Left ventricular ejection fraction using manual and semi-automated biplane method of discs in very preterm infants.

BACKGROUND: Biplane left ventricular ejection fraction (LVEF) is a valuable echocardiographic parameter for assessment of LV systolic pump efficiency in adults and children, but not often reported in preterm infants. The primary aim of this study was to longitudinally measure biplane LVEF in very preterm infants during the neonatal intensive care period. Secondary aim was to compare manual and semi-automatic determination of LVEF for agreement and variability.
METHODS: Stable preterm infants less than 30 weeks gestation were scanned on day 3, day 28, and at 36 weeks postmenstrual age. The LV endocardium was traced manually and semi-automatically using integrated speckle tracking software in apical 4-chamber and apical 3-chamber images to obtain end-diastolic volume and end-systolic volume, and calculate LVEF. Agreement between methods and variability within and between observers was determined using an interclass correlation coefficient (ICC) and Bland-Altman analysis.
RESULTS: Sixty-six preterm infants with a mean birth weight of 1100 (239) g were analyzed. The average manual biplane LVEF was 58 (3)%, 59 (3)%, and 55 (4)% at the three respective time points. Manual LVEF showed good agreement with semi-automatic LVEF (ICC 0.76) with a small bias of -1.5 (3.0)%. Interobserver variability of LVEF improved with semi-automatic tracing of the LV endocardial border (ICC manual 0.68 vs semi-automatic 0.80).
CONCLUSION: Left ventricular systolic pump efficiency in preterm infants remains stable during the neonatal intensive care period. Semi-automatic biplane LVEF has less interobserver variability and can be used interchangeably with manual biplane LVEF.