Clin, gadiol. (1979) 30, 95-103

The Radiological Evaluation of Atrial Situs joHN PARTRIDGE

From the Department o f Radiology, Yorkshire Regional Thoracic Centre, Killingbeck Hospital, York Road, Leeds

Visceral situs, as evidenced by the position of the stomach and liver on the plain film, does not predict atrial situs as accurately as does bronchial situs. In particular situs ambiguus of the atria is not reliably indicated by visceral situs, and its subdivision into either right and left isomerism, where both atria show either right or left morphology respectively, cannot be made. As a result serious abnormalities of venous connections, which are common in the isomerisms, can be missed. Bronchial situs indicates atrial situs much more reliably and can be determined from penetrated chest radiographs or tomograms since the right and left main bronchi normally show different lengths. A ratio of lengths of 2 : 1 or more diagnoses bronchial situs solitus if the longer bronchus is on the left or situs inversus if the longer bronchus is on the right. A ratio of 1.5 : 1 or less indicates an isomerism. Data are presented from which the type of isomerism can be deduced and ratios between 2 and 1.5 : 1 clarified as isomerism or no isomerism.

There is no need for the general radiologist to be farrdliar with the many pathologies encountered in congenital heart disease, especially since the plain film seldom yields a precise diagnosis. Moreover, many of the plain film changes are dramatic, e.g. dextrocardia, needing little skill in observation but needing specialised knowledge for their interpre- tation. In one area the radiologist's general experience is most helpful and that is the differentiation of plethora, oligaemia or interstitial oedema from normal or from other pulmonary pathologies, particu- larly in the infant. The radiologist should not be deterred from contributing in this respect simply because he or she does not have the specialised experience necessary for the transmission of his or her observations into diagnostic probabilities.

Until recently, the determination of atrial situs was reasonably straightforward. The left atrium was presumed to be on the same side as the stomach and the stomach bubble is easily enough identified not to need a radiologist's opinion. However,in the last few years it has been realised that not only is there a significant group of patients in whom atrial situs is indeterminate (situs ambiguus), but also that the situs of the abdominal organs does not follow atrial situs dosely enough to be diagnostically useful. Several authors have found that the best indicator of atrial situs is bronchial situs (Van Meirop et al., 1970; Landing et al., 1971; Partridge et al., 1975; Stanger et al., 1977; Macartney et al.. 1977). Since bronchial situs can be determined from plain films, and in view of the unfamilarity of most cardiologists with lung morphology, this communication will describe the techniques used in the hope that they will be of 0ccasional use to most radiologists. In addition some

of the rare but interesting malformations encountered will be presented.

TYPES OF ATRIAL SITUS AND THEIR IMPLI- CATIONS (Table 1)

The most important aspect of diagnosis in com- plex congenital heart disease is to establish exactly how the veins, atria, ventricles and great vessels are interconnected. It is beyond the scope of this article to review atrio-ventricular and ventriculo-artefial connections, and the interested reader is referred to the excellent review by Brandt and Calder (1977). Abnormal connections of the veins with the atria are just as important as any other abnormal connections downstream in the heart. It is true that the only way, at present, of establishing veno-atrial connec- tions with certainty is cardiac catheterisation and angiography. However it is worthwhile to try to predict veno-atrial connections from atrial situs because abnormal connections can so easily be over- looked at catheterisation if the investigator is not aware o f their probability.

In the normal, the atria are morphologically distinct. Unfortunately they are named after their position in the normal. In this article the terms right and left atria refer to their morphology, not their position. This is a recognised convention in cardio- vascular pathology.

When in situs solitus, the right atrium is on the right and the left atrium on the left. In situs inversus, the right atrium is on the left and the left atrium is on the right, In these two situations it is extremely rare for the systemic veins to drain anywhere else than the right atrium, i.e. the right-sided atrium in solitus and

96 CLINICAL RADIOLOGY

Table 1 - The major implications of the four types of atrial situs

Atrial situs

Solitus Inversus Right isomerism Left isomerism

Bronchial situs Usually solitus Usually inversus Usually right isomerism Abdominal situs Usually solitus Usually inversus Often asplenia Superior vena Drain to the Drain to the Often drain to both atria

cava(ae) right atrium right atrium Inferior vena cava Drains to the Drains to the ' May drain to either atrium

right atrium right atrium Ptdmonary veins Usually drain to the Usually drain to the Total anomalous pulmonary

left atrium left atrium venous drainage frequent Associated Normal spectrum of Wide spectrum but Transposition, pulmonary

pathologies heart d i sease transposition and stenosis or atresia, corrected transposition atrioventricular canal are more common

Usually left isomerism Often polysplenia Often drain to both atria

Usually interrupted

Often drain to ipsehteral atrium

Atrial and ventricular septal defects, atrio, ventricular canal

the left-sided atrium in inversus. The pulmonary veins will usually drain to the left atrium but are less predictable than the systemic veins and their connec- tions should not be presumed.

In the smaller group of situs ambiguus, nearly all cases can be classified as an isomerism of one of two sorts. In right isomerism, both atria have the mor- phology of a right atrium, in left isomerism both atria are morphologically left atria. Abnormal venous connections are very common in both groups, and the pattern of the abnormalities reflects the embryology of the situation.

In right isomerism (bilateral right sidedness) there is no left atrium, to which the pulmonary veins should anastomose. Not surprisingly, then, more than half these cases end up with total anomalous pulmonary venous drainage into a systemic vein.

The superior vena cava (SVC) in the normal drains to the RA. Persistent left SVC is not unusual in situs solitus atria, but it still drains to the RA via the coronary sinus. In right isomerism bilateral SVCs are often found but they drain separately into their ipselateral atrium and the coronary sinus is absent. The inferior vena cava (IVC) and the hepatic veins are normally strongly related to the RA and with bilateral right atria the IVC may drain either side, usually the side which has the bulk of the liver in it. The hepatic veins may drain directly into the nearest atrium.

In the left isomerism (bilateral left sidedness) the pulmonary veins are faced with left atrial tissue bilaterally and as a result they often drain separately to their ipselateral atrium. Whereas in right isomerism the pulmonary veins can drain to a systemic vein, in left isomerism the systemic veins join the pulmonary veins. The IVC has no right atrial tissue with which to anastomose preferentially and either enforces a connection with either atrium or, as is more often the

case, the IVC is absent. When the IVC is absent, venous blood from below the diaphragm passes through either the azygos or hemiazygos systems, which are consequently dilated. Which of these depends on which side is the SVC. One SVC is always present, on either side, draining to the ipselateral atrium. Often both SVCs persist, similar to right isomerism, and either can accept the azygos vein.

The intracardiac pathology which accompanies the isomerism is varied; the stronger association are listed in Table 1.

TYPES OF BRONCHIAL AND VISCERAL SITUS

The bronchi, spleen, liver and stomach are the other organs which show a right/left asymmetry which can be called situs. The position of the cardiac ventricles and the cardiac apex, and the side of the aortic arch, though they also show asymmetry, depends more on caridac malformations than on underlying disorders of body situs.

The bronchi show distinct differences in anotorny between the right and left lungs, due to their differing relationships with the branch pulmonary arteries (Fig. 1). These differences are reflected in the pattern of the segmental bronchi, and in the external lobulati0n of the lungs. However, the pattern of the fissures is not a reliable guide to the underlying anatomy; accessory fissures and congenital absences of the fissures are quite common. The lungs can be in situs inversus, with the morphologically right lung on the left and vice versa, or they can show right or left isomerism; there is no functional disturbance in any case.

In the abdomen situs solitus is evidenced by the larger lobe of the liver being on the right, the stornac~ bubble on the left, both signs being radiographically visible. The spleen is on the left, and the gut is

R A D I O L O G I C A L E V A L U A T I O N OF A T R I A L S I T U S 97

Fig. 1 - Relationship o f the main and branch pulmonary arteries to the tracheo-bronchial tree. Note tha t the left main bronchus is longer than the right to allow the whole of the left pulmonary artery to pass posterior to it before the first lobar bronchus is given off. On the right the pu lmonary artery passes posterior to the bronchus after the first lobar branch• T, trachea; RMB, LMB, right and left main bronchi; MPA, RPA, LPA, main, right and left pu lmonary arteries.

normally rotated. In situs inversus viscera, these positions are reversed• Again isomerisms may be found• Abdominal right isomerism is the syndrome of asplenia (Ivemark, 1955) because the spleen, which is an embryologically left-sided structure, is absent. A useful clue to this situation is the presence of Howell-Jolly bodies on the blood smear. With asplenia, the liver usually has right and left lobes of equal size and so appears midline. Malrotation of the gut is common and the stomach tends to be in its embryonic midline position, though it can be on either side. With left isomerism, the syndrome is polysplenia which strictly means multiple spleens on both sides of the abdomen• Midline liver and stomach, and gut malrotation are again common, but much less so than in asplenia. In particular, on the plain film the stomach and liver may be well latera- lised and falsely suggest situs solitus or inversus.

RELATIONSHIPS OF ATRIAL, BRONCHIAL AND ABDOMINAL SITUS

As a rule, all three compartments show the same situs, as illustrated in Fig. 2. However, differences in situs can exist between any of them, and the abdominal situs can vary from atrial and bronchial situs. For these reasons, and because the abdominal

8

Situs solitus Situs invert;us

Right isomerism

. 0 O i i

Left isomerism

Fig. 2 - The four basic arrangements o f body situs• In each o f these there is agreement between the three major compar t - ments , i.e. atrial, bronchial and visceral situs• Combinat ions can occur as described in the text_ RA, LA, morphologically right and left atria; RL, LL, morphologicaUy right and left lungs; L, liver; S, spleen, St, s tomach; Si splenunculi.

..A1

R ~ L

[

g 1 Clj ~ ..," F2 ".-.~ "~'~ .,Q ~ ~ - "

BI- ..... ' .......

Fig. 3 - Measurement of the bronchi. The upper end of a bronchus is marked by the intersection of the bronchial axis (BB1 or BB2) with the perpendicular f rom the carina. The lower end is marked by the perpendicular from either the tangent of the upper border of the bronchus or f rom the angle o f the lower border (whichever is the shortest distance if bo th are visible) intersecting with the bronchial axis.

98 CLINICAL R A D I O L O G Y

• = NORMAL" D = RIGHT AORTIC ARCH O ABNORMAL: NO ISOMERISM ~ = ISOMERISM

o

z o • DOOQ

~. Q [ l l lm •

• Q 9e l4mGOlO

0 • Q 0 0 0 0 e l • o e

....... ; ......... ~ .... ~ .... : .... ~ 2

RATIO B~:TWEEN BRONCHIAL LENGTHS

Fig. 4 - The distribution of bronchial length ratios (BLRs). Note that the ratios for a control group of patients with no bronchial isomerism, including those with right aortic arches and with abnormal cardiac positions, are above 1.5 and most lie above 2. The small group of patients with bronchial isomerism all lie at or below 1.5.

isomerisms as described above are n o t apparent on the plain film in many cases, abdominal situs is no longer taken as a reliable guide to atrial situs. For example Losekoot (1973) described cases o f clearly discordant atrial and abdominal situs.

As m e n t i o n e d in the in t roduc t ion the bronchial situs is more o f ten in agreement wi th atrial situs than is the abdominal situs. This is part icularly so wi th bronchial situs solitus and inversus, when atrial situs is rarely at variance wi th it.

There is only one case o f atrial inversus in the face of bronchial solitus in the l i terature (Clarkson et al., 1972).

Bronch ia l l eng th (c~,/

6 LEFT

0 R I G H T

# 4 + %,

: ' . r ' : 3 e, ° + *

oo o o o c~° °°

1~ ° e OOo o ~> ,% o o oO

o

' ") ' 4: ' 6

O O

. . ' . : . :

# #o Q ## ~ # ~, o j . . - - - -

J - f

° o o o

o°O ~ o o (D I

o

o

I/

Fig. 5 - The absolute lengths of the bronchi plotted against age in a group of normals. Using this diagram a bronchus of unknown morphology can be classified as right or left on its length alone. The solid lines are the calculated discriminant lines between the two groups of lengths; the.chance of a bronchus appearing on the wrong side of the line is much less than 1%.

When there is bronchial i somerism the atria usually also show isomerism o f the same sort. Except ions do occur but their f requency is uncertain. In path0. logical series, when atrial morpho logy can be directly inspected, they are rare (Stanger et aL, 1977). In clinical series, t hey are more f requent , probably because atrial situs is clinically de te rmined by the pa t te rn o f venoatrial connect ions which can be reasonably normal despite an under lying atrial is0- merism.

Fig- 6a - Case 1. Twelve-year-old girl with cyanosis. Chest film shows dextrocardia with left-sided stomach bubble suggesting situs solitus.

Fig. 6b - Case 1. Tomogram reveals bronical situs solitus (BLR 3.1). Catheterisation revealed atrial situs s'olitus als0 and corrected transposition.

RADIOLOGICAL EVALUATION OF ATRIAL SITUS 99

Fig. 7a - Case 2. Fourteen-year-old boy. Chest film shows left-sided stomach and cardiac apex, suggesting situs solitus, and pulmonary plethora.

Fig. 7b - Case 2. Tomograms showing carina (upper arrow) and upper lobe bronchi (lower arrows). There is left isomerism unsuspected from the plain chest film (BLR 1.1, right bronchus 3.5 era, left bronchus 3.8 cm). The IVC was inter- rupted, and the hepatic veins drained anomolously, Partial A/V canal.

Fig. 8a Case 3. Five-year-old child. Chest Film shows dex- trocardia, left-sided liver and right-sided stomach (confirmed by barium swallow).

Fig. 8b - Case 3. Tomogram reveals left isomerism (BLR 1.2).

100 C L I N I C A L R A D I O L O G Y

DETERMINATION OF BRONCHIAL SITUS

The segmental bronchial pattern is probably the most reliable indicator of bronchial situs, but since these patients are usually young children, tomo- graphy is often unsatisfactory and br0nchography not warranted. Because of this Partridge e t al. (1975) devised a method of determining bronchial situs from the lengths of the main bronchi since the anatomy of the normal bronchi is that the left is longer than the

Fig. 8c, 8d - Case 3. AP and latelal cine angiogram shows interruption of the IVC with a left azygous vein draining via a left SVC to a left-sided atrium.

Fig. 8e - Case 3. AP frame of injection into a right SVC which drains to the right-sided atr ium. Intracardiac anatomy was complex.

right (Fig. 1). This requires that the carina and main bronchi down to the upper lobe bronchus are visualised by a penetrated chest film or, if necessary. small angle tomography (zonography), Failure to achieve this is not uncommon in small infants.

The apparent lengths of the bronchi are measured according to the method shown in Fig. 3. If the ratio between the two is 2 : 1 or greater, isomerism is excluded; if the longer bronchus is on the left, there is situs solitus, and if it is on the right, situs inversus is diagnosed. If the ratio is 1.5 or less, isomerism is almost certain. These figures are based on data summarised in Fig. 4. When the ratio is between 2 and 1.5:1, or when isomerism is likely and it is desired to distinguish right isomerism from left isomerism, the absolute length of each bronchus can be compared to the data in Fig. 5 in order to type ,it as having right or left morphology. The situs follows from the combination of morptiologies so determined. To obtain the absolute length of a bronchus from its length on the film, a correction factor has to be applied. This is 0.95 for most plain films, be they AP or PA. Tomograms require more careful correction either by measurement or by the use of a standard object.

R A D I O L O G I C A L E V A L U A T I O N O F A T R I A L SITUS 101

Fig. 9a - Case 4. Twenty-year-old girl with cyanosis. Chest film shows dextracardia and left-sided stomach. Bilateral lesser fissures visible (arrowed), suggesting right isomerism.

Fig. 9b - Case 4. Tomogram confirms right isomerism. Both upper lobe bronchi (solid arrows) arise from short main bronchi close to the carina (open arrow). BLR 1.1, right bronchus 2.1 cm, left bronchus 2.5 cm.

Fig. 10a - Case 5. Three-year-old boy with cyanosis_ Chest film shows laevocardia, but left-sided liver and right-sided stomach suggesting situs inversus. However, Howel l -Jol ly bodies were repeatedly found in the blood, strongly suggesting asplenia.

Fig. 10b - Case 5. Tomogram unexpectedly reveals bronchial situs solitus (BLR 2.4) completely at variance with the abdominal situs. Catheterisation revealed that the atrial situs followed bronchial situs, i.e. solitus, and the cardiac patho- logy was limited to tetralogy o f Fallot.

102 C L I N I C A L R A D I O L O G Y

Fig. 1 la - Case 6. Six-year-old girl with cyanosis, dextr0- cardia and asplenia. The tomogram shows bilateral pte- eparterial bronchi (solid arrows) rising above the carina, (open arrow). This is consistent with right isomerism (see Fig. l lb ) .

Fig. l l b - Chance finding of a pre-eparterial bronchus in one of our normal control subjects. This is a variant of normal anatomy and only occurs in a morphologically right lung.

Fig. 11c - The pre-eparterial bronchus is sometimes known as bronchus suis, as it is a normal finding in the pig as shown by this post-morten bronchogram of a piglet. It clearly shows the bronchus arising well above the trachea on the right side.

worthwhile to obta in a penetra ted chest film on all first t ime catheter cases, as occasional cases of left isomerism can be found (e.g. case 2 below).

WHEN SHOULD BRONCHIAL SITUS BE EVALUATED?

Cases of congenital heart disease can be managed clinically wi thou t atrial situs being determined up to the point o f impending cardiac catheterisat ion. Then, in any case in which the plain film shows abnormal cardiac posi t ion or abdominal situs, and those cases with Howel l - Jo l ly bodies in the b lood, bronchia l situs should be determined. Even in straightforward cases, at Kil l ingbeck Hospital we have found i t

ILLUSTRATIVE CASES

Figs. 6 - 1 1 show examples of cardiac malpositions, with and without concomi tan t disorders of situs, and illustrate the usefulness of determining bronchial situs. Being a clinical series, one cannot be sure of the exact status of abdominal situs, part icularly with respect to polysplenia; however they do show that the posi t ion of the s tomach and liver can on occasions be a poor predic tor of broncho-atr ia l situs.

The il lustrations of the bronchial tomograms have all been reinforced, some heavily. This is n o t only because the in format ion was often on m o r e t h a n one cut bu t also patients were most ly too young to hold

R A D I O L O G I C A L E V A L U A T I O N OF A T R I A L SITUS 103

their b r e a t h adequa t e ly , and the i r small size d imin i shed the con t ras t o f t he image as a mul t i - sec t ion casse t te c/as used requir ing a h igh k i lovol tage .

Acknowledgements. Figs. 3,4, 5, 1 la, 1 lb are reproduced f~orn O'rculation (Partridge et al., 1975) by kind permission of the American Heart Association, Inc.

MY thanks to Mr D. Howard and the Department of g¢dical Illustration, St James's University Hospital, Leeds, for the majority of the illustrations; and to Mrs M. Illingworth and the radiographers at Killingbeck Hospital, who performed the tomography on many small and mobile patients; finally to B. Hobson for secretarial assistance.

REFERENCES Brandt, P- W. T. & Calder, A. L. (1977). Cardiac connections:

the segmental approach to radiologic diagnosis in con- genital heart disease. Current Problems in Diagnostic Radiology, 7, (3), 1-35 .

Clarkson, P. M., Brandt, P. W. T. & Barratt-Boyes, B. G. (1972)- 'Isolated atrial inversion.' Visceral situs solitus, visceroatrial discordance, discordant ventricular l-loop without transposition, dextracardia. Diagnosis and surgical correction. American Journal of Cardiology, 29, 877-881.

Ivemark, B. I. (1955). Implications of agenesis of the spleen in the pathogenesis of conotruncal anomalies in child- hood. Acta paediatrica scandinavia, Suppl. 104_

Landing, B. H., Tsun-Yee, K. L., Vaughn, C. P. & Wells, T. R. (1971). Bronchial anatomy in syndromes with abnormal visceral situs, abnormal spleen and congenital heart disease. American Journal of Cardiology, 2 8 , 4 5 6 - 462.

Losekoot, T. G. (1973). Mirror image dextroeardia with situs solitus of the abdominal organs and a normal heart. European Journal o f Cardiology, 1, 4 9 - 5 4 .

Macartney, F. J., Partridge, J. B., Shinebourne, E. A., Tyrian, M. J. & Anderson, R. H. (1978). Identification of atrial situs. In Paediatric Cardiology 1977, ed. Anderson, R. H. & Shinebourne, E. A. Churchill Livingstone, Edinburgh.

Partridge, J. B., Scott, O., Deverall, P. B. & Macartney, F. J. (1975). Visualisation and measurement of the main bronchi by tomography as an objective indicator of thoracic situs in congenital heart disease. Circulation, 51, 188-196.

Stanger, P., Rudolph, A. M. & Edwards, J. E. (1977). Cardiac malpositions: an overview based on a study of 65 necropsy specimens. Orculation, 56 ,159-172 .

Van Mierop, L. H. S., Eisen, S: & Schiebler, G. L. (1970). The radiographic appearance of the tracheobronchial tree as an indicator of visceral situs. American Journal of Cardiology, 26 ,432-435 .