Doppler echocardiographic evaluation of pulmonary artery flow after modified Fontan operation: importance of atrial contraction.

Doppler echocardiography was used to evaluate blood flow in the pulmonary artery in 14 patients 2 to 42 months (mean (SD) 17 (12) months) after a modified Fontan operation incorporating a direct atriopulmonary anastomosis. Preoperatively six patients had tricuspid atresia, six had a double inlet left ventricle, and two had pulmonary atresia with an intact ventricular septum. The postoperative rhythm was sinus in 11 patients, junctional in one, ventricular pacing in one, and atrioventricular sequential pacing in one. In one patient the Doppler trace was unsatisfactory for analysis. In all patients forward flow in the pulmonary artery had biphasic peaks related to both atrial and ventricular contraction. The mean (SD) peak flow velocity that was synchronous with atrial contraction was 80 (30) cm/s and that synchronous with ventricular contraction was 74 (23) cm/s. The atrial contribution to total pulmonary artery flow, assessed by velocity-time integrals, varied between 22% and 73% (mean (SD) 45 (14)%). In patients with tricuspid atresia the mean (SD) peak flow velocity with atrial contraction was 90 (27) cm/s and that with ventricular contraction was mean (SD) 68 (24) cm/s. In patients with double inlet left ventricle the mean (SD) peak flow velocity was 67 (36) cm/s with atrial contraction and 80 (25) cm/s with ventricular contraction. The atrial contribution to total pulmonary blood flow in patients with tricuspid atresia was significantly higher (53 (11)%) than in those with double inlet left ventricle (37 (14)%). Pulmonary artery flow after modified Fontan operation was biphasic and was related to both atrial and ventricular contraction. The atrial contribution to pulmonary blood flow is greater in patients with tricuspid atresia than in those with a double inlet left ventricle. The mechanism of the second peak related to ventricular contraction is unknown.     

Functional assessment of the Fontan operation: combined M-mode, two-dimensional and Doppler echocardiographic studies

Combined M-mode, two-dimensional and Doppler echocardiographic studies were used to assess the postoperative status of 33 patients who had undergone the modified Fontan procedure. Twenty-four patients had surgical repair with use of a simple direct right atrium to pulmonary artery anastomosis. The remaining patients had repair with use of a prosthesis or associated Glenn shunt. Twenty-seven patients were studied early in the postoperative period (2 months or less) and the remaining patients were studied up to 6 years postoperatively. A total of 36 examinations were performed. Of the 33 patients, 13 had tricuspid atresia, 12 had double inlet left ventricle with hypoplastic right ventricular outlet chamber and 8 had complex lesions with atrioventricular canal, double outlet right ventricle or a hypoplastic ventricle. Postoperative assessment by M-mode and two-dimensional echocardiography demonstrated normal or mildly reduced ventricular function (ejection fraction greater than 40%) in 22 patients. In 24 patients, a "normal" flow pattern was observed in the pulmonary artery by pulsed Doppler echocardiography, with predominant diastolic flow and accentuation by atrial systole somewhat similar to the venous flow pattern observed in the superior vena cava. "Abnormal" flow patterns (disorganized systolic flow, absence of atrial waves and little or no increase with inspiration) were observed in nine patients with reduced ventricular function or residual shunt. Continuous wave Doppler study also demonstrated mild dynamic subaortic obstruction in two patients. Combined pulsed and continuous wave studies showed atrioventricular valve insufficiency in 10 patients. Follow-up studies revealed a satisfactory clinical course in most patients. Three patients died approximately 4 to 8 months after their Fontan operation.     

Flow dynamics in the main pulmonary artery after the Fontan procedure in patients with tricuspid atresia or single ventricle

We analyzed the flow velocity pattern in the main pulmonary artery after Fontan operation in patients with tricuspid atresia (n = 10) or with single ventricle (n = 10) by means of a catheter-mounted velocity probe. The area underneath the velocity signal of the forward flow was integrated, and ratios of the portions during atrial systole and during the diastolic phase to the total area (Fa and Fd) were calculated. The Fa was 0.54 +/- 0.09 in patients with tricuspid atresia and 0.45 +/- 0.05 in those with single ventricle (p less than .01). Cardiac output, obtained by the thermodilution method, was 2.45 +/- 0.48 liters/min/m2 in patients with tricuspid atresia and 2.75 +/- 0.72 liters/min/m2 in those with single ventricle. The forward flow during atrial contraction, calculated by multiplying Fa by cardiac output, was 1.32 +/- 0.35 liters/min/m2 in patients with tricuspid atresia and 1.23 +/- 0.33 liters/min/m2 in those with single ventricle. The diastolic forward flow, calculated from Fd and cardiac output, was 0.99 +/- 0.25 liter/min/m2 in patients with tricuspid atresia and 1.52 +/- 0.45 liters/min/m2 in those with single ventricle (p less than 0.005). The sum of cross-sectional areas of the right and left pulmonary arteries normalized by body surface area (PA index) was 282 +/- 85 cm2/m2 in patients with tricuspid atresia and 462 +/- 65 cm2/m2 in those with single ventricle (p less than .005). The Fa was inversely correlated with the PA index in the whole group (r = -.69) and also in the tricuspid atresia group alone (r = -.87).(ABSTRACT TRUNCATED AT 250 WORDS)     

Early diastolic clicks after the Fontan procedure for double inlet left ventricle: anatomical and physiological correlates

M mode echocardiograms and simultaneous phonocardiograms were recorded in four patients with early diastolic clicks on auscultation. All had double inlet left ventricle and had undergone the Fontan procedure with closure of the right atrioventricular valve orifice by an artificial patch. The phonocardiogram confirmed a high frequency sound occurring 60-90 ms after aortic valve closure and coinciding with the time of maximal excursion of the atrioventricular valve patch towards the ventricular mass. One patient had coexisting congenital complete heart block. The M mode echocardiogram showed "reversed" motion of the patch towards the right atrium during atrial contraction. Doppler flow studies showed that coincident with this motion there was forward flow in the pulmonary artery with augmentation when atrial contraction coincided with ventricular systole. The early diastolic click in these patients was explained by abrupt cessation of the motion of the atrioventricular valve patch towards the ventricular mass in early diastole. In one patient atrial contraction led to a reversal of this motion and was associated with forward flow in the pulmonary artery.     

Early diastolic clicks after the Fontan procedure for double inlet left ventricle: anatomical and physiological correlates.

M mode echocardiograms and simultaneous phonocardiograms were recorded in four patients with early diastolic clicks on auscultation. All had double inlet left ventricle and had undergone the Fontan procedure with closure of the right atrioventricular valve orifice by an artificial patch. The phonocardiogram confirmed a high frequency sound occurring 60-90 ms after aortic valve closure and coinciding with the time of maximal excursion of the atrioventricular valve patch towards the ventricular mass. One patient had coexisting congenital complete heart block. The M mode echocardiogram showed "reversed" motion of the patch towards the right atrium during atrial contraction. Doppler flow studies showed that coincident with this motion there was forward flow in the pulmonary artery with augmentation when atrial contraction coincided with ventricular systole. The early diastolic click in these patients was explained by abrupt cessation of the motion of the atrioventricular valve patch towards the ventricular mass in early diastole. In one patient atrial contraction led to a reversal of this motion and was associated with forward flow in the pulmonary artery.     

Left ventricular function in double inlet left ventricle before the Fontan operation: comparison with tricuspid atresia

Fourteen patients with double inlet left ventricle and nine patients with tricuspid atresia had biplane left ventricular angiography with simultaneous measurement of left ventricular pressure by micromanometer. Age distribution, haemodynamic function, and previous palliative operation were similar in the two groups. Left ventricular volumes were calculated frame by frame throughout the cardiac cycle by Simpson's rule. The end diastolic volume index was similar in the two groups, but the ejection fraction was significantly lower in tricuspid atresia. Left ventricular peak filling and emptying rates were also lower in tricuspid atresia, although heart rates in the two groups were similar. End diastolic shape index was significantly higher in patients with tricuspid atresia, indicating a more globular shape, and changed less during systole, suggesting differences in the mechanism of ejection between the two groups. Analysis of pressure-volume loops showed normal phase relations between pressure and volume, but systolic stroke work was reduced in tricuspid atresia and correlated with stroke volume and shape change. Left ventricular function was impaired in patients with tricuspid atresia when compared with those with double inlet left ventricle and this finding may reflect structural differences caused by the absence of one atrioventricular connection.     

Left ventricular function in double inlet left ventricle before the Fontan operation: comparison with tricuspid atresia.

Fourteen patients with double inlet left ventricle and nine patients with tricuspid atresia had biplane left ventricular angiography with simultaneous measurement of left ventricular pressure by micromanometer. Age distribution, haemodynamic function, and previous palliative operation were similar in the two groups. Left ventricular volumes were calculated frame by frame throughout the cardiac cycle by Simpson's rule. The end diastolic volume index was similar in the two groups, but the ejection fraction was significantly lower in tricuspid atresia. Left ventricular peak filling and emptying rates were also lower in tricuspid atresia, although heart rates in the two groups were similar. End diastolic shape index was significantly higher in patients with tricuspid atresia, indicating a more globular shape, and changed less during systole, suggesting differences in the mechanism of ejection between the two groups. Analysis of pressure-volume loops showed normal phase relations between pressure and volume, but systolic stroke work was reduced in tricuspid atresia and correlated with stroke volume and shape change. Left ventricular function was impaired in patients with tricuspid atresia when compared with those with double inlet left ventricle and this finding may reflect structural differences caused by the absence of one atrioventricular connection.     

Tricuspid atresia: analysis of coronary artery distribution and ventricular morphology.

There is still disagreement concerning the precise nature of the anterior ventricular chamber in "tricuspid atresia". Some argue that it is a right ventricle, while our own previous studies have suggested it is comparable to the outlet chamber seen in classical "single ventricle". We have compared the morphology of the anterior ventricular chambers in 48 examples of tricuspid atresia (absent right atrioventricular connection), 24 hearts with double inlet to the left ventricular chamber, and 15 hearts with pulmonary atresia and intact ventricular septum. Since there is further disagreement concerning the nature of the coronary artery which delimits the posterior extent of the ventricular septum in these hearts, we analysed the position of this vessel relative to external reference points on the atrioventricular junction. No significant difference was found with respect to the morphometry of hearts with tricuspid atresia (absent right atrioventricular connection) and those with double inlet. In both groups, however, significant differences were shown between hearts with ventriculoarterial concordance and discordance. Coronary artery disposition was the same in both groups and different from that found in hearts with pulmonary atresia and intact septum. We conclude that the ventricular morphology is comparable in hearts with tricuspid atresia (absent right atrioventricular connection) and those with double inlet to a left ventricular chamber. The lesions are distinguishable by their atrioventricular connection which is nevertheless univentricular in both.     

Surgical management of subaortic obstruction in single left ventricle and tricuspid atresia

Subaortic obstruction caused by either a restrictive bulboventricular foramen in single left ventricle with an outflow chamber or by a restrictive ventricular septal defect in tricuspid atresia with transposition of the great arteries can lead to a hypertrophied, noncompliant ventricle and excessive pulmonary blood flow. This combination is disadvantageous to potential Fontan procedure candidates because they are dependent on good ventricular function and low pulmonary vascular resistance for survival. The results of surgical procedures to directly or indirectly relieve significant subaortic obstruction (gradient greater than 30 mm Hg) in 24 patients, 16 with single left ventricle and 8 with tricuspid atresia, were reviewed. Four patients had a left ventricular apex to descending aorta valved conduit; none survived. Seven patients had resection of subaortic tissue; four survived and four developed heart block at surgery. Adequate gradient relief was evident in only one of the four survivors. Thirteen patients had a main pulmonary artery to ascending aorta anastomosis or conduit; six survived. All survivors had adequate gradient relief. The overall survival was 42% (10 of 24). None of seven patients with a subaortic gradient greater than 75 mm Hg survived. These data show that: Surgical relief of established subaortic obstruction in patients with single left ventricle and tricuspid atresia carries a high mortality rate, especially if the subaortic gradient is greater than 75 mm Hg. The best procedure appears to be the pulmonary artery to ascending aorta anastomosis. A clearer understanding of the factors leading to the development of significant subaortic obstruction is necessary to prevent it or to devise improved therapeutic strategies.     

The modified Fontan operation for children less than 4 years old

Of 500 patients who had a modified Fontan operation at this institution between 1973 and 1987, 54 (33 boys and 21 girls) were less than 4 years old. This retrospective study related preoperative clinical and hemodynamic data to subsequent survival. Twenty patients less than 4 years old had tricuspid atresia, 13 had double inlet ventricle and 21 had other complex heart defects. There were 14 early deaths (less than 30 days after operation) and 6 late deaths. Multivariate analysis of survival for the entire group of 500 patients revealed the following factors to be significantly associated with poorer survival: absence of tricuspid atresia (p = 0.011), asplenia (p less than 0.001), age less than 4 years at operation (p = 0.042), atrioventricular valve dysfunction (p = 0.017), early calendar year of operation (p less than 0.001) and the presence of either one or more of the following: left ventricular ejection fraction less than 60%, mean pulmonary artery pressure greater than 15 mm Hg and pulmonary arteriolar resistance greater than 4 U.m2 (p less than 0.001). On the basis of this study of 500 patients, age less than 4 years at operation appears to be an independent risk factor for poorer survival after the modified Fontan operation.     

Effect of right ventricular anatomy on the cardiopulmonary response to exercise. Implications for the Fontan procedure

Incorporation of the right ventricle (RV) into the pulmonary circulation of patients with tricuspid atresia undergoing a Fontan procedure has been advocated. The consequences of this approach on the exercise function of these patients was studied by examining the effects of progressive and steady-state bicycle exercise tests performed by 11 patients with right atrial (RA)-RV Fontan anastomoses, seven patients with RA-pulmonary artery (PA) Fontan anastomoses, 13 patients after repair of tetralogy of Fallot, and 34 normal control patients. All patients were in New York Heart Association class I. The exercise function of the patients undergoing RA-RV and RA-PA Fontan procedures were similar. The achieved peak work loads 60% and 67% of control and peak oxygen consumptions 60% and 64% of control, respectively. Both groups also displayed excessive ventilation, elevated dead space/tidal volume ratios, and depressed cardiac output during steady-state exercise. In contrast, tetralogy of Fallot patients achieved peak work loads and oxygen consumptions 83% of control and maintained normal cardiac outputs and dead space/tidal volume ratios during steady-state exercise. These results suggest that the presence of an RV within the pulmonary circulation of the Fontan patient does not result in improved exercise function. This may be due to the development of obstructive gradients across the RA-RV conduits during exercise or to the RV's negative effect on left ventricular compliance. Moreover, in contrast with the postoperative tetralogy of Fallot patient, the hypoplastic RV of tricuspid atresia may not have sufficient myocardium to assume the active pumping function required by exercise.     

Surgical evaluation of the modified Fontan procedure.

From 1980 to 1990 152 patients underwent Fontan operation at our institution. The following patient groups were identified: 1. patients with tricuspid atresia (n = 82, 54.0%); 2. patients with single ventricle (n = 31, 20.3%); 3. patients with a wide variety of non correctable, complex cardiac malformations (n = 39, 25.7%). In 27.0% of the patients a primary Fontan operation was performed. 45.0% of the patients received a previous shunt to increase pulmonary blood flow and in 29.4% of the patients a pulmonary artery band was placed to reduce pulmonary flow. Overall mortality was not significantly different in patients with previous palliative procedures (19.4%, n = 18) as compared to 17.4% (n = 6) in patients with primary Fontan operation. Risk of death was high in the group with complex cardiac malformations (28.2%, n = 11) and in patients with single ventricle (19.4%, n = 6). Early mortality was considerably less in patients tricuspid atresia (8.5%, n = 7). Postoperatively patients with incorporation of the residual right ventricular chamber and pulmonary valve (Fontan-Bjoerk) showed a significant (p less than 0.05) lower incidence of pleural effusion as compared to patients with other modifications of the Fontan procedure. Actuarial survival rate of all patients is 83.8 +/- 3.1% (mean +/- SEM) at ten years. The modified Fontan procedures are providing an accepted surgical method for patients with otherwise non correctable cardiac malformations.     

Modified Fontan operation for complex cardiac anomalies--postoperative hemodynamics, cardiac function and clinical status

The purpose of this paper is to compare and define the postoperative hemodynamics, cardiac function and clinical status after the modified Fontan operation in patients with complex cardiac anomalies. Thirteen consecutive patients (6 with double-outlet right ventricle [DORV] [SLL : 4, SDL : 2], 5 with single ventricle [SV] [A-III : 3, B-III : 1, C-III : 1] and 2 with tricuspid atresia [TA] [Ib : 1, IIb : 1]) underwent the modified Fontan operation. These 13 patients ranging in age from 7 to 42 years of age (mean 18) were catheterized at 7 to 46 months (mean 8 months) postoperatively. Pressure tracings at rest demonstrated a dominant "a" wave in both the right atrium and the pulmonary artery suggesting a marked right atrial contraction after this operation. Left ventricular filling pressure was significantly decreased after the modified Fontan operation in DORV and SV. The ejection fraction and left ventricular end-diastolic volume index were unchanged or slightly increased after the modified Fontan operation. Cardiac index in 6 patients with DORV was significantly increased from rest to exercise by +48% (p less than 0.05) with a significant increase in the stroke volume (p less than 0.05) and with a slight increase in heart rate, but this index in 5 patients with SV and in 2 patients with TA was not significantly increased from rest to exercise. In the postoperative clinical status, 11 of 13 patients were in NYHA class I and 2 in class II at follow-up periods ranging from 4 to 75 months (mean 48 months). These results suggest that the modified Fontan operation can be of value and can provide excellent exercise tolerance for patients with complex cardiac lesions.     

Isolated ventricular inversion with double inlet left ventricle.

2 patients with viscero-atrial situs solitus, isolated ventricular inversion (IVI) and double inlet right-sided morphologic left ventricle are presented. Isolated ventricular inversion is a rare cardiac anomaly characterized by ventricular inversion, subpulmonary conus, and ventriculo-arterial concordance. Their angiocardiographic and pathologic features are presented, and the morphologic findings of the 9 patients in the literature with isolated ventricular inversion are reviewed. Of the 11 known patients with isolated ventricular inversion, levocardia was present in 10 and dextrocardia in 1; viscero-atrial situs solitus in 9 and inversus in 2; L-ventricular loop in 9 and D-loop in 2. The atrial septum was intact in 4. An intact ventricular septum was noted in only 2 patients while in 3, more than one ventricular septal defects were present, and 2 patients exhibited morphologic single ventricle. A solitary ventricular septal defect was noted in the remainder. Significant tricuspid valve abnormalities, including atresia, stenosis or hypoplasia with supravalvular fibrous ring were found in 7 patients. In 2 of these, both with significant obstruction at the tricuspid valve, both atrioventricular valves emptied into the morphologic left ventricle--thus isolated ventricular inversion with double inlet left ventricle. Pulmonary outflow tract obstruction was evident in only 3 patients. Total anomalous pulmonary venous return occurred twice and right juxtaposition of the atrial appendages once. Thus, while the patient with isolated ventricular iversion may present with clinical and hemodynamic features characteristic of classical transposition physiology the high frequency of significant associated anomalies would complicate this. Finally, the anomaly must be differentiated from the levo-transposition, isolated atrial inversion, and the anatomically corrected malpositions.     

Angiocardiographic evaluation of right ventricular size and morphology in tricuspid atresia

The potential for right ventricular growth and physiological repair in tricuspid atresia may influence the type of Fontan procedure. To evaluate preoperative right ventricular assessment, we compared the right ventricular size and morphology determined by selective right ventricular catheterization with axial left ventricular angiography. In seven consecutive patients with tricuspid atresia and ventriculo-arterial concordance, the right ventricular volume was 12.8 +/- 9.4 cc, with a predicted normal volume (based on body surface area) of 31 +/- 16 cc, 43% (range 24-78%) of normal. Right ventricular injection outlined a right ventricular area twice that visualized from an axial left ventricular injection (33.2 vs. 16.5 cm). All patients had a well developed but small trabecular portion of the right ventricle, often unopacified with left ventricular injection. Subinfundibular narrowing adjacent to the ventricular septal defect was invariably present, creating, in effect, a two-chambered right ventricle. Selective right ventriculography demonstrates the unique morphology of the right ventricle in patients with tricuspid atresia not visualized by axial left ventriculography.     

A quantitative analysis of hemodynamic effects of the right ventricle included in the circulation of the Fontan procedure

BACKGROUND. Right heart hemodynamics were analyzed with a catheter-mounted velocity meter in seven patients after direct atrioventricular anastomosis for Fontan procedure (RV group) and were compared with those obtained in eight patients after direct atriopulmonary anastomosis (RA group). METHODS AND RESULTS. In the RV group, cardiac output was 2.7 +/- 0.6 l/min/m2; mean right atrial and pulmonary artery pressures were both 13 +/- 3 mm Hg; mean pulmonary artery wedge pressure was 7 +/- 5 mm Hg; left ventricular end-diastolic volume, determined angiographically, was 129 +/- 40% of normal; and its ejection fraction was 0.50 +/- 0.09. In the RA group, data were similar to those of the RV group except that right heart pressure were lower in the RV group, which was related to the preoperative condition of the pulmonary circulation. In the RV group, the fraction of ventricular forward flow of the total forward flow in the main pulmonary artery ranged from 0.21 to 0.46 and was not correlated with cardiac output or with any other parameter. The backward flow into the inferior vena cava at ventricular systole was greater than the atrial flow in two patients in whom cardiac output was less than 2.2 l/min/m2, whereas caval backward flow at atrial contraction was greater than ventricular flow in the other five patients, of whom four had a cardiac output greater than 3.1 l/min/m2. CONCLUSIONS. We conclude that the inclusion of right ventricle in the circulation of the Fontan procedure does not necessarily improve overall hemodynamics in most patients.     

The potential value of transoesophageal evaluation of individual pulmonary venous flow after an atrial baffle procedure

Following a Mustard's procedure, transoesophageal echocardiography allowed the visualization of all 4 pulmonary veins in 7/12 patients (mean age 14.8 years) and of three veins in a further 2 patients. Both upper pulmonary veins could be visualized consistently. No patient had an isolated pulmonary venous stenosis. In 9 patients in sinus rhythm, computer analysis of Doppler tracings from the left upper pulmonary vein showed significantly lower systolic peak velocities (mean 0.39 +/- 0.10 m/s) and time velocity integrals (mean 6.9 +/- 1.66 cm) than in normal subjects (mean 0.6 +/- 0.09 m/s and 14.4 +/- 2.97 cm respectively; P less than 0.001). We postulate that this is due to compromised atrial relaxation and compliance. In contrast, patients in junctional rhythm (mean 10.7 vs. 7.7 cm in normal subjects). Flow reversal during early ventricular systole (due to tricuspid regurgitation or atrial contraction after retrograde conduction during junctional rhythm) was detected in 6/12 patients. These results confirm that the transoesophageal approach should allow the identification of isolated pulmonary venous obstruction after a Mustard procedure. In addition. detailed analysis of tracings of flow in the pulmonary veins can document the presence of compromised atrial relaxation and help to evaluate the severity of tricuspid regurgitation. It may provide a new index with which to assess impaired systemic ventricular function.     

Clinical and hemodynamic results of the Fontan operation for tricuspid atresia

Thirty patients aged 3 months to 26 years (median 9 years) with tricuspid atresia or severe tricuspid stenosis underwent a modified Fontan operation at Children's Hospital, Boston, between 1973 and 1980. Thirty-six palliative operations had been performed previously in 20 patients. Preoperative hemodynamic measurements included: mean pulmonary arterial pressure 6 to 18 mm Hg (median 9); pulmonary vascular resistance 0.3 to 2.2 Woods units/m² (median 1.5) and left ventricular end-diastolic pressure 3 to 14 mm Hg (median 7).There were five hospital deaths (17 percent), including two in the last 24 operations. The 25 survivors have been followed up for 4 to 85 months (median 23) with no late deaths. Twenty patients are in New York Heart Association class I and four in class II; one infant was not classified. Two patients have required reoperation, at, respectively, 71 and 26 months postoperatively; the former for late-developing conduit obstruction and residual bidirectional shunting; and the latter for a residual right to left shunt at atrial level. Early postoperative monitoring of hemodynamics in the intensive care unit revealed that a mean right atrial pressure of 17 mm Hg or greater, without obstruction of the anastomosis, was invariably associated with serious morbidity or mortality.Fifteen patients underwent elective cardiac catheterization 9 to 21 months (median 13) postoperatively. The mean right atrial pressure ranged between 5 and 12 mm Hg (median 8) and mean pulmonary arterial pressure between 4 and 11 mm Hg (median 7). The cardiac index (measured in 12 patients) was 1.9 to 4.0 liters/min per m² (median 3.2). The left ventricular ejection fraction measured in 11 patients did not change significantly postoperatively, remaining normal in 10. Three patients had a residual right to left shunt at atrial level while the arterial oxygen saturation in the remaining 12 patients was 93 to 97 percent. Thus the modified Fontan operation can be performed with excellent clinical and hemodynamic results in selected patients with tricuspid atresia.     

Left ventricular filling characteristics in pulmonary hypertension: a new mode of ventricular interaction.

OBJECTIVE--To examine the effects of pulmonary hypertension on left ventricular diastolic function and to relate the findings to possible mechanisms of interdependence between the right and left sides of the heart in ventricular disease. DESIGN--A retrospective and prospective analysis of echocardiographic and Doppler studies. SETTING--A tertiary referral centre for both cardiac and pulmonary disease. PATIENTS--29 patients with pulmonary hypertension (12 primary pulmonary hypertension, 10 pulmonary fibrosis, five atrial septal defect (ASD), and two scleroderma) were compared with a control group of 10 patients with an enlarged right ventricle but normal pulmonary artery pressure (six ASD, one after ASD closure, one ASD and pulmonary valvotomy, one tricuspid valve endocarditis and repair, and one pulmonary fibrosis). None had clinical or echocardiographic evidence of intrinsic left ventricular disease. MAIN OUTCOME MEASURES--M mode echocardiographic measurements were made of septal thickness, and left and right ventricular internal cavity dimensions. Doppler derived right ventricular to right atrial pressure drop, and time intervals were measured, as were isovolumic relaxation time, and Doppler left ventricular filling characteristics. RESULTS--The peak right ventricular to right atrial pressure gradient was (mean (SD)) 60 (16) mm Hg in pulmonary hypertensive patients, and 18 (5) mm Hg in controls. The time intervals P2 to the end of the tricuspid regurgitation, and P2 to the start of tricuspid flow were both prolonged in patients with pulmonary hypertension compared with controls (115 (60) and 120 (40) v 40 (15) and 45 (10) ms, p values less than 0.001). Pulmonary hypertensive patients commonly had a dominant A wave on the transmitral Doppler (23/29); however, all the controls had a dominant E wave. Isovolumic relaxation time of the left ventricle was prolonged in pulmonary hypertensive patients compared with controls, measured as both A2 to mitral valve opening (80 (25) v 50 (15) ms) and as A2 to the start of mitral flow (105 (30) v 60 (15) ms, p values less than 0.001). The delay from mitral valve opening to the start of transmitral flow was longer in patients with pulmonary hypertension (30 (15) ms) compared with controls (10 (10) ms, p less than 0.001). At the time of mitral opening there was a right ventricular to right atrial gradient of 12 (10) mm Hg in pulmonary hypertensive patients, but this was negligible in controls (0.4 (0.3) mm Hg, p less than 0.001). CONCLUSIONS--Prolonged decline of right ventricular tension, the direct result of severe pulmonary hypertension, may appear as prolonged tricuspid regurgitation. It persists until after mitral valve opening on the left side of the heart, where events during isovolumic relaxation are disorganised, and subsequent filling is impaired. These effects are likely to be mediated through the interventricular septum, and this right-left ventricular asynchrony may represent a hitherto unrecognised mode of ventricular interaction.     

Fontan procedure--indication and clinical results

Twenty-six patients with tricuspid atresia (15), univentricular heart (7), and single ventricle (4) underwent 27 Fontan or modified Fontan procedures between 1975 and 1981. The age of the patients varied between 4 and 26 years. Twenty patients had had a total of 33 palliative operations prior to correction. The original Fontan procedure was performed in 10 patients from 1975 to 1977. According to the various anatomical findings modifications of the Fontan procedure, such as direct anastomosis or implantation of a valveless conduit, were introduced in 1977. Early mortality among all the patients was 22% (6 patients died). Three deaths occurred in the initial period 1975 to 1977. Among the last 20 patients (1978 to 1981) there were 3 early deaths. Three patients with single ventricle survived, one died due to pulmonary failure. There were 2 late deaths (sepsis, sudden cardiac death). Postoperative cardiac catheterization performed in 17 patients revealed excellent results in 13 patients; the remaining 4 displayed diminished arterial oxygen saturation, three of them had Glenn palliation prior to corrective surgery. Postoperative right atrial mean pressure varied from 10 to 23 mmHg. The left ventricular parameters were within the normal range.