Author + information
- Received January 13, 1998
- Revision received April 16, 1998
- Accepted April 27, 1998
- Published online August 1, 1998.
- Julio Sandoval, MD, FACCa,* (, )
- Jorge Gaspar, MD, FACCa,
- Tomás Pulido, MDa,
- Edgar Bautista, MDa,
- Maria Luisa Martínez-Guerra, MDa,
- Marco Zeballos, MDa,
- Andrés Palomar, MDa and
- Arturo Gómez, MDa
- ↵*Address for correspondence: Dr. Julio Sandoval, Cardiopulmonary Department, Instituto Nacional de Cardiologı́a Ignacio Chávez, Juan Badiano No. 1, Tlalpan 14080, Mexico D.F., Mexico
Objectives. We sought to investigate the acute hemodynamic effects of graded balloon dilation atrial septostomy (BDAS) and to define the long-term impact of this procedure on New York Heart Association functional class and survival in adult patients with primary pulmonary hypertension (PPH).
Background. Current treatment strategies for patients with severe and refractory PPH are limited by either technical difficulties and high mortality or cost.
Methods. We studied 15 patients with severe PPH. BDAS was successfully performed in all patients by crossing the interatrial septum with a Brockenbrough needle, followed by progressive dilation of the orifice with a Mansfield balloon in a hemodynamically controlled, step-by-step manner.
Results. BDAS caused an immediate significant fall in right ventricular end-diastolic pressure and in systemic arterial oxygen saturation and an increase in cardiac index. One patient died, and 14 survived the procedure and significantly improved their mean functional class (from 3.57 ± 0.6 to 2.07 ± 0.3 [mean ± SD], p < 0.001). Exercise endurance (6-min test) also improved from 107 ± 127 to 217 ± 108 m (p < 0.001). Because of spontaneous closure, BDAS was repeated in four patients. The survival rate among patients who survived the procedure was 92% at 1, 2 and 3 years, which is better than that for historical control PPH patients (73%, 59% and 52%, respectively).
Conclusions. With careful monitoring, BDAS is a safe and useful palliative treatment for selected patients with severe PPH.
Treatment strategies for primary pulmonary hypertension (PPH) aim to modify the existing pulmonary microvascular obstruction as well as to alleviate right ventricular dysfunction. Current primary forms of intervention include anticoagulant and vasodilator therapy (i.e., calcium channel blocking agents), long-term continuous infusion of prostacyclin and lung transplantation (1–5). It appears clear now that a beneficial long-term response to vasodilator therapy in PPH is restricted to only 25% to 30% of the population (1,2). For the larger group of patients with severe disease, interventions such as the long-term infusion of prostacyclin and lung transplantation have proved to be of benefit (3–5), but they are limited by technical difficulties and cost.
Blade-balloon atrial septostomy (BBAS) as a palliative therapy for refractory PPH was first reported by Rich and Lamb in 1983 (6). Subsequent studies, particularly those of Nihill et al. (7)and Kerstein et al. (8), have shown that BBAS can be successfully performed in patients with advanced disease and can bring about significant clinical and hemodynamic improvement. However, in a recent review Rich et al. (9)stressed the significant mortality associated with this procedure.
In the past few years, we have offered and performed graded balloon dilation atrial septostomy (BDAS), a variant of BBAS, in patients with advanced PPH that is refractory to vasodilator therapy. To better define the role of BDAS in the treatment of PPH, we analyzed and compared our experience with that of published reports regarding atrial septostomy.
Between November 1994 and September 1997 we performed 22 procedures of graded BDAS in 15 patients (13 women, 2 men; mean age ± SD 33 ± 9 years, range 22 to 51) with an established diagnosis of severe PPH (Table 1). All patients were markedly symptomatic because of exercise-related shortness of breath (15 patients), chest pain (7 patients) and syncope or near syncope (7 patients). Eleven of the 15 patients had clinical evidence of systemic venous congestion (i.e., peripheral edema, ascites, hepatomegaly). At diagnostic catheterization, the mean hemodynamic profile was as follows: mean pulmonary artery pressure (PAP) 66 ± 13 mm Hg, cardiac index (CI) 2.23 ± 0.82 liters/min per m2, pulmonary vascular resistance index (PVRI) 33 ± 12 U/m2and mean right atrial pressure (RAP) 10.5 ± 4 mm Hg. During this diagnostic study, all patients had and did not respond to an acute vasodilator trial with adenosine and nifedipine (10,11). In three patients the procedure was done in a semiemergency setting because the patients were in refractory right ventricular failure and were receiving intravenous inotropic support. In the remaining 12 patients, the procedure was performed electively on the basis of severe pulmonary artery hypertension with right ventricular dysfunction or recurrent syncope despite maximal medical therapy (including diuretic drugs, anticoagulant therapy, digoxin and oxygen).
Before BDAS all patients underwent the previously published extensive diagnostic evaluation for PPH applied at our institution (12,13). Particular care was taken to rule out an anatomic interatrial communication through a two-dimensional echocardiogram, which was also useful to ensure well preserved left ventricular function. Also, patients considered candidates for the procedure had to have an arterial oxygen saturation >80% at rest and a hematocrit >35% to enable them to maintain adequate systemic oxygen transport after the procedure. Except for three very ill patients, the exercise endurance of the patients was assessed through a 6-min walk protocol (14)before and after (i.e., before discharge) the procedure. All procedures were approved by our institutional committee for clinical investigation. The risks involved and the potential benefits of BDAS were explained to the patients, and their written consent was obtained.
For BDAS, left heart catheterization was performed from the right femoral artery using a 6F pigtail catheter. Right heart catheterization was performed using a balloon flow-directed catheter. A second right femoral venous puncture was made for an 8F Mullins sheath and dilator. The transseptal puncture was performed through the Mullins sheath and dilator with a Brockenbrough needle through which a circular end Inoue guide wire was passed to the left atria. Over this guide wire, a first dilation with a 4-mm semirigid dilator was done and then exchanged for successive Mansfield balloons to perform a progressive dilation of the orifice in a step-by-step manner, ranging from 8 to 16 mm. At each step (4, 8, 12, 16 mm) we carefully assessed the concomitant changes that occurred in the following variables: right ventricular end-diastolic pressure (RVEDP), left ventricular end-diastolic pressure (LVEDP) and systemic arterial oxygen saturation (Sao2%). The final size of the defect was determined by the changes produced in the last two variables; as an end point we attempted to maintain Sao2% >75% and to keep LVEDP at <18 mm Hg. Cardiac output before and after the procedure was calculated through the indirect Fick principle using assumed oxygen consumption values.
All patients were monitored in an intensive care setting for at least 48 h after the procedure. In the absence of significant bleeding, heparin was started again 6 h after BDAS, and all patients were subsequently treated with oral anticoagulants to maintain an international normalized ratio of 2.5 to 3.0. They were also advised to use long-term nocturnal oxygen therapy as part of their treatment.
All survivors were followed up clinically and noninvasively through two-dimensional or transesophageal echocardiography at our outpatient clinic at regular intervals (i.e., every 3 months) through November 1997 (2 to 36 months after the initial BDAS). Exercise tolerance was reassessed at each of these intervals and during these evaluations, particular care was taken to detect spontaneous closure of the defect on the basis of any of the following findings: 1) reappearance of symptoms or signs of right ventricular failure, 2) spontaneous “improvement” of Sao2%, and 3) evidence of closure of the interatrial defect at two-dimensional or transesophageal echocardiography, or both. Transesophageal echocardiography was performed both at rest and during mild supine exercise in most patients (Fig. 1).
Comparisons between baseline and postprocedural data were made using the Student ttest for paired data. Significance was defined as a two-tailed p value <0.05. When applicable, the Bonferroni correction method (15)was used. Descriptive variables are presented as mean value ± SD. The Kaplan-Meier method (16,17)was used to estimate the overall survival distribution for patients who survived the procedure and for a historical control group consisting of 32 patients with PPH from our own institutional registry who were not treated with vasodilator therapy (12). For the analysis of survival after BDAS we used the date on which the procedure was performed. We also compared the survival of patients undergoing BDAS with their estimated probability of survival at 1, 2 and 3 years according to the prognostic equation proposed by the National Institutes of Health Registry on PPH (18).
Fourteen patients survived BDAS, and one patient (Patient 10) died 48 h after the procedure as a result of profound and refractory hypoxemia due to massive right to left shunt across the interatrial defect. Before the procedure, this patient was in severe right ventricular failure despite maximal vasoactive support. Another patient (Patient 8) who had had significant symptomatic improvement after the procedure, died 3 months later as a result of an unrecognized tracheal stenosis. This patient had received maximal cardiorespiratory support at the time of BDAS, including 8 days of mechanical ventilation. The patient eventually recovered and was able to perform normal daily activities. Five days before last admission she experienced fever, dry cough and progressive shortness of breath. She arrived at the hospital in severe respiratory failure, with an evident inspiratory stridor and increased work of breathing. It was not possible to perform intubation, and she died a few minutes later.
Thirteen patients are alive at 18 ± 13 months (range 2 to 36) after the procedure. Table 1shows the clinical outcome of these patients. There was a significant improvement in functional capacity of the survivors, as judged by a change in mean New York Heart Association functional class: from 3.57 ± 0.6 to 2.07 ± 0.3 (n = 14, p < 0.001) at 2 weeks to 1 month after septostomy and from 2.1 ± 0.3 to 1.5 ± 0.5 (n = 10, p < 0.05) at long-term follow-up. Exercise endurance, as defined by the 6-min walk test, increased from 107 ± 127 m before the procedure to 217 ± 108 m at 2 weeks to 1 month after septostomy (n = 14, p < 0.001) and from 191 ± 101 to 284 ± 73 m (n = 10, p < 0.05) over the long term (Fig. 2).
Spontaneous closure of the defect subsequently occurred in four patients (Patients 2, 3, 4 and 12) at 10, 12, 10 and 4 months, respectively, after initial septostomy. Except for older age (41 ± 7 vs. 31 ± 8 years, p < 0.05) and a smaller final size at initial septostomy (7.5 ± 1 vs. 10.8 ± 2.7, p < 0.05), no other clinical or hemodynamic variables were different in this group. All these patients underwent repeated BDAS. Patient 4 had a second closure, and Patient 2 has had a second and third spontaneous closure of the defect (see later). Second, third and fourth septostomies in these patients have been performed without complications.
Relevant hemodynamic variables for each patient before and immediately after BDAS are shown in Table 2and summarized in Table 3. In the total group, there was a significant decrease in mean RVEDP (from 15.5 ± 7 to 11 ± 7 mm Hg, p < 0.05), mean Sao2% (from 92 ± 3% to 83 ± 8%, p < 0.05) and mean PAP (from 59 ± 11 to 52 ± 8 mm Hg, p < 0.05). These changes were accompanied by a significant increase in mean LVEDP (from 5.5 ± 3 to 8.5 ± 2.5 mm Hg, p < 0.05) and mean CI (from 2.22 ± 0.46 to 3.00 ± 0.81 liter/min per m−2, p < 0.05). A separate analysis of patients in functional class III and IV showed that changes in mean RVEDP, mean PAP and mean CI were significant only in patients in functional class IV.
Hemodynamic changes for patients who had spontaneous closure of the BDAS are shown in Table 4. Although closure of the interatrial communication was well documented, follow-up catheterization in this group showed that a progressive decrease in the baseline (before the procedure) RVEDP had occurred in each of these patients.
As shown in Figure 3, the survival rate among patients who underwent and survived BDAS was 92% at 1, 2 and 3 years, which is significantly better than that of 73%, 59% and 52%, respectively, for historical control patients with PPH without vasodilator therapy from our own registry during the same period of time. The survival rate among patients who had atrial septostomy (92%) was also better than the 1-, 2- and 3-year survival rates of 61%, 49% and 38%, respectively, predicted by the equation developed from the National Institutes of Health Primary Pulmonary Hypertension Registry data (18).
Prognosis in PPH is largely dependent on the functional status of the right ventricle. Patients who are not responsive to vasodilator therapy and who have evidence of right ventricular failure (i.e., elevated RAP and low cardiac output) usually have a poor short-term prognosis (1). For these patients, the long-term continuous infusion of prostacyclin (epoprostenol) has been associated with improved exercise tolerance, quality of life and survival (3). However, prostacyclin is not available worldwide, and its widespread use is also limited by its high cost. Thus, other alternatives such as atrial septostomy need to be further evaluated.
Rationale for atrial septostomy in PPH
There are several experimental and clinical observations suggesting that the creation of a shunt at the interatrial level might be of benefit (19–22). In a canine model of right ventricular hypertension, Austen et al. (19)demonstrated that an atrial septal defect was beneficial in terms of both exercise performance and survival. From the clinical setting, it has been shown (20)that patients with congenital heart disease, pulmonary hypertension and Eisenmenger syndrome live longer than those with PPH. Likewise, patients with PPH and a patent foramen ovale live longer than those without this defect (21,22). At present, the rationale for atrial septostomy in these patients is based on the association of deterioration in symptoms, right heart failure and death with associated obstruction to systemic flow at the pulmonary circulation level and dilation and failure of the right ventricle (7,8). The presence of an atrial septal communication in this setting would allow right to left shunting to increase preload to the left ventricle and thus an increase in systemic output, which, despite the fall in Sao2%, produces a net increase in systemic oxygen transport. Also, the atrial septal defect would allow decompression of the right ventricle, alleviating its failure.
Previous experience with BBAS in patients with PPH
In 1983 Rich and Lam (6)were the first to use blade atrial septostomy as a palliative therapy in a patient with severe and refractory PPH. Unfortunately, this patient died 1 day after the procedure as a result of pulmonary edema and refractory hypoxemia. Although the number of subsequent reports on the use of BBAS in PPH is not abundant, experience gained with the its use as a palliative therapy for severe PPH has increased significantly in the past few years, and in this regard the work of Nihill et al. (7)and that of Kerstein et al. (8)have been particularly relevant given the sample size and the long-term follow-up in these series. These two studies have established that BBAS results in significant although variable clinical and hemodynamic improvement and in a trend toward improved survival in selected patients with severe PPH who have recurrent syncope and right heart failure. Both studies have also mentioned the risk of death involved in the use of this procedure. Patients with severe right heart failure and markedly elevated pulmonary vascular resistance do not tolerate atrial septostomy because massive right to left shunting may result in insufficient pulmonary blood flow and severe and refractory hypoxemia. The issue of a high procedure-related mortality rate (25%) in extremely ill patients was recently reviewed by Rich et al. (9). Based on a review of their experience along with that of Nihill et al. (7)and Kerstein et al. (8), they recommended that BBAS should be performed only by those institutions with an established track record in the treatment of advanced PPH, where blade septostomy is performed with low morbidity. They also recommended that atrial septostomy not be performed in patients with impending death and severe right ventricular failure and that it not be attempted in patients with a large pressure gradient between the right and left atria, especially in association with a reduced Sao2% (9).
Clinical and hemodynamic effects of BDAS
Our study demonstrates that graded BDAS, a variant of BBAS previously described by Hausknecht et al. (23)and Rothman et al. (24), can also be successfully performed in patients with advanced PPH. Similar to BBAS, BDAS results in significant clinical and hemodynamic improvement. Functional class improved in most patients after the procedure, and no patient in the present study experienced further syncope or had signs and symptoms of congestive right heart failure. Moreover, exercise endurance as assessed by the 6-min walk test increased significantly after the procedure, and there was further improvement in most patients at follow-up (Fig. 2). Other studies (7,8)have also documented that, for some as yet unknown reason, some patients experience further spontaneous clinical improvement after atrial septostomy.
In addition to symptomatic improvement, which has been described by others (23–26), our study demonstrated a trend toward improved survival in patients with severe PPH who underwent and survived BDAS. Although it is true that atrial septostomy may not alter the underlying disease process of PPH because the pulmonary vascular bed is unaffected by the procedure, atrial septostomy in these patients produces hemodynamic changes that may be of benefit: 1) Atrial septostomy reduces RAP and increases CI, two of the main variables clearly associated with survival in patients with PPH (18). Furthermore, in the study of Kerstein et al. (8), further spontaneous improvement of these two variables after septostomy was clearly demonstrated, suggesting some as yet unidentified form of long-term improvement of right ventricular function. 2) Although Sao2% decreases, cardiac output and oxygen delivery improve after the procedure because of right to left shunting at the atrial level (7,8). A further contribution to oxygen delivery on a long-term basis might be provided by the increase in hemoglobin level that occurs in most of these patients (Table 1). Thus, in addition to a possible improvement in right ventricular function, improved survival after atrial septostomy might be explained by modification of mechanisms involving oxygen delivery and peripheral oxygen utilization because it occurs in patients with congenital heart disease and Eisenmenger syndrome (27). The potential and separate role of long-term nocturnal oxygen therapy, as recommended by us, on the survival of our patients cannot be assessed in the present study.
With regard to hemodynamic changes immediately after BDAS in our study, modifications in most of the variables, albeit significant, appear not to be clinically relevant given the significant improvement in functional class and survival shown after the procedure. Similar hemodynamic results have been obtained after BBAS (7,8). To explain this apparent discrepancy, it is important to stress that the major physiologic impact of the septostomy is likely to be manifest when during exercise rather than at rest (9). Experimental evidence in animal models of right ventricular hypertension supports this statement (19,28). In our study, the follow-up assessment of atrial septostomy patency through transesophageal echocardiography at rest and during mild supine exercise (Fig. 1)also appears to support this concept for humans with PPH.
Procedure-related mortality associated with BDAS
The procedure-related mortality with BDAS in our study appears to be lower than that described for BBAS (9). After 22 procedures performed so far, only one death has occurred. This death may have resulted from use of a technique that allows step by step creation of a defect whose size would be appropriate for a given patient on the basis of careful monitoring of important variables, such as LVEDP and Sao2%. However, it could be argued that we probably apply this procedure in patients who are less sick or at an earlier stage than in patients undergoing BBAS. We do not believe that this is the case because baseline (before the procedure) variables, such as RVEDP and CI, in our patients were similar to those of Nihill et al. (7)and Kerstein et al. (8).
As has been established for BBAS (7–9), the risk of death during BDAS appears to be higher for patients with advanced disease who would benefit most from the procedure (Table 3). It should be stressed that the patient who died immediately after the procedure in our study did not have a marked elevation in calculated PVRI but had severe right ventricular failure (CI 1.57 liters/min per m2, RVEDP 30 mm Hg) and therefore was unable to achieve an increase in mean PAP. In this regard, the criteria for indicating or contraindicating atrial septostomy based only on PVRI may be misleading; therefore, the complete hemodynamic profile of the patient should be considered (9).
Spontaneous closure rate of BDAS
Although BDAS in our study appears to provide similar results in terms of clinical and hemodynamic improvement and to have a lower procedure-related mortality rate than that for BBAS, the rate of spontaneous closure of the defect is significantly higher than that described for BBAS. No patient in the series of Nihill et al. (7)and only one in that of Kerstein et al. (8), presented with closure of the interatrial communication, whereas in our study this event occurred in four patients and has recurred in at least in two of them (Table 4). By producing three to six blade incisions on the atrial septum, BBAS appears to be a more definitive and long-lasting procedure than BDAS. We do not know exactly why some of our patients developed spontaneous closure, but this event may be related to patient age and more importantly to the final size of the initial septostomy. However, it should be stressed that subsequent septostomies have been performed in these patients without any complications.
Another finding can be deduced from an analysis of these repeat septostomies. By evaluating the baseline (before the procedure) RVEDP at each subsequent procedure, it would appear that there is progressive improvement of right ventricular function (i.e., a progressive decrease in RVEDP). This improvement may only reflect the fact that we were careful enough to detect the closure at an early stage. However, it may also indicate that despite the closure, the time interval in which it remained open was sufficiently long enough for right ventricular function to improve, similar to that demonstrated by Kerstein et al. (8). A different treatment strategy for BDAS might be derived from this observation. In patients with PPH with severe hemodynamic compromise, those in whom procedure-related mortality is elevated, an initial small sized BDAS can be attempted to minimize profound hemodynamic deterioration. Then, after an interval of ≥2 weeks, completion of the defect to an optimal size should then be considered.
We used a historical control group nonresponsive patients with PPH from our own registry to assess the impact of BDAS on overall mortality in the present cohort with PPH. However, the present study represents a noncontrolled clinical trial, and the potential for selection bias may be significant. This fact is even more important in assessing the procedure-related mortality for BDAS versus BBAS. Accordingly, conclusions derived from this comparison may not be entirely valid and should be judged cautiously.
The results of the present study suggests that graded BDAS in the setting of severe PPH is comparable to BBAS for improving functional status, hemodynamic variables and survival. BDAS appears to reduce the risk of procedure-related mortality but is associated with a higher rate of spontaneous closure. Given the benefits of this procedure, we believe that it represents a viable alternative for patients with severe PPH refractory to vasodilator therapy, especially in developing countries where no other options exist.
- blade-balloon atrial septostomy
- balloon dilation atrial septostomy
- cardiac index
- left ventricular end-diastolic pressure
- pulmonary artery pressure
- primary pulmonary hypertension
- pulmonary vascular resistance index
- right atrial pressure
- right ventricular end-diastolic pressure
- systemic arterial oxygen saturation
- Received January 13, 1998.
- Revision received April 16, 1998.
- Accepted April 27, 1998.
- American College of Cardiology
- Rubin L.J
- Higenbottam T.W,
- Spiegelhalter D,
- Scott J.P,
- et al.
- Pasque M.K,
- Trulock E.P,
- Cooper J.D,
- et al.
- Kerstein D,
- Levy P.S,
- Hsu D.T,
- Hordof A.J,
- Gersony W.M,
- Barst R.J
- Sandoval J,
- Pulido T,
- Suarez J,
- Lopez S,
- et al.
- Schrader B.J,
- Inbar S,
- Kaufmann L,
- Vestal R.E,
- Rich S
- Sandoval J,
- Bauerle O,
- Palomar A,
- et al.
- Sandoval J,
- Bauerle O,
- Gomez A,
- et al.
- Riley M,
- McParland J,
- Stanford C.F,
- Nicholls D.P
- Wallenstein S,
- Zucker C.L,
- Fleiss J.L
- Pocock S.J
- STATA, version 3.00. 5th ed. Santa Monica (CA): Computing Resource Center, 1992.
- Rozkovec A,
- Montanes P,
- Oakley C.M
- Thanopoulos B.D,
- Georgakopoulos D,
- Tsaousis G.S,
- Simeunovic S
- Berman W Jr.,
- Wood S.C,
- Yabek S.M,
- Dillon T,
- Fripp R.R,
- Burstein R