Atrio-ventricular septal angle predicts successful septation repair of patients with straddling atrioventricular valves

Presenting Author(s)

• 

  Sara Santiaguel, MD

  Pediatric Cardiology Fellow
  The University of Texas Health Science Center at Houston
  Houston, Texas, United States

Primary Author(s)

• 

  Sara Santiaguel, MD

  Pediatric Cardiology Fellow
  The University of Texas Health Science Center at Houston
  Houston, Texas, United States

Co-Author(s)

• AV

  Ana L. Vasquez Choy, MD

  Pediatric Advanced Cardiac Imaging Fellow
  The University of Texas Health Science Center at Houston, Texas, United States
• 

  Mehul D. Patel, MD

  Assistant Professor
  The University of Texas Health Science Center at Houston
  Houston, Texas, United States

Background: Straddling atrioventricular valves (AVV) are rare and complex defects, managed either by single ventricle (SV) palliation or septation with closure of the ventricular septal defect. Abnormally wide atrioventricular septal angle (AVSA) has been described in heart specimens with a straddling tricuspid valve. AVSA using cardiac magnetic resonance (CMR), computed tomography (CT) or echocardiogram (ECHO) has not been described. We hypothesize that a narrow AVSA is associated with successful septation repair in patients with straddling AVV.

Methods: Retrospective cohort of pediatric patients with straddling AVV matched 1:1 with patients of similar anatomy. Demographics, diagnoses, procedures, CMR/CT/ECHO data, and outcomes were collected. The composite primary outcome was septation with survival by age 2. AVSA was measured as the angle between the interventricular and interatrial septal planes in the short axis (Figure 1) on CMR/CT/ECHO images. The straddling AVV area index (SAVVi) was measured on CMR/ECHO images and calculated as the ratio of the AVV area over the contralateral ventricle divided by the total AVV area. Measurements were performed by a fellow and experienced pediatric cardiologist to assess reproducibility.

Results: Twenty-four patients were included in the study; 50% had a straddling AVV, 58% underwent septation, and 33% SV palliation. Survivorship at age 2 of straddling AVV patients was 92%. Septated patients had significantly lower CMR/CT AVSA (mean 28.4° ± 18.4°, Table 1) than those with SV palliation/death (mean 56.2° ± 16.6°, p=0.002, Figure 2A). Findings were similar for ECHO AVSA: septation mean 29.5° ± 17.5° versus SV palliation/death mean 57.5° ± 16.9° (p< 0.001). ECHO and CMR/CT AVSA moderately correlated (r=0.72, p=< 0.001, Figure 2B) with overall good agreement (mean difference -2.1° ± 7.5°, Figure 2C). AVSA Inter-rater reproducibility was fair in CMR/CT (ICC 0.71, 95% CI 0.42-0.87) and good in ECHO (ICC 0.82, 95% CI 0.63-0.92). Among straddling AVV patients, septated cases trended towards a lower ECHO SAVVi (median 0.108, IQR 0.99-1.27, n=3) than those with SV palliation/death (median 0.662, IQR 0.584-0.740, n=9, p=0.064), but not statistically significant. Type of straddling AVV chordal connections, papillary muscle connections, or degree of nondominant ventricular hypoplasia were not associated with successful septation.

Conclusion: In patients with straddling AVV or similar cardiac defects, a narrow CMR/CT AVSA is associated with ventricular septation repair or survival by age 2. This non-invasive morphometric measure is feasible on both CMR/CT and ECHO, shows moderate correlation between modalities, and has fair reproducibility. Survival in straddling AVV is overall good, with no single anatomic feature consistently predicting successful septation. Combined with other morphologic features, AVSA and SAVVi may help guide counseling and surgical planning.

Table 1. Narrow CMR/CT AVSA and echocardiographic AVSA moderately correlate, have good Bland-Altman agreement, and are associated with successful ventricular septation and survival by age 2.

Unpaired T-test	N	T-Statistic	Effect Size	AVSA Mean	Standard Deviation	p-value
CT/CMR AVSA 2-V repair	12	-3.58	-1.58	28.4°	18.4°	0.002
CT/CMR AVSA SV/mortality	9			56.2°	16.6°
Echo AVSA 2-V repair	14	-3.92	-1.62	29.5°	17.5°	< 0.001
Echo AVSA SV/mortality	10			57.5°	16.9°
Pearson Correlation		Pearson’s r	p-value
Echo AVSA & CT/CMR AVSA	21	0.723	0.0001
Bland-Altman agreement		Mean difference estimate (SD)	95% CI upper limit of agreement	95% CI lower limit of agreement
Echo AVSA & CT/CMR AVSA	21	-2.1° ± 7.5°	-9.65	5.37
Mann-Whitney U (non-normative)	N	Statistic	Effect Size	Median	IQR	p-value
CT/CMR AVSA 2-V repair	3	3	0.778	0.108	0.99-1.27	p=0.06
CT/CMR AVSA SV/mortality	9			0.662	0.584-0.740

Figure 1. CMR images in short-axis plane at mid-ventricle (1A) and mid-atrium (1B) demonstrating the method of AVSA measurement. The interventricular septum (IVS, orange) is defined as the plane intersecting the superior and inferior septal insertion points regardless of septal flattening. The inferior angle between the plane of the interventricular septum and interatrial septum (IAS, blue) is measured.
Figure 2. Statistical analysis