Left Atrium Measurements via Computed Tomography Pulmonary Angiogram as a Predictor of Diastolic Dysfunction

Abstract

Purpose: Left atrium (LA) enlargement on echocardiography may be an indicator of diastolic dysfunction (DD). It is not well known if computed tomography pulmonary angiography (CTPA) can detect DD.

Methods: A total of 127 patients who underwent both CTPA and echo within 48 hours were analyzed retrospectively. Left atrium diameters from CTPA were correlated with echo and evaluated against degrees of DD. Computed tomography pulmonary angiography pulmonary artery (PA)/aorta ratio was analyzed as a tool to detect pulmonary hypertension.

Results: There were 42% of patients who had DD. There was a strong correlation between LA size on CTPA and echo (r = 0.78). An LA greater than 4.0 cm gave a sensitivity of 68.1% and specificity of 73.9% for DD detection. A PA/aorta cutoff greater than 0.84 yielded a sensitivity of 84% and specificity of 33% for pulmonary hypertension.

Conclusions: Computed tomography pulmonary angiography measurements of LA and PA/aorta ratio correlate strongly with equivalent findings on echo. We suggest that LA and PA/aorta measurements be included on chest CTPA reports.

Fig. 1. Inclusion and Exclusion Criteria

200 charts were originally selected for study based on inclusion criteria. 13 were eliminated as they were just outside of the 48-hour restriction, 45 had incomplete CTPA data. 15 of the remaining 142 charts were not included as they lacked LA measurements in their echo reports, leaving a total of 127 charts in the final analysis

Fig. 2. Left atrium and pulmonary artery measurements using CTPA

Panel A: CTPA left atrial measurement using the maximum anterior-posterior diameter of the midline in its middle 50%; Panel B: CTPA pulmonary artery measurement using the widest diameter perpendicular to the long axis at the pulmonary artery bifurcation. Ascending aorta measurement was taken at the same level.

Fig. 3. Left atrial sizes measured using CTPA correlated strongly with left atrial sizes measured with echo

Panel A: Left atrial sizes from CTPA were correlated against left atrial sizes gleaned from echo reports. There was a strong correlation between left atrial sizes using both methods, (r=0.78, n=127, 95% CI 0.70–0.84); Panel B: Further analysis with a Bland-Altman plot revealed a minimal average discrepancy between both modalities (n=127, bias=0.07, 95% limit of agreement= −0.56 to 0.71).

Fig. 4. Left atrial sizes measured on CTPA are significantly larger in patients with moderate or severe diastolic dysfunction

Panel A: LA sizes measured on CTPA were analyzed against diastolic dysfunction grades gleaned from echo reports; Panel B: Receiver operating curve of CTPA LA size and diastolic dysfunction on echo; Panel C: LA sizes measured on CTPA were correlated with estimation of filling pressures using E/E’ from echo reports.

Fig. 5. CTPA pulmonary artery to aorta ratios were significantly larger in patients with elevated pulmonary artery pressures determined by echo

Pulmonary artery size was measured on CTPA and standardized using the ascending aorta measurement for each patient. An estimated systolic pressure of >35 mmHg on echo was defined as elevated. There was a significant difference in the CTPA determined PA/Aorta ratio for patients with normal pressures, 0.89 (n=38, 95% CI 0.93–1.04), and those with elevated pressures, 0.98 (n=49, 95% CI 0.86–0.93).