121

© 2007, the AuthorsJournal compilation © 2007, Blackwell Publishing, Inc. Congenit Heart Dis. 2007;2:121–124

Blackwell Publishing IncMalden, USACHDCongenital Heart Disease1747-079X© 2007 The Authors; Journal compilation © 2007 Blackwell Publishing, Inc.? 200722121124Original ArticleFate of Equine Pericardial Roll Conduit for Rastelli OperationTakeuchi et al.

Fate of Equine Pericardial Roll Conduit for Rastelli Operation during Long-term Follow-up

Koh Takeuchi, MD,*† Arata Murakami, MD,* Akihiko Sekiguchi, MD,† Yasutaka Hirata, MD,* Katsuhide Maeda, MD,* Kazuo Kitahori, MD,* Yoshio Doi, MD,* and Shin-ichi Takamoto, MD*

*Department of Cardiac Surgery, University of Tokyo Graduate of Medical School, Tokyo, Japan; †Department of Cardiovascular Surgery, National Children’s Hospital, Tokyo, Japan

A B S T R A C T

Background. Right ventricular outflow tract obstruction is a frequent condition after Rastelli operation. Althoughseveral modifications have been reported elsewhere, ideal conduit has not been developed yet during long-termfollow-up. We reviewed our experiences over 15-year long-term follow-up with patients who underwent Rastellioperation using house-made equine pericardial roll graft.Methods. Since June 1981, 16 patients underwent Rastelli operation with the pericardial roll graft. Median follow-up time was 15.6 years (7.3–26.8 years).Results. Twelve out of 16 patients using pericardial roll graft with (n = 6) or without (n = 6) cusps underwent 13reoperations during the follow-up period. Median time from first Rastelli to re-do operation was 8.4 years withmedian time to reoperation of 8 years. Major indication for reoperation was conduit obstruction (n = 10), but notconduit regurgitation. Conduit problem includes kinking and compression of the graft. Reoperation proceduresinclude 7 Danielson procedures, 2 patch augmentations, 1 homograft replacement, 1 pericardial roll graft, 1expanded polytetrafluoroethylene tube graft replacement, and 1 patch closure for pulmonary artery aneurysm.Balloon angioplasty was not effective for pericardial roll conduit stenosis.Conclusion. We conclude that house-made equine pericardial roll graft was durable for certain time period, butconduit change may be inevitable. Because of excellent handling and wide application, further modification may bewarranted.

Key Words. RV Outflow Tract Obstruction; Conduit Stenosis; Rastelli Operation; Danielson Procedure

Introduction

ulmonary regurgitation after valveless repairof right ventricle outflow tract obstruction

(RVOTO) results in progressive right ventricle(RV) dilation and dysfunction.1 Valved conduitmay clear this problem to some extent, then valvedysfunction or valvular stenosis can occur. Con-duit stenosis can be another potential problemafter right ventricle outflow tract (RVOT) repair.Multiple technical modifications have beenreported to deal with these problems, but so farno such gold standard technique has beenestablished yet. Many modifications includinghomograft replacement, synthetic tube graftreplacement with or without prosthetic valve,2and pericardial patch augmentation with or with-out réparation à l’ótage ventriculaire (REV)procedure3 have been reported elsewhere.Recently, excellent short-term result with pericar-

Pdial roll graft has been reported. We have usedseveral conduits for Rastelli operation includinghouse-made pericardial roll graft with cusps as apreferred technique. During the over 15-yearmedian follow-up, we have experienced 16 patientswho underwent Rastelli operation with house-made pericardial roll graft. The purpose of thisstudy was to find the potential problems related tothe pericardial roll graft in Rastelli operation dur-ing the long-term follow-up and optimal repair fordeveloping RVOTO after Rastelli operation.

Materials and Methods

Since June 1981, 16 patients underwent Rastellioperation with house-made pericardial roll tubegraft with or without cusps for RV outflow tractrepair in University of Tokyo Hospital andNational Children’s Hospital (Figure 1). All

122 Takeuchi et al.

Congenit Heart Dis. 2007;2:121–124

patients have been followed by either Universityof Tokyo Hospital or National Children’s Hospi-tal. These patients include 8 male patients and 8female patients. Diagnoses at the first operationwere pulmonary atresia with ventricular septaldefect (PA/VSD; n = 7), truncus arteriosus (n = 1),double-outlet right ventricle (n = 3), tetralogy ofFallot with pulmonary stenosis (TOF/PS; n = 1),and corrected transposition of the great arterieswith VSD/PS (n = 4). Ten patients underwent pal-liative operation including 9 systemic to pulmo-nary shunt and 1 Brock procedure at median ageof 6.5 months and 5 patients among those under-went second palliation including 4 shunts and 1unifocalization. Age at the Rastelli operationranged from 1 year to 16 years with mean age of8.3 ± 4.5 years (mean ± SD) and median age of

7.3 years. Patient data were compiled by review ofthe clinical records including operative reports,preoperative angiographic, and 2-dimensionalechocardiographic studies. This study was ap-proved by the institutional ethics committee inUniversity of Tokyo and National Children’s Hos-pital on human research.

Results

Mean follow-up time for 16 patients were17.0 ± 6.8 years with median follow-up of15.6 years. Twelve patients underwent 13 reoper-ations and indications for re-do operation wereconduit stenosis (RVOTO) in 10 patients, infec-tious endocarditis (IE) in 1 patient, residual shunt/conduit stenosis/2-chambered RV in 1 patient,and left pulmonary artery aneurysm in 1 patient.Time to re-do operation for RVOTO from theRastelli operation ranged from 5 months to19.7 years with median time of 8.3 years. Age atre-do operation ranged from 13 to 28 years (meanage 18.6 ± 5.3 years). Age at last follow-up day was24.5 ± 2.8 years with median age of 25.1 yearsand mean follow-up time was 17.0 ± 6.8 years.Four patients have been free from reoperationwith median reoperation-free time of 14.2 years.Reoperation procedures include 7 Danielsonprocedures,4 2 patch augmentations, 1 homograftreplacement, 1 patch closure for pulmonaryartery aneurysm, 1 pericardial roll replacement,and 1 expanded polytetrafluoroethylene (ePTFE)tube graft replacement. One patient underwentpericardial roll replacement 69 months after thefirst Rastelli operation and then he underwentpatch augmentation for his repeated RVOTO56 months later. Concomitant procedures include1 tricuspid valve replacement/mitral valve plasty,1 VSD closure/infundibular muscle resection, 1infundibular muscle resection, and 1 VSD closure/graft replacement of the ascending aorta/resectionof vegetation. One patient whose diagnosis wasPA/VSD underwent Rastelli operation with peri-cardial roll at 12 years of age and then she devel-oped RVOTO, dehiscence of VSD patchassociated with IE. Because of fragile ascendingaorta based on IE, she required homograftreplacement of the ascending aorta, RV outflowtract repair with homograft, VSD patch closure,and resection of multiple vegetations, and then shedeveloped low cardiac output syndrome and diedon the first postoperative day.

Preoperative evaluation revealed 58 ± 12% ofleft ventricle (LV) ejection fraction, 52 ± 22 mL/

Figure 1. Method to create pericardial roll conduit withcusps or without cusp. Xenomedica® was used to create rollgraft. In conduit without cusp, the pericardium was sewntogether longitudinally with appropriate size and length.Proximal side was tailored to cobra-head shape. The sew-ing side was placed behind when the conduit was used asa right ventricle to pulmonary artery conduit (A,C). In con-duit with cusp, the pericardium was fold inside and themiddle edge was tagged to the pericardium to create oneof the commissures and then the roll conduit was createdwith the same fashion as conduit without cusp. Anotherpiece of pericardium was used as a hood to augment theproximal side (B,D).

A B

C D

Fate of Equine Pericardial Roll Conduit for Rastelli Operation 123

Congenit Heart Dis. 2007;2:121–124

m2 of LV end-diastolic volume index, and 13 ±4 mL/m2 of LV end-systolic volume index. Pres-sure gradient across the conduit was 60 ±30 mm Hg (median 55 mm Hg). This gradientimproved postoperatively to 22 ± 6 mm Hg(Figure 2). Two patients have been in class II statusof New York Heart Association classification(NYHA) and the remaining patients have been inclass I of NYHA at the last follow-up time.

Five patients required 12 times balloon dilationand 1 stent insertion during the follow-up timeperiod. In 1 patient, Danielson procedure was per-formed with pedicled autologous pericardium foranterior augmentation. He developed conduitstenosis and underwent multiple angioplasty andstent insertion during 6-year follow-up after re-doRVOT repair. This patient may need surgicalintervention at some point in near future. Kaplan-Meier estimated reoperation-free rate was around80% at 5 years, 50% at 10 years, and almost 0%at 20 years after surgery (Figure 3).

Discussion

Rastelli operation has been introduced for surgicalcorrection of transposition of great artery withVSD and subpulmonary stenosis and then becamean option for several anatomical variants in surgi-cal repair of congenital heart diseases. Because offrequent reoperation probability, indication forRastelli operation has been limited in these daysand we aimed not to do Rastelli operation as areconstruction of RVOT. But in some situation, itwould be an indication and some modification maybe indicated for longer reoperation-free interval.Selection of the conduit for RVOT repair mightbe a potential solution for freedom from reopera-tion. Several conduits are available in commercialbases and variable house-made conduits have been

reported. In Japan, homograft use has been verylimited and pericardial roll graft had been one ofthe choices. We had preferably used house-madepericardial roll graft 10–20 years ago in case thatpatient was young or relatively small-size conduitwas required or for the purpose of anticoagulationfree. And now we found the longevity of thishouse-made conduit was around 8 years. Multipleproblems were identified during the follow-uptime. These include compression of the conduitby chest wall, poor contour of the conduit, that is,mostly straight graft with less flexibility, and kink-ing of the graft. Kinking and compression mightbe major factors contributing to turbulent flow inthe graft and less longevity of the graft. Balloonangioplasty was not effective in this situation andstent insertion might be an option, although thismight be temporal relief. Similar technique usingpericardial roll that we used in our patients hasrecently been reported with excellent result.5However, number of patient was small and follow-up time was only up to 3 years in that study. In ourpatients, time to develop significant RVOTOranged from 5 to 236 months with median of9.5 years, and 12 patients out of 16 patients havealready required to replace them. So far, this tech-nique may not be our standard for Rastelli opera-tion because of frequent obstruction. Furthertechnical modification may solve these problemsand extend the longevity of the conduit.

Several right ventricle to pulmonary artery(RV-PA) conduit, including homograft and syn-thetic tube graft2 and other modification of RV-PAoutflow reconstruction, have been reported.6Homograft remains nonpopular conduit in Japan.A composite graft with synthetic tube graft and

Figure 2. Preoperative and postoperative pressure gradi-ent across the right ventricle (RV) outflow tract. It was60 ± 30 mm Hg (median 55 mm Hg) preoperatively andwas postoperatively improved to 22 ± 6 mm Hg.

Figure 3. Kaplan-Meier estimated reoperation-free rates.These were around 80% at 5 years, 50% at 10 years, andclose to 0% at 20 years after surgery.

Follow-up Time (months)

Reo

per

atio

n-F

ree

Rat

io

3002001000

1.0

.8

.6

.4

.2

0(11)(16)

50 150

(8)

Remaining number (2)

124 Takeuchi et al.

Congenit Heart Dis. 2007;2:121–124

mechanical valve or bioprosthetic valve has beenreported as an alternative technique for RV-PAreconstruction.5,7 Mean time for reoperation withmechanical valved conduit in our institution wasabout 15 years. However, it is not fair to comparewith pericardial roll conduit because of differentbackground, for example, age, anatomy.

Another alternative for RV outflow reconstruc-tion is direct anastomosis between the pulmonaryartery and RV with or without insertion of leftatrial appendage8 or pedicled own pericardium9 inthe posterior side. The direct anastomosisbetween the pulmonary artery and RV, that is,REV-type operation, has been reported frequentearly pulmonary artery stenosis.3 We found con-duit stenosis as frequent problem, but did not findsignificant RV dilatation due to regurgitationthrough the conduit. This might be associatedwith relatively short follow-up time. We appliedcusps in the conduit in some cases that did notcause any significant differences from noncuspgroup. Functioning cusp(s) may be potentiallyimportant in RVOT reconstruction in Rastellioperation as well as Danielson procedure. Mono-cusp ventricular outflow patch has been commer-cially available. However, it has been reported thattissue failure and degeneration of this patch wereinevitable.10 Monocusp created by own pericar-dium or xenopericardium is also known to lose thevalvular function within the next 6 months.11

Recently superiority of ePTFE monocusp valvehas been reported.12 Functioning cusps may beable to solve these problems related to conduitregurgitation in RVOT patch. Material of patchesand number of constructing cusps as well as tech-nical consideration for cusp implantation are curi-ous concern.

We have experienced 16 patients who under-went Rastelli operation with house-made per-icardial roll conduit over 15 years of medianfollow-up. Among the 16 patients, 12 patientsunderwent 13 reoperations with conduit stenosisas a major indication for reoperation. Based on ourfindings, we conclude that house-made pericardialroll graft has durable midterm result and it mayneed possible modification. Materials for RVOTpatch and cusps as well as technique to implantthese cusps may be curious concern in future.

Corresponding Author: Koh Takeuchi, MD, Depart-ment of Cardiac Surgery, University of Tokyo Hospital,7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Tel:(+81) 35800-8654; Fax: (+81) 35846-3989; E-mail:koutakeuchi-circ@umin.ac.jp

Accepted for publication November 12, 2006.

References

1 Rao V, Kadletz M, Hornberger LK, Freedom RM,Black MD. Preservation of the pulmonary valvecomplex in tetralogy of Fallot: how small is toosmall? Ann Thorac Surg. 2000;69:176–179.

2 Chun PK, Rocchini AP, Gibbs HR, Robinowitz M,Green D, Virmani R. Pannus formation in a Han-cock-valved conduit resulting in proximal intracon-duit obstruction: late complication of Rastelliprocedure for complete transposition of the greatvessels with ventricular septal defect and pulmonicstenosis. Am Heart J. 1981;101:855–857.

3 Black MD, Shukla V, Freedom RM. Direct neonatalventriculo-arterial connection (REV): early resultsand future implications. Ann Thorac Surg. 1999;67:1137–1141.

4 Danielson GK, Downing TP, Schaff HV, Puga FJ,DiDonato RM, Ritter DG. Replacement ofobstructed extracardiac conduits with autogenoustissue reconstructions. J Thorac Cardiovasc Surg.1987;93:555–559.

5 Iemura J, Oku H, Otaki M, Kitamura H, Matsu-moto T. Reconstruction of right ventricular outflowtract by pericardial valved conduit. Ann Thorac Surg.1997;64:1849–1851.

6 Aupecle B, Serraf A, Belli E, et al. Intermediate fol-low-up of a composite stentless procine valved con-duit of bovine pericardium in the pulmonarycirculation. Ann Thorac Surg. 2002;74:127–132.

7 Allen BS, El-Zein C, Cuneo B, Cava JP, Barth MJ,Ilbawi MN. Pericardial tissue valves and Gore-Texconduit as an alternative for right ventricular out-flow tract replacement in children. Ann Thorac Surg.2002;74:771–777.

8 Aeba R, Katogi T, Kashima I, et al. Left atrialappendage insertion for right ventricular outflowtract reconstruction. Ann Thorac Surg. 2001;71:501–506.

9 Kitagawa T, Katoh I, Chikugo F, et al. Techniquefor constructing the pulmonary trunk for tetralogyof Fallot with pulmonary atresia. Ann Thorac Surg.1995;59:1245–1248.

10 Morikawa M, Abe T, Takagi N, et al. Long-termresults of Rygg’s monocusp ventricular outflowpatch for the reconstruction of right ventricular out-flow tract in tetralogy of Fallot. Kyobugeka.2001;54:624–630.

11 Gundry SR, Razzouk AJ, Boskind JF, Bansal R,Bailey LL. Fate of the pericardial monocusppulmonary valve for right ventricular outflow tractreconstruction. J Thorac Cardiovasc Surg. 1994;107:908–912.

12 Turrentine MW, McCarthy RP, Vijay P, McCon-nell KW, Brown JW. PTFE monocusp valve recon-struction of the right ventricular; outflow tract. AnnThorac Surg. 2002;73:871–880.