Until recemtbf a subcostal, StermotQmy QP tboracoto .Aessary to impiamt the r&es used in fibrillator systemts (t-3). al ~ammfa~tmr-

ers are curremtly evaluating monthoracotomy lead systems two tramsvemo~s c&l electrodes that cam be used comjumction with a patch electrode placed sub-

cutaneously om the left chest wall, Successful implamtatiom of these electrode systems obviates the need for thoracot- omy, thereby decreasing morbidity, reducin hospital lemgth of stay and potentially lowering overall costs relative to thoracotomy4mplamted systems. At present, little is known about the factors contributing to the success or failure of cardioverter-defibrillator implantation using a momthoracotomy approach. These data would be helpful in plamnimg the implantation of such systems in individual patients. Hm the present study, B01 comsecutive patients

From the Cardiac Unit, Massachusetts General Hospital, Harvard MFg ical School, Boston, Massachusetts.

Manuscript received March II, 1993; revised manusctipt received July 14, 199

: Dr. Ross Brooks, Cardiac Unit, Bulfinch 019, Massachusetts General Hospital, Boston, Massachusetts 02114.

81993 by the American College of Cardiology

ring a ca~.rc8ioverter-deffbrilsator urn

to, itnplamt a momtho~a~oto~y system were ~ros~e~tive~y determined.

tiemts. A total of 101 consecutiv womea) were emtolled into this study underwent cardiac catheterization including echocardiography to deli abnormalities and left ventricular fum physiologic testing was performed dard techniques as previously desc imducible sustained ar arrhythmic drug t5al before cardioverter ~rnt~t~~m was at- tempted. Twenty-five patients (25%) were re~e~vim~ amtiar-

rhythmic drugs at the tismc J ~~~~amtat~om ~~m~m~d~m~ or procainamide [m = 161, mexiletime [m = 3]? sotalol Em = 41 or amiodarone [m = 2]), All ~ardiove~er~d~~b~~~ator ianplamta- tions were perfornred as primary surgical procedures.

1836 BRQOKS ET AL. PREDICTC)RS OF NONTHORACOTOMY DEFIBRILLATOR IMPLANTATION

JACC Vol. 22, No. 7 Decemkr 1993: 183~42

Tabk 1. Clinical Characteristics of the 101 Study Patients

61 + 13; 14 to 80 81120

96 of 101 (%6) 93 of ml (73) 67 of 73 (92) 36 of 73 (49) 19 of LO1 (17) 6 of iO1 (61

SIlOi (5)

49 (491 39 (39) II (11) 2 (21

33 (331 48 (48) 2 (2)

18 (18)

19 e?’ 2; 13 to 22 31 ” 2; 26 to 38

0.54 L!z O.an; 0.44 to Q.69 23 * 2; 18 to 29

Dm pmzntcd mz mean value * SD at~d r-aqe OF number (81 of parients. CAM = coronary ar!ery bypass grail surgery; LVlD = ten ventriculur intcrnrl dimension.

Patients were positioned on t access to the left chest within t clavicular pocket was opened,

Q anld connected to an externd ~~~~~~~~~t~rn~ device. Sensing and pacing ~~~$~~~~~ were

nd were saGsfactory in ail cases. A stepwise

ST&F TWQ

THORACOTOMY

STEP THREE

STEP FOt!R

51 ifP FIVE

STEP SIX

TRANSVENQUS COIL ELECTRODES ALONE

TRANsVENQuS CQIL ELECTRQDES

SLlBCepTANEOUS Ph=CH ELECTRODE

TRANSVbPIjOUS COIL ELECTRODES

SINGLE EPICARDIAL +PATCH ELECTRODE

TWO EPICARDIAL PATCH ELECTRODES

TWO EPICARDIAL PATCH ELECTRODES

ONE TRAWENOU+S COIL ELECTRORE

TWO EPICARDIAL PATCH ELECTRODES

TWO TRANSVENOLJ: COIL ELECTRODES

the patch was implante a lateral QF postero- tion on the chest supe he nQ~t~Qracotomy le

a satjs~actory de~~ril~atiQ~ thresh pite testing rn~~ti~~e configurations, a left thoracotomy was performed with addi- tional testing after placement of one or more epicardial patches as outlined in Figure 1. The pulse generator was subsequently tested for appropriate arrbyt~mia detection and defibrillator function and implanted in a left paraumb~~- ical subcutaneous pocket. In all patiemts, a cudioverter- dcfibrihator system was implanted during the same intra- operative session. Patients re uiring thoracotomy were admitted to the surgical in care unit postoperatively, but most patients with a racotomy implantation re- turned to a monitored med t bed after first going to the

finitions and measwre ents. The defibrillation thrcsh- old was defined as the lowest energy level used to terminate ventricular fibrillation on three consecutive attempts. Car- diac size was defined as the maximal diameter of the cardiac silhouette measured on a posteroanterior chest X-ray film. Thoracic size was defined as the maximal transverse diam- eter of the thorax measured on a posteroanterior chest X-ray film and cardiothoracic ratio as the ratio of cardiac to thoracic size. The anteroposterior diameter was measured ou a left lateral chest X-ray projection from the inside border of the sternum to the inside border of the ribs at the level of the dome of the right diaphragm. Left ventricular ejection fraction was measured by gated blood pool scan or left ventriculography. The echocardiographic dimensions were standard measurements as reported previously by Sahn et al. (5).

SD. Paired data were ~rop~rt~o~s were analyzed by the

model (4). Survival in the flonthoracotomy and t groups was compared with the use of life tab

eviees. A nonthoracotomy system was successfully implanted in 72 (71%) of 101 patients (Table 2). Seven different aonthora- cotomy configurations were used (Table a), with the most common configuration illustrated in Figure 2. of 101 patients required thoracotomy, many of whom also underwent testing using a higher defibrillating energy level. A total of seven different final thoracotomy configu were used, the most frequent “Rybrid” system as shown in patients, respective

generator. Patients with a received one of three approv

generators (Ventak MOO in = It& Ventak 1 Ventak 1,555 [n = 61). All implant ered monophasic pulses.

icatltrons were ob-

within the 1st man

hospital an4 have been followed up for a mean of 12 -C 7 months (range 3 to 56) after implantation (Table 7). Seven patients (7%) died during the follow-up period (range 1.3 to 6 months after implantation). The overaH actuarial survival rate at 1 year was 91 k 0.03% and did not differ between the two groups. A total of 30 patients (30%) received appropriate

therapy for one or more arr ic episodes during the

There are four main findings of our study. 1) Currently available nonthoracotomy cardioverter-defibrillator systems can be successfully implanted in >70% of patients who undergo this procedure as a first approach to treatment. 2) The two nonthoracotomy systems evaluated appear com- parable in their ability to defibrillate the heart. 3) Smaller cardiac size and female gender are the strongest positive predictors of successful nonthoracotomy implantation. 4) The midterm outcome of patients with nonthoracotomy- and thoracotomy-implanted sy

Success rate: of ~~~~bor~~~~~ ‘The results of our study are comparable to recent data pooled by the manufacturers (8) indicating success rates of 70% to 80%. Although our study group was not a consecutive series of patients undergoing cardioverter-defibrillator implantation,

it does represent a typical series of patients wit tbr~at~~i~g ventricular arrkythmias, serious cardiac disease and left ventricular dysfunction and is therefore similar to that in earlier series of patients u~dergo~mg “planned thora- cotomy” for ~ar~liovert~r-de~brj~lator impl;~~tatio~ (2,3,9- I I), Thus, the high success rate is not a reflection of patient selection with a ~re~o~dera~~e of patients at low risk, but rather is indicati~l~e of the true efficacy of these systems view is supported by our observation that 1 I (38%) of patients requiring thoracotomy required at least two epicar- dial patches, and 6 of the 11 patients required three or more electrodes, including two epicardial patches. The ~ro~o~~o~ of patients with a high d~fibri~latiom tbr~~hold (225 J) was 4%, which is similar’ to that described in most ““planned”’ thoracotomy series and suggests a $~rn~lar ~~~~tr~m of pa-

t RRoQKSETAL. JACC Voll. 2, No. 7 BaEgICTORSOFNONTNORAC~TOMYDEPlBRlLLATORlMPLANTATION Decemlrer PW93: fn5-42

Nonthoracotomy

Women [A] Men WI (n = 20, (n = 52)

59 f 16 43 t 9

49 ?: 19 38 d 13

5459 57 L 8

14 33

12 29

Tt 4xcotomy --1

Men [Cl (n = 29)

MO+ 15

40 z 16

61 2 10

9

9

A vs. B

NS

0,009

NS

NS

NS

p Value

w vs. C

NS

NS

N%

0.005

NS

success of dlefibrilbtion is probably a function of cardiac 4s fE), In most cases of cardiac enl

hem chamkrs uvolved and are mainly respons increased chest y cardiac size (13). enl ment is usually associated with thickness and heart w t wad is attributed to both hyper- trophy and dilation ( Chapman et d. (12) previously

V&Me Ctiac diieter (cm) 16.9 ” I.9 16.7 * 1.8 NS

diametP”(cm) 31.1 I 2.5 31.3 22.7 NS

0.54 t 0.06 0.54 f Q.06 NS

between ventricular fibrillation as the presenting arrhythmia and successful uonthoracotomy implantation is probably

ToMe 6. Complisations After Cantiaverter-Defibrillator Implantatian in 101 htknts

Thoracotomy Nonthoracotomy Total (n = 29) (n = 72) (n = 101)

SVC lead migradon

Trip&r undersensing

Trip&r high pacing threshold

Tripohr lead retraction

Tripolar lead perforation

Trip~!a wzse~~irr~

Subcutaneous patch

hematoma

7 (11) 10 (10)

Data presented are number (%) of patients.

atn preselsted are ~~~~ber (%I of patieurs or rneau value -c SD.

~sto~erat~ve recovery thoracotomy, who e treated with c~id~ra~

rativc days. Qn the basis of a ~ol~~arison of costs involving postoperative hospital Pength of stay, postoperative intensive care unit requirement and the need for epidural catheter insertion and maintenance, a minimal savings of $3,5Wpatient was realized in the nonthoracotomy patient group.

Limitations of the study+ There are several potential limitations of our study. 1) The study groplp was relatively small and was not completely unselected. 2) Our study was performed with existing nonthoracotomy systems, incorpo- rating pulse generators with monophasic pulse waveforms. Recent studies (16,17) have shown that bi~~as~c shock

forms can lower defibrillation energy requirements. Jt ely that newer systems utilizing biphasic waveforms

will be successful in a higher percent of patients witch LI

larger heart and that cardiac size itself might be less im- portant. 3) Because the transvenous leads were placed before the patient was taken to the operating room, adjust-

unsuccess

1. Brooks R. McGovern skin 9N. Cument treatment of patients surviving out-of-hospital cardiac arn3. 4AMA 1991;265: X-8.

2. Kelly PA. Canimi DS. Gam Ct A. Tk iWX+dC i~~p~antab~e cardiovener-defibritlalor: efficacy, bnlplications and survival in palients

with malignant venlricula~ arrhyhias. J Am Coil Cardiol 198R;I 1:1278- 86.

3, Winkle RA, Mead KH, Ruder MA, et al. Lon@~m o~lcome with the uulomatic ~rnp~a~l~ablc cardiovcner~defibrillator. J Am Coil Cardiol 1989;

, Pinkelstcin D, et al. Out-of-hospital cardiac arrest: logic testing in the predicti~m of long-term outcome

N En@ J Mod 1988;318:19-24. 5, Sahn DJ, DeMaria A, Kissto J, Weyman Wecommendations regard-

ing quantitation in M-Mode echocardio hy: results of a suwey of echocardiographic measurements. Circulation 1978;58:1012-83.

6. BMDP Statistical Software 1999. Program P2L. In: Brown MB, et al. Las Asgclcs: Univershy of California Pren, 19%

7. BMDP Statislical Software MO. PIL. In: BrOWfl MB, et d. La+ An&%: University of California Press, 1999.

8. Bhandari AK, USA Endofak investigator Group. Inlraopcra~ivc defibril- lation efficacy of endocardial subcutaneous lead canfrguratioas [abstraell, Circulation 1992;86 Suppl J:I-799).

9. Manolis AS. Tan-DeGuzman W, Lee MA, et al. Clinical experience in sevcn~y-seven patients with the automatic ~mpla~f~ble cardiovertcr defibrillator, Am Heart J 1989;l l8:445-9.

IO. Tchou PJ, Kadri N, Anderson JV C.acercs JA, Janyeri M, Akhtar Automatic jmplantablc cardiove~cr/de~hrillat~r~ aad survival Of ~a~~~~~~ with left ventricular dysfuuction and malignant wcntricul~r smhythmias. Ann l[nlcrn Med 1988;109:529-34.

I I, Epstein AE, E~~efl~~~en MA, Kirk KA, et al. Chnical C~~~~~C~~~~~~C~ itnd outCOme of patients with high defibrillation thresholds. ~~rc~~tion 1992;

12. Chapman PD, Sagar KB, Wetherbee MD, Troup PJ. Relationship Of left ventricular mass to defibrillation threshold for the implantable defi.