Review
Expert Review of Cardiovascular Therapy Downloaded from informahealthcare.com by Nyu Medical Center on 06/06/15 For personal use only.
Approaching atrial septal defects in pulmonary hypertension Expert Rev. Cardiovasc. Ther. 13(6), 693–701 (2015)
Markus Schwerzmann* and Jean-Pierre Pfammatter Center for Congenital Heart Disease, University Hospital Inselspital, Bern, Switzerland *Author for correspondence: Tel.: +41 316 327 859 Fax: +41 316 328 050 [email protected]
Atrial septal defects (ASDs) are one of the most frequent congenital cardiac malformations, accounting for about 8–10% of all congenital heart defects. The prevalence of pulmonary arterial hypertension (PAH) in adults with an ASD is 8–10%. Different clinical PAH scenarios can be encountered. At one end of the spectrum are adults with no or only mild pulmonary vascular disease and a large shunt. These are patients who can safely undergo shunt closure. In the elderly, mild residual pulmonary hypertension after shunt closure is the rule. At the other end of the spectrum are adults with severe, irreversible pulmonary vascular disease, shunt reversal and chronic cyanosis, that is, Eisenmenger syndrome. These are patients who need to be managed medically. The challenge is to properly classify ASD patients with PAH falling in between the two ends of the spectrum as the ones with advanced, but reversible pulmonary vascular disease amenable to repair, versus the ones with progressive pulmonary vascular disease not responding to shunt closure. There are concerns that adults with progressive pulmonary vascular disease have worse outcomes after shunt closure than patients not undergoing shunt closure. Due to the correlation of pulmonary vascular changes and pulmonary hemodynamics, cardiac catheterization is used in the decision-making process. It is important to consider the hemodynamic data in the context of the clinical picture, the defect anatomy and further noninvasive tests when evaluating the option of shunt closure in these patients. KEYWORDS: ASD closure . atrial septal defect . cardiac catheterization . Eisenmenger syndrome . pulmonary hypertension . selective pulmonary vasodilator therapy
Atrial septal defect & pulmonary hypertension
Pulmonary hypertension is defined as a mean pulmonary artery (PA) pressure ‡25 mmHg at rest, measured during right heart catheterization at end-expiration, irrespective of its etiology [1]. According to Ohm’s law modified to apply to fluid mechanics, pulmonary hypertension can be caused by pulmonary vascular disease or vasoconstriction, leading to increased pulmonary vascular resistance (PVR), by high pulmonary blood flow (e.g., due to the left-toright shunt at an atrial level), by an increased pulmonary capillary wedge pressure (PCWP) or due to a combination of these factors. The term pulmonary arterial hypertension (PAH) implies an increase in PVR due to the presence of pulmonary vascular disease and describes a subpopulation of patients with PH characterized by the presence of an endinformahealthcare.com
10.1586/14779072.2015.1047763
expiratory PCWP £15 mmHg and PVR >3 Wood units (WU) [1]. In atrial septal defects (ASDs) patients, this has to be distinguished from shunt-related pulmonary hypertension (e.g., a mean PA pressure ‡25 mmHg) with normal PVR but high pulmonary blood flow. We need to know PVR to properly classify pulmonary hypertension in ASD patients as occurring due to increased pulmonary blood flow or pulmonary vascular disease. In this review, we use the term PAH exclusively for patients with increased PVR. ASD in childhood
ASDs are one of the most frequent congenital cardiac malformations, accounting for about 8–10% of all congenital heart defects. Rarely does an ASD lead to symptoms in infancy or childhood, and very often the defect will close spontaneously within months after birth. This is especially true for smaller defects with a
2015 Informa UK Ltd
ISSN 1477-9072
693
Expert Review of Cardiovascular Therapy Downloaded from informahealthcare.com by Nyu Medical Center on 06/06/15 For personal use only.
Review
Schwerzmann & Pfammatter
diameter 60 years, whereas in the younger adults the defect leading to pulmonary hypertension was typically on a ventricular level. In the province of Quebec, Canada, the prevalence of PAH was 6% among >38,000 adults with a diagnosis of CHD, and 8% among the 7000 adults with an ASD [18]. A similar prevalence of PAH was also reported in adults undergoing cardiac catheterization for an isolated ASD or sinus venosus defect at the Mayo Clinic between 1953 and 1978. Of 702 patients, 40 (6%; 34 women and six men) were found to have PAH, defined as a total pulmonary resistance index >7 WU * m2. Total pulmonary resistance is calculated by dividing mean PA pressure to pulmonary blood flow [19]. It ignores left atrial pressures and thus does not accurately reflect PVR, especially in clinical scenarios with high left atrial filling pressures. In an average-sized adult and assuming normal left atrial filling pressures, a total pulmonary resistance index >7 WU * m2 is comparable with the current definition informahealthcare.com
of PAH with a PVR >3 WU. Total pulmonary resistance is a historical measure and not widely used anymore. The European Heart Survey reported a prevalence of pulmonary hypertension of 12% in adults with a closed ASD and of 33% with an open ASD [20]. In this registry, pulmonary hypertension was defined as a systolic PA pressure >40 mmHg on echocardiography. A systolic PA pressure of >40 mmHg corresponds to a mean PA pressure threshold of 25 mmHg, as used to define pulmonary hypertension [21]. However, due to the lack of catheterization data, patients cannot be divided into the ones with PAH due to increased PVR and the ones with pulmonary hypertension due to increased pulmonary blood flow or elevated left atrial filling pressures. Therefore, this study has important limitations when estimating the proportion of adults with an ASD and PAH due to progressive pulmonary vascular disease. Clinical scenarios
In adults with CHD and PAH, four clinical scenarios can be encountered (TABLE 1) [22]. At the one end of the spectrum (scenario 1) are the cyanotic adults with shunt reversal and irreversible pulmonary vascular disease, i.e., the Eisenmenger syndrome. In these patients, therapeutic options include selective pulmonary vasodilators, and supportive and preventive measures to minimize complications of chronic cyanosis. The challenge is to correctly identify progressive pulmonary vascular disease in noncyanotic patients with a large shunt, but not yet the clinical picture of an Eisenmenger syndrome. These are likely not candidates for shunt closure, as pulmonary vascular disease will progress independent of shunt closure (scenario 2). Another challenge is to correctly identify an ASD as too small to induce pulmonary vascular disease. These are patients with PAH and a coincidental defect (scenario 3). We do not have accepted definitions of what constitutes a ‘small’ defect never leading to PAH throughout life. With aging of the CHD patients, the shunt volume through an ASD can increase due 695
Expert Review of Cardiovascular Therapy Downloaded from informahealthcare.com by Nyu Medical Center on 06/06/15 For personal use only.
Review
Schwerzmann & Pfammatter
to a rise in left-sided filling pressures secondary to the development of hypertensive or ischemic heart disease. In other words, also an initially not relevant ASD (without hemodynamic importance) can lead to right ventricular volume overload or pulmonary hypertension with aging. In the elderly, an ASD with 1 cm or more of effective diameter on echocardiography can be hemodynamically important [23]. In a recent report of 96 adults, aged 60–84 years, undergoing ASD closure mainly for correction of right heart overload, the mean defect diameter on transesophageal echocardiography was 15 ± 5 mm, and the corresponding Qp:Qs was 2:0 [24]. The prognosis of PAH in CHD depends on the underlying clinical scenario. In a recent Italian study of 192 adults, 90 (47%) had Eisenmenger syndrome, 48 (25%) had shuntinduced PAH, 10 (5%) had PAH in the setting of a small defect and 44 (23%) had PAH after corrective surgery. Outcome was worst for patients with PAH after corrective surgery [25]. Ten-year Kaplan–Meier survival estimates were 89% (95% CI: 79–94%) for Eisenmenger patients, 93% (95% CI: 76–99%) for shunt-induced pulmonary hypertension, 88% (95% CI: 39–98%) for PAH with small defects (half of them had a small ASD, half of the had a small ventricular septal defect) and 65% (95% CI: 43–80%) for the patients with PAH after corrective surgery. As comparison, 10-year survival was 46% (95% CI: 38–54%) in a group of adults with idiopathic PAH treated at the same period in the center. These data suggest that closing an ASD in an adult with advanced shunt-induced pulmonary hypertension resulting in persistent or recurrent PAH after shunt closure is hazardous for the patient (scenario 4). In all other scenarios, there is a communication between the pulmonary and the systemic circulation, which allows maintaining cardiac output by acting as relief valve, at the cost of central cyanosis. Assessment of pulmonary vascular disease
In the past, histology of lung tissue obtained from biopsies during surgical repair of cardiac defects was used to estimate the severity of pulmonary vascular disease. Rabinovitch et al. showed that the morphology of pulmonary vascular changes could be used to predict regression of pulmonary hypertension late after surgery [26]. In her study, there was also a positive correlation between preoperative mean PA pressure and the severity of morphological changes. In all children undergoing cardiac repair before age 9 months, pulmonary hemodynamics became normal late after surgery, independent on the initial severity of the pulmonary vascular changes. However, in children aged >2 years at the time of surgery, only the ones with mild histological lesions could normalize PA pressure after corrective surgery, but none of the children with severe lesions. Nowadays, lung biopsy is very rarely used for deciding on the operability of children and adults with CHD and PAH. The procedure has bleeding risks, and the accurate histological interpretation of the biopsy requires special expertise, not widely available anymore. However, lung biopsy can still provide helpful information in very selected cases and is also still used for research purposes [15]. Studies with lung biopsies led to the current concept of how shunt-induced PAH 696
evolves. Pulmonary hypercirculation and other stimuli induce a cascade of histological and functional changes in the pulmonary vasculature and lead from an initially reversible to an irreversible form of obstructive pulmonary vascular disease with the Eisenmenger syndrome as final clinical destination. Due to the correlation of pulmonary vascular changes and pulmonary hemodynamics, cardiac catheterization has replaced lung biopsy in the decision making regarding the feasibility of defect closure in adults with an ASD and PAH (FIGURE 1). Some aspects deserve attention: hemodynamic evaluation in ASD patients with PAH is prone to errors [27]. If the Fick method is used to calculate pulmonary blood flow, repeated sampling of oxygen saturations in the PA and pulmonary veins is necessary to reliably estimate pulmonary blood flow. In the presence of a left-to-right shunt, PA saturations are high at rest. In this situation, a minor change in PA saturations can have a major effect on the calculated amount of pulmonary blood flow. Pulmonary blood flow measurement is the Achilles heel in quantifying PVR in ASD patients. Cardiac MRI is a valuable alternative to measure pulmonary blood flow. Combining the results of cardiac catheterization with the results of a cardiac MRI exam can be helpful if uncertainties exist about the true amount of the left-to-right shunt. Despite the lack of solid data, acute pulmonary vasodilator testing is commonly used in cases with a baseline PVR index of 6–9 WU * m2 to test the residual dilatative capacity of the pulmonary vascular bed. A decrease of 20% in PVR, a decrease of 20% in PVR:SVR ratio, a final PVR index 9 WU * m2 PVR: SVR > 0.5 Clinical findings: Bi-directional shunting Desaturation during exercise
Clinical findings: Left-to-right shunting No desaturation during exercise
Acute vasodilator testing Decrease in PVR > 20% Final PVR < 6 WU * m2 Final PVR: SVR < 0.33 Individual discussion including – Clinical findings – Anatomy
Consider shunt closure
Consider selective pulmonary vasodilators
Hemodynamic reevaluation after 3 to 6 months of therapy In case of A: consider shunt closure In case of B-C: no shunt closure
Figure 1. Suggested treatment algorithm for patients with ASD and PAH. PVR: Pulmonary vascular resistance; PVRi: Pulmonary vascular resistance index; SVR: Systemic vascular resistance; WU: Wood units.
Recommendations for ASD closure in PAH patients
A PVR cutoff that precludes defect closure in the absence of an Eisenmenger syndrome has not been established [14,15]. In the study of Steele et al. including 26 adults with PAH (defined as total pulmonary resistance >7 WU * m2) undergoing cardiac surgery, 22 had a total pulmonary resistance
Expert Review of Cardiovascular Therapy Downloaded from informahealthcare.com by Nyu Medical Center on 06/06/15 For personal use only.
Approaching atrial septal defects in pulmonary hypertension Expert Rev. Cardiovasc. Ther. 13(6), 693–701 (2015)
Markus Schwerzmann* and Jean-Pierre Pfammatter Center for Congenital Heart Disease, University Hospital Inselspital, Bern, Switzerland *Author for correspondence: Tel.: +41 316 327 859 Fax: +41 316 328 050 [email protected]
Atrial septal defects (ASDs) are one of the most frequent congenital cardiac malformations, accounting for about 8–10% of all congenital heart defects. The prevalence of pulmonary arterial hypertension (PAH) in adults with an ASD is 8–10%. Different clinical PAH scenarios can be encountered. At one end of the spectrum are adults with no or only mild pulmonary vascular disease and a large shunt. These are patients who can safely undergo shunt closure. In the elderly, mild residual pulmonary hypertension after shunt closure is the rule. At the other end of the spectrum are adults with severe, irreversible pulmonary vascular disease, shunt reversal and chronic cyanosis, that is, Eisenmenger syndrome. These are patients who need to be managed medically. The challenge is to properly classify ASD patients with PAH falling in between the two ends of the spectrum as the ones with advanced, but reversible pulmonary vascular disease amenable to repair, versus the ones with progressive pulmonary vascular disease not responding to shunt closure. There are concerns that adults with progressive pulmonary vascular disease have worse outcomes after shunt closure than patients not undergoing shunt closure. Due to the correlation of pulmonary vascular changes and pulmonary hemodynamics, cardiac catheterization is used in the decision-making process. It is important to consider the hemodynamic data in the context of the clinical picture, the defect anatomy and further noninvasive tests when evaluating the option of shunt closure in these patients. KEYWORDS: ASD closure . atrial septal defect . cardiac catheterization . Eisenmenger syndrome . pulmonary hypertension . selective pulmonary vasodilator therapy
Atrial septal defect & pulmonary hypertension
Pulmonary hypertension is defined as a mean pulmonary artery (PA) pressure ‡25 mmHg at rest, measured during right heart catheterization at end-expiration, irrespective of its etiology [1]. According to Ohm’s law modified to apply to fluid mechanics, pulmonary hypertension can be caused by pulmonary vascular disease or vasoconstriction, leading to increased pulmonary vascular resistance (PVR), by high pulmonary blood flow (e.g., due to the left-toright shunt at an atrial level), by an increased pulmonary capillary wedge pressure (PCWP) or due to a combination of these factors. The term pulmonary arterial hypertension (PAH) implies an increase in PVR due to the presence of pulmonary vascular disease and describes a subpopulation of patients with PH characterized by the presence of an endinformahealthcare.com
10.1586/14779072.2015.1047763
expiratory PCWP £15 mmHg and PVR >3 Wood units (WU) [1]. In atrial septal defects (ASDs) patients, this has to be distinguished from shunt-related pulmonary hypertension (e.g., a mean PA pressure ‡25 mmHg) with normal PVR but high pulmonary blood flow. We need to know PVR to properly classify pulmonary hypertension in ASD patients as occurring due to increased pulmonary blood flow or pulmonary vascular disease. In this review, we use the term PAH exclusively for patients with increased PVR. ASD in childhood
ASDs are one of the most frequent congenital cardiac malformations, accounting for about 8–10% of all congenital heart defects. Rarely does an ASD lead to symptoms in infancy or childhood, and very often the defect will close spontaneously within months after birth. This is especially true for smaller defects with a
2015 Informa UK Ltd
ISSN 1477-9072
693
Expert Review of Cardiovascular Therapy Downloaded from informahealthcare.com by Nyu Medical Center on 06/06/15 For personal use only.
Review
Schwerzmann & Pfammatter
diameter 60 years, whereas in the younger adults the defect leading to pulmonary hypertension was typically on a ventricular level. In the province of Quebec, Canada, the prevalence of PAH was 6% among >38,000 adults with a diagnosis of CHD, and 8% among the 7000 adults with an ASD [18]. A similar prevalence of PAH was also reported in adults undergoing cardiac catheterization for an isolated ASD or sinus venosus defect at the Mayo Clinic between 1953 and 1978. Of 702 patients, 40 (6%; 34 women and six men) were found to have PAH, defined as a total pulmonary resistance index >7 WU * m2. Total pulmonary resistance is calculated by dividing mean PA pressure to pulmonary blood flow [19]. It ignores left atrial pressures and thus does not accurately reflect PVR, especially in clinical scenarios with high left atrial filling pressures. In an average-sized adult and assuming normal left atrial filling pressures, a total pulmonary resistance index >7 WU * m2 is comparable with the current definition informahealthcare.com
of PAH with a PVR >3 WU. Total pulmonary resistance is a historical measure and not widely used anymore. The European Heart Survey reported a prevalence of pulmonary hypertension of 12% in adults with a closed ASD and of 33% with an open ASD [20]. In this registry, pulmonary hypertension was defined as a systolic PA pressure >40 mmHg on echocardiography. A systolic PA pressure of >40 mmHg corresponds to a mean PA pressure threshold of 25 mmHg, as used to define pulmonary hypertension [21]. However, due to the lack of catheterization data, patients cannot be divided into the ones with PAH due to increased PVR and the ones with pulmonary hypertension due to increased pulmonary blood flow or elevated left atrial filling pressures. Therefore, this study has important limitations when estimating the proportion of adults with an ASD and PAH due to progressive pulmonary vascular disease. Clinical scenarios
In adults with CHD and PAH, four clinical scenarios can be encountered (TABLE 1) [22]. At the one end of the spectrum (scenario 1) are the cyanotic adults with shunt reversal and irreversible pulmonary vascular disease, i.e., the Eisenmenger syndrome. In these patients, therapeutic options include selective pulmonary vasodilators, and supportive and preventive measures to minimize complications of chronic cyanosis. The challenge is to correctly identify progressive pulmonary vascular disease in noncyanotic patients with a large shunt, but not yet the clinical picture of an Eisenmenger syndrome. These are likely not candidates for shunt closure, as pulmonary vascular disease will progress independent of shunt closure (scenario 2). Another challenge is to correctly identify an ASD as too small to induce pulmonary vascular disease. These are patients with PAH and a coincidental defect (scenario 3). We do not have accepted definitions of what constitutes a ‘small’ defect never leading to PAH throughout life. With aging of the CHD patients, the shunt volume through an ASD can increase due 695
Expert Review of Cardiovascular Therapy Downloaded from informahealthcare.com by Nyu Medical Center on 06/06/15 For personal use only.
Review
Schwerzmann & Pfammatter
to a rise in left-sided filling pressures secondary to the development of hypertensive or ischemic heart disease. In other words, also an initially not relevant ASD (without hemodynamic importance) can lead to right ventricular volume overload or pulmonary hypertension with aging. In the elderly, an ASD with 1 cm or more of effective diameter on echocardiography can be hemodynamically important [23]. In a recent report of 96 adults, aged 60–84 years, undergoing ASD closure mainly for correction of right heart overload, the mean defect diameter on transesophageal echocardiography was 15 ± 5 mm, and the corresponding Qp:Qs was 2:0 [24]. The prognosis of PAH in CHD depends on the underlying clinical scenario. In a recent Italian study of 192 adults, 90 (47%) had Eisenmenger syndrome, 48 (25%) had shuntinduced PAH, 10 (5%) had PAH in the setting of a small defect and 44 (23%) had PAH after corrective surgery. Outcome was worst for patients with PAH after corrective surgery [25]. Ten-year Kaplan–Meier survival estimates were 89% (95% CI: 79–94%) for Eisenmenger patients, 93% (95% CI: 76–99%) for shunt-induced pulmonary hypertension, 88% (95% CI: 39–98%) for PAH with small defects (half of them had a small ASD, half of the had a small ventricular septal defect) and 65% (95% CI: 43–80%) for the patients with PAH after corrective surgery. As comparison, 10-year survival was 46% (95% CI: 38–54%) in a group of adults with idiopathic PAH treated at the same period in the center. These data suggest that closing an ASD in an adult with advanced shunt-induced pulmonary hypertension resulting in persistent or recurrent PAH after shunt closure is hazardous for the patient (scenario 4). In all other scenarios, there is a communication between the pulmonary and the systemic circulation, which allows maintaining cardiac output by acting as relief valve, at the cost of central cyanosis. Assessment of pulmonary vascular disease
In the past, histology of lung tissue obtained from biopsies during surgical repair of cardiac defects was used to estimate the severity of pulmonary vascular disease. Rabinovitch et al. showed that the morphology of pulmonary vascular changes could be used to predict regression of pulmonary hypertension late after surgery [26]. In her study, there was also a positive correlation between preoperative mean PA pressure and the severity of morphological changes. In all children undergoing cardiac repair before age 9 months, pulmonary hemodynamics became normal late after surgery, independent on the initial severity of the pulmonary vascular changes. However, in children aged >2 years at the time of surgery, only the ones with mild histological lesions could normalize PA pressure after corrective surgery, but none of the children with severe lesions. Nowadays, lung biopsy is very rarely used for deciding on the operability of children and adults with CHD and PAH. The procedure has bleeding risks, and the accurate histological interpretation of the biopsy requires special expertise, not widely available anymore. However, lung biopsy can still provide helpful information in very selected cases and is also still used for research purposes [15]. Studies with lung biopsies led to the current concept of how shunt-induced PAH 696
evolves. Pulmonary hypercirculation and other stimuli induce a cascade of histological and functional changes in the pulmonary vasculature and lead from an initially reversible to an irreversible form of obstructive pulmonary vascular disease with the Eisenmenger syndrome as final clinical destination. Due to the correlation of pulmonary vascular changes and pulmonary hemodynamics, cardiac catheterization has replaced lung biopsy in the decision making regarding the feasibility of defect closure in adults with an ASD and PAH (FIGURE 1). Some aspects deserve attention: hemodynamic evaluation in ASD patients with PAH is prone to errors [27]. If the Fick method is used to calculate pulmonary blood flow, repeated sampling of oxygen saturations in the PA and pulmonary veins is necessary to reliably estimate pulmonary blood flow. In the presence of a left-to-right shunt, PA saturations are high at rest. In this situation, a minor change in PA saturations can have a major effect on the calculated amount of pulmonary blood flow. Pulmonary blood flow measurement is the Achilles heel in quantifying PVR in ASD patients. Cardiac MRI is a valuable alternative to measure pulmonary blood flow. Combining the results of cardiac catheterization with the results of a cardiac MRI exam can be helpful if uncertainties exist about the true amount of the left-to-right shunt. Despite the lack of solid data, acute pulmonary vasodilator testing is commonly used in cases with a baseline PVR index of 6–9 WU * m2 to test the residual dilatative capacity of the pulmonary vascular bed. A decrease of 20% in PVR, a decrease of 20% in PVR:SVR ratio, a final PVR index 9 WU * m2 PVR: SVR > 0.5 Clinical findings: Bi-directional shunting Desaturation during exercise
Clinical findings: Left-to-right shunting No desaturation during exercise
Acute vasodilator testing Decrease in PVR > 20% Final PVR < 6 WU * m2 Final PVR: SVR < 0.33 Individual discussion including – Clinical findings – Anatomy
Consider shunt closure
Consider selective pulmonary vasodilators
Hemodynamic reevaluation after 3 to 6 months of therapy In case of A: consider shunt closure In case of B-C: no shunt closure
Figure 1. Suggested treatment algorithm for patients with ASD and PAH. PVR: Pulmonary vascular resistance; PVRi: Pulmonary vascular resistance index; SVR: Systemic vascular resistance; WU: Wood units.
Recommendations for ASD closure in PAH patients
A PVR cutoff that precludes defect closure in the absence of an Eisenmenger syndrome has not been established [14,15]. In the study of Steele et al. including 26 adults with PAH (defined as total pulmonary resistance >7 WU * m2) undergoing cardiac surgery, 22 had a total pulmonary resistance
