Aus der Klinik
fUr Herz-, Thorax-, Transplantations-, und Gefdl3chirurgie (Arztlicher Direktor Prof. Dr. h.c. Axel Haverich)
des Zentrums Chirurugie
der Medizinischen Hochschule Hannover
Clinical Outcome of Patients 20 Years after Fontan Operation -Effect of Fenestration on Late Morbidity-
Dissertation zuf
Erlangung des Doktorgrades der Medizin der Medizinischen Hochschule Hannover
Vorgelegt von:
Masamichi Ono aus Aomori
Hannover 201'0
Angenommen vom Senat der Medizinischen Hochschule Hannover am 14.O3.2011
Gedruckt mit Genehmigung der Medizinischen Hochschule Hannover Prasident: Prof. Dr. med. Dieter Bitter-Suermann
Betreuer: PD Dr. med. Christian Hagl Referent: Prof. Dr. Armin Wessel
Korreferent: Prof. Dr. med. Gerd Peter Meyer Tag der mundlichen PrUfung: 14.03.2011
Promotionsausschussmitglieder: Prof. Dr. med. Tobias Welte, PD Dr. Kinan Rifai, PD Dr. Frank Goss6
Meinen Eltern Toyoshige und Sumako Ono
ffi
ELSEVIER.uR0r[4\" puN\L .tr-
CANDIO.THOfu\CIC SURGERY European Journat of Cardio-thoracic Surgery 30 (2006) 923 929
www.etseviercom/loaate/ejcts
Clinical outcome of patients 20 years after Fontan operation - effect of fenestration on late morbidity*
Masamichi Ono u'*, Dietmar Boethig b, Heidi Goerter a, Melanie Lange u, Mechthitd Westhoff- Bleck', Thomas Breymann u
d Division ol Thoracic and cardiavosculor Surgery, Hannover ll,edicol School, Hannover, Germany oDivision ol Pediatric Cordiology dnd Inteffive Cote lledicine, Honnover hledicol School, HannoveL Germany
c Division ol catdiolow, Honnovet hledical School, Hannover, Gemony
Received 27 March 2006; received in revised form 6 August 2006; accepted 14 August 2006; Avaitable online 30 October 2006
Abstract
Objective: The Fontan operation has been proposed as definitive palLiation for an increasing variety of hearts with comptex univentricutar anatomy, but tate morbidity after Fontan operation is stitl a matter of concern. This retrospective study evaluates the late outcome in patients wjth Fontan circllation. Methods: We inctuded 121 consecutive patients that underwent Fontan operation between 1984 and 2004. Modifications of the Fontan operation inctuded atriopulmonary anastomosis (APA; n = 28), totat cavopuLmonary connection (TCPC; n = 63), and fenestrated ICPC (f-TCPC; n = 30). Mean age was 5.8 + 5.5 years. Post operative mortality, morbidity, hemodynamics, and somatic devetopment were anatyzed. Results: ActuariaI survival was 87% at 20 years after Fontan operation. There were 10 earty deaths, 5 tate deaths, and 2 takedowns fottowed by successful conversion and heart transpLantation. Among 108 early-survivors with Fontan circutation, 19 underwent reoperation, jnctudjng 3 conversions of APA to TCPC. Freedom from reoperation was 76% at 20 years. Freedom from intervention was 34% at 20 years. Freedom from tachyarrhlthmia or pacemaker implantation was23% and 77%, respectivetyat 20years. Heterotaxyand atrioventricular valve anomaLywere rjsk factors for tate failufe and tachyarrhythmias. Patients with fenestrated TCPC had reduced incidence of tate tachyarrhythmias, and patients wjth APA who devetoped cottatera [s showed tow i ncidence of late tachyarrythmia. Postoperative sinus node dysfunction or tachya rrhyth mias was associated with significantty tower cardiac index. Somatic development was graduatty compensated after Fontan operation. Weight normatized comptetety 15 years postoperativety. Conclusions: Long-term survivat after Fontan procedure js encouraging, but tate morbidity remains suboptimat. During foltow-up, emerging comptications shoutd be managed by surgicat and interventional procedures. Fenestration in Fontan cjrcutation provided better cardiac output and lower incidence of late tachyarrh]thmias, suggesting a benefit of fenestration for late outcome.
acl 2006 Etsevier B.V Att rights reserved.
l(eywordii Fontan procedure; Fenestration; Tachyarrhythmia
1 . lntroduction
The Fontan operation was first performed in patientswith tricuspid atresia as a functionat repair [1]. Modifications of the Fontan procedure and staging with bidirectionaL cavoputmonary shunt have extended the indicatjons for this operation now apptied to a wide range of congenitat heart defects unsuitabte for biventricutar repair [2,3].
Furthermore, creation of fenestration in Fontan pathway reduces earty mortaLity and morbjdity even in high-risk patients for Fontan comptetion [4]. Despite the improve-
" Presented at thejoint 20th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 14th Annual Meeting of the European Society of Thoracjc Surgeons, Stockhotm, Sweden, September 10-11, 2006,
' Coftespondjng author Address: Division of Thoracic and Cardiovascular surgery Hannover Medical SchooL, Carl'Neuberg-Str 1, 30625 Hannover, G e r m a n y . T e l . : + 4 9 5 1 1 5 3 2 9 3 9 7 ; f a x : + 4 9 5 1 1 5 3 2 9 8 3 2 .
E moil dddress: Ono.Masamichi@MH-HannoverDE (1y1. Ono).
1010-7940/5 - see front matter O 2006 Etsevier B.V Atl rights reserved.
d o i : 1 0 . 1 0 1 6 / j . e j c r s . 2 0 0 6 . 0 8 . 0 2 5
ments in surgjcal techniques that reduce perioperatiye mortatity [5-7], late deterioration in functionat status can be observed with longer duration of fol.tow-up [8-10]. Late morbidity is mainLy defined by pathway obstruction [11], atriat arrhythmia [12-14], and cyanosis because of systemic venous cottateraUzation [15], thrombus formation ['16], protein-tosing enteropathy [17], and putmonary arteriove- nous matformations [18]. Postoperative morbidity with the need of hospitatization atso inctudes surgicaI and catheter- based re-interventions. Somatic growth failure was docu- mented after Fontan operation [19]
The reason why these various morbidities occur in time- dependent manner after Fontan operation is stitl unclear.
Francis Fontan himsetf [8] stated that this operation has a pa(tiative nature in itsetf. A recent study demonstrated the evidence of progression of putmonary vascutar disease [20] in patients with fajting Fontan circutation. Neurohormonal abnormatities [21 ,22] and putmonary endothetiaL dysfunction
l . Ono et oL / European Journal ol Cardio-thoracic furgery 30 (2N6) 923-929
[23] are atso reported in patients late after Fontan operation.
These findings suggest undertying mechanisms in the putmonary circulation that induce Late morbidities after Fontan operation.
ln this retrospectjve study, we evaluated tate mortaUty, tate morbidity, postoperative hemodynamics, and somatic devetopment during long-term fottow-up after Fontan-type operations.
2. Patients and methods 2.1- Potients
Between 1984 and 2004, 121 patients with functionatty univentricutar hearts underwent Fontan-type procedures at our institution. Patients' characteristics are shown in Tabte '1 . Battoon atrial septostomy has been performed in 11 patients.
ln our institute, bidirectional Gtenn shuntswere introduced in 1989 and performed in high-risk patientsfor Fontan operation.
Since 1997, bidirectional Gtenn shunts has been performed in atl patients prior to Fontan comptetion. Mean age at the time of bidirectiona( Gtenn shunt was '1 .8 + 1.5 years, and mean period prior to Fontan comptetion was 1.3 + 0.6 years.
Mean age at Fontan operation was 5.8 + 5.5 years, ranging from 0.6 to 32 years. There were ?0 patients who underwent Fontan procedure at the age of 10 or otder. Seventeen of 20 patients were operated before 1994. The main cause was late referral from other institutions. Modifications of the Fontan operation included atrioputmonary anastomosis (APA;
n=28), totat cavoputmonary connection (TCPC; n=63),
Tabte 1
Patients characteristics (n = 12t)
and fenestrated TCPC (f-TCPC; n = 30). Concomitant proce- dures were tricuspid vatve ctosure in six patients, mitral vatve ctosure in two, atrioventricutar vatve ptasty in two, left ventricutar outflow tract obstruction retiefin seven, repair of total anomatous pulmonary venous return in one, and pulmonary artery reconstruction in three patients.
2. 2. Su rgi cal p rocedu res
APA was performed by connection of the atriat appendage to the main pulmonary artery. In three patients, a homograft was inserted between the atriat appendage and main putmonary artery and jn the others APA was performed by direct anastomosis. TCPC was performed as taterat tunnel technique, described by de Leval. et at. [2]. As for the material for atriat patch, pericardium or Gore-tex patches were used. Fenestrations were created using a 4mm diameter punch. Onty one patient underwent extracardiac TCPC using a Gore-tex tube graft.
2-3. Doto couection and analysis
Pre. and postoperative data were cottected from patients' records of Hannover Medical Schoot. For Datients fottowed etsewhere, data were cottected by fax transmission. Post- operative hemodynamic data were cotlected from cardiac catheterization reports. Somatic devetopment after Fontan procedure was investigated using weight and height gain retated to standardized percentites. These parameters were evatuated preoperativety and 1 ,2, 5,7, 10, 12, and 15 years postoperativety, and expressed in percentites. Late mortaUty and Late morbidity was evatuated using the Kaplan-Meier method.
2. 4. Statisti cat analysi s
Vatues are expressed as mean + standard deviation. Data were anatyzed using 5P55 statistical software (SPSS Inc., Chicago, lL). Estimated actuariaI survival and freedom from late morbidities were determined by the Kaptan-Meier method and analyzed with the tog-rank test. Muttivariate risk anatysis was performed wjth Cox regression test. A vatue of p < 0.05 was considered to be statisticatty significant.
3, Results
3.1. ttlortolity ond Iate Fontan lailure
Ten patients died within the first 30 postoperative days (earty death): causes of death were tow output syndrome in eight, intractable atriat tachycardia in one and intraput- monary bteeding in one. There were fiye tate deaths: one patient died 'l month after operation because of sepsis and massive pteural effusion. Two patients died of tow output syndrome 2 and 5 months after the operation. Two patients with heterotaxy syndrome died 29 and 66 months post- operativety because of mutti-organ faiture and sudden death at home, Two Datients underwent Fontan takedown 4 and 30 months after Fontan operation, both of them are ative and doing wett: in one patient extra-cardiac TCPC was performed
Diagnosis Tricuspid atresia
Other predominant left ventrjcle Predominant right ventricle Transposition of great arteries Heterotaxy {atrial isomerjsm) Atrioventricutar valve anomaty Systemic venous anomaLy Putmonary venous anomaly Paltiative procedure
Atrioseptectomy Coarctation repair B T s h u n t CentraI shunt Original Glenn shunt Bidirectional Glenn 5hunt PA banding
PA reconstruction Others
Perioperative varjative E C C > 1 2 0 m i n MC > 60 min Age at Fontan procedure
A g e < 3 3 < a g e < 1 0 1 0 < age
27 39 58 1 3
z8
9 7
64 42
4 1
60 20 1 7
9
sz (63)' 24 1261,
1 3 1 1 3 ( 1 6 ) ' 1 3
14.1 43.0 1 9 . 8 0.9 25.6'l'1.6 10.7 10.7 22.3 45.5 32.2 47.9 10.7 23.1 7.4 5.8
20.7 52.9 34.7 33.9 49.6 1 6 . 5 ' Number of procedures,
M. Ono et ol. / European Journol al Cordio.thoro.i. Surgery 30 (2006) 923-929
6 years after takedown, in the other received heart transptantation 6 years after takedown. Overat( surviyat in Fontan circulation was 87% at 20 years after Fontan operation (Fig. 1A). Actuariat survival in patients with APA was 75% at 10 years, which was significantty lower than that of TCPC (90%) and f-TcPc (90%) 1p = 9.941 1P1t. ''u,.
3.2. Late morbidities
Focusing on the late events and risk anaLysis, post- operative tong-term morbidity was evatuated using 108 patients who survjved more than 6 months in Fontan circuLation (exctuding '10 earty deaths and 3 hospitat deaths).
The fottow-up period was 10.9 + 5.2 years. At the time of their tast visit, 100 patients were in NYHA ctass I and 8 were in NYHA ctass ll.
3.2.1. Reoperotions
Nineteen patients (16%) underwent reoperations, inctud- ing conversion from APA to TCPC in three, pericardectomy in six, retief of teft ventricutar outflow tract obstruction in three, pulmonary artery reconstruction in two, tricuspid valve ctosure in two, homograft exptantation between atrjaI appendage and pulmonary artery in one, aortic vatve reptacement in one, and atriat patch revision in two. At|'
0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 (A) TimeAfter Fontan Procedure (Years)
(B) TimeAfter Fontan Procedure {Years)
Fig. 1. Estimated probabitity of survivaL after Fontan procedure. Survival curves are demonstrated for att patients (A) and in each type of Fontan procedure (B). APA, atrio pulmonary anastomosis; TCPC, totat cavopulmonary connectionj f'TCPC, fenestrated totaL cavopulmonary connection.
patients survived reoperations. overatl freedom from reoperation was 76% at 20 years. Freedom from reoperation at 10 years was 67% in APA, 86% in TCPC, and 92% in f-TCPC.
The differences were not statisticatty significant between the types of Fontan procedure (Fig. 2A).
3,2.2. Catheter i nterventions
Thirty-four patients (31%) required the fottowing catheter interventions: 16 catheter abtations, 11 fenestration cto- sures, 9 veno-venous coltateraI ctosures, 8 arterio-venous cotLateral ctosures, 6 battoon diLatations of putmonary artery 2 stent imptantations in the putmonary artery 1 fenestration creation, and 1 ctosure of persistent teft superior vena cava.
Freedom from intervention was 34% at 20 years. Freedom from intervention at 10 years was 63% in APA, 77% inTCPC, and 49% jn f-TCPC. The differences were not statistjcatty significant between the types of Fontan procedure (Fig. 28).
3 - 2 - 3 - Tochyar rhy thmio
Tachyarrhythmia occurred in 40 patients, atria[ flutter or fibritl.ation in '15, supraventricutar tachycardia in 26, and ventricutar fibrittation in 4. Two patients requjred lmptan- table Cardioverter DefibritLation (lCD) imptantation. Free- dom from tachyarrhythmia after Fontan operation was 23% at 20 years. Freedom from tachyarrhythmia at t0 years was 76%
in APA, 66% in TCPC, and 90% in f-TCPC. Patients who underwent fenestrated TCPC showed Lower incidence of tate tachyarrhythmia (Fig. 2C). Interestingty, among 21 late survivors in APA, patients who devetoped cottaterat formatjon had a Low incidence of tachyarrhythmia (1 of 6), compared with those who did not (10 of 15) (Fig. 3). This difference is statisticaLty significant (p=0.03). In tong-term fottow up, patients with cottaterat formation showed lower mean putmonary artery pressure (10.5+2.8mmHg) and smatter inferior vena cava diameter (1.7+0.3cm) than those without cottaterat formation ('12.2 + 2.7 mmHg and 2.1 +0.4cm), but these vatues were not sjgnificantty different ( p = 0.24 and p = 0.'l 9, respectivety).
3.2.4. Bradyarrhythmio
Bradyarrhythmia occurred in 27 patients, 20 of which underwent pacemaker imptantation, Freedom from pace- maker imptantation was 80% at 10 years and 77% at 20 years.
Freedom from pacemaker imptantation at 10 years was 76% in APA, 82% in TCPC, and 89% in f-TCPC. There was no significant difference between the types of Fontan procedure (Fig. 2D).
3.3. Other morbidities or observations
Thromboembotic comptications occurred in 11 patients, 3 patients developed protein losing enteropathy. Putmonary arteriovenous matformations occurred in 2 patients, Cottat- eral formation occurred in 21 cases inctuding arterio-venous cottaterats in 18 and veno-venous cotLaterats in 15 patients.
3.4. Risk factor anolvsis for late failure and lote morbidities
Univariate and multjvariate analysis have been performed to find risk factors for tate Fontan faiture and late morbidity.
Univariate analysis showed that heterotaxy was a risk factor
1.00
.9 .eo
= 3 .40
o-
s
a* 6 0
€ .qo o-
1 . 0 0
1 0 1 2 1 4 1 6
926 (A) 1
0.0
( c ) r.o
T C P C 1 8 1 6 51 48
T C P C 1 7
1 6 1
l . Ono et ol. I Europeon Jouftml ol Cardio-thorocic furgery 30 Gn6) 923-929
5 . o
6
E . 2
g
ILE
o-I
r ! E
cI
( B ) 1 . 0
(D) 1
f-TcPc E9 . 8
t " ' . 6
l ! 6
E >
E S P "
a!
t F
o q b *= F
x s
P O -o-
1 9 1 9 1 4 1 2
11 9 2 4 6 8 1 0 1 2 1 4 1 6
Time Aftef Fontan Procedure (Years)
1 3 1 2 6 4 1
50 48 45 36 2A
1 3 7 6 2
46 44 4A 29 10 1
l A l U 1 2 1 2 9 3 1
1 6 1 0 4 7 2
45 45 44 40 30 11 1
l 8 1 6 1 5 1 5 1 2 1 2 9 2 1
1 4 1 4 7 6 1 A P A
4 7 4 1 3 9 3 2 1 8 7 1 1 8 1 6 1 5 1 5 1 2 1 1 9 2 1
53 20
4 6 8 1 0 1 2 1 4 1 6 1 8 Time Aflef Fontan Procedure (Years)
e
E E ( 5 9 FdI
TCPC 18 50
Fig. Z. Estimated probabitities of freedom from reoperatjon (A), intervention (B), tachyarrhythmja (C), and pacemaker imptantation {D) in each type of Fontan procedure. APA, atrio putmonary anastomosis; TCPC, total cavopulmonary connection; f-TCPC, fenestrated tota( cavoputmonary connection,
4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 Time After Fontan Procedure (Years)
for late Fontan faiture and tate tachyarrhythmia. Atrioven- tricutar regurgitation was a risk factor for late Fontan failure and tate intervention. Pattiative procedures, perioperative variabtes, or age at Fontan operation were not riskfactorsfor tate mortatity or tate morbidity (Tabl.e 2).
3. 5. P ostope r ative he modynami cs
Postoperative hemodynamics were evatuated by cardiac catheterization in 89 patients for 200 times at our institute.
o 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 Time After Fonian Procedure (Years)
Mean fotlow-up after Fontan comptetion was 66.0 + 40.0 (6-'184) months. ArteriaI oxygen saturation (SaO2), Cardiac Index (Cl), mean putmonary artery pressure (mPAP), and end-diastotic pressure of the systemic ventricte (sVEDP) were evatuated in each type of Fontan procedure. In patients after fenestrated TCPC, SaO2 was significantty tower compared with APA and TCPA, but Cl was significantty higher than that of other procedures. Average mPAP and SVEDP were not significantty different between the types of Drocedure.
Univariate analysis showed that previous postoperative sinus node dysfunction and tachyarrhythmia were riskfactors for a lower cardiac index. Patients who received coarctation repair and putmonary artery banding were atso at risk for higher mean putmonary artery pressure after Fontan comptetion (Tabte 3).
3.6. Somatic develoDment
Before undergoing the Fontan procedure, the patients ranged between the 20th and 30th percentite.
After the Fontan procedure, somatic devetopment was compensated concerning both weight and height gain.
Weight normatized comptetety '15 years after Fontan operation (Fig. 4A), but height did not reach aver- age normal [eve[ even 15 years postoperativety (Fis. aB).
a 2 4 6 3 r 0 t 6 l 3 2 0
Time After Akiopulmonary Anastomosis Procedure (Years) Fig. 3. Estimated probabi(jties of freedom from tachyarrhythmia jn patients with atrioputmonary anastomosis.
5 5 4 1 1 3 3 3
l 3 1 2 r 0 t 0 3 1 '
M. Ono et al. I Ewopeon Journal of Cardio-thoracic Surgery 30 (2006) 923-929 Table 2
Risk factor anatyses for tate Fontan failire and morbiditier (n = 108)
Variabte n Late failure Re-OP Inter-vention Tachyarrhythmia Pacemaker implantation
927
Diagnosis
Tricuspid atresia 23 0.76
37 0.51
51 0.36
11 q.q1'
24 0.04
Atnoseptectomy 15 0.85
5V {RV Type) TGA
PaL(iative procedure CoA repair B T s h u n t C e n t r a l s h u n t BDG PA bandine
MC > 60 min Age at Fontao
Age < 3 year 3 < a g e < 1 0 1 0 < a g e
9 0.82
4 8 0 . 1 3 2 3 0 . 1 5
79 0.26
1 3 0 . 5 0
38 0.97
41 0.94
60 0.67
20 0.64
0.71 0.77
0.38 0.22
0.53 0.08
o.71 0.23
0.40 0.005'
0.64 0.20
0.71 0.06
0.90 0,64
0.66 0.57
0.09 0.17
0.93 0.87
0.37 0.62
0.06 0.99
o.91 0.82
o.71 0.79
o.97 0.37
0.54 0.49
0.42 0.88
0.61 0.83 0.58 0.02 0.004.
o.62 0.62 0.46 0.18 0.06 0.98 0.06 o.47 o . 1 4 0 . 1 4 0_ 10 0.78 0 . 1 1
0 . 1 6 0.07 0.62 0.54 0.47 0.21 o.79 0.27 0 . 1 9 0.75 0 . 1 6 0.59 0.25 o.76 0.49 0 . 1 1 0.49 0.30
PA reconstruction 13 0.88
Perioperative variative
P n m a r y F o n t a n 2 1 0 . 1 7
ECC > 120 min 57 0.92
BDG, bidirectionaL Glenn shunt. The p-vatues are reported in each risk factor.
' Significant risk factor in multivariate anatysis. UnderLined = tog |ank test.
4. Discussion tion of the modifications of the Fontan procedure, the
incidence of morbidities has tended to decrease, but this After 20 years of foltow"up, the number of late deaths is remains to be defined for more recent modifications.
timited and survivat is satisfactory but there is a retevant tate The main findings of this study are: patients after morbidity to be observed after Fontan procedure. Consider- fenestrated TCPC showed significantty lower incidence of abte tate morbidities have been demonstrated cteartyfor the late tachyarrhythmias, and patients after atrioputmonary atriopulmonary types of Fontan circutation. After introduc- anastomosis that developed cottateraLs showed lower mean
Post"operative hemodynamics ln = 89)
5aO2 c l SVEDP
Type of Fontan Nonfenestrated TCPC Fenestrated TCPC Univariate analysis
Tricuspid atresia
5V (mr' type)
TGA fleterotaxy Atrioseptectomy CoA repair B T s h u n t Cent|at shunt BDG PA banding PA reconstruction Reoperation lntervention Sinusnode d!s.
Tachyarrhythmia Col(ateraI formation
1 8 50 21 1 9 48 1 1 1 9 1 3 36 20 t t 12 1 7 1 7 37 24
93.7 r4.9 93.6 f 4.1 90.2 +3,2' o.74 0.82 0 . 1 9 0 . 1 5
0.n
0.61 0.48 0.81 0 . 1 8 0.04 0.20 o.z9 0.14 0.81 0.59 0.08 o.67
2 . 9 + 1 . 1 3.2 r 0.8 4 . 2 + 1 . 8 ' 0.93 0 . 5 1 0.96 0.59 0 , 1 7 0.79 0.35 0.69 0 . 1 1 o . 1 2 0,28 0,69 0,06 0.83 0.001 0.03 0.79
12.2.i:2.9 1 1 . 9 + 2 . 6 1 3 . 5 + 3 . 1 0.65 0.62 0.37 0.89 0 . 1 8 0.20 0.01 0.42 o.71 o.24 0.02 0.74 o.75 0.28 0.28 0 . 1 8 0.73
6.9 ,r 3.3 6 . 9 f 3 . 8 7 . 2 t 2 . 8 0.40 0.64 0.42 0.30 0.58 0.86 0 . 1 1 0.48 0.67 0.09 0.53 0.70 0.03 0.74 0.06 0.20 0.61 APA, atrio putmonary anastomosis; TCPC, totat cavopulmonary connection; BDG, bidirectionat ctenn shunt. The p,vatues are reported in each risk factor Undertined ' p < 0 . 0 5= unpaired 7-test.
T . r -
- r - r ? ? I
r - L l * - - - I
' I -
14, Ono et al. / Eurcpeon Journol ol Cordio-thorocic Sugery 30 em6) 923-929 (A)
N = s 0 e o 3 s u u u , u o u r ' t i " / ' ' N = s 0 e o 3 e . . , , . o . , ' I l ' " ' Fig. 4. Somatic development after Fontan procedure. Weight (A) and height (B) gain pre Fontan and post Fontan operation, Parameters are shown in percentites (normaU 50%) before and 1, 2, 5, 7, 10, 12, 15 years after Fontan oDeration. Data are exDressed as mean t standard error
puLmonary artery pressures, smatter venous pathway dia- meters, and a significantty tower incidence of tate tachyar- rhythmias than those who did not devetop coltaterats.
Furthermore, patients that underwent fenestrated TCPC demonstrated a higher cardiac index than other types of Fontan procedure, and patjents who devetoped tate tachyar- rhythmias showed a significantty lower cardiac index. These resutts demonstrate the importance of preputmonary venous decompression by increased fenestration shunting particu- larty under exercise load to prevent tate onset of tachyar"
rhythmias. A possibte exp(anation for tower incidence of tate tachyarrhythmias of fenestrated patients could be: an adequate shunt votume jnto the systemic ventricte, which bypasses the putmonary circutation, contributes to maintain cardiac output, and prevents the heart from ditatation by avolding extensive preputmonary venous toad under exer- cise,
These findings also impticate the latent progression of putmonary vascutar disease after Fontan operation. As it is difficuLt to access putmonary vascutar resistance after Fontan procedure directl.y, Mitchetl et at. [20] demonstrated the evidence of progression of pulmonary vascutar disease in Datients with heart transDtantation after Fontan faiture.
Atthough it seems experimentatty [24] and ctinicatty [2]
proven that putsatitity is not necessary for putmonary circulation in Fontan circuit and sinus rhythm is important for maintaining systemic ventricu(ar function, long term nonputsatite flow of putmonary circutation itsetf induces neurohormonal abnormatity [21,22] causing putmonary vascutar remodeting tate after cavoputmonary shunt. These changes may progress very stowty, but are likety to have unavoidabte impact on Fontan circutation severat years postoperativety leading to late comptications.
The concept of baffle fenestration ofthe Fontan operation was introduced by Bridges et at. [4] to decrease operative mortatity for high-risk Fontan candidates. Nowadays, this modification has been widely accepted and contributed to improve short-term outcome even in standard-risk patients [7]. However, tong-term effects of baffle fenestration are not ctear. The necessity, proper timing, and ctinicat benefit of fenestration ctosure remain unknown. Goff et at. [25]
described improved oxygenation and reduced need for anticongestive medjcation, but an increase in use of
antiarrhythmics during a mean fottow-up of 3.4 years after fenestration ctosure.
In this study, simitar to the previous reports ['12-14], tachyarrhythmia continues to devetop with time in patients with atrioputmonary anastomosis and nonfenestrated TCPC.
In this context, fenestration ctosure, which estabtishes comptete separation of putmonary and systemic circutation, may contribute to late tachyarrhythmias, even when this procedure js performed several years after Fontan operation.
Because of the beneficiat tong-term effect of fenestration, the indication for fenestration ctosure shoutd be reconsid- ered at teast for some cohort of patients with single ventricte physiotogy after Fontan comptetion despite the risk of thoromboembotism. As for somatic devetoDment after Fontan operation, there is no significant difference between fenestrated TCPC and the other Drocedures.
As for bidirectional Gtenn shunt, it was introduced in 1989 and was performed for setected patients untit 1997, then it has been routinety performed before Fontan comptetion. Among 31 patients, there were no hospital deaths at the time of Fontan procedure, but 2 patients died 78 and 180 days after Fontan comptetion because of cardiac dysfunction, which made no statistical difference for tate mortatity. In our overatl risk anatysis, bjdirectionat Gtenn shunt showed no beneficial effect on late morbidity.
Exptanation may be the increased preoperative risks for Fontan procedure in this cohort. The beneficiat effects of this shunt on Late morbidity shoul.d be addressed using case matched studies in future.
The timitation of this study anatysis is that this study is hypothesis-generating onty and that significances do not mean that a difference was proven.
In summary tong-term survivaI after Fontan procedure is satisfactory but tate morbidity remains suboptimat. During the fottow-up, cumutative occurrence of late comptications can and shoutd be managed by surgicat and medical interventions. Fenestration in Fontan circuit provided better cardiac output and lower incidence of late tachyarrhythmia, suggesting a benefit of fenestration for late outcome.
Acknowledgements
This study was supported in parts by Grant from The Uehara Memorial Foundation and The Mochida Memoriat Foundation for Medicat and Pharmaceutical Research. The authors thank Dr Hikaru Matsuda, Professor emeritus, osaka University Graduate School of medicine, for his arrangement of this research work.
References
[ 1] Fontan F, Baudet E. Surgicat repair of tricuspid atresia. Thorax 1 971 ;26:
240-8.
[2] de Levat MR, Kjtner PK, Cewitlig M, Butt C. Total cavopulmonary connec' tion: a Logicat atternative to atrioputmonary connection for complex Fontan operations, Experimental studies and early clinicat experience.
J Thorac Cardiovasc Surg 1988;96:682-95.
[3] Bridges ND, Jona5 M, Mayer JE, Flanagan ME Keane JE Castaneda AR.
Bidirectionat cavoputmonary anastomosis as interm pattation for high- risk Fontan candidates: earty resutts. Circulation 1990;82(5 SuppL):
tv170-6.
(B)
f I , r T I
It. Ono et ol. t Europeon Journol al Cordio lhoroci. Surgery 30 12006) 923 929 929
l4l Bridges ND, Lock JE, Castaneda AR. Baffle fenestration with subsequent transcatheter ctosure, Modification of the Fontan oDeration for oatients at increased risk. Circutation 1990;82:1681 9.
[5] Kauhtz R, Zjemer G, Luhmer l, Kattfelz HC. Modified Fontan operation in functionatly unjventricutar heartsr preoperative risk factors and inter- mediate resuLts. J Thorac Cardiovasc Surg 1996i112:658-64.
[6] GenttesTL, Ltayer JE, Gauvreau K, NewburgerJW LockJE, Kupfe6chmid JP, Burnett J, Jonas M, Castaneda AR. Fohtan operation in five hundred consecutive patients: factors influencing earty and tate outcome. J Thorac Cardiovasc Surg 1 997;'1141376 91 .
[7] Lemter lr{S, Scott WA, Leonard SR, Stromberg D, Ramaciotti C. Fenestra- tion improves clinicat outcome of the Fontan procedure. A prospectjve, randomized study. Cirdrtation 2002;105:207-12.
l8l Fontan E Kirklin JW, Fernandez G, Costa E Naftel DC, Tritto f, Blackstone H. Outcome after a "Perfect" Fontan operation. Circulation 1990;81:
1 5 2 0 - 3 6 .
[9] Gentles TL, Gauvreau l(, Fhhberger SB, Burnett J, Colan 5D, Newburger JU Wernovsky G. Functionat outcome afterthe Fontan operation: factors influencing tate morbidity. J Thorac Cardiovasc Surg 19971114:392 4O3.
[10] A(phonso N, Baghai M, Tutloh R, Austin C, Anderson D. Interrnediate-term outcome fottowing the Fontan opeation: a survivat, functional and risk- factor anatysis. Eur J Cardiothorac Surg 2005i28:529-35.
[11] Petko M, Myung RJ, WernovskyG, CohenMl, RychikJ, Nicotson 5C, Gaynor JU Spray TL. SurgicaL reinterventions fotLowing the Fontan procedure.
Eur .J Cardiothorac Surg 2003i241255-9.
[12] Fishberger 58, WernovskyG, GentlesTL, Gauvreau K, BurnettJ, MayerJE, Walsh EP. Factors that influence the devetopment of athat flutter after the Fontan operation. J Thorac Cardiovasc Surg 1997;111:80 6.
[1]l Durongpi5itkut K, PorterC, Cetta F, Offord Ke SLezak JM, Puga FJ, Schaff HV Danietson CK, Driscolt DJ. Predictors of early- and late-onset supra.
ventricular tachyarrhlthmias after Fontan operation. Circutation 1998;
9 8 : 1 0 9 9 - 1 0 7 .
[14] Wejpert .1, Noebauer C, Schreiber C, Kostotny M, Zrenner B, Wacker A, Hess J, Lange R, occLrrrence and management of atrial arrhythmia after long-term Fontan circutation. Thorac Cardiovasc Surg 2004i127:457 64.
[15] hutitz R, Ziemer G, Paul I Peuster M, Bertrum H, Hausdorf G. Fontan- type procedures: residuat tesjons and tate interventions. AnnThorac Surg 2002i74t778-85.
[16] Kautitz R, ZiemerG, Rauch R, Girisch M, Bertrum H, WessetA, HofbeckM.
Prophylaxis of throm boemboljc complications after the Fontan operation (total cavopulmonary anastomosis). J Thorac Cardiovasc Surg 2005;129:
569 75.
[17] Mertens L, Hagter DJ, Sauer U, Somervilte J, Gewitlig M. Protejn-tosing enteropathy after the Fontan operation; an internationaL multicenter study. J Thorac Cardiovasc Surg 1998;115:1063-73.
[18] Kawashima Y Cavopulmonary shunt and putmonary arteriovenous mal- formation. Ann Thorac Surg 1997;63:930-2.
[19] CohenMl, Bush DM, Ferry RJ, SprayTL, MoshangJrl Wernovsky c, Vetter VL. Somatic growth faiture after the Fontan operation. Cardiol Young 2000;10:447 -57 .
l20l Mitchetl MB, Campbett DN, lvy D, Boucek i\,\M, Sondheimer HM, Pietra B, Das BB, Col[ JR. Evidence of putmonary vascular disease after heart transptantation for Fontan circutation failure. J Thorac Cardiovasc Surg 20O4i178:693-702.
[21] onoM, Fukushuma N,lchikawa H, hhizakal SawaY Matsuda H. Elevatioo of plasma angiotensin with the devetopment of pulrnonary arteriovenous malformations after cavoputmonary shunt. J Thorac Cardiovasc Surg 2 0 0 5 ; 1 1 0 : 8 8 5 - 7 .
[22] HjortdatVE, Stenbog EV, Ravn HB, Emmertsen K, Jensen KT, Pedersen E8, Otsen KH, Hansen OK. Sorensen KE. Neurohormonal activation late after cavoputmonary connection. Heart 2000;83:439-41.
[23] Khambadkone 5, Li J, de LevatMR, Cullen 5, Deanlield JE, Redington AN- gasal putmonary vascuLar resistance and nitric oxide responsiveness tate after Fontan-type operation. circulation 2001;107:3204-8.
[24] Matsuda H, Kawashima Y Takano H, Miyamoto K, Mori T. Experimentat evaluation of atriaL function in right atrium-pulmonary artery conduit operation for tricuspid atresia. J Thorac Cardiovas.Sutg1981t81:762 7.
[25] Goff DA, gtume ED, Gauveau K, Mayer JE, Lock JE, Jenkins KJ. Clinjcat outcome of fenestrated Fontan patients after closure, The first 10 years, Circutation 2000;102:2094-9.
Appendix A. Conference discussion
Dr R. Jonos lwoshington, DC, U5,4): Can you tett us how you defined cotlaterats, and whether you quantitated the hemodynamic impact of lhose cottaterats? That's one que5tion.
Atso, do you think it's important that the fenestration remain patent iong term or is it simply an earty effect? And if you think tong-term patency is important, how did you measure that? What are your thoughts about device ctosure of a fenestration?
Dr ono: A fenestrated Fontan was done in 30 patients, Therewereabout 60 patieots in our study 8roup. lt is not so gross number
About 10 patients, the fenestration is ctosed spontaneously, or it was catheter closure in 5 or 6 patients. These patients aLso did not devetop lhe lale tachyarrhlthmias untiL now Though we analyze the effcacy or the hemodynamic change of effect of ctosure of fenestration, but itt nol significantty different in our sna[ number of our study.
ln my opinion, though, I think Goff et at., had reported tong.term results aftercathetercLosure afterfenestration in Circulation 2000. Thev said that !he hemodynamics change sjgnificantly because of saturation, but though smatt risk of medication for antiarrhythmic drugs. So I think after ctosure of fenestration that these risks for tachyarrhythmias for tate phase may occur after tong term after Fontan circutation,
Dl" Jonar: What about the collaterats?
U Vouhe (Paris, Fronce): Yes, what about the coltaterats? What is your defr nition of cottaterals?
Dr ono: Coltaterats were defined by cardiac catheterization.
U Vouhet Yes, but how do you quantify the cotlateral circutation?
Dr ono: These are cardjac catheterization resutts, no? That defines the cotlaterat formation. This is our definition, not cLear for me.
fu 6. Ziemer (Tuebingen, 6ermony): Actualty, I've operated abolt 80% of these patients when I was in Hannover, so, Rjchard, I may answer you.
We did not have the occtusion devices availabLe in the early'90s when y/e started the fenestrations. lt obviousty was invented in Boston, and I picked it up right away. We started with smaller hotes and hoped that they woutd close on their own. Therefore we usuatly used the 2.8 punch in the beginnjng, and onty in larger patients, who in the early '905
were not too manyanymore, aswe operated then aLreadyour Fontan candidates before 3 years of age.
W h a t l w o u l d b e i n t e r e s t e d i n , D r O n o , i s t h e i m p a c t o f t h e G l e n n s t a g e . w e had the feeljng at the beginning that those patients who got the Glenn in the staging process, did worse by means of tachyarrhythmias earty postopera- tively. So do you have an idea whether tate postoperativety there is a difference between those patientswho had a Gtenn stage and who did not have a Gtenn stage as far as rhythm probtems are concerned?
Dr Ono: Ou. bidirectional Gtenn shunt had no effect, or a benefrcial effect, for tate tachyarrhythmia. But fenestration had a significantly lower incidence for late iachyarrhythmia.
1. lnhaltsverzeichnis
1. Inhaltsvezeichnis ..
. . . 2 0
- t -
2. Publikation
"Clinical Outcome of Patients 20 Years after Fontan Operation - Effect of Fenestration on Late Morbidity,,
Ono M, Boethig D, Goerler H, Lange M, Westhoff-Bleck M, Breymann T.
Eur J Cardiothorac Surg, 2006 Dec;30(6):923-929.
Presented at 20th Annual Meeting of the European Society of Thoracic Surgeons, Stockholm, Sweden, September 10.13, 2006.
3. Zusammenfassung
3.{. Einleitung
Aktuell werden funf bis acht von 1000 Lebendgeborenen mit einem angeborenen Hezfehler geboren. Dabei sind zu circa 90 ok multifaktorielle Ursachen, vor allem Chromosomenanomalien, teratogene Noxen und Infektionen, anzunehmen.
Je nach Schweregrad des Vitiums werden die Kinder zu unterschiedlichen ZeitpunKren symptomatisch.
Gerade Neugeborene mit einem funktionell univentrikuliiren Kreislaufsystem kdnnen sich wegen der Schwere der Vitien nur unzureichend adaptieren und werden bereits in der Neonatalperiode durch eine starke Zyanose, Tachypnoe oder Trinkschwilche auffiillig.
Haufig ist dann schon in den ersten Lebenstagen oder -wochen eine palliative Operation notwendig [1].
Zu einem spdteren Zeitpunkt erfolgt die definitive Palliation eines funktionell univentrikuldren Hezens meistens mit einer kreislauftrennenden Operation nach Fontan [2]. Bis zur erstmaligen Durchfuhrung der Fontan-Operation an Patienten im Jahre 1968 gab es keine therapeutische Alternative fUr Kinder mit einem funktionell univentrikul:iren Kreislaufsvstem.
3.1.1. Das funktionell univentrikul?ire Herz
Ein bis drei Prozent aller Patienten mit angeborenem Hezfehler haben ein funktionell univentrikuleres Hez. Das funktionell univentrikuliire Hez ist durch das Vorliegen eines dominanten Ventrikels charakterisiert. Dieser steht mit dem anderen hypoplastisch oder rudimentar angelegten Ventrikel iiber einen Ventrikelseptumdefekt in Verbindung. Die Verbindung der Vorhofe mit dem dominanten Ventrikel erfolgt uber eine gemeinsame oder zwei AV-KIappen.
Hierbei kommt es zu einer Parallelschaltung von System- und Lu ngenkreislauf, was eine erhebliche Volumenbelastung der singuliiren Kammer zur Folge hat. Aufgrund einer weitgehenden Durchmischung des system- und pulmonalvendsen Blutes auf Ventrikelebene
besteht eine systemarterielle Sauerstoffuntersdttigung, wobei das AusmaB der Hypoxiimie wesentlich vom Anteil des pulmonalen Minutenvolumens abhangt [31.
Ob ein morphologisch linker (Double inlet left ventricle) oder rechter (Double inlet right ventricle) Ventrikel vorliegt, erkennt man unter anderem an der Myokardtrabekularisation.
- 3 -
Die hiiufigsten Hezfehler aus dieser Gruppe sind die Trikuspidalatresie (TA), der,,double inlet left ventricle" (DILV) und das hypoplastische Linkshezsyndrom (HLHS).
Abb. 1: Univentrikulare atrioventrikulare Verbindung
D o u b l e I n l e t S i n g l e I n l e t Common Inlet adolescents, 5th Edition) (Moss and Adams. Heart disease in infants. children. and
Die Auspriigung des Herzfehlers und die Art der Fehlbildung sind sehr vielfiiltig, so dass der klinische Zustand bei den Kindern sehr stark variieren kann. In der Regel sind bei diesen Patienten zwei bzw. drei Operationen erforderlich, bis die Kreisleufe letztlich getrennt werden k6nnen: Der Lungenwiderstand muss ausreichend niedrig sein, und das sauerstoffreiche Blut muss mit ausreichender Kraft in den Korper gepumpt werden konnen. Die erste Operation ist im Siiuglingsalter notwendig, die abschlieBende Operation zur Kreislauftrennung (sog.
Fontan-Zirkulation) findet meist im Alter von 2-3 Jahren statt.
3.{.2. Operationen nach dem Fontan-Prinzip
'1971 beschrieb Francis Fontan zum ersten Mal eine Operationsmethode zur Behandlung der Trikuspidalatresie an drei Kindern [2]. Ziel dieser Methode ist es, zwei eigenstandige
Kreisliiufe zu entwickeln, ohne dass sich weiterhin venoses Blut mit oxygeniertem vermischt.
lm Gegensatz zum nati.rrlichen Blutfluss, bei dem ein Ventrikel das Blut durch die Lunge und der andere es durch den Korper pumpt, nach etablierter Fontan-Zirkulaiion ein einziger Ventrikel das Blut zunechst durch den Kdrper, von wo aus es mit dem verbleibenden Druck passiv auch noch durch die Lunge flieBt, bis es vom Ventrikel wieder in den Korper befctrdert wird.
R A . l l L A
- 4 -
