Ann Thorac Surg 2000;69:1873-1879
© 2000 The Society of Thoracic Surgeons
Original articles: Cardiovascular
Evolution of the Fontan procedure in a single center
Ko Bando, MDa,
Mark W. Turrentine, MDa,
Hyung Joo Park, MDa,
Thomas G. Sharp, MDa,
Vincent Scavo, MDa,
John W. Brown, MDa
a Section of Cardiothoracic Surgery, Riley Hospital for Children and Indiana University Medical Center, Indianapolis, Indiana, USA
Address reprint requests to Dr Bando, Department of Cardiovascular Surgery, National Cardiovascular Center, 5–7-1, Fujishirodai, Suita, Osaka, 565–8565, Japan
e-mail: kobando{at}hsp.ncvc.go.jp
Presented at the Thirty-sixth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 31–Feb 2, 2000.
 |
Abstract
|
|---|
Background. Surgical approaches to single ventricle variants include staged, fenestrated, and completed Fontan operations. This study compares outcomes with these modifications of the Fontan operation at a single center.
Methods. Preoperative risk factors and operative results were analyzed by multivariate techniques in 129 patients undergoing modified Fontan operations since March 1988.
Results. Overall early and late mortality was 5.4% and 0.8%, respectively. Before 1993, completed Fontan operation using right atrial to pulmonary artery anastomosis without fenestration was performed in the majority of patients (44 of 58; 76%). During this period, 10 of 17 patients at high risk had completed Fontan with three takedowns. In 1994, the staged hemi-Fontan and modified Fontan with a lateral tunnel anastomosis and with or without small fenestration (2.5 to 4 mm) were introduced. The majority of patients at high risk during this period underwent hemi-Fontan followed by fenestrated Fontan with no takedowns. Late atrial dysrhythmias occurred in 6 patients (4.7%), generally with larger fenestrations or right atrial to pulmonary anastomoses. Three patients (2.3%) had a stroke, 2 with large (
4 mm) fenestrations. Of 38 fenestrations, 32 (84%) closed spontaneously by 1 year. No protein-losing enteropathy occurred. Most patients (118 of 121) were in New York Heart Association class I/II 4.5 years postoperatively. By multivariate analysis, only Down’s syndrome (p < 0.001) predicted early mortality, whereas both Down’s syndrome and a systemic right ventricle decreased late survival (p < 0.006).
Conclusions. Proper selection of patients for modifications of the Fontan procedure resulted in excellent early and late survival with a low incidence of atrial dysrhythmia and stroke. Midterm functional outcomes were excellent.
|
Introduction
|
|---|
Since the first successful Fontan operation for tricuspid atresia was reported in 1971, numerous modifications have been proposed for an increasing variety of complex cyanotic congenital heart defects with a functional single ventricle [1–5]. More recent surgical modifications have focused on staging the separation of systemic and pulmonary circulations and the use of intraatrial communication in an attempt to reduce the early mortality and morbidity associated with the modified Fontan operation in patients with higher risk factors [6–8].
Early survival has improved from 75% to 83% in the 1970s to more than 90% in the current era, despite application of the operation to patients with complex forms of single ventricle and to those with hemodynamic or other factors previously considered to carry higher risk [9–12]. As more patients survive the operation and the duration of follow-up increases, the continuing risk of late failure and a poor functional outcome in some long-term survivors has become clear [11, 13–15]. Atrial dysrhythmias are common, exercise capacity is reduced, and some patients develop thromboembolic events or a problematic protein-losing enteropathy.
The purpose of this study was to review a single-center experience that has evolved over a decade and to determine the patient- and procedure-related risk factors influencing early and late outcomes after the Fontan procedure.
|
Material and methods
|
|---|
Patient population
The 129 consecutive patients who underwent various modifications of the Fontan procedure at the James Whitcomb Riley Hospital for Children between March 1988 to June 1998 were identified from the databases of the Divisions of Cardiac Surgery and Pediatric Cardiology. Patients who underwent only a superior vena cava–pulmonary anastomosis were excluded. Also excluded were patients who had undergone the first stage of a Norwood procedure or cavopulmonary anastomosis at other centers.
Surgical technique
The surgical technique of the modified Fontan procedure has evolved over the last 10 years at our institution. Initially (before 1993), the Fontan procedure was completed without a prior hemi-Fontan using a right atrial appendage to pulmonary artery anastomosis and no fenestration. In 1994, the first-stage hemi-Fontan and modified Fontan with a lateral tunnel anastomosis (with or without a small fenestration of 2.5 to 4 mm) were introduced. Our current surgical technique is as follows.
The hemi-Fontan procedure
Using median sternotomy, an arterial cannula is placed in the aorta. A straight venous cannula is placed in the right atrium and a small right-angled tube is inserted high in the superior vena cava. Immediately after initiating bypass, the Blalock shunt (if present) is ligated. The azygos vein is dissected free, doubly ligated, and divided. The superior vena cava is divided between clamps 3 to 4 mm above the atriocaval junction. Using continuous polydioxyanone sutures, an end-to side anastomosis is performed between the cephalic end of the divided superior vena cava and a longitudinal arteriotomy placed in the right medial superior aspect of the pulmonary artery. After snaring the tourniquet around the inferior vena cava cannula, the right atrium is opened (parallel to the atrioventricular groove) and a small patch of PTFE is sutured approximately 1 cm below the superior margin of the cardiac end of the divided superior vena cava. The cardiac end of the superior vena cava is then anastomosed to the pulmonary artery (Fig 1).
The modified Fontan procedure with small fenestration
After repeat median sternotomy venous cannulation is accomplished by placement of the cannulas into the superior vena cava and at the junction of the inferior vena cava and the right atrium. An incision is then made high in the atrium, again parallel to the atrioventricular grove. The previously placed polytetrafluoroethylene (PTFE) patch is removed and a shoe-shaped baffle is cut from a sheet of 0.4 mm cardiovascular patch PTFE. The baffle is sewn into position with running PTFE suture, beginning at the superior margin and carrying the suture line to the nontrabeculated tissue at the entrance site of the superior vena cava, beneath the crista terminalis. The suture line is then continued around the entrance of the inferior vena cava into the atrium, leaving the coronary sinus on the pulmonary venous side of the baffle (Fig 2). If the baffle is fenestrated, a small incision is made in the medial aspect of the baffle, and the fenestration is created with an aortic punch 2.5 to 4 mm. Finally, the remainder of the baffle is sandwiched between the free edges of the atrium and included with the atriotomy closure (Fig 3).
View larger version (36K):
[in this window]
[in a new window]
|
Fig 2. Technique of lateral caval tunnel anastomosis. The size and location of the atriotomy is exaggerated for artistic demonstration of the patch placement.
|
|
Criteria for fenestration
In general, fenestration was used in patients with any combination of the following factors: elevated pulmonary artery pressures, abnormal pulmonary artery anatomy, systemic right ventricle, or significant atrioventricular valve regurgitation.
Strategy of postoperative anticoagulation
Our current regimen for postoperative anticoagulation after either a staged or nonstaged Fontan procedure includes aspirin 10 mg/kg per day and warfarin 0.5 to 1 mg/kg per day. The International Normalized Ratio is measured only during the initial admission for a Fontan procedure and is adjusted to approximately 1.5. For the last 2 years, subcutaneous low molecular weight heparin (1 mg/kg every 12 hours) was used for 2 days immediately after surgery (but not before 1996). Moreover, the majority of patients after 1994 received ACE inhibitors.
Definition of outcome events
Early mortality was defined as death during the initial hospitalization or within 30 days of operation. Late mortality was defined as death after discharge after the initial hospitalization. Atrial dysrhythmia was defined as the new onset of supraventricular tachycardia, including atrial flutter and atrial fibrillation or need for permanent pacemaker implantation, at any time after the Fontan operation.
Statistical analysis
Early outcome was treated as a binary response variable. Univariate analyses for categoric predictor variables (eg, diagnostic group) (see Table 1) were carried out by means of 
2 and Fisher’s exact tests. Cut points for continuous variables were constructed. For example preoperative pulmonary vascular resistance was classified as either low or high (< 3 WU vs 3 WU). Baffle fenestration was classified as present or absent. To assess the simultaneous effects of perioperative characteristics on early outcome, variables that were significant at the 0.1 level in a univariate analysis were included in a multivariate logistic regression model. A significance level of 0.05 was required for retention in the multivariate model.
In the analysis of late outcomes, time to failure or time to occurrence of an atrial arrhythmia were the response variables of interest. The effect of early mortality was not eliminated. Patients in whom the treatment did not fail were censored as of the time of last follow-up. For categorical predictor variables, survival estimates were obtained for each level of the variable by the Kaplan-Meier method. A significance level of 0.05 was again required for retention in the model.
|
Results
|
|---|
Patient population
Age at operation in the 129 patients ranged from 0.8 to 41.4 years (mean 5.0 ± 5.2 years, median 4.53 years), and the mean age at operation declined from 6.8 ± 6.1 years (median 6.1 years) in the early experience (1988 to 1993) to 3.9 ± 4.2 years (median 3.2 years) in the late experience (1994 to 1998). The relative proportions of the six diagnostic groups changed over the study period, as the number of cases with more complex forms of single ventricle including hypoplastic left heart syndrome, complex double outlet right ventricle, and pulmonary atresia with intact ventricular septum increased (Table 2). Of the patients, 73% had undergone at least one prior palliative procedure. The most common palliative procedures were Blalock-Taussig shunt in 72 and pulmonary arterial banding in 19 patients. A total of 64 had undergone a previous hemi-Fontan procedure and 18 patients with hypoplastic left heart syndrome had undergone a stage 1 Norwood procedure. In addition, 73 patients had undergone two or more palliative procedures, the majority before 1993. The early operative technique involved direct atriopulmonary anastomosis (n = 50) changing more recently to the total cavopulmonary anastomosis (n = 79). A fenestration was placed in the intraatrial baffle in 52 patients primarily in the last 5 years (Table 3). Initially a large fenestration (4 to 5 mm) was created; more recently, however, the size of fenestration has been reduced to 2.5 mm. In the early period a majority of patients received the single-stage Fontan operation, whereas in the last 5 years a single-stage Fontan was selectively performed primarily for patients with simple tricuspid atresia without other risk factors (Table 3).
View this table:
[in this window]
[in a new window]
|
Table 2. Trends of Preoperative Diagnosis in Early and Late Experience With Fontan Procedure at a Single Institution
|
|
Early outcome and predictors of early death
Early failure occurred in 7 patients and no patient survived after early takedown. Thus early mortality was 5% (7 of 129). Univariate analyses demonstrated that preoperative diagnosis, other than tricuspid atresia (odds ratio 4.1, 95% confidence interval 3.0 to 5.2, Down’s syndrome: odds ratio 0.93, 95% confidence limit 0.87 to 0.99 and systemic right ventricle: odds ratio 0.37, 95% confidence limit 0.28 to 0.64) was associated with early mortality. Multivariate analysis demonstrated a significant association between early death and patients with Down’s syndrome (Table 4).
Late outcome and predictors of late death
Two patients died late, making the 5- and 10-year actuarial survival rate 93.5% ± 1.0% (Fig 1). Univariate analyses identified: preoperative diagnosis other than tricuspid atresia (odds ratio 4.2, 95% confidence interval 3.5 to 5.1); Down’s syndrome (odds ratio 0.66, 95% confidence limit 0.58 to 0.74); and functional right ventricle (odds ratio 0.66, 95% confidence limit, 0.38 to 0.94) as possible predictors of poor late outcome. Multivariate analysis demonstrated that Down’s syndrome (p < 0.001) and systemic right ventricle (p = 0.004) were significant risk factors for late death (Table 4). At the most recent follow-up, 97% of patients (118 of 121) were in New York Heart Association class I or II.
Late atrial dysrhythmias occurred in 6 patients, all with a right atrial appendage to pulmonary artery anastomosis. No patients with lateral caval tunnel Fontan have experienced significant dysrhythmias. Actuarial freedom from atrial dysrhythmia at 10 years was 92.0% ± 0.4% (Fig 2). Thromboembolic events occurred in 3 patients, 2 with large fenestrations ( > 4 mm) and 1 without fenestration. All of these patients received aspirin and low-dose warfarin as previously described.
Fate of fenestration
Of the 46 patients with baffle fenestrations, 2 patients (1 patient with a 6-mm and 1 with a 2.5-mm fenestration) died (Table 5). Only 64% of fenestrations greater than 4 mm closed spontaneously within 1 year after Fontan, whereas 90% of small (2.5 mm) fenestrations closed spontaneously 1 year after surgery. Device closure was undertaken in 1 patient and 2 other patients underwent surgical closure of the fenestration.
|
Comment
|
|---|
This study represents a single center experience with modifications of the Fontan procedure that have evolved over the last 10 years. During this time, the indications for the modified Fontan operation were extended to include younger patients with more complex lesions, reducing the relative percentage of patients with the widely accepted criteria for low-risk operation (eg, low pulmonary arterial pressure, low pulmonary vascular resistance, minimal pulmonary distortion, minimal atrioventricular valve regurgitation, and age > 4 years) (Table 6). These changes in criteria were associated with the modifications in surgical technique including the use of a staged hemi-Fontan procedure with lateral tunnel anastomosis (rather than the right atrium to pulmonary artery anastomosis) with or without fenestration.
In the current study we analyzed early and late outcome separately to determine which factors increase the risk of operative failure and which factors predispose to late failure of the Fontan procedure. It was our particular interest whether the factors previously associated with a higher risk for a complete Fontan procedure have been neutralized by the introduction of the staged hemi-Fontan procedure or fenestrated Fontan operation.
Elevated pulmonary artery pressure, higher pulmonary vascular resistance, and pulmonary artery distortion were widely recognized risk factors for early failure [14, 15]. In this study, neither elevated pulmonary arterial resistance (> 3 Wood Units) higher pulmonary arterial pressure (mean, > 15 mm Hg), nor distortion of pulmonary artery was an independent predictor of early death. Improved preoperative medical management and surgical palliation reduced the development of elevated pulmonary arterial pressure, with only 7 of 106 patients demonstrating significant pulmonary hypertension. It remains unclear whether the staging of the cavopulmonary connection is actually responsible for neutralizing these risk factors, as these variables had a large number of missing values that limited the power of the analysis.
Younger age at operation was previously reported as an independent risk factor for early failure [11, 14–16]. However, the impact of age on operative outcome has been neutralized in the last 10 years of this study. In our analysis of the data, neither age less than 2 years nor age less than 4 years at the time of a Fontan procedure was associated with increased early mortality (data not shown). Although Fontan candidates of younger age have smaller anatomical structures and a more reactive pulmonary vascular bed after cardiopulmonary bypass, early Fontan has several advantages, including (1) preservation of ventricular function through relief of chronic volume overload and chronic hypoxemia, and (2) protection of the pulmonary vasculature by removing systemic to pulmonary arterial shunts [17].
(Figures 4 and 5) The only risk factor for early mortality by multiple regression analysis was Down’s syndrome. Children with Down’s syndrome are known to have more reactive pulmonary vasculature and tend to be require longer endotracheal intubation. However, the retrospective nature of this study and the fact that there were only 3 patients with Down’s syndrome would not allow us to analyze this further.
View larger version (10K):
[in this window]
[in a new window]
|
Fig 4. Actuarial survival after the modified Fontan procedure. Dotted lines indicate 95% confidence limits.
|
|
Baffle fenestration in the Fontan procedure has been shown to be beneficial in patients with increased pulmonary vascular resistance, impaired ventricular function, young age, or other risk factors, especially in the early postoperative period [7, 8]. It may improve mid-late term cardiac output and reduce the incidence of protein-losing enteropathy. Possible disadvantages of baffle fenestration include the risk of thromboembolism through an atrial septal defect placed in prosthetic intraatrial material and prolonged desaturation from a persistent right to left shunt. In our experience, thromboembolic events were observed in two patients who had a large (> 4 mm) fenestration despite the use of aspirin. Learning from these experiences, we have reduced the size of the fenestration to a 2.5-mm diameter in the last 4 years. The rate of spontaneous closure of the fenestrations was directly related to their size. In fact, 90% of 2.5 mm fenestrations closed despite the use of aspirin and low dose of warfarin. This size of fenestration never resulted in thromboembolic events but still seems to have been beneficial to the high-risk patients in their immediate postoperative recovery.
By multiple regression analysis, only Down’s syndrome and the presence of a systemic right ventricle seemed to be risk factors for late death after the modified Fontan procedure. Pulmonary artery distortion, preoperative pulmonary arterial pressure, and pulmonary vascular resistance were not associated with higher a risk for late death. No significant differences were identified between late outcomes in patients with either staged Fontan or fenestration and those with single-stage Fontan. These results may be due to the fact that the majority of the high-risk patients who had elevated pulmonary arterial pressure or resistance and who had pulmonary arterial distortion were treated with either a staged Fontan or fenestration, or with both.
Atrial dysrhythmias and poor functional outcome have been major concerns during late follow-up after a Fontan procedure. Two important studies from Boston and New York have shown that both the staged approach to a Fontan and total cavopulmonary connection are associated with sinoatrial node dysfunction early after Fontan type procedures [18, 19]. Both studies suggested that repeated dissection in the region of sinoatrial node during staged Fontan procedures resulted in an additive risk of abnormal sinoatrial node function and reported a 23% to 50% incidence of late atrial dysrhythmias in these patients. Our current results with a 92% freedom from late atrial arrhythmias at 10 years after the Fontan procedure compares favorably with these earlier reports and is consistent with their hypothesis that avoiding repeat dissection would improve late outcomes. In our series, all of the patients with late arrhythmias had right atrial to pulmonary arterial anastomosis and no patient with a hemi-Fontan has yet developed a significant late dysrhythmia. With this technique only the superior vena caval flow is committed to the pulmonary artery, as with a standard bidirectional Glenn procedure, and the subsequent intervention may be done completely within the atrium, avoiding dissection in the region of the SA node or its blood supply [18]. Another consideration is that in our technique of fenestrated Fontan, the lateral tunnel patch is placed in such a way as to avoid suturing to, or near, the crista terminalis. Additionally, the suture line is performed with multiple shallow bites rather than broad, deep advances. This may also help to reduce the incidence of late dysrhythmias after the modified Fontan procedure.
This review of a single center experience with the Fontan procedure has several limitations. First, the mean duration of follow-up (especially for the staged hemi-Fontan and subsequent completion of Fontan with lateral tunnel and baffle fenestration) is only 4.5 years. Thus, it may be too early to draw any definite conclusions regarding late outcomes, as the most statistically powerful determinant of morbidity was a longer duration of follow-up [20]. Second, because data regarding pulmonary vascular resistance and the size of the pulmonary arteries were missing in some patients because of the retrospective nature of the study as well as the low incidence of either death or late arrhythmias, the statistical power may not have been sufficient to identify all major risk factors in a sample of 129 patients.
Despite these limitations, we conclude that with proper selection and preoperative management of patients, modifications of the Fontan procedure result in excellent early and late survival with a low incidence of atrial dysrhythmia and stroke. Although midterm functional outcome is good, long-term follow-up is certainly warranted to continue the assessment of these operations.
|
References
|
|---|
-
Fontan F., Baudet E. Surgical repair of tricuspid atresia. Thorax 1971;26:240-248.[Abstract/Free Full Text]
-
Kawashima Y., Kitamura S., Matsuda H., Shimazaki Y., Nakano S., Hirose H. Total cavopulmonary shunt operation in complex cardiac anomalies. J Thorac Cardiovasc Surg 1984;87:74-81.[Abstract]
-
Puga F.J., Chiavarelli M., Hagler D.J. Modifications of the Fontan operation applicable to patients with left atrioventricular valve atresia or single atrioventricular valve. Circulation 1987;76(Suppl):III53-III60.
-
De Leval M.R., Kliner P., Gewillig M., Bull C. Total cavopulmonary connection. J Thorac Cardiovasc Surg 1988;96:682-695.[Abstract]
-
Mayer J.E., Helgason H., Jonas R.A., et al. Extending the limits for modified Fontan procedures. J Thorac Cardiovasc Surg 1986;92:1021-1028.[Abstract]
-
Lamberti J.L., Spicer R.L., Waldman J.D., et al. The bidirectional cavopulmonary shunt. J Thorac Cardiovasc Surg 1990;100:22-30.[Abstract]
-
Laks H., Haas G.S., Pearl J.M., et al. The use of an adjustable intra-atrial communication in patients under going the Fontan and definitive right heart procedure. Circulation 1988;78(Suppl):II357.
-
Bridges N.D., Lock J.E., Castaneda A.R. Baffle fenestration with subsequent transcatheter closure. Circulation 1990;82:1681-1689.[Abstract/Free Full Text]
-
Mair D.D., Hagler D.J., Julsrud P.R., Puga F.J., Schaff H.V., Danielson G.K. Early and late results of the modified Fontan procedure for double-inlet left ventricle. J Am Coll Cardiol 1991;18:1727-1732.[Abstract]
-
Sanders S.P., Wright G.B., Keane J.F., Norwood W.I., Castaneda A.R. Clinical and hemodynamic results of the Fontan operation for tricuspid atresia. Am J Cardiol 1982;49:1733-1740.[Medline]
-
Gentles T.L., Mayer J.E., Gauvreau K., et al. Fontan operation in five hundred consecutive patients. J Thorac Cardiovasc Surg 1997;114:376-391.[Abstract/Free Full Text]
-
Jacobs M.L., Norwood W.I. Fontan operation. Ann Thorac Surg 1994;58:945-952.[Abstract]
-
Gentles T.L., Gauvreau K., Mayer J.E., et al. Functional outcome after the Fontan operation. J Thorac Cardiovasc Surg 1997;114:392-403.[Abstract/Free Full Text]
-
Mayer J.E., Jr, Bridges N.D., Lock J.E., Hanley F.L., Jonas R.A., Castaneda A.R. Factors associated with marked reduction in mortality for Fontan operations in patients with single ventricle. J Thorac Cardiovasc Surg 1992;103:444-451.[Abstract]
-
Fontan F., Fernadez G., Costa F., et al. The size of the pulmonary arteries and the results of the Fontan operation. J Thorac Cardiovasc Surg 1989;98:811-819.
-
Knott-Craig C.J., Danielson G.K., Schaff H.V., Puga F.J., Weaver A.L., Driscoll D.J. The modified Fontan operation. J Thorac Cardiovasc Surg 1995;109:1237-1243.
-
Pearl J.M., Lack H.L., Drinkwater D.C., Capouya E.R., George B.L., Williams R.G. Modified Fontan procedure in patients less than 4 years of age. Circulation 1992;86(Suppl II):II100-II105.
-
Manning P.B., Mayer J.E., Wernovsky G., Fishberger S.B., Walsh E.P. Staged operation to Fontan increases the incidence of sinoatrial node dysfunction. J Thorac Cardiovasc Surg 1996;111:833-840.[Abstract/Free Full Text]
-
Kavey R.E.W., Gaum W.E., Byrum C.J., Smith F.C., Kveselis D.A. Loss of sinus rhythm after total cavopulmonary connection. Circulation 1995;92(suppl II):II304-II308.
-
Cyran S.E. Commentary. J Thorac Cardiovasc Surg 1997;114:404-405.[Free Full Text]
Related Article
-
Discussion
Ann. Thorac. Surg. 2000 69: 1879.
[Extract]
[Full Text]
[PDF]
This article has been cited by other articles:
|
|
|
|
 
C. S.D. Almond, J. E. Mayer Jr, R. R. Thiagarajan, E. D. Blume, P. J. del Nido, and D. B. McElhinney
Outcome After Fontan Failure and Takedown to an Intermediate Palliative Circulation
Ann. Thorac. Surg.,
September 1, 2007;
84(3):
880 - 887.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. Schreiber, J. Horer, M. Vogt, J. Cleuziou, Z. Prodan, and R. Lange
Nonfenestrated Extracardiac Total Cavopulmonary Connection in 132 Consecutive Patients
Ann. Thorac. Surg.,
September 1, 2007;
84(3):
894 - 899.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. E. Mitchell, R. F. Ittenbach, J. W. Gaynor, G. Wernovsky, S. Nicolson, and T. L. Spray
Intermediate outcomes after the Fontan procedure in the current era
J. Thorac. Cardiovasc. Surg.,
January 1, 2006;
131(1):
172 - 180.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. C. Jenkins, M. F. Flanagan, J. D. Sargent, C. E. Canter, R. E. Chinnock, K. J. Jenkins, R. N. Vincent, G. T. O'Connor, and A. N. A. Tosteson
A comparison of treatment strategies for hypoplastic left heart syndrome using decision analysis
J. Am. Coll. Cardiol.,
October 1, 2001;
38(4):
1181 - 1187.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
