This study shows that left ventricular diastolic function improves towards normal values after ASD closure in children by increasing the hydraulic force, as evaluated by AVADi. Although AVADi improved after ASD closure, it was still lower than in controls 7 months thereafter. The decreased left ventricular filling may contribute to decreased exercise capacity [8, 17]. It may be hypothesized that earlier ASD closure, improving diastolic filling, improves long-term outcome. Therefore, in patients where atrioventricular area difference is low, ASD closure may help avoid diastolic function deterioration and improve outcome. This is likely important also in patients with small shunt volumes, who currently do not undergo ASD closure. Changes in clinical routine may be considered pending larger outcome studies.

The current study shows that both children with ASD and controls have negative AVADi in the beginning of diastole suggesting a hydraulic force acting towards the ventricle. Early in diastole, AVADi becomes positive leading to a change of direction of the hydraulic force, which during the main part of diastole thus works towards the atrium. This means that the movement of the atrioventricular plane towards the base of the heart is facilitated and filling of the left ventricle during diastole is augmented. Although this mechanism is valid both in children with ASD and in controls, the current study also shows that AVADi at end diastole is lower in patients after ASD closure as compared to controls. In children with ASD, the time with positive AVADi and thus a hydraulic force towards the base of the heart augmenting diastolic filling, was shorter during diastole than in controls. Together with a higher heart rate in children with ASD, these mechanisms suggest that diastolic dysfunction in children with ASD is likely augmented by a reduced hydraulic force.

The anatomical reason for a decreased AVADi in patients with ASD is mainly the small left ventricle, since the size of the atrium is normal, as confirmed in the current study. This is in line with previous studies showing that patients with ASD have diastolic impairment and increased sympathetic tone, which is likely related to the underfilled left ventricle [4, 8, 18]. The atrial function also affects the AVAD, reflected by an increase in AVAD at late diastole caused by the atrial contraction, as shown in the examples in Fig. 3.

After ASD closure, the left ventricle increased in both end-systolic and end-diastolic volumes as preload increased, in line with previous studies [19]. Low left ventricular compliance and abnormal filling properties have been reported several years after closure in this context [20], which might also explain the decreased exercise capacity these patients experience decades after ASD closure [2, 8]. In patients with ASD left ventricular volume decreases with age with a concomitant decrease in stroke volume and diastolic function but disproportionate to left ventricular mass [21]. These age-related changes, in addition to left ventricular remodeling due to LV underfilling related to the left-to-right shunt in patients with ASD, predispose the low AVADi before ASD closure. In the current study AVADi was still lower in patients after ASD closure than in controls despite that the patients had normal atrial and ventricular volumes after ASD closure. This suggests that the shape of the left ventricle and/or atrium might also be abnormal as AVAD is based on the area differences and not volumes.

In the current study of children and adolescents, the ASD lesion was mainly an isolated defect, and only in rare cases associated with partial anomalous pulmonary venous drainage without obstruction. There was no sign of clinically apparent arterial or pulmonary hypertension based on clinical assessments, including echocardiography. Further, there was no additional right-sided heart lesion that otherwise could have altered right atrial pressure, nor any left-sided obstructive lesions such as mitral stenosis. These relatively young patients had no left atrial enlargement, which otherwise seems do develop over time [19, 22]. A dilated left atrium in addition to a small left ventricle would diminish the hydraulic force even more and might thus further increase diastolic dysfunction over time. A small restrictive left ventricle could also explain the increased atrial pressure seen at exercise, and often already at rest [20], which predisposes to the increased risk for atrial enlargement and thus atrial fibrillation with increasing age.

The independent prognostic value of AVADi in pediatric patients with ASD and after ASD closure remains to be investigated. Nevertheless, a decrease in AVAD and thus a decrease in hydraulic force, assessed by echocardiography in adults with preserved systolic left ventricular function but with diastolic dysfunction, was linked to increased overall mortality at 5-years follow-up with a hazard ratio of 20–33% [23]. Decreased hydraulic force was also shown to have adverse prognostic value beyond the common measures of diastolic dysfunction in that population of more than 5,000 patients [23]. Thus, it may be hypothesized that AVAD will add prognostic information also in patients with ASD.

According to the current guidelines for ASD closure, the indication relies on the echo-assessed size of the right atrium and right ventricle based on echocardiography. Including AVAD data for decision making could be of particular importance in patients with smaller ASD’s and borderline right atrial and right ventricular sizes. The current study suggests that ASD closure may be beneficial when AVADi is low even if the left-to-right shunt does not cause right ventricular dilatation and, according to the current guidelines, is considered hemodynamically non-significant [24]. One child in this study had a shunt volume below treatment threshold and was therefore excluded from the case group analysis. A separate analysis, however, showed lower AVADi (8.5 cm2/m) than in controls, which suggests that the left ventricle actually was affected in this child. Since the hydraulic force contributes to diastolic filling, ASD closure leading to an increase in AVAD as shown in the current study may be assumed to substantially improve diastolic function. This also agrees with the findings of Maagaard et al. showing reduced cardiac index during exercise in adults with unrepaired ASD and small or spontaneously closed ASD [25]. As of today there are no pediatric reference values for AVADi but the current study suggests, although the population is limited in numbers and spread of age, that the lower reference value is approximately 10 cm2/m, which means that it is likely similar to that of adults [5]. The current results also support ASD closure at a young age to minimize the duration of an underfilled left ventricle. Future longitudinal studies might reveal if AVAD has additional prognostic information in patients with ASD.