- Email: [email protected]
QRS axis in isolated perimembranous ventricular septal defect and influences of morphological factors on QRS axis
Journal of Electrocardiology Vol. 34 No. 3 2001
QRS Axis in Isolated Perimembranous Ventricular Septal Defect and Influences of Morphological Factors on QRS Axis
H. Ercan Tutar, MD, Semra Atalay, MD, Sadi Tu¨rkay, MD, and Ayten Imamoglu, MD
Abstract: To detect the frequency of left axis deviation (LAD) in isolated perimembranous ventricular septal defects (VSD) we retrospectively analyzed electrocardiograms of 59 patients, aged 8 months to 15 years. Patients were grouped into those having ventricular septal aneurysm (VSA) formation (n:20) and those who did not have VSA (n:39). Patients with VSA were then stratified into 2 groups according to the presence of left ventricular-to-right atrial (LV-RA) shunt. Four hundred healthy children served as the control group. We found that 12 (20.3%) of 59 patients with isolated perimembranous VSD had a LAD. Five of 6 patients with perimembranous outlet VSD and 6 with perimembranous inlet VSD had abnormal LAD with a qR pattern in I and aVL and rS in aVF. Abnormal LAD was more prevalent in patients with VSA (40%) than without VSA (7.7%) (P ⬍ .01). We also found that mild right ventricular hypertrophy (RVH) with the rsRⱊ or rSRⱊ pattern in V1 was more frequent in patients with VSA, especially those who had LV-RA shunt. However, we could not find significant difference between patients with or without LV-RA shunt for the incidence of abn LAD and mild RVH. Localization of perimembranous VSD was not found to have an effect on frequency of abnormal LAD and mild RVH in this patient group. In patients with clinical findings of VSD, the existence of abnormal LAD especially if it is associated with mild RVH, should raise the possibility of perimembranous VSD with VSA formation. Key words: Perimembranous ventricular septal defect, ventricular septal aneurysm, left ventricular-to-right atrial shunt, left axis deviation.
Abnormal left axis deviation (abn LAD) of QRS axis in frontal plane is uncommon in childhood and is usually associated with complex congenital heart
disease such as atrioventricular septal defect, tricuspid atresia due to posterior and inferior displacement of the conduction system resulting in inferior initial activation and a larger superior terminal activation (1). Left axis deviation frequency in healthy infants and children was reported in a range of 0.4% to 1.4% (2– 4) was found in 6.2% to 10.1% of patients with ventricular septal defect (VSD) (5,6). We noticed that abn LAD was frequent in patients who had left ventricular-to-right atrial (LV-RA) shunt associated with perimembranous
From the Ankara University, Medical School, Department of Pediatric Cardiology, Ankara, Turkey. Reprint requests: H. Ercan Tutar, MD, Defne sitesi 8. Blok No: 37, Umitkoy, 06530, Ankara, Turkey; e-mail: [email protected]. Copyright © 2001 by Churchill Livingstone® 0022-0736/01/3403-0003$35.00/0 doi:10.1054/jelc.2001.24763
197
198 Journal of Electrocardiology Vol. 34 No. 3 July 2001 VSD. We, therefore, retrospectively analyzed electrocardiograms of patients with isolated perimembranous VSD to detect the frequency of abn LAD, especially in a subgroup of patients with LV-RA shunt.
Patients and Methods Electrocardiograms of 59 patients (29 girls, 30 boys), aged 8 months to 15 years (6.4 ⫾ 4.0 years) who had perimembranous VSD were retrospectively analyzed. Patients with any associated congenital heart disease and/or Down’s syndrome were excluded. Cardiac catheterization and left ventricular angiography were performed in 39 patients. Eighteen patients with isolated perimembranous VSD who had pulmonary hypertension were also excluded. Pulmonary hypertension was defined as: 1) Mean pulmonary artery pressure ⬎ 25 mm Hg during cardiac catheterization; or 2) In patients without cardiac catheterization, pulmonary artery systolic pressure ⬎ 35 mm Hg, which was calculated from peak systolic continuous wave Doppler pressure gradient of interventricular shunt through the ventricular septal defect. Echocardiographic studies were performed with Toshiba Sonolayer SSH 140A machine (Toshiba Corp, Japan) with 3.75 MHz and 5 MHz transducers. Standard and modified parasternal, apical, and subcostal echocardiographic windows were used. The classification of perimembranous VSDs according to their exten-
Fig. 1. Two-dimensional echocardiogram (parasternal four-chamber view) shows perimembranous inlet ventricular septal defect and ventricular septal aneurysm (arrows). The septal leaflet of the tricuspid valve is involved in the formation of the ventricular septal aneurysm. LV, left ventricle; RV, right ventricle; LA, left atrium; RA, right atrium.
sions, as inlet, outlet, and trabecular was based on the previous reports (7–9). Ventricular septal aneurysm (VSA) formation of perimembranous VSD was defined as echocardiographic evidence of deposition of tissue around the margins of VSD, producing a sac-like structure with distinct margins that seemed to partially occlude the flow through the VSD (Fig. 1). LV-RA shunt was defined as echocardiographic evidence of a high velocity turbulence moving from left ventricle, through the area of VSD directly into right atrium (10,11). The peak velocity of the LV-RA and interventricular shunt was measured by continuous wave Doppler. The pressure gradient of LV-RA shunt was equivalent to the pressure difference between systolic left ventricular and right atrial pressures and could be differentiated from tricuspid regurgitation by the high velocity of the LV-RA shunt in the absence of elevated right ventricular pressure (11). M-mode demonstration of tricuspid valve systolic flutter was used as a supportive evidence of LV-RA shunting (11–13). Standard 12-lead electrocardiograms were obtained with a three-channel electrocardiographic recorder (Hewlett Packard, model 4745 A; Hewlett Packard Inc, Camas, WA). QRS axis in the frontal plane was calculated and recorded with the other features of the tracings. QRS axis less than 0° was defined as LAD (1). Abn LAD or abnormal superior QRS axis was defined as an inferior initial activation and a larger superior terminal activation (qR pattern in I and aVL, rS pattern in aVF). Patients were divided into 2 groups according to the existence of
Abnormal QRS Axis VSD •
Tutar et al. 199
Table 1. Demographic and Hemodynamic Parameters and Distribution of Ventricular Septal Defect Types According to the Patient’s Group in Patients With Isolated Perimembranous Ventricular Septal Defect
VSA (⫹) (n:20) LV-RA shunt (⫹) (n:8) LV-RA shunt (⫺) (n:12) VSA (⫺) (n:39) P
Trabecular
Mean PA Pressure (mmHg)
Qp/Qs
1 (5) 1 (12.5) ⫺ 2 (5) ⬎.05
16.8 ⫾ 4.6 (n:12) 16.0 ⫾ 5.1 (n:6) 17.7 ⫾ 4.5 (n:6) 18.7 ⫾ 4.2 (n:27) ⬎.05
1.50 ⫾ 0.39 (n:12) 1.65 ⫾ 0.34 (n:6) 1.34 ⫾ 0.41 (n:6) 1.65 ⫾ 0.37 (n:27) ⬎.05
Perimembranous VSD, n (%)
Age (years)
Sex (M/F)
Inlet
Outlet
5.1 ⫾ 4.0 5.2 ⫾ 3.2 5.0 ⫾ 4.6 7.0 ⫾ 3.9 ⬎.05
9/11 3/5 6/6 21/18 ⬎.05
7 (35) 2 (25) 5 (42) 15 (38.5) ⬎.05
12 (60) 5 (62.5) 7 (58) 22 (56.5) ⬎.05
VSA, ventricular septal aneurysm; LV-RA, left ventricular-right atrial; Qp, pulmonary blood flow; Qs, systemic blood flow; PA, pulmonary artery; (⫹) denotes presence, (⫺) denotes absence.
echocardiographic signs of VSA formation. Patients with VSA were then divided into 2 groups whether they had LV-RA or did not have LV-RA shunt. Electrocardiograms were obtained from 400 (182 girls, 228 boys) healthy children aged 5 years to 18 years (10.2 ⫾ 2.4 years) who served as control group. All data are expressed as mean ⫾ standard deviation. Chi-square test, Fisher’s exact test, unpaired Student’s t-test and Kruskall Wallis One-Way ANOVA were used where appropriate for statistical comparisons. A two-tailed P value ⬍ .05 was considered statistically significant.
Results Demographic data and hemodynamic parameters of the patients and distribution of perimembranous VSD types according to their location are shown in Table 1. Echocardiographically 22 (37.3%) patients had perimembranous inlet, 34 (57.6%) patients had perimembranous outlet, and 3 (5.1%) had perimembranous trabecular VSD. No patient had cleft septal tricuspid valve leaflet. There was no significant difference in mean age, mean pulmonary artery pressure, and the mean pulmonary/
systemic blood flow ratio between patient groups. We also did not find difference in distribution of VSD types between patient groups. The electrocardiographic features of patient groups and control subjects are summarized in Table 2. The mean QRS axis in patients with VSD especially in with VSA was lower, but the difference did not reach the statistically significant level (P ⫽ .07). Twelve (20.3%) of patients with VSD had LAD in which 11 had abn LAD, but only 6 (1.5%) of children in control group had (P ⬍ .001). In patients with LAD, QRS axis in the frontal plane ranged from ⫺5° to ⫺60°. Except for 1 patient who had perimembranous outlet VSD without VSA formation, all patients had abn LAD in the frontal plane. Abn LAD was found in 8 (40%) of patients with VSA, but in 3 (7.7%) of patients without VSA (P ⬍ .01). Mild right ventricular hypertrophy (RVH) (rsR⬘, rSR⬘ pattern in V1) was also frequent in patients with VSD. It was found in 15 (25.4%) patients with VSD, but in 7 (1.75%) healthy children (P ⬍ .001). We found that 11 (55%) patients with VSA had mild RVH, but only 4 (10.2%) patients without VSA had it (P ⬍ .001). We could not find significant difference between patients with or without LV-RA shunt for the incidence of abn LAD or mild RVH (P ⫽ .64, P ⫽ .19, respectively).
Table 2. Electrocardiographic Findings of Controls and Patients With Isolated Perimembranous Ventricular Septal Defect According to the Patient’s Group
VSA (⫹) (n:20) LV-RA shunt (⫹) (n:8) LV-RA shunt (⫺) (n:12) VSA (⫺) (n:39) Control (n:400)
QRS Axis (°)
Abn LAD n (%)
Mild RVH n (%)
Abn LAD ⫹ Mild RVH n (%)
RAE n (%)
LVH n (%)
BVH n (%)
28.5 ⫾ 55.7 20 ⫾ 60.8 34.2 ⫾ 54.1 46.9 ⫾ 37.9 60.9 ⫾ 18.1
8 (40) 4 (50) 4 (33.3) 3 (7.7) 6 (1.5)
11 (55) 6 (75) 5 (42) 4 (10.2) 7 (1.75)
4 (20) 2 (25) 2 (16.7) 1 (2.5) 1 (0.25)
6 (30) 5 (62.5) 1 (8.3) 1 (2.5) ⫺
5 (25) 2 (25) 3 (25) 11 (28.2) ⫺
2 (10) ⫺ 2 (16.7) 1 (2.5) ⫺
VSA, ventricular septal aneurysm; LV-RA, left ventricular-right atrial; abn LAD, abnormal left axis deviation; RVH, right ventricular hypertrophy; RAE, right atrial enlargement; LVH, left ventricular hypertrophy; BVH, biventricular hypertrophy; (⫹) denotes presence, (⫺) denotes absence.
200 Journal of Electrocardiology Vol. 34 No. 3 July 2001
Fig. 2. The electrocardiogram of a 4-year-old patient with perimembranous outlet ventricular septal defect and left ventricular-toright atrial shunt shows abnormal left axis deviation and mild right ventricular hypertrophy (rsR⬘s⬘ pattern in V1).
Electrocardiographic finding of right atrial enlargement (tall, peaked P waves in DII, V1) was more frequent (62.5%) in patients with LV-RA shunt than other patient groups; patients with VSA but without LV-RA shunt (8.3%, P ⬍ .05), patients without VSA (2.5%, P ⬍ .001). Twenty percent of the patients with VSA had both abn LAD and mild RVH (Fig. 2); this association was only found in 2.5% of patients without VSA (P ⬍ .001). Most of the patients with isolated perimembranous VSD did not show electrocardiographic evidence of ventricular hypertrophy; 25% of our patients showed slight left ventricular hypertrophy and 5% of patients showed biventricular hypertrophy (Table 2). There was no significant difference in the frequency of abn LAD, mild RVH, and right atrial enlargement among the perimembranous VSD types (Table 3). Detailed electrocardiographic data of patients with LAD are shown in Table 4.
Discussion Although more defects in the muscular ventricular septum close spontaneously than perimembranous VSDs, a good proportion of the latter show a tendency to diminish in size to close (14). This closure or diminishing in size is usually a result of VSA formation. Ventricular septal aneurysm forma-
tion was found in up to three fourths of patients with perimembranous VSD (15). Both pathologic (16) and echocardiographic studies (17) showed that in most perimembranous VSDs, the tissue forming VSA, derived from tricuspid valve leaflets. In a large series of 930 patients with isolated perimembranous VSD, LV-RA shunt was shown in approximately 15% of patients after the occurrence of VSA formation (15). In those patients, the mechanism responsible for the occurrence of LV-RA shunt was usually described as widening of anteroseptal commissure of tricuspid valve. Malformation, cleft, and perforation of the tricuspid valve were also reported as responsible mechanisms for the occurrence of LV-RA shunt (13,15,18).
Table 3. Electrocardiographic Findings of Patients With Isolated Perimembranous Ventricular Septal Defect According to the Ventricular Septal Defect Types
Perimembranous inlet (n:22) Perimembranous outlet (n:34) Perimembranous trabecular (n:3) P value
Abn LAD n (%)
Mild RVH n (%)
RAE n (%)
QRS Axis (°)
6 (27.3)
6 (27.3)
2 (9.1)
30.2 ⫾ 48.2
5 (14.7)
8 (23.5)
4 (11.8) 47.3 ⫾ 42.1
0 (0) .34
1 (33.3) .90
1 (33.3) 41.6 ⫾ 59.2 .47 .36
Abn LAD, abnormal left axis deviation; RVH, right ventricular hypertrophy; RAE, right atrial enlargement.
Abnormal QRS Axis VSD •
Tutar et al. 201
Table 4. Electrocardiographic Characteristics of 12 Patients With Left Axis Deviation According to the Ventricular Septal Defect Types Type of VSD Perimembranous inlet
Perimembranous outlet
Patient No.
QRS Axis (°)
Abn LAD
aVF Pattern (V)
Mild RVH
RAE
LVH
1 2 3 4 5 6 7 8 9 10 11 12
⫺50 ⫺20 ⫺40 ⫺15 ⫺50 ⫺40 ⫺5 ⫺40 ⫺10 ⫺60 ⫺20 ⫺10
⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹
rS (150/800) rS (400/600) rS (200/700) rS (300/500) rS (400/1000) rS (250/800) qrS (100/300/400) rS (500/900) rS (450/550) rS (450/900) rS (300/400) rS (600/700)
⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫺
⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺
⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺
Abn LAD, abnormal left axis deviation; RVH, right ventricular hypertrophy; RAE, right atrial enlargement; LVH, left ventricular hypertrophy; (⫹) denotes presence, (⫺) denotes absence.
Electrocardiographic evidence of right atrial enlargement with a prolonged PR interval, mild RVH, and biventricular hypertrophy were all frequently reported in patients with LV-RA communication (18,19). We also found right atrial enlargement (62.5%) and mild RVH (75%) frequently in the electrocardiograms of our patients with LV-RA shunt. This type of defect results in volume overload of right heart chambers; right atrial enlargement and mild RVH are the predictable electrocardiographic findings. Abn LAD was reported in about 25% of patients with LV-RA shunt usually in association with combined ventricular hypertrophy in the atrioventricular canal type of defect (18 –20). In our patient group, abn LAD was found in 4 (50%) of 8 patients with LV-RA shunt (Table 2). Left axis deviation was reported in 0.4% to 1.4% of healthy children and adolescents (2– 4). In our control group of 400 healthy children we found LAD in 1.5% incidence (Table 2). Left axis deviation was seen in up to 10% of patients with isolated ventricular septal defect (5,6). In a few studies concerning the relation between the localization of VSD and LAD, authors pointed out that LAD was more common when VSD was in the inlet portion of the ventricular septum or affected this area (5,6). However, Shaw et al. (5) showed that 41% of patients with LAD had the defect at a site other than the perimembranous inlet septum. In our study group, 11 (18.6%) patients with isolated perimembranous VSD had abn LAD. In 6 of 12 patients, the defect was classified as perimembranous inlet and the other 5 as perimembranous outlet. Although, in patients with perimembranous inlet VSD, the frequency of abn LAD (27.3%) was slightly higher than with other defect locations, the difference did not reach the statistically significant level (Table 3).
Abn LAD in patients with atrioventricular septal defect is a well-known feature and is believed to be associated with posteroinferior displacement of the atrioventricular conduction system as a result of anatomical characteristics of defect (21,22). Whether the anatomical disconfiguration of atrioventricular conduction system, like in atrioventricular septal defect, exists in VSDs associated with abn LAD is an area of interest of pathological studies. Although we did not find any relation between abn LAD and defect location, we found significant difference between patients with or without VSA formation in the frequency of abn LAD. The distribution of defect location was also not different between the patients with or without VSA. We could find only 1 study that explained the relation between VSA formation and LAD frequency in the English literature (23). Farru-Albohaire et al. (23) found in two thirds of patients with VSA had LAD; however, none without VSA had it (23). In that study, the occurrence of LAD was a new finding, which appeared after the development of VSA in 10 patients. We do not know whether or not the abn LAD was present before the VSA development in our patients. Authors then concluded that the appearance of LAD in a patient with perimembranous VSD indicated a spontaneous reduction in the defect by VSA formation (23). Unfortunately, there was no information about the classification of perimembranous VSDs in that study. We as well as Farru-Albohaire et al. (23) have not been able to explain the development of abn LAD after the occurrence of VSA by the adherence of tricuspid valve leaflets to the rim of the defect. Ueda and Becker (24) investigated the anatomy of atrioventricular conduction system in perimembranous VSDs and showed a close relation
202 Journal of Electrocardiology Vol. 34 No. 3 July 2001 between the conduction bundle and the rim of the defect and the anomalous tricuspid valve attachments in 2 heart specimens. In one specimen, conduction bundle extraordinarily infiltrated the leaflet tissue (24). So, one can speculate that if the tricuspid valve leaflet which contributes to the VSA formation contains the conduction tissue, this may be responsible for the development of abn LAD. However, we could not show the appearance of abn LAD after the development of VSA. So, abn LAD might have been present before the occurrence of septal aneurysm. Although the QRS axis tends to be more leftward than is normal for the patient’s age with left ventricular hypertrophy, abn LAD is rare in childhood (1). Abnormal LAD in our patients cannot be attributed to left ventricular hypertrophy because only small percent of these patients showed electrocardiographic signs of mild left ventricular hypertrophy. The mild RVH pattern was found in electrocardiograms of normal children in up to 7% of the cases (1). In our control group, we found 1.75% of children had mild RVH. The frequency of this pattern was reported 20% to 30% of patients with VSD (6,25). Mild RVH was present in 75% of our cases with LV-RA shunt as expected. However, we showed almost half of the patients with VSA without LV-RA shunt had also mild RVH pattern. This pattern was documented only 10.2% of patients without VSA. Farru-Albohaire et al. (23) showed that mild RVH was present in 34% of patients with VSA, but only in 6.8% of without VSA similar to our results. We found that half of our patients with VSA and abn LAD also had mild RVH (Table 2). Farru-Albohaire et al. (23) also reported that 48% of their patients with VSA and LAD had mild RVH but none without VSA. Although abn LAD associated with mild RVH pattern is considered a distinct electrocardiographic feature of atrioventricular septal defect (26), it can also be found in patients with VSD associated with VSA, like in ours. We believe that mild RVH (rsRⱊ or rSRⱊ pattern in V1) is the result of volume overload of right ventricle in our cases. We showed that abn LAD was frequent in patients with LV-RA shunt. However, this finding was not limited to this group. Abn LAD was also observed more frequently in patients with VSA without LV-RA shunt than in patients without VSA. Mild RVH was also a frequent finding especially in patients with VSA associated with LV-RA shunt. We suggest that in patients with clinical findings of VSD, the existence of abn LAD, especially if it is
associated with mild RVH, should raise the possibility of perimembranous VSD with VSA formation.
References 1. Garson AJ: Electrocardiography, in Garson AJ, Bricker JT, Fisher DJ, (eds): The Science and Practice of Pediatric Cardiology. Williams & Wilkins, Baltimore, MD, 1998, p 735 2. Gup AM, Franklin RB, Hill JE: The vectorcardiogram in children with left axis deviation and no apparent heart disease. Am Heart J 69:619, 1965 3. Calcaterra G, Puglisi R: Left axis deviation in healthy infants and children. Int J Cardiol 24:236, 1989 4. Pearl W, Stafford EM: Left axis deviation in children. Int J Cardiol 34:119, 1992 5. Shaw NJ, Godman MJ, Hayes A, et al: Superior QRS axis in ventricular septal defect. Br Heart J 62:281, 1989 6. Ba¨ckman H: Influence of structural and functional features of ventricular septal defect on frontal plane QRS axis of the electrocardiogram. Br Heart J 34:274, 1972 7. Soto B, Becker AE, Moulaert AJ, et al: Classification of ventricular septal defects. Br Heart J 43:332, 1980 8. Sutherland GR, Godman MJ, Smallhorn JF, et al: Ventricular septal defects: Two dimensional echocardiographic and morphological correlations. Br Heart J 47:316, 1982 9. Helmcke F, Souza A, Nanda NC, et al: Two-dimensional and colour doppler assessment of ventricular septal defect of congenital origin. Am J Cardiol 63: 1112, 1989 10. Leung MP, Mok CK, Lo RNS, et al: An echocardiographic study of perimembranous ventricular septal defect with left ventricular to right atrial shunting. Br Heart J 55:45, 1986 11. Wu MH, Wu JM, Chang CI, et al: Implication of aneurysmal transformation in isolated perimembranous ventricular septal defect. Am J Cardiol 72:596, 1993 12. Nanda NC, Gramiak R, Manning JA: Echocardiography of the tricuspid valve in congenital left ventricular-right atrial communication. Circulation 51:268, 1975 ¨ zkutlu S, Pas¸aog˘lu I: Echocardiographic 13. Sarac¸lar M, O diagnosis of congenital left ventricular-right atrial communication. Jpn Heart J 29:671, 1988 14. Shirali GS, Smith EO, Geva T: Quantitation of echocardiographic predictors of outcome in infants with isolated ventricular septal defect. Am Heart J 130: 1228, 1995 15. Wu MH, Chang CI, Wang JK, et al: Characterization of aneurysmal transformation in perimembranous ventricular septal defects: An adhered anterior leaflet of tricuspid valve predisposes to the development of left ventricular-to-right atrial shunt. Int J Cardiol 47:117, 1994
Abnormal QRS Axis VSD • 16. Anderson RH, Lenox CC, Zuberbuhler JR: Mechanisms of closure of perimembranous ventricular septal defect. Am J Cardiol 52:341, 1983 17. Magherini A, Urciuolo A, Tommassini CR, et al: Restrictive tissue in the area of perimembranous ventricular septal defect. Cross-sectional and doppler echocardiographic study. Eur Heart J 11:601, 1990 18. Riemenschneider TA, Moss AJ: Left ventricular-right atrial communication. Am J Cardiol 19:710, 1967 19. Riemenschneider TA: Left ventricular-right atrial communication, in Adams FH, Emmanouilides GC (eds): Moss Heart Disease in Infants, Children and Adolescents. Williams & Wilkins Co, Baltimore, MD, 1983, p 154 20. Shakibi JG, Aryanpur I, Paydar M, et al: The vectorcardiogram as an aid to diagnosis in left ventricularright atrial communication. J Electrocardiol 10:337, 1977 21. Feldt RH, DuShane JW, Titus JL: The atrioventricular conduction system in persistent common atrioventricular canal defect. Circulation 42:437, 1970
Tutar et al. 203
22. Boineau JP, Moore EN, Patterson DF: Relationship between the ECG, ventricular activation, and the ventricular conduction system in ostium primum ASD. Circulation 48:556, 1973 23. Farru-Albohaire O, Arcil G, Hernandez I: An association between left axis deviation and an aneurysmal defect in children with a perimembranous ventricular septal defect. Br Heart J 64:146, 1990 24. Ueda M, Becker AE: Morphological characteristics of perimembranous ventricular septal defects and their surgical significance. Int J Cardiol 8:149, 1985 25. Heller J, Hagege AA, Besse B, et al: “Crochetage” (notch) on R wave in inferior limb leads: A new independent electrocardiographic sign of atrial septal defect. J Am Coll Cardiol 27:877, 1996 26. Feldt RH, Porter CJ, Edwards WD, et al: Atrioventricular septal defect, in Emmanouilides GC, Riemenschneider TA, Allen HD (eds): Heart Disease in Infants, Children, and Adolescents. Williams & Wilkins Co., Baltimore, MD, 1995, p 704
QRS Axis in Isolated Perimembranous Ventricular Septal Defect and Influences of Morphological Factors on QRS Axis
H. Ercan Tutar, MD, Semra Atalay, MD, Sadi Tu¨rkay, MD, and Ayten Imamoglu, MD
Abstract: To detect the frequency of left axis deviation (LAD) in isolated perimembranous ventricular septal defects (VSD) we retrospectively analyzed electrocardiograms of 59 patients, aged 8 months to 15 years. Patients were grouped into those having ventricular septal aneurysm (VSA) formation (n:20) and those who did not have VSA (n:39). Patients with VSA were then stratified into 2 groups according to the presence of left ventricular-to-right atrial (LV-RA) shunt. Four hundred healthy children served as the control group. We found that 12 (20.3%) of 59 patients with isolated perimembranous VSD had a LAD. Five of 6 patients with perimembranous outlet VSD and 6 with perimembranous inlet VSD had abnormal LAD with a qR pattern in I and aVL and rS in aVF. Abnormal LAD was more prevalent in patients with VSA (40%) than without VSA (7.7%) (P ⬍ .01). We also found that mild right ventricular hypertrophy (RVH) with the rsRⱊ or rSRⱊ pattern in V1 was more frequent in patients with VSA, especially those who had LV-RA shunt. However, we could not find significant difference between patients with or without LV-RA shunt for the incidence of abn LAD and mild RVH. Localization of perimembranous VSD was not found to have an effect on frequency of abnormal LAD and mild RVH in this patient group. In patients with clinical findings of VSD, the existence of abnormal LAD especially if it is associated with mild RVH, should raise the possibility of perimembranous VSD with VSA formation. Key words: Perimembranous ventricular septal defect, ventricular septal aneurysm, left ventricular-to-right atrial shunt, left axis deviation.
Abnormal left axis deviation (abn LAD) of QRS axis in frontal plane is uncommon in childhood and is usually associated with complex congenital heart
disease such as atrioventricular septal defect, tricuspid atresia due to posterior and inferior displacement of the conduction system resulting in inferior initial activation and a larger superior terminal activation (1). Left axis deviation frequency in healthy infants and children was reported in a range of 0.4% to 1.4% (2– 4) was found in 6.2% to 10.1% of patients with ventricular septal defect (VSD) (5,6). We noticed that abn LAD was frequent in patients who had left ventricular-to-right atrial (LV-RA) shunt associated with perimembranous
From the Ankara University, Medical School, Department of Pediatric Cardiology, Ankara, Turkey. Reprint requests: H. Ercan Tutar, MD, Defne sitesi 8. Blok No: 37, Umitkoy, 06530, Ankara, Turkey; e-mail: [email protected]. Copyright © 2001 by Churchill Livingstone® 0022-0736/01/3403-0003$35.00/0 doi:10.1054/jelc.2001.24763
197
198 Journal of Electrocardiology Vol. 34 No. 3 July 2001 VSD. We, therefore, retrospectively analyzed electrocardiograms of patients with isolated perimembranous VSD to detect the frequency of abn LAD, especially in a subgroup of patients with LV-RA shunt.
Patients and Methods Electrocardiograms of 59 patients (29 girls, 30 boys), aged 8 months to 15 years (6.4 ⫾ 4.0 years) who had perimembranous VSD were retrospectively analyzed. Patients with any associated congenital heart disease and/or Down’s syndrome were excluded. Cardiac catheterization and left ventricular angiography were performed in 39 patients. Eighteen patients with isolated perimembranous VSD who had pulmonary hypertension were also excluded. Pulmonary hypertension was defined as: 1) Mean pulmonary artery pressure ⬎ 25 mm Hg during cardiac catheterization; or 2) In patients without cardiac catheterization, pulmonary artery systolic pressure ⬎ 35 mm Hg, which was calculated from peak systolic continuous wave Doppler pressure gradient of interventricular shunt through the ventricular septal defect. Echocardiographic studies were performed with Toshiba Sonolayer SSH 140A machine (Toshiba Corp, Japan) with 3.75 MHz and 5 MHz transducers. Standard and modified parasternal, apical, and subcostal echocardiographic windows were used. The classification of perimembranous VSDs according to their exten-
Fig. 1. Two-dimensional echocardiogram (parasternal four-chamber view) shows perimembranous inlet ventricular septal defect and ventricular septal aneurysm (arrows). The septal leaflet of the tricuspid valve is involved in the formation of the ventricular septal aneurysm. LV, left ventricle; RV, right ventricle; LA, left atrium; RA, right atrium.
sions, as inlet, outlet, and trabecular was based on the previous reports (7–9). Ventricular septal aneurysm (VSA) formation of perimembranous VSD was defined as echocardiographic evidence of deposition of tissue around the margins of VSD, producing a sac-like structure with distinct margins that seemed to partially occlude the flow through the VSD (Fig. 1). LV-RA shunt was defined as echocardiographic evidence of a high velocity turbulence moving from left ventricle, through the area of VSD directly into right atrium (10,11). The peak velocity of the LV-RA and interventricular shunt was measured by continuous wave Doppler. The pressure gradient of LV-RA shunt was equivalent to the pressure difference between systolic left ventricular and right atrial pressures and could be differentiated from tricuspid regurgitation by the high velocity of the LV-RA shunt in the absence of elevated right ventricular pressure (11). M-mode demonstration of tricuspid valve systolic flutter was used as a supportive evidence of LV-RA shunting (11–13). Standard 12-lead electrocardiograms were obtained with a three-channel electrocardiographic recorder (Hewlett Packard, model 4745 A; Hewlett Packard Inc, Camas, WA). QRS axis in the frontal plane was calculated and recorded with the other features of the tracings. QRS axis less than 0° was defined as LAD (1). Abn LAD or abnormal superior QRS axis was defined as an inferior initial activation and a larger superior terminal activation (qR pattern in I and aVL, rS pattern in aVF). Patients were divided into 2 groups according to the existence of
Abnormal QRS Axis VSD •
Tutar et al. 199
Table 1. Demographic and Hemodynamic Parameters and Distribution of Ventricular Septal Defect Types According to the Patient’s Group in Patients With Isolated Perimembranous Ventricular Septal Defect
VSA (⫹) (n:20) LV-RA shunt (⫹) (n:8) LV-RA shunt (⫺) (n:12) VSA (⫺) (n:39) P
Trabecular
Mean PA Pressure (mmHg)
Qp/Qs
1 (5) 1 (12.5) ⫺ 2 (5) ⬎.05
16.8 ⫾ 4.6 (n:12) 16.0 ⫾ 5.1 (n:6) 17.7 ⫾ 4.5 (n:6) 18.7 ⫾ 4.2 (n:27) ⬎.05
1.50 ⫾ 0.39 (n:12) 1.65 ⫾ 0.34 (n:6) 1.34 ⫾ 0.41 (n:6) 1.65 ⫾ 0.37 (n:27) ⬎.05
Perimembranous VSD, n (%)
Age (years)
Sex (M/F)
Inlet
Outlet
5.1 ⫾ 4.0 5.2 ⫾ 3.2 5.0 ⫾ 4.6 7.0 ⫾ 3.9 ⬎.05
9/11 3/5 6/6 21/18 ⬎.05
7 (35) 2 (25) 5 (42) 15 (38.5) ⬎.05
12 (60) 5 (62.5) 7 (58) 22 (56.5) ⬎.05
VSA, ventricular septal aneurysm; LV-RA, left ventricular-right atrial; Qp, pulmonary blood flow; Qs, systemic blood flow; PA, pulmonary artery; (⫹) denotes presence, (⫺) denotes absence.
echocardiographic signs of VSA formation. Patients with VSA were then divided into 2 groups whether they had LV-RA or did not have LV-RA shunt. Electrocardiograms were obtained from 400 (182 girls, 228 boys) healthy children aged 5 years to 18 years (10.2 ⫾ 2.4 years) who served as control group. All data are expressed as mean ⫾ standard deviation. Chi-square test, Fisher’s exact test, unpaired Student’s t-test and Kruskall Wallis One-Way ANOVA were used where appropriate for statistical comparisons. A two-tailed P value ⬍ .05 was considered statistically significant.
Results Demographic data and hemodynamic parameters of the patients and distribution of perimembranous VSD types according to their location are shown in Table 1. Echocardiographically 22 (37.3%) patients had perimembranous inlet, 34 (57.6%) patients had perimembranous outlet, and 3 (5.1%) had perimembranous trabecular VSD. No patient had cleft septal tricuspid valve leaflet. There was no significant difference in mean age, mean pulmonary artery pressure, and the mean pulmonary/
systemic blood flow ratio between patient groups. We also did not find difference in distribution of VSD types between patient groups. The electrocardiographic features of patient groups and control subjects are summarized in Table 2. The mean QRS axis in patients with VSD especially in with VSA was lower, but the difference did not reach the statistically significant level (P ⫽ .07). Twelve (20.3%) of patients with VSD had LAD in which 11 had abn LAD, but only 6 (1.5%) of children in control group had (P ⬍ .001). In patients with LAD, QRS axis in the frontal plane ranged from ⫺5° to ⫺60°. Except for 1 patient who had perimembranous outlet VSD without VSA formation, all patients had abn LAD in the frontal plane. Abn LAD was found in 8 (40%) of patients with VSA, but in 3 (7.7%) of patients without VSA (P ⬍ .01). Mild right ventricular hypertrophy (RVH) (rsR⬘, rSR⬘ pattern in V1) was also frequent in patients with VSD. It was found in 15 (25.4%) patients with VSD, but in 7 (1.75%) healthy children (P ⬍ .001). We found that 11 (55%) patients with VSA had mild RVH, but only 4 (10.2%) patients without VSA had it (P ⬍ .001). We could not find significant difference between patients with or without LV-RA shunt for the incidence of abn LAD or mild RVH (P ⫽ .64, P ⫽ .19, respectively).
Table 2. Electrocardiographic Findings of Controls and Patients With Isolated Perimembranous Ventricular Septal Defect According to the Patient’s Group
VSA (⫹) (n:20) LV-RA shunt (⫹) (n:8) LV-RA shunt (⫺) (n:12) VSA (⫺) (n:39) Control (n:400)
QRS Axis (°)
Abn LAD n (%)
Mild RVH n (%)
Abn LAD ⫹ Mild RVH n (%)
RAE n (%)
LVH n (%)
BVH n (%)
28.5 ⫾ 55.7 20 ⫾ 60.8 34.2 ⫾ 54.1 46.9 ⫾ 37.9 60.9 ⫾ 18.1
8 (40) 4 (50) 4 (33.3) 3 (7.7) 6 (1.5)
11 (55) 6 (75) 5 (42) 4 (10.2) 7 (1.75)
4 (20) 2 (25) 2 (16.7) 1 (2.5) 1 (0.25)
6 (30) 5 (62.5) 1 (8.3) 1 (2.5) ⫺
5 (25) 2 (25) 3 (25) 11 (28.2) ⫺
2 (10) ⫺ 2 (16.7) 1 (2.5) ⫺
VSA, ventricular septal aneurysm; LV-RA, left ventricular-right atrial; abn LAD, abnormal left axis deviation; RVH, right ventricular hypertrophy; RAE, right atrial enlargement; LVH, left ventricular hypertrophy; BVH, biventricular hypertrophy; (⫹) denotes presence, (⫺) denotes absence.
200 Journal of Electrocardiology Vol. 34 No. 3 July 2001
Fig. 2. The electrocardiogram of a 4-year-old patient with perimembranous outlet ventricular septal defect and left ventricular-toright atrial shunt shows abnormal left axis deviation and mild right ventricular hypertrophy (rsR⬘s⬘ pattern in V1).
Electrocardiographic finding of right atrial enlargement (tall, peaked P waves in DII, V1) was more frequent (62.5%) in patients with LV-RA shunt than other patient groups; patients with VSA but without LV-RA shunt (8.3%, P ⬍ .05), patients without VSA (2.5%, P ⬍ .001). Twenty percent of the patients with VSA had both abn LAD and mild RVH (Fig. 2); this association was only found in 2.5% of patients without VSA (P ⬍ .001). Most of the patients with isolated perimembranous VSD did not show electrocardiographic evidence of ventricular hypertrophy; 25% of our patients showed slight left ventricular hypertrophy and 5% of patients showed biventricular hypertrophy (Table 2). There was no significant difference in the frequency of abn LAD, mild RVH, and right atrial enlargement among the perimembranous VSD types (Table 3). Detailed electrocardiographic data of patients with LAD are shown in Table 4.
Discussion Although more defects in the muscular ventricular septum close spontaneously than perimembranous VSDs, a good proportion of the latter show a tendency to diminish in size to close (14). This closure or diminishing in size is usually a result of VSA formation. Ventricular septal aneurysm forma-
tion was found in up to three fourths of patients with perimembranous VSD (15). Both pathologic (16) and echocardiographic studies (17) showed that in most perimembranous VSDs, the tissue forming VSA, derived from tricuspid valve leaflets. In a large series of 930 patients with isolated perimembranous VSD, LV-RA shunt was shown in approximately 15% of patients after the occurrence of VSA formation (15). In those patients, the mechanism responsible for the occurrence of LV-RA shunt was usually described as widening of anteroseptal commissure of tricuspid valve. Malformation, cleft, and perforation of the tricuspid valve were also reported as responsible mechanisms for the occurrence of LV-RA shunt (13,15,18).
Table 3. Electrocardiographic Findings of Patients With Isolated Perimembranous Ventricular Septal Defect According to the Ventricular Septal Defect Types
Perimembranous inlet (n:22) Perimembranous outlet (n:34) Perimembranous trabecular (n:3) P value
Abn LAD n (%)
Mild RVH n (%)
RAE n (%)
QRS Axis (°)
6 (27.3)
6 (27.3)
2 (9.1)
30.2 ⫾ 48.2
5 (14.7)
8 (23.5)
4 (11.8) 47.3 ⫾ 42.1
0 (0) .34
1 (33.3) .90
1 (33.3) 41.6 ⫾ 59.2 .47 .36
Abn LAD, abnormal left axis deviation; RVH, right ventricular hypertrophy; RAE, right atrial enlargement.
Abnormal QRS Axis VSD •
Tutar et al. 201
Table 4. Electrocardiographic Characteristics of 12 Patients With Left Axis Deviation According to the Ventricular Septal Defect Types Type of VSD Perimembranous inlet
Perimembranous outlet
Patient No.
QRS Axis (°)
Abn LAD
aVF Pattern (V)
Mild RVH
RAE
LVH
1 2 3 4 5 6 7 8 9 10 11 12
⫺50 ⫺20 ⫺40 ⫺15 ⫺50 ⫺40 ⫺5 ⫺40 ⫺10 ⫺60 ⫺20 ⫺10
⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹
rS (150/800) rS (400/600) rS (200/700) rS (300/500) rS (400/1000) rS (250/800) qrS (100/300/400) rS (500/900) rS (450/550) rS (450/900) rS (300/400) rS (600/700)
⫹ ⫺ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫺
⫹ ⫺ ⫹ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺
⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫺
Abn LAD, abnormal left axis deviation; RVH, right ventricular hypertrophy; RAE, right atrial enlargement; LVH, left ventricular hypertrophy; (⫹) denotes presence, (⫺) denotes absence.
Electrocardiographic evidence of right atrial enlargement with a prolonged PR interval, mild RVH, and biventricular hypertrophy were all frequently reported in patients with LV-RA communication (18,19). We also found right atrial enlargement (62.5%) and mild RVH (75%) frequently in the electrocardiograms of our patients with LV-RA shunt. This type of defect results in volume overload of right heart chambers; right atrial enlargement and mild RVH are the predictable electrocardiographic findings. Abn LAD was reported in about 25% of patients with LV-RA shunt usually in association with combined ventricular hypertrophy in the atrioventricular canal type of defect (18 –20). In our patient group, abn LAD was found in 4 (50%) of 8 patients with LV-RA shunt (Table 2). Left axis deviation was reported in 0.4% to 1.4% of healthy children and adolescents (2– 4). In our control group of 400 healthy children we found LAD in 1.5% incidence (Table 2). Left axis deviation was seen in up to 10% of patients with isolated ventricular septal defect (5,6). In a few studies concerning the relation between the localization of VSD and LAD, authors pointed out that LAD was more common when VSD was in the inlet portion of the ventricular septum or affected this area (5,6). However, Shaw et al. (5) showed that 41% of patients with LAD had the defect at a site other than the perimembranous inlet septum. In our study group, 11 (18.6%) patients with isolated perimembranous VSD had abn LAD. In 6 of 12 patients, the defect was classified as perimembranous inlet and the other 5 as perimembranous outlet. Although, in patients with perimembranous inlet VSD, the frequency of abn LAD (27.3%) was slightly higher than with other defect locations, the difference did not reach the statistically significant level (Table 3).
Abn LAD in patients with atrioventricular septal defect is a well-known feature and is believed to be associated with posteroinferior displacement of the atrioventricular conduction system as a result of anatomical characteristics of defect (21,22). Whether the anatomical disconfiguration of atrioventricular conduction system, like in atrioventricular septal defect, exists in VSDs associated with abn LAD is an area of interest of pathological studies. Although we did not find any relation between abn LAD and defect location, we found significant difference between patients with or without VSA formation in the frequency of abn LAD. The distribution of defect location was also not different between the patients with or without VSA. We could find only 1 study that explained the relation between VSA formation and LAD frequency in the English literature (23). Farru-Albohaire et al. (23) found in two thirds of patients with VSA had LAD; however, none without VSA had it (23). In that study, the occurrence of LAD was a new finding, which appeared after the development of VSA in 10 patients. We do not know whether or not the abn LAD was present before the VSA development in our patients. Authors then concluded that the appearance of LAD in a patient with perimembranous VSD indicated a spontaneous reduction in the defect by VSA formation (23). Unfortunately, there was no information about the classification of perimembranous VSDs in that study. We as well as Farru-Albohaire et al. (23) have not been able to explain the development of abn LAD after the occurrence of VSA by the adherence of tricuspid valve leaflets to the rim of the defect. Ueda and Becker (24) investigated the anatomy of atrioventricular conduction system in perimembranous VSDs and showed a close relation
202 Journal of Electrocardiology Vol. 34 No. 3 July 2001 between the conduction bundle and the rim of the defect and the anomalous tricuspid valve attachments in 2 heart specimens. In one specimen, conduction bundle extraordinarily infiltrated the leaflet tissue (24). So, one can speculate that if the tricuspid valve leaflet which contributes to the VSA formation contains the conduction tissue, this may be responsible for the development of abn LAD. However, we could not show the appearance of abn LAD after the development of VSA. So, abn LAD might have been present before the occurrence of septal aneurysm. Although the QRS axis tends to be more leftward than is normal for the patient’s age with left ventricular hypertrophy, abn LAD is rare in childhood (1). Abnormal LAD in our patients cannot be attributed to left ventricular hypertrophy because only small percent of these patients showed electrocardiographic signs of mild left ventricular hypertrophy. The mild RVH pattern was found in electrocardiograms of normal children in up to 7% of the cases (1). In our control group, we found 1.75% of children had mild RVH. The frequency of this pattern was reported 20% to 30% of patients with VSD (6,25). Mild RVH was present in 75% of our cases with LV-RA shunt as expected. However, we showed almost half of the patients with VSA without LV-RA shunt had also mild RVH pattern. This pattern was documented only 10.2% of patients without VSA. Farru-Albohaire et al. (23) showed that mild RVH was present in 34% of patients with VSA, but only in 6.8% of without VSA similar to our results. We found that half of our patients with VSA and abn LAD also had mild RVH (Table 2). Farru-Albohaire et al. (23) also reported that 48% of their patients with VSA and LAD had mild RVH but none without VSA. Although abn LAD associated with mild RVH pattern is considered a distinct electrocardiographic feature of atrioventricular septal defect (26), it can also be found in patients with VSD associated with VSA, like in ours. We believe that mild RVH (rsRⱊ or rSRⱊ pattern in V1) is the result of volume overload of right ventricle in our cases. We showed that abn LAD was frequent in patients with LV-RA shunt. However, this finding was not limited to this group. Abn LAD was also observed more frequently in patients with VSA without LV-RA shunt than in patients without VSA. Mild RVH was also a frequent finding especially in patients with VSA associated with LV-RA shunt. We suggest that in patients with clinical findings of VSD, the existence of abn LAD, especially if it is
associated with mild RVH, should raise the possibility of perimembranous VSD with VSA formation.
References 1. Garson AJ: Electrocardiography, in Garson AJ, Bricker JT, Fisher DJ, (eds): The Science and Practice of Pediatric Cardiology. Williams & Wilkins, Baltimore, MD, 1998, p 735 2. Gup AM, Franklin RB, Hill JE: The vectorcardiogram in children with left axis deviation and no apparent heart disease. Am Heart J 69:619, 1965 3. Calcaterra G, Puglisi R: Left axis deviation in healthy infants and children. Int J Cardiol 24:236, 1989 4. Pearl W, Stafford EM: Left axis deviation in children. Int J Cardiol 34:119, 1992 5. Shaw NJ, Godman MJ, Hayes A, et al: Superior QRS axis in ventricular septal defect. Br Heart J 62:281, 1989 6. Ba¨ckman H: Influence of structural and functional features of ventricular septal defect on frontal plane QRS axis of the electrocardiogram. Br Heart J 34:274, 1972 7. Soto B, Becker AE, Moulaert AJ, et al: Classification of ventricular septal defects. Br Heart J 43:332, 1980 8. Sutherland GR, Godman MJ, Smallhorn JF, et al: Ventricular septal defects: Two dimensional echocardiographic and morphological correlations. Br Heart J 47:316, 1982 9. Helmcke F, Souza A, Nanda NC, et al: Two-dimensional and colour doppler assessment of ventricular septal defect of congenital origin. Am J Cardiol 63: 1112, 1989 10. Leung MP, Mok CK, Lo RNS, et al: An echocardiographic study of perimembranous ventricular septal defect with left ventricular to right atrial shunting. Br Heart J 55:45, 1986 11. Wu MH, Wu JM, Chang CI, et al: Implication of aneurysmal transformation in isolated perimembranous ventricular septal defect. Am J Cardiol 72:596, 1993 12. Nanda NC, Gramiak R, Manning JA: Echocardiography of the tricuspid valve in congenital left ventricular-right atrial communication. Circulation 51:268, 1975 ¨ zkutlu S, Pas¸aog˘lu I: Echocardiographic 13. Sarac¸lar M, O diagnosis of congenital left ventricular-right atrial communication. Jpn Heart J 29:671, 1988 14. Shirali GS, Smith EO, Geva T: Quantitation of echocardiographic predictors of outcome in infants with isolated ventricular septal defect. Am Heart J 130: 1228, 1995 15. Wu MH, Chang CI, Wang JK, et al: Characterization of aneurysmal transformation in perimembranous ventricular septal defects: An adhered anterior leaflet of tricuspid valve predisposes to the development of left ventricular-to-right atrial shunt. Int J Cardiol 47:117, 1994
Abnormal QRS Axis VSD • 16. Anderson RH, Lenox CC, Zuberbuhler JR: Mechanisms of closure of perimembranous ventricular septal defect. Am J Cardiol 52:341, 1983 17. Magherini A, Urciuolo A, Tommassini CR, et al: Restrictive tissue in the area of perimembranous ventricular septal defect. Cross-sectional and doppler echocardiographic study. Eur Heart J 11:601, 1990 18. Riemenschneider TA, Moss AJ: Left ventricular-right atrial communication. Am J Cardiol 19:710, 1967 19. Riemenschneider TA: Left ventricular-right atrial communication, in Adams FH, Emmanouilides GC (eds): Moss Heart Disease in Infants, Children and Adolescents. Williams & Wilkins Co, Baltimore, MD, 1983, p 154 20. Shakibi JG, Aryanpur I, Paydar M, et al: The vectorcardiogram as an aid to diagnosis in left ventricularright atrial communication. J Electrocardiol 10:337, 1977 21. Feldt RH, DuShane JW, Titus JL: The atrioventricular conduction system in persistent common atrioventricular canal defect. Circulation 42:437, 1970
Tutar et al. 203
22. Boineau JP, Moore EN, Patterson DF: Relationship between the ECG, ventricular activation, and the ventricular conduction system in ostium primum ASD. Circulation 48:556, 1973 23. Farru-Albohaire O, Arcil G, Hernandez I: An association between left axis deviation and an aneurysmal defect in children with a perimembranous ventricular septal defect. Br Heart J 64:146, 1990 24. Ueda M, Becker AE: Morphological characteristics of perimembranous ventricular septal defects and their surgical significance. Int J Cardiol 8:149, 1985 25. Heller J, Hagege AA, Besse B, et al: “Crochetage” (notch) on R wave in inferior limb leads: A new independent electrocardiographic sign of atrial septal defect. J Am Coll Cardiol 27:877, 1996 26. Feldt RH, Porter CJ, Edwards WD, et al: Atrioventricular septal defect, in Emmanouilides GC, Riemenschneider TA, Allen HD (eds): Heart Disease in Infants, Children, and Adolescents. Williams & Wilkins Co., Baltimore, MD, 1995, p 704