One of the earliest references to an “infundibulum” came from Keith’s (1909) description of “malformations of the heart” to the Royal College of Surgeons in England, in part to try to explain the development of a ventricular septal defect in the context of pulmonic stenosis (Keith 1909). Our current understanding on the development of double-outlet right ventricle still relies on theories of conotruncal development, particularly the segmentation and rotation, as well as growth and resorption of the conus, an embryologic structure that, in the normal heart, persists as the subpulmonic infundibulum (Jonas 2004a; Restivo et al. 2006).

During the fifth week of embryologic development, there is a partitioning of the truncus arteriosus and the conus arteriosus. The neural crest cells develop in a spiral pattern, ultimately leading to the development of outflow tracts and the great arteries. The pulmonary conus develops and becomes the infundibulum in the normal heart, separating the pulmonic valve from the atrioventricular valves with the muscular infundibulum. Resorption of the subaortic conus leaves the aortic valve in fibrous continuity with the atrioventricular valves. The two primary theories of the development of DORV are taken from Lev (1972) and Van Praagh et al. (1970) and differ in the process by which DORV forms (Lev 1972; Van Praagh et al. 1970).

In Lev’s theory, failure of the spiral septation of the conotruncus leads to a parallel arrangement of the great arteries (transposition of the great arteries), and proper septation leads to the proper arrangement of the great arteries. Any intermediate arterial position along this spectrum (tetralogy of Fallot, DORV) occurs due to partial spiraling (Jonas 2004a; Van Praagh et al. 1970).

Van Praagh proposed that the development of tetralogy of Fallot (TOF) develops due to underdevelopment of the subpulmonic infundibulum (Jonas 2004a; VanPraagh 1970). In this theory, development of the conus during embryogenesis brings the pulmonary artery anteriorly toward the right ventricle and also raises it above the level of, and out of fibrous continuity with, the other three cardiac valves. Underdevelopment of the subpulmonic conus would leave the pulmonic valve in a more posterior and leftward position and the aorta more anterior and rightward. This also would cause malalignment of the conal septum with the ventricular septum, leading to obstruction from an anteriorly malaligned conal septum relative to the ventricular septum. From this theory, we can derive that DORV would involve further underdevelopment of the subpulmonic infundibulum and possibly overdevelopment or underabsorption of the subaortic infundibulum, leading to both arteries arising from the right ventricle. Here you would see bilateral coni with both semilunar valves separated from the fibrous continuity of the atrioventricular valves. The far end of this spectrum would be transposition of the great arteries (TGA), where a subaortic conus exists without a subpulmonic conus, and the pulmonic valve is over the left ventricle and in fibrous continuity with the atrioventricular valves, while the aortic valve is outside of fibrous continuity and has been pushed anteriorly by growth of the subaortic conus.

No one theory completely explains the spectrum of disease, leading to controversies over criteria for diagnosis of DORV. For instance, while pathologic assessment of hearts with a diagnosis of DORV seems to support Van Praagh’s theory, only 9 of 24 (37.5 %) of hearts with DORV studied by Howell et al. (1991) had bilateral coni, a criteria commonly used to distinguish DORV from either TOF or TGA (Howell et al. (1991)). These theories are, in any case, useful for conceptualization of the development of disease and for the categorization of these lesions. Whether certain features are regarded as strict “criteria” or simply as “descriptors” imply the stringency with which the diagnosis is judged.

About 86 % of patients with double-outlet right ventricle have atrioventricular concordance (Walters et al. 2000); and between 55 and 77 % of patients with double-outlet right ventricle are able to undergo a two-ventricle repair (Bradley 2007; Kleinert 1997). Patients in which a poorly formed ventricle, prohibitive atrioventricular valve anatomy, prohibitive papillary muscle or chordal configuration, or other complex anatomy precludes single-ventricle palliation will not be covered here. Single-ventricle palliation is discussed in other sections.

There are various classification systems for DORV. Lev (1972) initially categorized double-outlet right ventricle (DORV) based on the position of the ventricular septal defect (Lev 1972a). Kirklin and Barratt-Boyes (1993), as well as other authors, have asserted that this classification does not correlate to the surgical management required for repair, which makes sense when considering the multiple associated anomalies that can impact physiology, clinical presentation, and surgical repair (Kirklin 1993). A more clinically relevant classification of DORV for anesthesiologist is that proposed by the Society of Thoracic Surgeons and the European Society for Thoracic Surgery, which classifies DORV into five different subtypes (PS = pulmonic stenosis) (Spaeth 2014; Lacour-Gayet 2008):

This classification allows for better conceptualization and understanding of physiologic and surgical goals.

Roughly 50 % of those with DORV have a subaortic VSD (Walters et al. 2000). Generally, the subaortic VSD is perimembranous, and the defect is separated a variable distance from the aortic valve and can extend to the annulus of the tricuspid valve, which puts it in closer proximity to the conduction system (Walters et al. 2000).

In one autopsy series, 77 % of those with a subaortic VSD had bilateral coni, with the remaining 23 % having a subpulmonic conus only (Walters et al. 2000). This underscores the variability of conal configuration.

Another 10 % of those with DORV have a doubly committed VSD. Generally there is either a deficient conus bilaterally with a deficient conal septum or there can be a single conus under both great arteries (Walters et al. 2000). Some have dubbed this lesion “double-outlet both ventricles,” as it may be difficult to distinguish between DORV and DOLV (double-outlet left ventricle) where a side-by-side arrangement of the great arteries overrides the VSD (Tchervenkov 2000).

In the series by Aoki et al. (1994), 48 % of those with a subaortic VSD, and 40 % of those with a doubly committed VSD, had PS (Aoki 1994). In their 20-year experience, Brown et al. (2001) also reported a 49 % and 33 % rate of PS in those with a subaortic and doubly committed VSD, respectively (Brown 2001). Most of the pulmonary outflow tract obstruction that occurs in DORV is subvalvar, highlighting its similarity to tetralogy of Fallot (TOF) (Spaeth 2014).

One of the controversies in the classification of DORV is its distinction from TOF (Spaeth 2014; Jonas 2004a; Mahle et al. 2008; Walters et al. 2000). Based on theories of the embryogenesis of DORV, the amount of aortic override and fibrous continuity between the aortic valve (AV) and the atrioventricular valves (AVV) has been used. Most authors advocate the use of the 50 % rule, where >50 % override of the aorta into the right ventricle is required for the diagnosis of DORV (Raju 2013; Bashore 2007; Walters et al. 2000; Kleinert 1997). But this may be difficult to distinguish both by echocardiography and by direct inspection by the surgeon and has little significance for management. With regard to fibrous continuity between the AV and the AVV, a significant percentage of patients with a subaortic VSD will have a subpulmonic conus only, preserving fibrous continuity between the AV and the AVV (Walters et al. 2000).

Nonetheless, DORV with a subaortic VSD and PS is part of the spectrum of DORV that poses similar management concerns to TOF for the anesthesiologist (Spaeth 2014).

Subpulmonic VSD is present in roughly one-third of DORV cases (Walters et al. 2000; Aoki et al. 1994). Generally, the VSD is unrestrictive. The Taussig–Bing anomaly, as defined by Van Praagh et al. (1968), is a subset of DORV with a subpulmonic VSD in which there are bilateral coni with the semilunar valves at the same level (fibrous continuity between the semilunar and AVV), a side-by-side arrangement of the great arteries (L-malposed), and no PS (Van Praagh et al. 1968).

In the series by Aoki et al. (1994), of those with a subpulmonary VSD, 15 % had aortic outflow tract obstruction and 52 % had aortic arch obstruction of some kind, similar to the series by Soszyn et al. (2011), where incidence was 16 % and 51 %, respectively (Soszyn 2011; Aoki 1994). Arch hypoplasia is present in up to 78 % of DORV with subpulmonic VSD. Pulmonary outflow tract obstruction is very uncommon in TGA-type DORV (Soszyn 2011; Artrip 2006; Brown 2001; Walters et al. 2000; Aoki 1994).

Another controversy in the classification of DORV is its distinction from TGA (Jonas 2004a; Walters et al. 2000). Again, the amount of aortic override and fibrous continuity between the pulmonic valve (PV) and the atrioventricular valves (AVV) has been used (Jonas 2004a). As in distinguishing between DORV and TOF, most authors advocate the use of the 50 % rule, where >50 % override of the pulmonary artery into the right ventricle is required for the diagnosis of DORV (Jonas 2004a; Walters et al. 2000; Aoki 1994). But this may be difficult to distinguish and may not impact management. Additionally, some patients with a subpulmonic VSD will have a subaortic conus only, preserving fibrous continuity between the PV and the AVV (Walters et al. 2000; Kirklin 1993).

DORV with remote (noncommitted) VSD represents 10–20 % of the spectrum of DORV (Walters et al. 2000). This is defined as a lesion where the distance between the VSD and the great arteries is at least the diameter of the aortic valve (Lacour-Gayet 2008). The VSD tends to be either muscular or an inlet VSD, can be restrictive, and may be part of an associate AVSD (Spaeth 2014; Walters et al. 2000). Frequently, there are multiple VSDs, making surgical management difficult (Spaeth 2014; Lacour-Gayet 2008; Walters et al. 2000). While some authors suggest that these are predominantly single-ventricle lesions, at least one series by Artrip et al. (2006) reports that 70 % underwent a two-ventricle repair (Artrip 2006).