The segmental approach to the diagnosis of congenital heart disease is based on an understanding of the morphologic and segmental anatomy of the heart. The morphologic anatomy of the heart is summarized in Chapter 3 . The segmental anatomy of the heart is presented here. The segment-by-segment or step-by-step approach to diagnosis greatly simplifies the diagnostic problem posed even by the most complex forms of congenital heart disease.
The Cardiac Segments
The cardiac segments are the anatomic and developmental “building blocks” out of which all hearts—normal and abnormal—are made.
The three main cardiac segments are the atria, the ventricles, and the great arteries.
The two connecting cardiac segments are the atrioventricular canal or junction and the infundibulum or conus.
Hence, there are five diagnostically and surgically important cardiac segments: (1) the atria, (2) the atrioventricular canal, (3) the ventricular sinuses, (4) the infundibulum or conus, and (5) the great arteries.
Atrial Situs
Where is the morphologically right atrium (RA)? Where is the morphologically left atrium (LA)? To answer these diagnostic questions, an understanding of the two main types of visceroatrial situs is helpful ( Fig. 4.1 ):
- 1.
In visceroatrial situs solitus , the liver is predominantly right sided, the stomach and spleen are left sided, the right lung is trilobed, the left lung is bilobed, the right bronchus is eparterial, the left bronchus is hyparterial, the RA is right sided, and the LA is left sided.
- 2.
In visceroatrial situs inversus , the liver is predominantly left sided, the stomach and spleen are right sided, the left-sided lung is trilobed, the right-sided lung is bilobed, the left-sided bronchus is eparterial, the right-sided bronchus is hyparterial, the RA is left sided, and the LA is right sided.
- 3.
In “situs ambiguus” or visceral heterotaxy —often but not always associated with the asplenia or the polysplenia syndrome—the liver can be bilaterally symmetrical; the stomach may be left sided, right sided, or midline because of a common gastrointestinal mesentery (in which the stomach is not “tacked down”); the lungs may be bilaterally trilobed with bilaterally eparterial bronchi—often with the asplenia syndrome; or the lungs may be bilaterally bilobed with bilaterally hyparterial bronchi—often with the polysplenia syndrome; and the superior venae cavae may be bilateral.
In Fig. 4.1 , note the visceroatrial situs concordances , the situs (pattern of anatomic organization) of the viscera (abdominal and thoracic) and of the atria usually being the same: both solitus (usual, ordinary, customary—therefore normal), both inversus (a mirror image of situs solitus), or both ambiguus—as can occur with the heterotaxy syndromes (with asplenia, polysplenia, or occasionally with a normally formed spleen).
“Situs ambiguus” or visceral heterotaxy is not a third type of visceroatrial situs.
There are only two basic types of visceroatrial situs (see Fig. 4.1 ): situs solitus and situs inversus. “Situs ambiguus” (visceral heterotaxy) is the poorly lateralized or abnormally symmetrical pattern of visceral organization that can occur with the asplenia and polysplenia syndromes. However, on careful study, the heterotaxy syndromes appear to have either basically situs solitus of the viscera and atria or basically situs inversus of the viscera and atria. “Situs ambiguus” is now much less ambiguous than it used to be.
Visceral heterotaxy and “situs ambiguus,” when not otherwise qualified, indicate that the basic types of visceral situs and atrial situs have not been diagnosed. These are cases awaiting diagnosis of their anatomic types of visceroatrial situs.
The concepts of atrial isomerism and of atrial appendage isomerism , as mentioned in Chapter 3 , are wrong. To summarize, bilaterally right atria or bilaterally left atria have never been documented. In the heterotaxy syndromes of asplenia and polysplenia, in terms of size, shape, and position, the atrial appendages often are not mirror image; that is, atrial appendage isomerism (mirror imagery) often is not present. Also, it should be understood that partial isomerism (e.g., of the atrial appendages only) is not isomerism. For example, D-glucose and L-glucose would not be isomers if only some of their atoms were mirror images but others were not. Isomerism is like pregnancy; it is either present or absent—but not partial. Partial isomerism (e.g., of the atrial appendages only) is a contradiction in terms—an error in logic. The realization that the old concept of atrial-level isomerism is erroneous is important, because this understanding then “opens the door” to diagnosing the atrial situs. Atrial-level “isomerism” masks the fact that the atrial situs is undiagnosed.
However, it should also be emphasized that whenever one encounters two atrial appendages that both look “rightish” (broad, triangular, pyramidal), one should immediately think of the asplenia syndrome as a strong diagnostic possibility. Similarly, whenever one sees two atrial appendages that both look “leftish” (long, thin, finger-like), one should immediately think of the polysplenia syndrome as a diagnosis to consider. This bilateral right sidedness or bilateral left sidedness appearance of the atrial appendages is thus diagnostically very helpful.
But one should also understand that the patient really does not have two right atrial appendages (one on each side) nor two left atrial appendages (one on each side). Instead, the foregoing are diagnostically helpful anatomic appearances , not real anatomic facts . For example, the polysplenia syndrome can occur in visceroatrial situs inversus. The morphologically RA is left sided, but it may well have a finger-like tip to its appendage. Nonetheless, this is a left-sided RA, not a left-sided atrium that is mostly RA but with an LA tip of its appendage. If the latter situation were in fact the case, then this left-sided atrium would be a chimera: mostly RA, but with an LA tip of its appendage. Instead, the LA (including its appendage) is right sided, and the RA is left-sided, just as one would expect in visceroatrial situs inversus, despite the “leftish” appearance of both atrial appendages.
Thus, Dr. Jesse Edwards’ “bilateral right sidedness” and “bilateral left sidedness” concepts are very helpful when correctly understood as teaching mnemonics to help one to remember the various features of the asplenia and polysplenia syndromes, respectively. However, these ideas should not be “oversold” at the atrial level as anatomic facts, because they are not. Each human being has only one RA and one RA appendage, just as each human being has only one LA and one LA appendage.
In the asplenia syndrome, for example, when we are unable to diagnose the morphologic anatomic identity of the atria and hence are unable to diagnose the anatomic type of atrial situs, we make the diagnosis of situs ambiguus of the atria , which means that we do not know what the atrial situs is. The latter is an honest statement, which we think is vastly preferable to the diagnosis of “right atrial isomerism” or “right atrial appendage isomerism,” because the latter diagnoses are erroneous in terms of literal anatomic accuracy.
Anatomic accuracy is the “gold standard” that we have endeavored to employ throughout this study. Accuracy we regard as the basic principle of science. Congenital heart disease is so complex that the only way to avoid confusion is by anatomic accuracy. Accurate pathologic anatomy is the basis of diagnosis in congenital heart disease.
An accurate physiologic diagnosis is also of great importance. But in congenital heart disease, the pathologic anatomy usually determines the pathophysiology . Consequently, accurate pathologic anatomy remains the basis of diagnosis in congenital heart disease. For example, cyanosis with decreased pulmonary blood flow and venoarterial shunting can have many different anatomic causes, which must be diagnosed accurately and treated effectively. Hence, the anatomic details are of paramount practical importance.
Occasionally in the heterotaxy syndromes, the situs of the abdominal viscera and the situs of the atria can be different, that is, visceroatrial situs discordance can occur. ,
Although visceroatrial situs concordance is the rule in situs solitus and in situs inversus (see Fig. 4.1 ), visceroatrial situs discordance can occur in the heterotaxy syndromes.
Inversion is defined in anatomy as mirror imagery. In situs inversus , there is right–left reversal but without antero-posterior or superoinferior change (see Fig. 4.1 ), as in a mirror image. The mirror is the sagittal plane between the diagrams of situs solitus and situs inversus in Fig. 4.1 .
The first of the main cardiac segments is thus the viscera and the atria—not just the atria. This is why the diagnosis of the anatomic type of visceroatrial situs (see Fig. 4.1 ) usually is helpful for atrial localization. If the plain posteroanterior chest x-ray shows situs solitus (the usual arrangement) of the viscera—with liver shadow to the right and stomach bubble to the left ( Fig. 4.2 ), then very probably the atria are also in situs solitus—with RA to the right and LA to the left.
Conversely, if the liver shadow is to the left and the stomach bubble is to the right ( Fig. 4.3 ), situs inversus of the viscera strongly suggests that the atria also will be in situs inversus.
But if the liver shadow is bilaterally symmetrical ( Fig. 4.4 ), this strongly suggests “situs ambiguus,” or the heterotaxy syndrome of the viscera and of the atria.
Although the diagnosis of the anatomic type of visceroatrial situs (solitus, inversus, or “ambiguus,” see Fig. 4.1 ) is a useful first approximation concerning atrial identification, in complex cases one must go much further:
- 1.
Where is the inferior vena cava (IVC)? Is it right sided or left sided? Does it switch sides at the level of the liver? The atrium to which the IVC connects directly, in our experience, always has been the RA.
- 2.
Where is the ostium of the coronary sinus? The atrium into which the ostium of the coronary sinus opens always has proved to be the RA. (A coronary sinus septal defect—or an unroofed coronary sinus—must not be mistaken for the right atrial ostium of the coronary sinus.)
- 3.
What is the shape and size of the atrial appendage? Is the appendage large, triangular, and anterior? If so, this is typical of the RA. Is the appendage small, finger-like, and posterior? If so, this is typical of the LA.
- 4.
What is the morphology of the atrial septal surface? Is it characterized by septum secundum’s superior and inferior limbic bands? If so, this is typical of the RA. Is septum primum well seen on the atrial septal surface? If so, this is typical of the LA.
(Please see Chapter 3 for more details concerning the morphologic anatomy of the RA and the LA.)
For convenience and brevity, situs solitus of the viscera and atria may be abbreviated as S . Situs inversus of the viscera and atria may be symbolized as I . “Situs ambiguus” of the viscera and atria may be abbreviated as A .
The three main cardiac segments—the atria, the ventricles, and the great arteries—may be regarded as the elements of a set. The standard mathematical symbol meaning “the set of” is braces: {}.
The segmental situs set is written in sequential, blood-flow order: {atria, ventricles, great arteries}. The three elements of the set are separated by commas, this being conventional set notation.
The segmental situs set thus may begin as {S,-,-} or as {I,-,-} or as {A,-,- —meaning situs solitus of the viscera and atria (S), or situs inversus of the viscera and atria (I), or “situs ambiguus” of the viscera and atria (A), respectively.
The diagnostic questions concerning atrial localization (where is the morphologically RA, and where is the morphologically LA?) are answered by determining the anatomic type(s) of visceral situs and atrial situs that is (are) present (see Fig. 4.1 ). In situs solitus and in situs inversus, there virtually always is visceroatrial situs concordance: the situs of the viscera and the situs of the atria are both the same, both solitus, or both inversus. But, as noted heretofore, in the heterotaxy syndromes (asplenia, polysplenia, and heterotaxy with a normally formed spleen), there can be visceroatrial situs discordance. The situs of the viscera and the situs of the atria can be different —a fact that is relevant to diagnostic atrial localization. To the best of our present knowledge, the atrium with which the IVC connects always is the morphologically RA. (Please see Chapter 3 for the other anatomic features of the morphologically right and left atria.)
Ventricular Situs
Where is the morphologically right ventricle (RV) and where is the morphologically left ventricle (LV)? The answers to these diagnostically important questions are largely determined by establishing the type of ventricular loop that is present ( Fig. 4.5 ).
With D-loop ventricles , the RV typically is right-sided and somewhat anterior relative to the LV (see Fig. 4.5 ).
With L-loop ventricles , the RV usually is left sided relative to the LV (see Fig. 4.5 ). Whether the RV is anterior, side by side, posterior, or superior to the LV is variable.
For example, with superoinferior ventricles and crisscross atrioventricular relations, the spatial relationship between the RV and the LV cannot be described in right–left terms. For example, with superoinferior ventricles in which the ventricular septum is approximately horizontal, the RV is superior to the LV; the right-versus-left approach does not work.
This is why we proposed chirality or handedness in 1980. The D-loop RV is right handed in the following sense ( Fig. 4.6 ): The thumb of the right hand passes through the tricuspid valve, the fingers of the right hand pass through the RV outflow tract, the palm of only the right hand faces the right ventricular septal surface, and the dorsum of only the right hand faces the right ventricular free wall.
Conversely, the L-loop RV is left handed in the following sense ( Fig. 4.7 ): The thumb of your left hand passes through the left-sided tricuspid valve, the fingers of your left hand pass through the RV outflow tract, the palm of only the left hand faces the right ventricular septal surface, and the dorsum of only the left hand faces the right ventricular free wall.
Hence, the D-loop RV and the L-loop RV are stereoisomers, as are the right and left hands.
Chirality works just as well for the LV as it does for the RV.
The D-loop LV is left-handed: With the thumb through the mitral valve and the fingers in the left ventricular outflow tract, the palm of only the left hand faces the left ventricular septal surface, and the dorsum of only the left hand faces the left ventricular free wall.
The L-loop LV is right handed: With one’s thumb (literally or figuratively) through the right-sided mitral valve and one’s fingers in the left ventricular outflow tract, the palm of only the right hand faces the left ventricular septal surface, and the dorsum of only the right hand faces the left ventricular free wall.
When not otherwise specified, chirality or handedness should be assumed to be referring to the RV (see Figs. 4.6 and 4.7 ), even though chirality works just as well for the LV.
Parenthetically, chirality also applies to the atria: In situs solitus of the viscera and atria, the RA is right handed. The thumb of the right hand passes (literally or figuratively) into the superior vena cava and the fingers of the right hand pass out into the right atrial appendage. The palm of only the right hand faces the right atrial septal surface, and the dorsum of only the right hand faces the right atrial free wall.
Conversely, in situs inversus of the viscera and atria, the RA is left handed. The thumb of the left hand passes into the superior vena cava, the fingers of the left hand pass out into the right atrial appendage, the palm of only the left hand faces the right atrial septal surface, and the dorsum of only the left hand faces the right atrial free wall.
Similarly, in visceroatrial situs solitus, the LA is left handed. The thumb of the left hand passes into the pulmonary veins, the fingers of the left hand pass out into the left atrial appendage, the palm of only the left hand faces the left atrial septal surface, and the dorsum of only the left hand faces the left atrial free wall.
In visceroatrial situs inversus, the LA is right handed. The thumb of the right hand passes into the pulmonary veins; the fingers of the right hand pass out into the left atrial appendage; the palm of only the right hand faces the left atrial septal surface; and the dorsum of only the right hand faces the left atrial free wall.
To return to the ventricles, the important point is that D-loop ventricles and L-loop ventricles refer to the situs of the ventricles—that is, to the pattern of anatomic organization of the ventricles—not to their spatial relations . For example, with superoinferior ventricles , the RV is above, the LV is below, and the ventricular septum is approximately horizontal. Since the RV extends both to the right and to the left, and the LV also extends both to the right and to the left, what type of ventricular loop is present: D-loop ventricles or L-loop ventricles?
One must not confuse spatial relations (such as right–left relations) with situs (handedness). What you really want to know is the situs of the ventricles (see Figs. 4.6 and 4.7 ), no matter what their spatial relations may be.
So how do you diagnose the ventricular situs (D-loop/L-loop) with superoinferior ventricles? Just look at the RV. Is it right handed (see Fig. 4.7 )? That is the answer.
With crisscross atrioventricular relations , , the RV may start on the right side but may extend far to the left, and the LV may begin on the left but may proceed far to the right. What kind of ventricular loop is that? Again, remember that you are not concerned with spatial relations such as right–left. What you want to know is the intrinsic pattern of anatomic organization of the ventricle—the ventricular situs, no matter how bizarre their spatial relations may be. So, look at the RV. Is it right handed and therefore a D-loop (see Fig. 4.6 )? Or is it left handed and therefore an L-loop (see Fig. 4.7 )?
Consider single RV , with no identifiable vestige of LV anywhere. How can you prove where the absent LV should have been in order to establish what type of ventricular loop is present? Again, just look at the big RV. Is it a right-handed RV (see Fig. 4.6 )? If so, the ventricular situs is solitus, that is, D-loop ventricles. Is it a left-handed RV (see Fig. 4.7 )? If so, one is dealing with L-loop ventricles that are assumed to be inverted.
Although it is quick, easy, and convenient to describe D-loop ventricles and L-loop ventricles in terms of their spatial relations, as we did at the outset (D-loop ventricles = right-sided RV, L-loop ventricles = left-sided RV), one must understand that this is just a handy short cut. It usually works (in uncomplicated cases). But this convenient shortcut breaks down in complex cases, such as those aforementioned (superoinferior ventricles, crisscross AV relations, and single RV). It is the complex cases that teach one the difference between ventricular spatial relations and ventricular situs (D-loop/L-loop ventricles). It is the complex cases that teach one that ventricular spatial relations and ventricular situs are really two different variables . It is the complex cases that teach one to use ventricular chirality (see Figs. 4.6 and 4.7 ) to make the diagnosis of D-loop ventricles or L-loop ventricles—and not to be confused by ventricular spatial relations.
Why do we not just talk about situs solitus of the ventricles (instead of D-loop ventricles) and situs inversus of the ventricles (instead of L-loop ventricles)?
D-loop ventricles are, to the best of our knowledge, always solitus ventricles, that is, as in situs solitus totalis.
L-loop ventricles appear to be truly inverted ventricles when they occur as part of situs inversus totalis. For the ventricles to be truly inverted, one must postulate right–left reversal of the ventricular primordia.
Consider a straight heart tube. Label the right side x and the left side y (see Fig. 4.5 ). D-loop formation places x on the greater curvature of the loop that bulges convexly to the right, whereas y is located on the lesser curvature to the left.
If the same straight heart tube folds convexly to the left, y occupies the greater curvature bulging convexly to the left, while x occupies the lesser curvature to the right (see Fig. 4.5 ).
Now compare this D-loop and this L-loop (see Fig. 4.5 ). Labeling the sides x and y makes it clear that the L-loop is not really a mirror image of the D-loop. For this L-loop to be a true mirror image of this D-loop, one would have to postulate right–left reversal of the anlagen at (or before) the straight tube stage, prior to L-loop formation. If x and y were right–left reversed at (or before) the straight tube stage, only then would the L-loop be a true mirror image of the D-loop. Only then would this L-loop be truly inverted compared with this D-loop (see Fig. 4.5 ).
If x and y were not right–left reversed at or before the straight tube stage, then this L-loop is not a true mirror image of this D-loop, as comparison of the x and y labels makes clear.
So then it became obvious, at least in theory, that there could be two very different kinds of L-loop ventricles (see Fig. 4.5 ) :
- •
those that are truly inverted and
- •
those that are only apparently inverted, that is, solitus heart tubes that have undergone malrotation with looping to the left instead of to the right.
What do the anatomic data suggest? Mirror-image dextrocardia, as part of situs inversus totalis, appears to have truly inverted ventricles. If one photographs these ventricles, reverses the negatives, and prints them reversed, then these “uninverted” images do look like normal noninverted ventricles. Or more simply, take your slides of the ventricles in mirror-image dextrocardia, front-back reverse them, and project them reversed. Again, these “uninverted” images look like ordinary solitus D-loop ventricles. If there is a ventricular septal defect (VSD), the atrioventricular conduction system runs below the VSD—in the normal and in the inverted normal way.
If one does the same thing to photos of the ventricles in classical corrected transposition of the great arteries, the “uninverted” images of the ventricles often do not look like normal solitus D-loop ventricles. This might be because the ventricles in corrected transposition often are abnormal; so their “uninverted” images cannot look normal.
But there may well be more to it than that. For example, in classical corrected transposition, the atrioventricular bundle of His typically runs above and in front of a VSD—instead of behind and below the VSD, which is normal both in situs solitus and in situs inversus totalis. In this respect, the ventricles in corrected transposition clearly are not a mirror image of D-loop ventricles.
In the right-sided LV of classical corrected transposition, the superior papillary muscle often is antero medial (instead of antero lateral , which is normal). The inferior papillary muscle is often postero lateral (instead of postero medial , which is normal).
These observations can be explained by postulating that the ventricular sinuses are not inverted (right–left reversed) but upside down, because a solitus heart tube underwent L-loop formation (instead of D-loop formation, which is normal in the development of a solitus straight heart tube). In classical corrected transposition, if the right-sided LV were upside down, this would explain why the AV conduction system is above the VSD, why the upper papillary muscle is paraseptal (antero medial ), and why the lower papillary muscle is remote from the septum (postero lateral ).
If the ventricles in classical corrected transposition are malrotated with L-loop formation (but not really inverted), then the RV sinus is also upside down (as is the LV sinus). Perhaps this is why the left-sided RV sinus is so often malformed in classical corrected transposition. If one remembers that the development of the tricuspid valve is intimately related to the development of the RV sinus, this may help to explain the very high prevalence of left-sided tricuspid valve anomalies in classical corrected transposition.
The upside-down orientation of the RV sinus in classical corrected transposition may be masked by the downgrowth of the subaortic conal musculature from above, making it look as though the left-sided RV is right side up.
If this hypothesis is true, then the subaortic conus would be connecting with the bottom of the RV sinus rather than with the top of the RV sinus, as occurs normally.
How could this happen? Perhaps this connection of the conus with the caudal aspect of the RV sinus (instead of with the cephalic end of the RV sinus, which is normal) might be explained by remodeling in utero. The conotruncus may twist upward and migrate from a caudal to a cephalic position (similar to the caudocranial migration of the embryonic subclavian arteries) in order for the conotruncus to originate above the upside-down ventricular sinuses.
We are not sure that this hypothesis is true. Instead, the point is that the ventricles in classical corrected transposition may not be truly inverted . This may apply to all discordant L-loop ventricles, that is, to all cases with situs solitus of the viscera and atria, with discordant L-loop ventricles. The point is that in classical corrected transposition, for example, the ventricular part of the heart clearly has looped in a leftward direction: an L-loop definitely is present. But whether these L-loop ventricles are truly inverted—or only apparently inverted—remains unknown at the present time.
Consequently, in our diagnosis and designation of the ventricular part of the heart in congenital heart disease, we have deliberately chosen to use the designations D-loop ventricles and L-loop ventricles, defined anatomically in terms of chirality (see Figs. 4.6 and 4.7 ), because these terms are thought to be anatomically and embryologically accurate. We have deliberately chosen not to use the designations noninverted ventricles and inverted ventricles, because the anatomic and developmental accuracy of the latter term appears to be very uncertain. Discordant L-loop, in particular, may or may not prove to be inverted. Hence, in the interests of anatomic accuracy, we thought it wise to avoid the ventricular noninversion/inversion hypothesis because of its highly uncertain validity.
Other problems associated with ventricular noninversion/inversion were that these terms were usually employed in an atrial-situs-dependent sense. This meant that ventricular inversion meant one thing in visceroatrial situs solitus (L-loop ventricles), the opposite thing in visceroatrial situs inversus (D-loop ventricles), and nothing in visceroatrial situs ambiguus—because the atrial situs, which was the frame of reference, was itself uncertain or unknown.
Others used ventricular inversion not in a visceroatrial-situs-dependent sense. For them, ventricular inversion meant that the pattern of ventricular anatomic organization was as in situs inversus totalis.
Consequently, we realized in 1964 that the designation ventricular inversion suffered from confusion in usage: Some used this diagnosis in an atrial-situs-dependent sense, whereas others did not. We thought it would be absurd for us to tell other investigators what they should mean by ventricular inversion .
This is why we introduced the diagnostic terms atrioventricular concordance and atrioventricular discordance in 1964. For example, in physiologically corrected transposition in situs inversus, whether one wished to regard these (D-loop) ventricles as inverted for situs inversus or as noninverted in situs inversus , everyone would agree that these ventricles are discordant relative to the atria, because the RA opened into the LV (both left sided) and the LA opened into the RV (both right sided).
In order to facilitate diagnostic data analysis in large series of complex cases of congenital heart disease, one needs to have simple anatomic terms that do not change in meaning or become meaningless, depending on the type of visceroatrial situs that coexists. This is why we introduced D-loop and L-loop ventricles in 1964. The meanings of D-loop ventricles (see Fig. 4.6 ) and of L-loop ventricles (see Fig. 4.7 ) never change or become meaningless; this immutability of meaning greatly facilitates diagnostic data analysis.
The really new feature of the segmental approach to diagnosis , , was that the ventricles and the great arteries are diagnosed and designated per se rather than being expressed in terms of the atrial situs. Consequently, the ventricular and great arterial diagnoses do not change or become meaningless, depending on the atrial situs.
Why not express the ventricular situs (D-loop ventricles/L-loop ventricles) in terms of atrioventricular concordance/discordance?
First, like ventricular noninversion/inversion , atrioventricular concordance/discordance is an atrial-situs-dependent approach. Consequently, the meaning of (for example) atrioventricular discordance changes or becomes meaningless, depending on the type of visceroatrial situs that coexists. For example, in visceroatrial situs solitus, AV discordance means L-loop ventricles . In visceroatrial situs inversus, AV discordance means D-loop ventricles . In visceroatrial situs ambiguus, AV discordance is meaningless ( Fig. 4.8 ).
