DISCUSSION
Fontan circulation had been evolved to overcome situations in which two distinct ventricular chambers are not readily available to pump blood into pulmonary and systemic circulations in a parallel fashion. In fact, in early 1940s, it was speculated that pulmonary vascular pressure was lower than systemic venous pressure and this pressure gradient may be sufficient to propel blood to the lungs without a power source (i.e. right ventricle) at least in animal models [10]. Beginning from the initial description of the FP in human in early 1970s, the surgical evolution of the surgical procedures in single ventricle physiology may be classified in 4 generations: i. First generation – atriopulmonary Fontan Kreutzer procedure; ii. Second generation – lateral tunnel Fontan procedure; iii. Third generation – the extracardiac conduit Fontan procedure; Fourth generation – the intra/extracardiac conduit with fenestration [11]. In addition to single ventricle situations, in some cardiac malformations, the biventricular approach is avoided and the patients are treated with construction of a Fontan pathway such as inadequate ventricles (eg. hypoplastic left ventricle in Shone’s syndrome) or AV valves (eg. tricuspid stenosis with pulmonary atresia/intact ventricular septum), c-TGA, unbalanced AV canal defect, DORV with non-committed VSD or heterotaxy syndromes with complex ventricular relationships [12]. For the purpose of clarifying the nomenclature, the term ‘functionally single ventricle’ is preferred rather than ‘single ventricle’ since some abovementioned cardiac anomalies have well developed but non-septatable ventricles [13]. In our current practice, we prefer to perform intra-extracardiac FP in cases with unusual systemic venous and pulmonary artery relationships as well as patients whom a fenestration may be necessary. In our patient population, we operated on three patients with well-developed two ventricles in order to construct a Fontan circulation, two patients with c-TGA and one patient with DORV and criss-cross AV connection.
Inherent to Fontan circulation, chronic elevation of systemic venous pressure and absence of a power source for pumping blood to the pulmonary vascular bed, low pulmonary artery pressure and vascular resistance as well as an optimal systemic ventricular function are essential ingredients of successful long-term results [14]. Beginning from the traditional ten commandments, AV valves have been repaired, subaortic and aortic arch obstructions have been relieved, pulmonary vascular geometry has been optimized, encouraging results have been reported in patients less than 4 years of age and arrhythmias have been taken under control, therefore these 10 factors have been reduced to two in mid 2000s: preoperatively impaired ventricular function and elevated pulmonary artery pressure [7,15]. Moreover, the pulmonary vascular impedance is stated to be the single most important factor limiting cardiac output in some reports [16]. Interestingly, only 15-20% of total pulmonary compliance is determined by the proximal pulmonary arteries [17].
At least five papers in English literature emphasize the elevated MPAP (>15 mmHg) as an important predictor of long-term results in Fontan patients [14,18-21]. On the other hand, in a recent study of Tran and colleagues, elevated pulmonary artery pressure, not pulmonary vascular resistance (PVR) is reported to be associated with short-term morbidity, in patients with bidirectional cavopulmonary connection [22]. In our preoperative evaluation, we did not routinely perform pulmonary vascular resistance calculation. Another parameter with regard to pulmonary artery architecture is emphasized by Itatani and co-workers in which the lower limit for pulmonary artery (Nakata) index is stated to be 110 mm2/m2 [23]. In our patient series, the mean pulmonary artery (Nakata) index was 293 ± 72 mm2/m2 (range 171 to 456 mm2/m2). On the other hand, in the original paper published by Fontan and colleagues, a McGoon ratio of 1.8 and a pulmonary artery index of 200 mm2/m2 is defined as a necessary criterion for the indication of a Fontan procedure [24].
Another important concern in patients with Fontan physiology is chronic pulsatile flow deprivation in the pulmonary circulation leading to an impaired endothelial function and nitric oxide release, reduced vascular recruitment and impaired lung growth, all leading to progressive elevation in PVR [25,26]. Nevertheless, tachycardia in a normal circulation may increase pulmonary blood flow by up to 35% without changing the diameter of impedance of the pulmonary vasculature and this mechanism is lacking in Fontan patients [27,28]. At least three exogenously administered pulmonary vasodilators are used to improve exercise capacity and myocardial performance in Fontan patients: iloprost (inhaled form of prostacyclin), sildenafil citrate (phosphodiesterase-5 inhibitor) and bosentan (endothelin receptor antagonist) [28]. In a recent meta-analysis, the oral forms of pulmonary vasodilators significantly and safely improved the hemodynamics of Fontan patients, reduced the NYHA functional class and increased 6-minute walking distance [29]. In our clinical practice, we use iloprost infusion at the early postoperative period in Fontan patients when the central venous pressure rises over 10 mmHg, especially during the period of extubation at doses between 0.5-2 ng/kg/min. We administered oral sildenafil citrate and bosentan on the postoperative first day routinely in this patient population. We did not have to administer inhaled nitric oxide in our patients in the postoperative period.
One of the important short-term morbidities following the operation of a Fontan patient with high MPAP is prolonged pleural effusions. The necessity for continued chest tube drainage often causes pain and decreases ambulation of the patients. Prolonged pleural effusion is defined as either effusion lasting longer than 14 days or an effusion requiring an intervention for reaccumulated pleural fluid following the removal of the chest tubes [30]. When this definition is considered, we did not encounter any patients with prolonged pleural effusion in our patient population, although elevated preoperative MPAP is reported to be the single most determinant for this situation in some series [30]. Mean duration of drainage was 3.9 ± 5.3 days in our patients. We used the criterion of 2cc/kg/day drainage for chest tube removal as recommended in literature [31]. We agree with the factors reported by Arsdell and colleagues in order to minimize pleural drainage: utilization of the extracardiac conduits, acceptable periods of aortic cross clamping and cardiopulmonary bypass, modified ultrafiltration and institution of inotropes and vasodilators when necessary for an optimal intravascular volume and cardiac output and early postoperative extubation [32]. In our patients, we deliberately used diuretics and angiotensin receptor blockers in the postoperative period. Temporary inotropic support was infused only in 3 of our patients. We think that angiotensin receptor blockers are highly effective in order to control afterload for an optimum cardiac output of the single ventricle as stated in literature [33]. In patients whom the daily chest tube drainage exceeded 5cc/kg, we administered steroids (first three doses through intravenous route followed by oral tablets at a dose of 3 mgr/kg methylprednisolone divided in three doses). The oral form of steroids was gradually weaned and ceased after discharge; we did not administer oral steroids more than one month although Rothman et al. reported a weaning period over 3 – 6 months [34]. At extreme cases of uncontrolled pleural effusion, bleomycin or talc slurry may be used in order to perform pleurodesis [35,36].
Baffle fenestration, initially hypothesized for right-to-left decompression and a smooth postoperative course is an important modification in the history of Fontan physiology [5]. Bridges et al. emphasized the benefits of fenestration including maintenance of optimal cardiac output that may also reduce the incidence and duration of pleural effusions. The authors assigned the patients to a fenestrated or nonfenestrated groups. The fenestrated group had a higher risk profile with regard to MPAP, pulmonary artery distortion and higher ventricular filling pressures. Interestingly, the rate of Fontan failure was equal in both groups but the fenestrated patients significantly had fewer days with pleural effusion. Similar results were reported in a randomized study published by Lemler and colleagues [37]. A recent meta-analysis, a total of 4806 Fontan patients were evaluated in order to analyze the effect of fenestration on Fontan procedure outcomes [38]. The fenestrated group (a total of 2727 patients) had significantly lower need for pleural drainage with a lower MPAP (-0.99 mmHg mean difference) and oxygen saturation (-3.07% mean difference). Importantly, there was no significant difference in stroke occurrence between the fenestrated and nonfenestrated groups. Fenestration (a right-to-left shunt) helps to increase preload, stroke volume and cardiac output for the functioning ventricle at the expense of mild cyanosis. Nevertheless, there’s still no general consensus about routine use of fenestration [38]. In our patient population, we did not routinely perform fenestration during the Fontan procedure. The resting room oxygen saturation levels were above 90% in our fenestrated Fontan patients. However, two essential factors should be outlined about this procedure: the fenestration size and postoperative anticoagulation protocol. We perform fenestration with a 4 or 4.5 mm punch on the extracardiac conduit. A larger shunt is known to result in hypoxemia induced acid base disturbance and increased pulmonary vascular resistance with clinically overt cyanosis [39]. This vicious cycle may result in decreased cardiac output. Secondly, although different protocols are available avoiding chronic use of warfarin and administering aspirin alone, we put all the patients on oral anticoagulation and antiaggregant treatment following Fontan procedure at least for one year [40,41]. At the end of the first year, we decide to stop warfarin individually, however patients with fenestration continue to use oral anticoagulants. We did not encounter any stroke or conduit thrombosis in our patient group. Li et al. compared the long-term results of fenestration on systemic oxygen saturation in a meta-analysis including 1929 Fontan patients and reported that although the early postoperative SaO2 was lower in fenestrated patients, the late postoperative SaO2 levels did not differ [42]. Therefore, some centers insist on routine fenestration in all Fontan patients, but we still decide individually depending on the perioperative MPAP value [43].
Staged approach for achieving a successful Fontan circulation have been advised in early 1980s [44]. Hopkins and associates recommended to perform a bidirectional cavopulmonary anastomosis before a Fontan completion. We adopted this approach in our clinic and all of the patients in our patient population had a stage II Glenn anastomosis before FP. In fact, this algorhythm has some important advantages. Patients with a prior bidirectional cavopulmonary anastomosis tolerate stage III FP better. Secondly, any additional intracardiac interventions and/or optimization of a sufficient pulmonary vascular architecture becomes possible. We performed pulmonary artery patch augmentation, AV valve repair and atrial septectomy as concomitant procedures in stage II. On the other hand, after the stage II palliation, systemic to pulmonary artery shunts that are mandatorily interposed as stage I interventions are closed. Early closure of a systemic to pulmonary artery shunt is important, since the diastolic run off steals blood from coronary circulation which may lead to deprived performance of the future single ventricle. Moreover, bidirectional Glenn procedure is a more effective way of pulmonary gas exchange and provides better hemodynamic performance when compared to MBTS [45] . We uneventfully performed bilateral bidirectional Glenn procedure in three patients as stage II palliation, which is speculated to have a worse outcome tendency in Fontan patients especially in association with pulmonary artery bifurcation stenosis [46].
Kreutzer and co-workers reviewed the reflections on five decades of the Fontan Kreutzer procedure and grouped the factors that jeopardize the late outcome into three categories: i. suboptimal surgical approach, ii. ventricular dysfunction and iii. increases in pulmonary vascular resistance [47]. On the other hand, Vigano and colleagues suggest that in the modern era of congenital cardiac surgery, either the‘ten commandments’ of Choussat [7] or the ‘two commandments’ of the Birmingham-UK group [14] are helpful for identifying the ‘high risk’ candidates for Fontan completion [48]. They report that there is no actual difference in perioperative outcome in a mean of 7 years follow up. Our results are consistent with their findings.
Mean pulmonary artery pressure, transpulmonary gradient and PVR along with ventricular function are still the most important parameters addressing the long-term outcome of Fontan circulation [49]. However, even the cardiac catheterization at increased altitudes may present variability [49]. In fact, determination of the exact MPAP value is not always reliable, since the pressure measurements may differ when the patient is under local anesthesia and sedation at the catheterization lab or under general anesthesia with neuromuscular blockage in the operation room (17.5 ± 2.1 mmHg vs 16.5 ± 1.8 mmHg in our patient population). Therefore, we recommend routine evaluation of the PVR and transpulmonary pressure gradient in these patients. In future, more real time monitorization tools for pulmonary artery pressures may become available on routine basis and provide improved management of pulmonary hemodynamics in Fontan patients [50]. Obviously, the story of seeking for a perfect Fontan candidate and research focusing on stem cells and optimal medical management strategies will not end in the following decades [15].
In our limited patient population, we think that Fontan procedure may be performed with satisfactory mid-term results in patients with a preoperative mean pulmonary artery pressure over 15 mmHg. These patients should be carefully followed up after the operation for the well-known complications in long term course of Fontan physiology.