GUEST LECTURES IN FULL

THE FONTAN CIRCULATION:  RESULTS, LATE FOLLOW-UP AND MANAGEMENT

James L. Wilkinson, MB ChB, FRCP (London), FACC
Cardiology Department, Royal Children's Hospital, Victoria, Australia

  Early History and "misconceptions"
  Later Modifications
  Case Selection
  Survival Analysis and Hazard Function
  Analysis of Results:  Boston and Melbourne
  Conclusions 
  References

Early History and "misconceptions"

   Since the first “Fontan” operation was performed in April 19681 the procedure has undergone many modifications (Figures 1-7). Every 3-5 years significant changes have been introduced, many of which have been widely adopted2-8. 

   Initially most surgeons assumed that the right atrium would provide useful contractility and could become “ventricularised”, especially if hypertrophied, to support the pulmonary circulation. This would be aided by the placement of valves to maintain forward flow1; 3. Hence it was usual, during the early years, for a valve to be placed between the right atrium and the pulmonary arteries (Figures 2-4).  In addition valves were often implanted in the caval orifices to prevent blood being ejected into the cavae during atrial systole3 (Figures 1-2).  The valves that were so placed were often Homograft valves, which may have functioned initially but probably only for limited periods of time and would not grow – hence needing replacement to allow for patient growth. Modifications introduced by Kreutzer2 (Figure 3) and Bjork4 (Figure 4) made use of the patient’s own pulmonary valve in the atrio-pulmonary connection.  The Kreutzer modification involved translocation of the native pulmonary valve and pulmonary artery, which were anastomosed directly to the right atrium.  Bjork’s modification involved an anastomosis between the right atrium and the hypoplastic right ventricle, which was enlarged with a pericardial roof and provided a largely non-contractile conduit between atrium and pulmonary arteries, via the native pulmonary valve.  Both these approaches depended on the presence of an adequate native pulmonary valve and in many patients, where the valve was too small, a graft valve was still necessary.

Later Modifications

   Within a decade from the first Fontan operation it became evident that valves were not essential to success of the procedure and that the right atrium acted largely as a passive conduit for venous blood to enter the pulmonary circulation.  Doty introduced a modification that involved the construction of a large non-valved connection between the upper pole of the right atrium and the pulmonary arteries8 (Figure 5).  This “Atrio-Pulmonary” Fontan became a popular alternative to the earlier options – avoiding the problems associated with growth and the potential necessity for revision/valve replacement.  Nonetheless, the development over several years of progressive dilatation of the right atrium, atrial arrhythmias and increasing effort intolerance led to the search for other alternatives.  DeLeval identified the hydro-dynamic disadvantages of the various Fontan options that employed the entire right atrium in the circuit and proposed the adoption of a lateral tunnel to ensure more laminar flow from the IVC to the lungs5 (Figure 6).  This was often performed in patients who had had a previous Bi-directional Glenn operation (often described as a ‘BCPC’ or Bi-directional Cavo-Pulmonary Connection).  The construction of a lateral tunnel, to complete the Fontan, in such patients became known as “TCPC” or Total Cavo-Pulmonary Connection.

   At much the same time, as surgeons were moving from the Atrio-Pulmonary Fontan to the TCPC, many were also adopting a regular routine of a “Two Stage” Fontan (BCPC first followed by TCPC later)7.  To some extent this was a return to the original Fontan philosophy, as many of the early Fontan procedures were performed on patients who had previously had a Glenn procedure.  In between however it became the norm, in suitable patients, to proceed to the Fontan operation as a single stage procedure, rather than using a Glenn as a first stage.

   In contrast to the re-invention of the two-stage Fontan, another change that was introduced during the late 80s was a reversal of earlier policy with the deliberate creation of a small right to left shunt at atrial level, or ‘fenestration’. In the early years of the Fontan procedure the presence of a residual right to left shunt was regarded as highly undesirable and many patients with low saturations post operatively were taken back to the operating room for further surgery to close any residual atrial communication.  Laks and others became convinced that the presence of a small residual right to left shunt, with mild desaturation, was advantageous during the early post-operative period as it reduced systemic venous pressure and ameliorated the frequent problem of prolonged pleural effusions that often kept patients in hospital for lengthy periods of time9.  Such “fenestrations” could be closed later, often employing a trans-catheter “device” (such as the “Clamshell” device being developed at that time for closure of atrial septal defects)10.

   Despite all these changes, the frequent development of intractable atrial arrhythmias11-13 in survivors of the Fontan procedure after ten years or longer led to efforts to minimize or eliminate completely suture lines within the right atrium and segments of the atrium that remained at high pressure in the “Fontan Circuit”.  Marceletti proposed the use of an extracardiac prosthetic conduit to carry IVC blood to the pulmonary arteries (Figure 7) and this led during the 1990s to the widespread adoption of the so-called “Extracardiac Fontan”14. 

   Yet another approach has been the use of a so-called “11/2 Pump Repair”, with the SVC blood being diverted to the pulmonary circulation via a BCPC and IVC blood being directed through the right ventricle.  This option has been employed for patients with such defects as Pulmonary Atresia with Intact Septum and a hypoplastic Right Ventricle that is inadequate for a “2 pump repair” and also for some patients with Ebstein’s Anomaly in whom the anatomy precludes a complete repair15.

Case Selection

   Case selection generally for Fontan procedures has evolved over the 35 years since this operation was first performed16. In the early years most patients considered for such surgery had Tricuspid Atresia, though other malformations such as Double Inlet Left Ventricle were soon tackled also.  Criteria for suitability were carefully defined as described by Choussat and Fontan in 197817.  Such criteria included normal Pulmonary Artery pressure and resistance, as well as a number of anatomic and other considerations, including patient age and heart rhythm.  These criteria came to be known as the “Ten Commandments” and were rapidly discarded by many cardiologists and cardiac surgeons who regarded them as being unnecessarily restrictive18.

The Ten Commandments of Choussat17:

 1.  Age > 4 years and < 16 years
 2.  Sinus rhythm 
 3.  Low pulmonary artery pressure (mean <15 mmHg) 
 4.  Low pulmonary vascular resistance (< 4 u.m2)
 5.  Normal venous drainage 
 6.  Normal right atrial size 
 7.  Normal left ventricular function 
 8.  Adequate pulmonary artery size 
 9.  No atrio-ventricular valve regurgitation 
10. No distortion of pulmonary arteries from prior shunt surgery 

   Over the years since the introduction of the “Fontan” operation the concept has been extended to include many diagnostic groups. Almost all patients whose anatomy is such as to make a 2 pump repair difficult, high risk, or impracticable have been considered for the procedure and some surgeons have carried it out in infancy or in adult life and for patients with significantly elevated pulmonary artery pressure or resistance16;18-21. 

   Such bold attempts have often led to sub-optimal results and to a high rate of both early and late problems/complications.  Even with careful patient selection and adherence to restrictive criteria such as those defined by Choussat and colleagues, there is a disappointingly high incidence of late problems, leading to significant morbidity. These have included atrial arrhythmias, the development of thrombus within the Fontan circulation, or elsewhere – leading to thrombo-embolic complications, gradually progressive ventricular dysfunction/failure, protein losing enteropathy, increasing AV valve regurgitation, etc18; 21-25. 

Survival Analysis and Hazard Function 

   Reviewing results in patients who have undergone Fontan procedures presents problems in that the type of surgery has changed during the period under review, as have the malformations being subjected to the procedure, as also have the criteria for selection. Thus, almost all large cohorts of Fontan patients/survivors are heterogenous. They exhibit significant early and late morbidity and mortality, with a range of ongoing problems that contribute to symptoms and/or Fontan failure.

   Fontan and Kirklin analyzed the published results of the procedure and noted an increasing hazard function after about ten years. They constructed a theoretical survival curve for patients without adverse features, which they described as representing survival after “The Perfect Fontan”. This composite actuarial graph showed a survival at 15 years of around 73%26.

   Although such survival appears rather disappointing, it certainly corresponded to the experience in many centers, where both early and late mortality remained a serious concern. Fortunately, with careful case selection and good post-operative management, early mortality has been substantially reduced, although some patients continue to struggle with troublesome and protracted pleural effusions after their surgery and occasional patients need to have their Fontan “taken down” (with return to a circulation that is dependent on a systemic to pulmonary (e.g. modified Blalock) or a BCPC (Glenn) shunt). Other patients with “Fontan Failure” (either early or late) may need to be considered for transplant. With such options the concept of “Fontan Failure”, which used to be almost synonymous with “death” has a somewhat less sinister implication and a small number of patients who fail to achieve useful improvement following Fontan surgery may be kept alive with alternative forms of palliation14; 27-30. By the use of these strategies the actuarial survival curve has improved somewhat – albeit the “Failure Free” survival curve (e.g. patients alive with a Fontan circulation) may still be fairly close to the “Perfect Fontan” survival curve.

Analysis of Results:  Boston and Melbourne 

   The Boston group published their data on “Functional outcome after the Fontan operation” in 1997, looking specifically at factors influencing late morbidity.  Their study group comprised all patients undergoing the Fontan procedure in Boston in the years 1973 – 1991 (a total of 500 patients)20; 21.  

  Professor NS Kim, then a research fellow at the Royal Children’s Hospital, conducted an analysis of long-term survival of Fontan patients in Melbourne in 1996/97.  His study group comprised all patients having a Fontan operation between 1980 and 1995 (totaling 266 patients).  Because of the complexity of the data compiled, his results have not yet been published (the data is currently being brought up to date and re-analyzed).  However a number of conclusions are clear from these two studies.  Early Fontan failure has become progressively less frequent with a fall in the Melbourne series from 7.3% in the years 1980-1987 to 2.9% between 1988 and 1995.  The Boston early failure rate dropped progressively from 27% in the early 70s to 7.5% in the late 80s.  Overall late failure rate was 1.8% per year in Melbourne and 1.5% per year in the Boston series, though actuarial analysis of the Melbourne data does appear to show significant improvement in the late failure rate amongst the survivors in the second time period (1988-1995), when compared with the earlier cohort (1980-1987) (p = 0.042).

   The overall failure-free survival in the Melbourne cohort (totaling 266 patients), when analyzed actuarially, is very similar to the predicted “Perfect Fontan” survival graph, with 73% of patients being free of Fontan failure at 16 years, with early failure being rather better in the Melbourne series and intermediate failure being a little worse such that the graphs become almost identical at about ten years of follow-up.

   Factors contributing to early and late failure were analysed in the two studies and were broadly very similar.

   Certain diagnostic categories were clearly at higher risk – most notably the presence of “Right Isomerism” (Melbourne) or “Heterotaxy” (Boston).  These two categories clearly correspond closely to one another – representing different terminologies rather than different malformations. Hypoplastic left heart syndrome was a risk factor for early failure (Boston) and for late failure (Melbourne and Boston) as also for a poor functional outcome (Boston).  Elevated PVR and high pulmonary artery pressure (>19mm) were, not surprisingly, risk factors for early (Melbourne and Boston) and late failure (Melbourne).  Young age at the time of surgery was another adverse feature, as was surgery in an earlier time period.

   Several other factors were associated, in the Boston data, with late failure or with a poor functional outcome.  These included prior coarctation repair, the need for a Damus or VSD enlargement, requirement for a pacemaker prior to Fontan and necessity for atrial septectomy.  Several of these are linked with associated defects or with procedures, which are likely to have an adverse affect on ventricular function (pacemaker; VSD enlargement, Damus, coarctation) or an increase in pulmonary resistance (requirement for atrial septectomy).

   The presence of a dominant LV with normally connected great arteries (as in Tricuspid Atresia) was protective against late failure.  The presence of a dominant RV did not seem to have an adverse effect, except in patients with hypoplastic left heart syndrome and those with isomerism/heterotaxy.
 

Conclusions 

   A range of factors contribute to the success or failure of the Fontan procedure. Modifications to the procedure, including the adoption of regular fenestration and the so-called extracardiac Fontan may have contributed to the decline in early mortality / failure.  It is probable that the fall in early mortality (in part a reflection of improved case selection, better post-operative care, improved strategies for dealing with “Fontan failure” and other measures that reduce ongoing morbidity), may be rewarded with improving “failure-free survival”.

   Paradoxically the multiple changes in indications for and technical modification of the Fontan procedure have resulted in real difficulties in establishing which of the many changes have led to significant improvement in outcomes.  There is a perception, which is widely held amongst pediatric cardiologists and cardiac surgeons, that the move from Atrio-Pulmonary Fontans to Total Cavo Pulmonary Connection and hence to  extracardiac Fontans, coupled with the use of a Fenestration in many/most patients in recent years, has led to improved results. The only part of this for which evidence exists is that Fenestration reduces the duration of pleural effusions and length of hospital stay. 

   For the rest we will have to continue to accumulate data and analyze the results, in the hope and expectation that the various changes in strategy that have been adopted will be shown to be produce real benefits for our patients. ¨

References 

1.  Fontan F, Baudet E: Surgical repair of tricuspid atresia. Thorax 1971;26:240-248.

2. Kreutzer G, Galindez E, Bono H, de Palma C, Laura JP: An operation for the correction of     Tricuspid Atresia. J.Thorac.Cardiovasc.Surg. 1973;66:613 (Abstract).

3. Yacoub MH, Radley-Smith R: Use of a valved conduit from right atrium to pulmonary artery for     “correction” of single ventricle. Circulation 1976;54:III63-III70.

4. Bjork VO, Olin CL, Bjarke BB, Thoren CA: Right atrial-right ventricular anastomosis for     correction of tricuspid atresia. J.Thorac. Cardiovasc. Surg. 1979; 77: 452-458.

5. de Leval MR, Kilner P, Gewillig M, Bull C: Total cavopulmonary connection: a logical     alternative to atriopulmonary connection for complex Fontan operations. Experimental studies     and early clinical experience. J.Thorac. Cardiovasc. Surg. 1988; 96:682-695.

6. Marcelletti C, Corno A, Giannico S, Marino B: Inferior vena cava-pulmonary artery extracardiac     conduit. A new form of right heart bypass. J.Thorac. Cardiovasc. Surg. 1990;100:228-232.

7. Giannico S, Iorio FS, Carotti A, Marcelletti C: Staging toward the Fontan operation. Semin.     Thorac. Cardiovasc. Surg. 1994;6:13-16.

8. Doty DB, Marvin WJ, Jr., Lauer RM: Modified Fontan procedure. Methods to achieve direct     anastomosis of right atrium to pulmonary artery. J.Thorac. Cardiovasc. Surg. 1981;81:470-475.

9. Laks H, Pearl JM, Haas GS, Drinkwater DC, Milgalter E, Jarmakani JM, Isabel-Jones J,     George BL, Williams RG: Partial Fontan: advantages of an adjustable interatrial     communication. Ann. Thorac. Surg. 1991;52:1084-1094.

10. Bridges ND, Lock JE, Castaneda AR: Baffle fenestration with subsequent transcatheter     closure. Modification of the Fontan operation for patients at increased risk. Circulation     1990;82:1681-1689.

11. Balaji S, Gewillig M, Bull C, de Leval MR, Deanfield JE: Arrhythmias after the Fontan     procedure. Comparison of total cavopulmonary connection and atriopulmonary connection.     Circulation 1991; 84:III162-III167.

12. Gewillig M, Wyse RK, de Leval MR, Deanfield JE: Early and late arrhythmias after the Fontan     operation: predisposing factors and clinical consequences. Br.Heart J. 1992;67:72-79.

13. Greason KL, Dearani JA, Theodoro DA, Porter CB, Warnes CA, Danielson GK: Surgical     management of atrial tachyarrhythmias associated with congenital cardiac anomalies: Mayo     Clinic experience. Semin. Thorac. Cardiovasc. Surg. Pediatr. Card Surg. Annu. 2003;6:59-71.

14. Marcelletti CF, Hanley FL, Mavroudis C, McElhinney DB, Abella RF, Marianeschi SM, Seddio     F,Reddy VM, Petrossian E, de la TT, Colagrande L, Backer CL, Cipriani A, Iorio FS, Fontan F:     Revision of previous Fontan connections to total extracardiac cavopulmonary anastomosis: A     multicenter experience. J. Thorac. Cardiovasc.Surg. 2000;119:340-346.

15. Kreutzer C, Mayorquim RC, Kreutzer GO, Conejeros W, Roman MI, Vazquez H, Schlichter AJ,     Kreutzer EA: Experience with one and a half ventricle repair. J.Thorac.Cardiovasc.Surg.     1999;117:662-668.

16. Barlow A, Pawade A, Wilkinson JL, Anderson RH: Cardiac anatomy in patients undergoing the     Fontan procedure. Ann.Thorac.Surg. 1995; 60: 1324-1330.

17. Choussat A, Fontan F, Besse P, Vallot F, Chauve A, Bricand H: Selection Criteria for     Fontan’s Procedure., in Anderson RH, Shinebourne EA (eds): Pediatric Cardiology 1977.     Edinburgh,Churchill Livingstone, 1978, pp 559.

18. Mayer JE, Jr., Helgason H, Jonas RA, Lang P, Vargas FJ, Cook N, Castaneda AR: Extending     the limits for modified Fontan procedures. J.Thorac. Cardiovasc. Surg. 1986;92:1021-1028.

19. Weber HS, Gleason MM, Myers JL, Waldhausen JA, Cyran SE, Baylen BG: The Fontan     operation in infants less than 2 years of age. J.Am.Coll.Cardiol. 1992;19:828-833.

20. Gentles TL, Gauvreau K, Mayer JE, Jr., Fishberger SB, Burnett J, Colan SD, Newburger JW,     Wernovsky G: Functional outcome after the Fontan operation: factors influencing late     morbidity.  J.Thorac. Cardiovasc. Surg. 1997;114:392-403.

21. Gentles TL, Mayer JE, Jr., Gauvreau K, Newburger JW, Lock JE, Kupferschmid JP, Burnett     J,Jonas RA, Castaneda AR, Wernovsky G: Fontan operation in five hundred consecutive     patients: factors influencing early and late outcome. J.Thorac.Cardiovasc.Surg.     1997;114:376-391.

22. Wilson DG, Wisheart JD, Stuart AG: Systemic thromboembolism leading to myocardial     infarction and stroke after fenestrated total cavopulmonary connection. Br.Heart J.     1995;73:483-485.

23. Gardiner HM, Dhillon R, Bull C, de Leval MR, Deanfield JE: Prospective study of the incidence     and determinants of arrhythmia after total cavopulmonary connection. Circulation     1996;94:II17-II21.

24. Warnes CA, Feldt RH, Hagler DJ: Protein-losing enteropathy after the Fontan operation:     successful treatment by percutaneous fenestration of the atrial septum. Mayo Clin.Proc.     1996;71:378-379.

25. Monagle P, Karl TR: Thromboembolic problems after the Fontan operation. Semin. Thorac.     Cardiovasc. Surg. Pediatr. Card Surg. Annu. 2002;5:36-47.

26. Fontan F, Kirklin JW, Fernandez G, Costa F, Naftel DC, Tritto F, Blackstone EH: Outcome     after a “perfect” Fontan operation. Circulation 1990;81:1520-1536.

27. de Leval MR: The Fontan circulation: What have we learned? What to expect? Pediatr. Cardiol.     1998;19:316-320

28. Mavroudis C, Backer CL, Deal BJ, Johnsrude CL: Fontan conversion to cavopulmonary     connection and arrhythmia circuit cryoblation. J.Thorac. Cardiovasc. Surg. 1998;115:547-556.

29. Hsia TY, Khambadkone S, Redington AN, de Leval MR: Effect of fenestration on the sub-     diaphragmatic  venous hemodynamics in the total-cavopulmonary connection. Eur. J.     Cardiothorac. Surg. 2001;19:785-792.

30. Mavroudis C, Deal BJ, Backer CL: Arrhythmia surgery in association with complex congenital     heart repairs excluding patients with fontan conversion. Semin. Thorac. Cardiovasc. Surg.     Pediatr. Card Surg. Annu. 2003;6:33-50.