Rapid growth of new treatment modalities in the field of Interventional Cardiology during recent years has now expanded to structural heart disease. New percutaneous devices are now available which safely treat and even cure congenital heart disorders. It is now feasible to percutaneously close atrial communications, thereby completely altering the sequelae of abnormal shunting. The safety and simplicity of these new catheter-based modalities over open heart surgery or life-long anticoagulation have resulted in wide acceptance of percutaneous atrial septal defect (ASD) and patent foramen ovale (PFO) closure. Over the past two decades, 400,000-500,000 percutaneous ASD/PFO device closures have been performed in this country and Europe. The absence of proven benefit of medical therapy and the permanence of the complications of disorders of the atrial septum (arrhythmia, hypoxemia, pulmonary injury, heart failure, and disabling stroke) have led many expert clinicians to believe that mechanical closure is a reasonable treatment option in selected high risk patients. Given new demonstrations of low long term risk of Amlatzer™ ASO and Amplatzer™ PFO devices, wider application of these treatment modalities now seems likely and a better understanding of specific anatomy, physiology, and indications for treatment is now required.

Basic to all device medical therapy of structural heart disease is a clear definition of anatomy, structure, and related pathophysiology. When clear definitions are absent, there is significant confusion and conflict which results in insufficient scientific evaluation and poor patient care.

Nowhere is the absence of clear definition more evident than in the management of adults with defects of the atrial septum, particularly PFO closure. For example, a clear or uniform definition of a PFO is lacking in the majority of published PFO studies1, in documents calling for the completion of randomized trials2,3, and in all randomized trials themselves. The most contemporary PFO-Stroke trial, REDUCE, defines a PFO as “any bubble seen at rest or with Valsalva.” Despite catheterization laboratory validation studies indicating that PFO are not present in patients with less than 30-100 bubbles4, significant PFO continue to be diagnosed with less than 10 bubbles5,6,7. With the exception of Intermountain Health Care (MED-Tech policy; 2006) there is not one single insurance payor which has adopted a contemporary scientific policy of definition of defects such as atrial septal defect, fossa ovalis defect, or patent foramen ovale.

During the past decade, new understandings of atrial septal anatomy and physiology derived from trans-esophageal echocardiography, intracardiac echocardiography, and device implantation observations have challenged previous concepts. These new understandings replace antiquated ideas of atrial septal anatomy and physiology from over 40 years ago which were derived from autopsy specimens and formalin fixed hearts. Although this information has been rapidly accruing over the past decade, its adoption in medical practice has been very limited.

Given the availability of well-established new information, Utah Regional PFO/ASD Study Group seeks to document the current scientific and evidence-based understanding of these defects and to define anatomy as it relates to percutaneous catheter treatment options.

Defects of the Atrial Septum Amenable to Catheter-Based Closure

The embryologic development of the atrial septum is very complex and a discussion of such embryology is beyond the scope of this document. Please refer to Moss and Adams’ Heart Disease in Infants, Children, and Adolescence, 7th Edition for more information. The document is specific to fossa ovalis defects: ostium secundum defects-ASDII, PFO, and fenestrations, which are amenable to catheter-based closure.

Other atrial communications such as AV cushion defects (“primum” ASD), superior sinus venosus defects, inferior sinus venosus defects, and coronary sinus defects all remain within the realm of surgical therapy and will not be addressed in this document.

Fossa Ovalis Defects (FOD)

fossa ovalis shown by pathological specimenTraditionally, secundum atrial septal defects or ASD II have been considered to be due to the re-absorption of the septum primum resulting in abscense of the septum. In reality these defects are due to variable size of the ostium secundum and its coincidence with the larger fossa ovalis. In mid gestation, an interatrial partition, as a thin mobile pocket-like structure, develops. This “flap valve” is the basic floor of the atrium and constitutes the fossa ovalis. The upper portion of the primum then breaks down and then the “ostium secundum,” which is the opening of the pocket into the left atrium, forms. This flap, which is often called the flap valve of the foramen ovale or valvulus foraminis ovalis (VFO) is the basic structure which results in defects of the fossa ovalis. In mammals, the septum secundum actually occurs much later as an in-folding rather than true septum. Its importance lies in the degree to which it guards the ostium secundum.

Blom N et al and others8-11 have evaluated autopsied hearts with documented secundum ASD. Although secundum ASD have been considered solely due to reabsorption, in adults the defects also appear to be due to deficient flap (insufficient VFO) of the fossa ovalis as this type of defect accounted for >80% of lesions in one series.8 The Blom classification of four types of fossa ovalis defects is representative of these four studies:

Type 1a

Type 1a consists of a significant absence of and reduction of the VFO resulting in a defect. However, membrane edge overlap is often present and this study confirms that the demonstration of mild flap or edge separation by transesophageal echo or intracardiac echo does not exclude a defect from being an ASD; i.e. membrane presence may occur with ASD or PFO.

type 1a fossa ovalis defect (fod)

Type 1b

These defects are due to a deficiency of the VFO, which is variable, resulting in a semicircular defect. These defects accounted for the majority of secundum defects (80%) seen in adults and represent a continuum from complete overlap resulting in a competent valve (PFO) and varying amounts of deficient flap resulting in continuous flow throughout the cardiac cycle (ASD). Cardiac catheterization using small balloon catheters and intracardiac echocardiography (ICE) mimics the technique used in autopsy specimens to define deficient VFO type fossa ovalis defects.12 Therefore, laboratory evaluation can distinguish these types of defects by a simple catheterization method.

Traditional large ASD balloon sizing techniques as well as smaller balloon techniques permit in-vivo definition of Blom 1a or 1b fossa ovalis defects using intracardiac echocardiography (ICE).12

type 1b fossa ovalis defect            

Type 1c

These defects are associated with perforations within the VFO. The perforations may be multiple and associated with aneurysms of the fossa ovalis. Such perforations of the fossa ovalis also commonly occur in conjunction with competent flap defects (PFO).

type 1c fossa ovalis defect

Type 1d

These defects (not shown) are rare posterior superior defects. These are not the same as sinus venosus ASD but clinically cannot be distinguished from siunus venosus defects and are treated by surgery.

Andrew Cook, internationally acknowledged as one of the foremost structural heart anatomists, has provided an extensive discussion of the development of the atrial septum and the phenomenon related to deficient flap valves. Within this textbook, it is recommended that we abandon the term secundum ASD since these defects involve the fossa ovalis. He states:

True atrial septal defects are then characterized by further deficiencies within the confines of the fossa ovalis. In the past, such defects have been known as “secundum defects” since they are present at the site of the secondary embryonic foramen. Although it is worth noting from the preceding discussions that they are due to deficiencies in the primary atrial septum, not the secondary septum. This is our reason for choosing to call them fossa ovalis defects. Often the deficiency is a result of the flap valve or its rims being of insufficient size to overlap one another in the superior caval or retro-aortic margins. The rims of the defect seen echocardiographically will then be the infolded wall of the oval fossa cranially and the solid partition forming the oval fossa inferiorly. Alternatively there may be perforations within the body of the oval fossa, either solitary or multiple. In this situation, the apparent margins of the defect visible by ultrasound would consist of tissue derived from the oval fossa and not to the infolded rims of the oval fossa itself. Of particular note, such folds within the oval fossa are commonly oval rather than circular, making assessment by echocardiography using multiple planes of interrogation important or else balloon sizing may be necessary.13

Clift P and Thorn S concur with the description of the “secundum atrial septal defect” as follows:

Following birth and the act of breathing, pulmonary venous return increases pressure of the left atrium pushing the septum primum against the secundum septum closing the fossa ovalis. Failure of this to occur results in failure of the septum primum or secundum to overlap which leads to a secundum defect.14

This also relates to the definition of the PFO as the distinction between primum overlap and absence of overlap depends on the redundancy of the septum.

Porter CJ and Edwards WD 7th Edition of Moss and Adams’ Heart Disease in Infants, Children, and Adolescence Including the Fetus and Young Adult (Lippencott, Williams, Wilkins 2008) provide the most contemporary definition of atrial septal defects and their relationship to patent foramen ovale.15

This is the most accepted contemporary textbook of congenital heart disease. The atrial septal defect definition is as follows:

Any opening in the atrial septum, other than a competent foramen ovale, is an atrial septal defect (ASD). Defects at the level of the fossa ovalis presumably result from deficiency, perforation, or absence of the septum primum (the valve of the fossa ovalis). Because the ostium secundum appears enlarged or unguarded, these defects are labeled as secundum type.15

Acquired Atrial Septal Defect (ASD)

Porter et al. also described the variability and alteration of defects which may change the definition of defects. Changes of atrial anatomy may alter a defect such that its definition changes. Accordingly, a PFO may become an ASD with changes which occur related to other illness or other developmental changes:

If atrial dilation occurs among individuals with a patent foramen ovale, the limbus may become so stretched that the ostium secundum (valve of the fossa ovalis) may no longer be covered by the limbic ledge. The result is a valvular incompetent patent foramen ovale that allows interatrial shunting throughout the cardiac cycle and thus constitutes an acquired ASD.15

acquired atrial septal defectValvular incompetence resulting in ASD physiology occurs for three reasons: 1) stretching of the superior limbus of the fossa ovalis with atrial dilatation leading to lack of apposition, 2) aneurysmal formation of the septum primum preventing complete closure of the interatrial communication, and 3) deficiency of the septum primum (Blom 1a or 1b).11, 15

Fifty years ago Kuzman22 reported observations of a distinct clinical entity: the acquired ASD. Subsequently, Tandon and Eduards20 evaluated 72 necropsy specimens of children dying with documented ASD. In sixty seven of the specimens, the defect was at the fossa ovalis with 7% being due to fenestrations and 93% demonstrating that the VFO (flap or primum) was of inadequate size to guard the foramen ovale. In other words, the atrial ostium secundum coincided in space with the foramen ovale resulting in a through-and-through defect. In contrast, a PFO is characterized by valvular competence which derives from proportionate VFO/foramen ovale size and the absence of flow due to the fact that the ostium secundum and fossa ovalis do not coincide in space. A PFO may become an ostium secundum defect if other conditions result in stretching or distortion of the VFO (atrial septal aneurysm, posture, volume overload, upright posture, cardiac enlargement). As this happens, the ostium secundum becomes pulled back and then coincides in time and space with the foramen ovale.

Acquired ASD cannot be distinguished from congenital primum deficiencies. Both acquired ASD and Blom 1b defects may be difficult to diagnose by echocardiography and are often confused with PFO. Since VFO incompetence defines and distinguishes PFO from ASD, shunt evaluation alone may suffice. Any demonstration of VFO incompetence would suggest ASD rather than PFO: continuous flow, significant rest flow, or bidirectional flow.

This view is supported by evaluation of 50 consecutive patients with severe rest and provoked shunt by power M-mode TCD evaluated by intra-cardiac echocardiography. Over 90% of patients demonstrated clear left-to-right shunt by Doppler color flow (Heart Brain Symposium 2011). Similarly, nearly 10% of ASD patients in the PIVOTAL Amplatzer ASO trial (basis for FDA approval) had bidirectional shunting with minor left-to-right shunt or only right-to-left shunt (3 patients).26 In a recent study of “PFO” patients with level 5 TCD shunt evaluated by intra-cardiac echocardiography and an in vivo balloon evaluation of VFO coincidence,12 fifty percent of patients were found to have a Blom 1b ASD and underwent closure using an Amplatzer™ ASO device. Of the remaining patients, sixty six percent had severe rest shunting and intra-cardiac in vivo VFO/FO evaluation found significant differences (compatible with co-incidence) compared with patients without rest shunt.25

Patent Foramen Ovale (PFO)

Considering the previous observations,8-11, 13-15 a patent foramen ovale involves a VFO (flap valve) that completely covers the septum secundum resulting in complete closure and only latent or potential shunt. The American Academy of Neurology Practice Parameter for PFO16 also describes a PFO as a failure of sealing of atrial septa resulting in “potential shunt between the right and left atria.” Gray’s Anatomy defines PFO as “uni-directional flap valves.” Considering the possibility of a PFO becoming an acquired ASD and the current definition of ASD, a patent foramen ovale can only be considered as a defect which is competent. This means that left-to-right color flow indicating bidirectional flow, or rest right-to-left shunting indicating continuous flow, are properties of fossa ovalis defects and not of a patent foramen ovale.

patent foramen ovale (PFO)

Fossa Ovalis Defect Diagnosis

These new evidence-based observations are relevant to the current techniques utilized for diagnosing septal defects. While transesophageal echo- cardiography (TEE) has been considered the gold standard for evaluating congenital heart disease due to the ability to define anatomy, it falls short in evaluating and defining shunt. The physiology of shunt appears to be more important in predicting risk and defining defect physiology.17,18 The combination of TCD and transthoracic echocardiography (TTE) is a reasonable alternative to TEE and can be used to define fossa ovalis defects based upon definition of continuous flow. Membrane separation by TEE (primum/secundum separation) has long been considered to be indicative of a PFO, may occur without shunt, and as an edge or “shelf” phenomenon associated with fossa ovalis defects. Clearly hybrid defects and deficient flap defects may have edge separation. ICE imaging with defect morphologic evaluation by catheter manipulation seems a more reliable and consistent way of defining specific anatomy.8, 12


Patent Foramen Ovale (PFO)

A PFO is a defect of the atrial septum characterized by: 1) complete overlap of the septum primum (VFO) and the septum secundum and 2) right-to-left shunting which is latent: provocable by Valsalva or respiratory strain only. A PFO may become an “acquired ASD” if atrial morphology change results in flap deficiency or shunt throughout the cardiac cycle. TCD is the best method for shunt severity risk stratification.27

Fossa Ovalis Defects (ASD)

Defects of the fossa ovalis may include reabsorption defects (ASD II, single or multiple perforations or fenestrations), primum absence, or most commonly deficient flap or VFO with or without aneurysmal change. Defect characteristics may be defined by imaging with contrast bubble studies by TTE, TEE, ICE, TCD, cardiac MRI, or any combination thereof.

The diagnosis of a fossa ovalis defect (formerly ASD II) is as follows:

a. Exclusion of AV cushion, sinus venosus, coronary sinus defects

b. Exclusion of a competent PFO

c. Demonstration of provoked right-to-left shunt plus one of the following:

i. Demonstration of color flow left-to- right shunt; i.e. bidirectional shunt

ii. Demonstration of any perforation or fenestration of the fossa ovalis

iii. ICE demonstration of deficient flap by Forbes / Spence maneuver (1b FOD)12

iv. Demonstration of significant supine or upright rest shunt

Acquired ASD

Atrial communications with permanent or continuous shunt because of alteration of septal anatomy due to:

a. Defect stretching or alteration due to other comorbidities such as pulmonary hypertension, tricuspid regurgitation, cardiomyopathy, aortic enlargement, or postural change (orthodeoxia)

b. Septal device placement with residual shunt

c. Surgical alteration of the septum

d. Interatrial aneurysm with VFO incompetence

Shunt Differentiation of PFO and ASD II (FOD) by TCD


(>90% have left-to-right color flow by ICE)


All ASD/PFO pathology defect evaluations use metallic probes (Hagen, Blom, Tandon)
The Spence maneuver uses a PCW/SG catheter in vivo to mimic pathology probe use.



Left atrial ICE imaging of bubble appearance after selective pulmonary bubble injections

1) Useful to evaluate pulmonary be for shunt for suspected pAVM
2) Useful if septum cannot be crossed in a patient with severe TCD/ECHO shunt

  1. Woods TD, Patel A. A critical review of patent foramen ovale detection using saline contrast echocardiography: When bubbles lie. J Am Soc Echocardiogr 2006; 10: 215-222.
  1. O’Gara PT et al. Percutaneous device closure of patent foramen ovale for secondary stroke prevention. A call for completion of randomized clinical trials. Circulation 2009; 119: 2743-47.
  1. Pinto Slottow TL et al. Overview of the 2007 FDA circulatory system devices panel meeting on patent foramen ovale devices. Circulation 2007; 116: 677-682.
  1. Jeserem J et al. Diagnosis of secondary source of right-to-left shunt with balloon occlusion of patent foramen ovale and power M-Mode transcranial Doppler. JACC Cardiov Interv 2009; 2: 561-7.
  1. Homma S et al. Effect of medical treatment in stroke patients with patent foramen ovale. Patent Foramen Ovale in Cryptogenic Stroke Study (PICSS). Circulation 2002; 105: 2625-2631.
  1. Meissner I et al. Patent foramen ovale: Innocent or guilty? Evidence from a prospective population-based study. JACC 2206; 47: 440-5.
  1. Rundek T et al. Patent foramen ovale and migraine. A cross sectional study from the northern Manhattan Study (NOMAS). Circulation 2008; 118: 1419-142.
  1. Blom NA, Ottencamp J, Jongeneel TH et al. Morphogenic differences of secundum atrial septal defects. Pediatric Cardiology 2005; 26: 338-343.
  1. Chan KC, Goldman MJ. Morphologic variations of fossa ovalis atrial septal defects (secundum): feasibility for transcutaneous closure with clam- shell device. Br Heart J 1993; 69: 52-55.
  1. Ferreira SM, Ho SY, Anderson RH. Morphological study of defects of the atrial septum within the fossa ovalis: Implications for transcatheter closure of right-to-left shunt. Br Heart J 1992; 67: 316-320.
  1. Patel AR, D’Alessandro LCA, Weinberg Anatomy of the atrial septum. Chapter 1: pages 3-15. Transcatheter Closure of ASDs and PFOs: a comprehensive assessment. Cardiotext 2010; Hijazi ZM Ed.
  1. Spence MS, Kahn AA, Mullen MJ. Balloon assessment of patent foramen ovale morphology and modification of tunnels using a balloon de- tunnelization technique. Catheter Cardiovasc Interv 2008; 71: 222-228.
  1. Cook AC. Anatomy of the atrial septum. Pages 3-20 in Percutaneous Device Closure of the Atrial Septum. Brecker SJD Ed., InformaHealthcare UK 2006.
  1. Clift P, Thorne S. Clinical presentation of the atrial septal defect in adults. Pages 21-40;
  1. Porter CJ, Edwards WD. Chapter 30; Atrial Septal Defects; pages 632-645. Moss and Adams’ Heart Disease in Infants, Children, and Adolescents, Including Infants and Young Seventh Edition 2008; Lippincott, Williams and Wilkins.
  1. Messe S et al. Practice parameter: Recurrent stroke with patent foramen ovale and atrial septal aneurysm. Neurology 2004; 6: 1042-50.
  1. Zito C et al. Patent foramen ovale: Comparison among diagnostic strategies in cryptogenic stroke and migraine. Echocardiography 2009; 26: 495- 503.
  1. Sastry S et al. Transcranial Doppler detection of venous-to-arterial circulation shunts: Criteria for patent foramen ovale. J Clinical Ultrasound 2009; 37: 276-86.
  1. Stranger P, Silverman NH, Foster E, Diagnostic accuracy of pediatric echocardiograms performed in adult laboratories. Am J Cardiol 1999;83:908
  1. Tandon R, Eduards JE, Atrial septal defects in infancy. Common association with other anomalies. Circulation 1974; 49:1005-10
  1. Scheuerle A. Clinical differentiation of Patent Foramen Ovale and Secundum Arial Septal Defect, A survey of pediatric cardiologists in Dallas Texas. J Registry Manag 2011; 38:4-8 
  1. Kuzman WJ, Yuskis AS. Acquired atrial septal defect. A distinct clinical entity. Circulation 1964; 29:432
  1. Spencer MP, Moehring MA, Jesurum, Gray WA et al, Power M-Mode transcranial Doppler for diagnosis of patent foramen ovale and assessing trans-catheter closure. J Neuroimaging 2004;14:342-349
  1. Schuchlenz HW, Weihs W, Beitz A, Transesophageal echocardiography for quantifying size of patent foramen ovale in patients with cryptogenic cerebrovascular events. Stroke 2002; 33:293-6
  1. Sorensen SG, Spruance SL, Smoutz R, Horn S. Transcranial Doper quantification of residual shunt after percutaneous patent foramen ovale closure: correlation of device efficacy with intra-cardiac measures. J Interven Cardiol 2012; 25:304-312
  1. Du ZD, Hijazi ZM, Kleinman CS, et al, Comparison between trans-catheter and surgical closure of ASD in children and adults, JACC 2002; 39:1836-44 
  1. Wessler BS et al, The RoPE Score and right-to-left shunt severity by transcranial Doppler in the CODICIA Study. Cebrovasc Dis 2015;40:52-8