We read with great interest the recent report by Alvarez et al , which demonstrated that left ventricular (LV) systolic strain rate (SR) has a positive force-frequency relationship (FFR) in children aged 8 to 18 years and a neonatal-infant porcine model. The investigators concluded that SR is a valid measure of LV contractility in children and neonates because it “demonstrates a positive FFR which is a fundamental characteristic of myocyte response to chronotropy.” The use of SR as a noninvasive and reproducible measure of inherent myocardial contractility that is independent of loading conditions would be a valuable tool in the clinical field. This would be of particular importance in premature infants shortly after birth, when significant changes in loading conditions occur. The FFR evident with SR has not been previously explored in the premature infant population. Because of the wide range of physiologic heart rate (HR) values present in premature infants over the first 24 hours of age, its influence on SR to demonstrate a positive FFR can be examined. Our group recently validated LV deformation measurements using both tissue Doppler and speckle-tracking echocardiographic techniques in premature infants <29 weeks’ gestation over the first 48 hours of age, demonstrating acceptable intra- and interrater reproducibility, and we published serial reference ranges for those measurements. We did not explore the relationship between HR and those deformation parameters in this cohort. We therefore aimed to investigate whether SR possesses a FFR in extremely premature infants in the first 24 hours of life by reexamining the echocardiography database of this retrospective cohort of premature infants.
We reexamined the data collected on day 1 of age to assess the relationship between SR and HR. One hundred fourteen infants with median gestation and birth weight of 26.8 weeks (interquartile range [IQR], 25.1–28.0 weeks) and 880 g (IQR, 738–1,141 g), respectively, were included. Eighteen (16%) were small for gestational age (SGA, <10th centile). None of the infants in this cohort were in receipt of inotropes or nitric oxide at the time of the scan. Their median 5-min Apgar score and first pH were 8 (IQR, 7–9) and 7.33 (IQR, 7.29–7.37), respectively. Echocardiography was carried out at 10 hours (IQR, 7–13 hours) of age. We derived LV basal strain and systolic SR using Doppler tissue imaging, as previously described. In addition, we simultaneously obtained surrogate measures of preload (LV end-diastolic diameter and echocardiography-measured stroke volume) to determine the interplay among HR, SR, and loading conditions. We observed a modest positive correlation between SR and HR in our cohort ( Figure 1 ). There was no relationship between LV basal strain and HR. There was no significant association between LV end-diastolic diameter or stroke volume and HR ( Figure 1 ). In a multivariate linear regression model devised to assess the independent effect of HR in addition to other potential confounders (gestational age, use of antenatal steroids, mean airway pressure at the time of the scan, LV end-diastolic diameter, and SGA) on SR, HR was the only significant variable associated with SR (standardized β = 0.42, P < .001).
