25:109-14 112

Wei-Ru Chiou et al.

Fig ure 2. (A)Cor o nary angiogram in the right an te rior oblique

cra nial pro jec tion showed a 40% ste no sis of the mid dle LAD (ar row)

and a 58% ste no sis of the dis tal LAD (bold ar row) with TIMI - 2 flow.

(B) Cor o nary angiogram in the right an te rior oblique cau dal pro jec tion

showed a fis tula from the prox i mal LAD into pul mo nary ar tery (bold

ar row) and a 48% ste no sis of the ob tuse mar ginal branch of the LCX

(ar row). (C) Cor o nary angiogram in the right an te rior oblique cra nial

pro jec tion showed a fis tula from the prox i mal right cor o nary ar tery into

pul mo nary ar tery (bold ar row) with TIMI - 2 flow.

A B

C

phatidylcholine changes the so dium chan nel ki net ics,

and then causes length en ing of repolarization by a noni -

nactivation of so dium chan nel.6 All the pre vi ous find -

ings sug gest that myo car dial ischemia may in crease in -

ward so dium cur rent and de crease out ward po tas sium

cur rent, which in duces EADs, QT pro lon ga tion and pre -

cip i tates TdP. Fur ther more, the num bers of a1-adren -

ergic re cep tors, the in creased ef fi ciency of effector-re -

cep tor cou pling, and the re lease of norepinephrine from

lo cal nerve end ings are in creased by ischemia and all of

them aug ment the net a1-adrenergic ac ti va tion.3 EADs

are fa cil i tated by a1-adrenergic ac ti va tion.3 Kenigsberg

et al.7 re ported QTc pro lon ga tion de vel oped dur ing cor o -

nary angioplasty and early ischemia in a se ries of all 74

pa tients un der go ing se rial ECGs. QTc pro lon ga tion may

oc cur dur ing the early phase of ischemia.

The key word cor o nary arteriovenous fistulae or co r -

o nary fistulae com bined with Tor sade de Pointes or

poly mor phic ven tric u lar tachy car dia was searched by

Medline, but no re lated case re port was found. This case

may be the first re port ing CAVF-in duced TdP. CAVF

with cor o nary steal phe nom e non is an un com mon etio -

logy of myo car dial ischemia, and myo car dial ischemia

is not caus ative of TdP fre quently. The in fer ence of the

ex tremely un com mon ap pear ance with re gard to CAVF-

in duced TdP is rea son able.

The nat u ral his tory of CAVF is vari able; some have

long pe ri ods of sta bil ity and oth ers have sud den on set or

grad ual pro gres sion of symp toms. Spon ta ne ous clo sure

has been re ported in chil dren but is un com mon in adults.

There is a gen eral con sen sus that re pair ing the fis tula is

rec om mended for symp tom atic pa tients, and for those

asymp tom atic pa tients at risk for fu ture com pli ca tions

in clud ing cor o nary steals, aneurysms, and large shunts

(Qp/Qs greater than 1.5:1).8 Oth er wise, the prog no sis

among asymp tom atic pa tients is good and a con ser va tive

fol low-up is rec om mended. The need for endocarditis

pro phy laxis among un treated pa tients re mains con tro ver -

sial. Al though our pa tient’s Qp/Qs was 1.26, it is rea son -

able to as sume that her myo car dial ischemic was in -

duced by steal phe nom e non of CAVF com bined with

cor o nary ar tery stenoses.

Sur gi cal clo sure usu ally has low mor tal ity and mor -

bid ity, and the long-term out comes are ex cel lent. The

po ten tial ef fi cacy and safety of transcatheter clo sure

make it an al ter na tive method other than sur gery.9 The

use of implantable coils is cur rently con sid ered as the

method of choice in the ma jor ity of cases.10 Sev eral fac -

tors must be taken into ac count to choose be tween sur gi -

cal and percutaneous treat ment of CAVF, in clud ing age,

con com i tant car diac de fects, cor o nary ath ero scle ro sis,

the anat omy of the fis tula, and the ex per tise of the op er -

a tor.9 Transcatheter oc clu sion may be safely per formed

in older pa tients with a higher sur gi cal risk and a fa vor -

able anat omy, and sur gery should be con sid ered for pa -

tients hav ing as so ci ated con gen i tal de fects or con sid er ing

cor o nary ar tery by pass and with a com plex anat omy.10

CONCULSION

CAVF can cause cor o nary steal phe nom e non, which

may cause myo car dial ischemia in the pres ence of sig -

nif i cant or in sig nif i cant CAD and in duce long QT syn -

drome and TdP. Clo sure of CAVF by ei ther sur gi cal

method or transcatheter oc clu sion should be ap plied in

such symp tom atic pa tients.

REF ER ENCES

1. Levin DC, Fel lows KE, Abrams HL. Hemodynamically sig nif i -

cant pri mary anom a lies of the cor o nary ar ter ies: angiographic as -

pects. Cir cu la tion 1978;58:25-34.

2. Chang SN, Lee WJ, Chiang FT, Tseng CD. Im ag ing of the cor o -

nary arterio-ve nous fistulae from left main cor o nary ar tery to pul -

mo nary ar tery with multidetector com puted to mog ra phy. Acta

Cardiol Sin 2007;23:177-80.

3. Tan HL, Hou CJY, Lauer MR, Sung RJ. Electrophysiologic

mech a nisms of the long QT in ter val syn dromes and tor sade de

pointes. Ann In tern Med 1995;122:701-14.

4. Shivkumar K, Deutsch NA, Lamp ST, et al. Mech a nism of

hypoxic K loss in rab bit ven tri cle. J Clin In vest 1997;100:1782-8.

5. Ju YK, Saint DA, Gage PW. Hypoxia in creases per sis tent so dium

cur rent in rat ven tric u lar myocytes. J Physiol 1996;497:337-47.

6. Undrovinas AI, Fleidervish IA, Makielski JC. In ward so dium

cur rent at rest ing po ten tials in sin gle car diac myocytes in duced by

the ischemic me tab o lite lysophosphatidylcholine. Circ Res 1992;

71:1231-41.

7. Kenigsberg DN, Khanal S, Kowalski M, et al. Pro lon ga tion of the

QTc in ter val is seen uni formly dur ing early transmural ischemia.

J Am Coll Cardiol 2007;49:1299-305.

8. Urrutia-S CO, Falashi G, Ott DA, Cooley DA. Sur gi cal man age -

113 Acta Cardiol Sin 2009;25:109-14

CAVF with Cor o nary Ar tery Stenoses In duces Poly mor phic Ven tric u lar Tachy car dia

ment of 56 pa tients with con gen i tal cor o nary ar tery fistulae. Ann

Thorac Surg 1983;35:300-7.

9. Armsby LR, Keane JF, Sherwood MC, et al. Man age ment of cor -

o nary arterio-ve nous fistulae: pa tient se lec tion and re sults of

transcatheter clo sure. J Am Coll Cardiol 2002;39:1026-32

10. Fiocca L, Clerissi J, Bronzini R, et al. Myo car dial ischemia due to

a cor o nary-pul mo nary fis tula treated with coil embolization. Ital

Heart J 2004;5:551-3.

Acta Cardiol Sin 2009;25:109-14 114

Wei-Ru Chiou et al.

********************************************************************************

Errata

This is hereby to express our sincere apology to the readers.

The third word of the first sentence of the discussion section on page 33 was misspelled in the March issue, 2009 (Vol.

25 / No. 1). The correct spelling should be “venomous” instead of “enomous”.

Yang HP, Chen FC, Chen CC, et al. Acta Cardiol Sin 2009;25:31-5.

The title on page 7 is incorrect in the March issue, 2009 (Vol. 25 / No. 1). It should read as follows: “NT-ProBNP but

not High Sensitivity CRP Independently Predicts Abnormal Exercise Duke Score in Well-Controlled Hypertension and

Pre-Hypertension – A study of Subjects Undergoing Health Evaluation”.

Hung CL, Liu CC, Yeh HI, et al. Acta Cardiol Sin 2009;25:7-17.

********************************************************************************