Early Human Development 90 (2014) 829–835

Contents lists available at ScienceDirect

Early Human Development journal homepage: www.elsevier.com/locate/earlhumdev

Assessment of myocardial performance in preterm infants less than 29 weeks gestation during the transitional period Adam T. James a, John D. Corcoran a, Amish Jain b,c,d, Patrick J. McNamara d,e, Luc Mertens f, Orla Franklin g, Afif F. EL-Khuffash a,h,⁎ a

Department of Paediatrics, The Rotunda Hospital, Dublin, Ireland Department of Paediatrics, Mount Sinai Hospital, Toronto, Canada Department of Physiology, University of Toronto, Toronto, Canada d Physiology and Experimental Medicine, Hospital for Sick Children, Toronto, Canada e Department of Neonatology, The Hospital for Sick Children, Toronto, Canada f The Labatt Family Heart Centre, The Hospital for Children, Toronto, Canada g Department of Cardiology, Our Lady's Children's Hospital Crumlin, Dublin, Ireland h Department of Paediatrics, Royal College of Surgeons in Ireland, Dublin, Ireland b c

a r t i c l e

i n f o

Article history: Received 7 July 2014 Received in revised form 3 September 2014 Accepted 7 September 2014 Available online xxxx Keywords: Preterm infant Strain Strain rate Tissue Doppler imaging TAPSE FAC Transitional circulation

a b s t r a c t Background: The transitional circulation and its effect on myocardial performance are poorly understood in preterm infants. Aims: We assessed myocardial performance in infants less than 29 weeks gestation in the first 48 h of life using a comprehensive echocardiographic assessment. Design: Infants b 29 weeks gestation were prospectively enrolled. Small for gestation, infants on inotropes and/or inhaled nitric oxide and septic infants were excluded. Conventional echocardiography, left ventricular (LV), septal and right ventricular (RV) tissue Doppler imaging (TDI) and tissue Doppler-derived strain and strain rate (SR), tricuspid annular plane systolic excursion (TAPSE) and global RV fractional area change (FAC) were assessed at a median of 10 and 45 h post-delivery. Results: Fifty-four infants with a median [IQR] gestation and birth weight of 26.5 weeks [25.8–28.0 weeks] and 915 g [758–1142 g] were included. There was no change in shortening or ejection fraction across the two time points. Systolic and diastolic TDI of the LV, septum and RV increased across the two time points (all p values ≤ 0.01). There was an increase in septal peak systolic and early diastolic SR (p = 0.002). Septal systolic strain and late diastolic SR did not change. With the exception of RV strain and early diastolic SR, all RV functional parameters including SR, late diastolic SR, TAPSE, and FAC increased across the two time points (all p values b 0.01). Conclusion: Describing the normal hemodynamic adaptations in stable preterm infants during the transitional period provides the necessary information for the assessment of those parameters in various disease states. © 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Cardiovascular adaptation during the transitional period and its effect on myocardial performance is poorly described in preterm infants. During the early phase, preterm infants are particularly susceptible to morbidities such as intraventricular hemorrhage (IVH), hypotension, and ventilator dependency, which may be partially related to hemodynamic instability [1,2]. Monitoring the cardiovascular status of preterm ⁎ Corresponding author at: The Rotunda Hospital, Dublin, Ireland, Honorary Clinical Senior Lecturer, Royal College of Surgeons in Ireland, Dublin, Ireland. Tel.: + 353 1 817 1700. E-mail address: afi[email protected] (A.F. EL-Khuffash).

http://dx.doi.org/10.1016/j.earlhumdev.2014.09.004 0378-3782/© 2014 Elsevier Ireland Ltd. All rights reserved.

infants remains a challenge due to the insensitivity of clinical indicators in defining systemic perfusion [3], and the limitations of conventional echocardiography functional parameters such as shortening fraction (SF) and ejection fraction (EF) in assessing left ventricular (LV) function [4]. Moreover, data on the assessment of right ventricular (RV) function in preterm infants are still limited. Recent advances in echocardiography have led to the development of techniques that directly measure global and regional myocardial function, rather than depend on changes in cavity dimensions. Tissue Doppler imaging (TDI) and myocardial deformation measurements (myocardial strain rate and strain) may provide more accurate information on systolic and diastolic myocardial function [5–7]. Tissue Doppler imaging (TDI) and myocardial deformation based on

830

A.T. James et al. / Early Human Development 90 (2014) 829–835

tissue Doppler derived strain and strain rate (SR) are emergent techniques [8]. The most significant value of those techniques is the ability to detect subclinical local myocardial dysfunction before the appearance of clinically apparent ventricular impairment. Quantitative assessment of RV function can be obtained using TDI, strain and strain rate, in addition to RV specific markers of performance including tricuspid annular plane systolic excursion (TAPSE) and fractional area contraction (FAC). These modalities may possess better sensitivity in detecting changes in myocardial performance during the early preterm neonatal period, and provide more insight into the adaptations that occur during the transitional period. In this study, we aimed to document changes in myocardial performance at two discrete time points over the first 48 h of life in stable preterm infants less than 29 weeks gestation using a comprehensive echocardiographic assessment.

Electric, version 112 revision 1.3) for later offline analysis. All offline analysis was carried out by a single investigator (ATJ). We obtained the following echocardiography measurements in study infants using methods previously described [10]: diastolic septal wall thickness, LV internal diameter in diastole (LVID), LV posterior wall diameter in diastole (LVPWD) (LV dimension parameters all obtained using M-mode); left ventricular output (LVO); LV shortening fraction (SF) based on M-mode; ejection fraction measured by Simpson's biplane method [9]; PDA diameter in 2D measured at the pulmonary end; direction of flow and PDA shunt gradient, pulmonary artery acceleration time and RV ejection time. LV length was measured in diastole as the distance from the closed mitral valve to the apex in the four-chamber view. The presence of a patent foramen ovale and the shunting across it and the presence of tricuspid valve regurgitation (TR) were also noted. In addition, we calculated LV wall stress using the following formula: [1.35 × (mean arterial pressure) × (LVID)] / [4 × (LVPWD) × (1 + LVPWD/LVID)] [11].

2. Methods 2.4. Tissue Doppler imaging (TDI) 2.1. Study population This was a prospective observational study carried out in the neonatal intensive care unit (NICU) of the Rotunda Hospital Dublin, Ireland (a tertiary maternity unit which caters for over 9000 deliveries per annum). This was a nested study within a larger ongoing cohort study designed to define the natural history of patent ductus arteriosus (PDA) in preterm infants less than 29 weeks gestation. Infants were excluded if they: were small for gestational age (weight less than the 10th centile for given gestation); received inotropes or inhaled nitric oxide (iNO) in the first 48 h of life; died within the first 7 days of life; had a suspected or definite chromosomal abnormality; culture confirmed sepsis or congenital heart disease other than a PDA and patent foramen ovale (PFO) identified antenatally or on the initial echocardiogram. Our unit currently adopts a conservative approach to PDA treatment. Prophylactic indomethacin is not used at this institution and medical treatment of the PDA with non-steroidal anti-inflammatory drugs is not provided in the first 7 days of life. High frequency oscillation (HFO) is only used as a rescue mode of ventilation. Hypotension is treated with inotropes if blood pressure is lower than the 3rd centile for any given gestation in addition to clinical and laboratory signs of hemodynamic compromise. The results of the two research scans were not communicated to the medical team caring for the infants unless they specifically requested a clinically indicated echocardiographic assessment or if congenital heart disease was identified. Written parental informed consent was obtained from all participants and ethical approval was obtained from the Hospital Ethics Committee prior to recruitment. 2.2. Clinical demographics Antenatal, birth and neonatal characteristics were collected. In addition clinical cardio-respiratory characteristics during the two echocardiography assessments were collected and included: systolic, diastolic and mean blood pressure, heart rate, mean airway pressure, mode of ventilation, oxygen requirements, oxygen saturation, volume of fluid intake and pH. 2.3. Echocardiographic assessment Echocardiography was performed on day 1 of life at a median of 10 h (echo 1) and at day 2 of life at a median of 45 h (echo 2) using the Vivid I echocardiography system and 10 MHz multi-frequency probe (GE Medical, Milwaukee, USA). All studies were conducted using a standardized functional protocol adapted from recently published guidelines [9]. All infants were in a supine position at the time of the scan. The scans were all stored as raw data in an archiving system (EchoPac, General

Tissue Doppler velocities were obtained from the apical four-chamber view using a pulsed wave Doppler sample gate of 2 mm at the level of the annuli and the basal part of the intraventricular septum. We aligned the pulsed wave cursor with the longitudinal plane of motion at all times. On the tissue Doppler traces we measured peak systolic (s′), early diastolic (e′) and late diastolic (a′) velocities. If the e′ and a′ wave were fused, we measured the single wave as an a′ wave. The LV, septal and RV systolic velocities were normalized to LV and RV lengths accordingly using the following formula: normalized s′ = s′/ventricular length in cm. Isovolumic contraction (IVCT) and relaxation (IVRT) times and left ventricular systolic and diastolic times were also measured. The systolic to diastolic time ratio (SD ratio) was derived from the tissue Doppler traces. The myocardial performance index (MPI) was calculated from TDI as the sum of IVCT and IVRT divided by LV systolic time using the following formula: MPI = (IVCT + IVRT)/LV systolic time. 2.5. Tissue Doppler-derived strain and strain rate The four-chamber view was used to acquire color-tissue Doppler images of the LV and RV free walls and the septum. Sector width was narrowed to maximally increase the frame rates. Offline analysis was performed to measure longitudinal peak systolic strain, peak systolic strain rate (SRS), early diastolic strain rate (SRE) and late diastolic strain rate (SRA) in the basal segments of the LV and RV free wall and the IVS. RV strain was obtained from the free wall following angling towards the RV to obtain a clearer image of the wall. Image quality was assessed visually prior to analysis and only images of sufficient 2D quality were used. A single elliptical region of interest (ROI) was determined with a width of 2 mm and length of 1 mm. Strain length (the computational distance) was set at 6 mm. Those settings have been demonstrated to be the most reliable in extremely premature infants [12–14]. Linear drift compensation and 40 ms Gaussian smoothing was used. Event timing, including aortic and mitral valve opening and closure, was determined using the electrocardiogram and pulsed wave Doppler of the flow across those valves. Strain, SR, SRE and SRA were manually determined by averaging the results of three cardiac cycles (Fig. 1). Two cardiac cycles were used if measurement artifact was present in one cycle. If two cycles contained measurement artifact then the study was excluded from analysis. If E and A wave fusion was present in diastole then the single wave was reported as an A wave. 2.6. RV functional and dimension measurements Tricuspid annular plane systolic excursion (TASPE) is a measure of movement of the tricuspid annulus from base to apex during systole and reflects global RV function. TAPSE was measured based on

A.T. James et al. / Early Human Development 90 (2014) 829–835

831

Fig. 1. Tissue Doppler-derived measurement of strain and strain rate. During aortic valve opening (AVO), the wall segment strain is at baseline as indicated in (a). Peak systolic strain (a) is identified at the time of aortic valve closure (AVC). (b) demonstrates peak systolic strain rate occurring in mid systole (SRS). Early (SRE) and late (SRA) diastolic strain rates occur between mitral valve opening (MVO) and mitral valve closure (MVC).

M-mode echocardiography through the tricuspid annulus [7]. TAPSE was also normalized to RV length which was measured as described below. Global fractional area change (FAC) is a measure of the change in RV cavity area from diastole to systole in two planes. This measurement is obtained by averaging FAC from the apical four-chamber and three-chamber views. The RV three-chamber view is acquired by rotating the transducer anticlockwise from the standard apical four-chamber view until the LV is no longer visible followed by anterior tilting of the probe. The RV inflow and outflow should both be visible along with a cross sectional view of the aorta between those two structures (Fig. 2a). RV dimensions were measured at end diastole from the apical four-chamber view. Tricuspid valve annular diameter (TVAD) was measured as the distance between the two hinge points of the two visible valve leaflets. Basal diameter (RVBD) was measured as the maximal distance between the RV lateral wall and the septum parallel to the annular diameter. RV length (RVL) was measured from the midpoint of the annular diameter to the RV apex and finally, the RV mid cavity (RVMC) was measured as a line parallel to the basal diameter at the midpoint of the RV length (Fig. 2b).

2.7. Reliability analysis The reliability of tissue Doppler velocity measurements in the preterm population was recently demonstrated by our group [2,15]. Intra and inter-observer variability of strain, SR and the RV-specific function and dimension parameters were assessed using 30 randomly selected studies from the cohort. For intra-observer variability, one investigator (ATJ) performed two offline analyses 12 weeks apart to avoid recall bias while inter-observer variability was assessed by a second investigator (AK) who was blinded to the measurements of the first investigator. Intra- and inter-observer agreement was tested using Bland–Altman (BA) analysis and is presented as mean bias and 95% confidence intervals. In addition, the intraclass correlation coefficient (ICC version 2.1) was used to assess agreement. 2.8. Statistical analysis Continuous data were presented as means (standard deviation, SD) for normally distributed variables, or medians [inter-quartile range,

Fig. 2. Right ventricular three-chamber view and right ventricular dimension measurement in the four chamber view. (a) demonstrates the RV three-chamber view with all the identifiable structures. (b) demonstrates the points at which the various aspects of RV four chamber dimensions during diastole were measured. TV: tricuspid valve; RA: right atrium; Ao: aorta; PA: pulmonary artery; PV: pulmonary valve; TAVD: tricuspid valve annular diameter; RVBD: right ventricular basal diameter; RVMC: right ventricular mid cavity diameter; RVL: right ventricular length.

832

A.T. James et al. / Early Human Development 90 (2014) 829–835

IRQ] for skewed data unless otherwise stated. Categorical variables are presented as proportions. Results from day 2 scans were compared with day 1 using paired Student's t-tests, Wilcoxon signed-rank test, Chi square or Fisher's exact test as appropriate. Two group comparisons were performed using the Student's t test or a Mann–Whitney U test as appropriate. We accepted a p value of b0.05 as significant. We used SPSS (Version 21, IBM Corp.) for analysis. 3. Results 3.1. Population details and clinical parameters Seventy-three infants less than 29 weeks gestation were considered for inclusion during the study period. Seven were excluded due to investigator unavailability, one refused consent, three had weights less than the 10th centile, four received inotropes during the study period, and four died in the first week of life. Fifty-four infants were included whose median [IQR] gestation and weight at birth were 26.5 weeks [25.8–28.0 weeks] and 915 g [758–1142 g] respectively. Thirty-one infants (57%) were male. Twenty-seven infants (50%) were singleton births with nine (33%) twin pairs and three (17%) triplet sets. Fortyfive (83%) received a full course of antenatal steroids with seven (13%) receiving one dose and two (4%) receiving none. Their median [IQR] 1 and 5 min Apgar scores were 6 [5–8] and 9 [7–9] respectively. The mean cord gas of the cohort was 7.33 (0.05). All infants received early surfactant prior to the first echo. The median time between the two scans was 34 h [27–41 h]. There were small but significant changes in heart rate, blood pressure oxygen requirements and pH. However, those were not felt to be clinically relevant. There was a significant reduction in the number of infants invasively ventilated (Table 1). 3.2. Conventional echocardiography measurements None of the infants had a PDA diameter of less than 1.5 mm during echo 1. Only four infants had a PDA diameter of b1.5 mm during echo 2. There was a decrease in proportion of neonates with bidirectional PDA shunts between days 1 and 2. In addition peak flow velocity and pressure gradient of the left-to-right transductal shunt increased and pulmonary artery acceleration time increased (Table 2). The flow velocity through the duct increased in the absence of a significant decrease in PDA size. There was no change in LV shortening fraction, ejection fraction, or wall stress between the two time points, but LVO increased. Twenty infants (37%) had a visible TR jet on the first scan and 13 (24%) had a visible TR jet on the second scan. Right ventricular systolic

Table 1 Clinical parameters at the time of the echocardiogram.

Hours of life Heart rate Systolic BP (mm Hg) Diastolic BP (mm Hg) Mean BP (mm Hg) Mean airway pressure (cm H2O) Oxygen Invasive ventilation Fluid Intake (ml/kg/day) pH

Table 2 Conventional echocardiography parameters and markers of pulmonary vascular resistance.

Left ventricular output (mls/kg/min) Shortening fraction (%) Ejection fraction (%) Left ventricular wall stress (g/cm2) Number (%) patent ductus arteriosus Patent ductus arteriosus diameter Number (%) with bidirectional shunt Peak pressure gradient (mm Hg) Number (%) with PFO Patent foramen ovale shunt (m/s) Pulmonary artery acceleration time (ms) Right ventricular ejection time (ms)

Echo 1

Echo 2

p

171 (60) 36 (6) 60 (5) 23.3 [16.5–32.7] 51 (94) 2.4 [2.1–2.9] 19 (35) 4.7 [2.8–10.9] 39 (72) 0.44 [0.35–0.56] 42 (10)

219 (77) 38 (5) 62 (5) 27.7 [19.4–34.6] 45 (83) 2.9 [2.2–3.2] 7 (13) 7.7 [5.1–16.4] 45 (83) 0.64 [0.47–0.82] 45 (12)

b0.001 0.1 0.05 0.1 0.1 0.1 0.02 0.002 0.2 b0.001 0.04

151 (21)

157 (29)

0.2

Data are presented as medians [inter-quartile range], means (standard deviation) or proportions (percentages).

pressure was not calculated from those TR jets as the majority of those were trivial. 3.3. Feasibility and reliability of strain, strain rate, TAPSE and FAC Strain measurement was feasible in the majority of infants except for 10 LV (9%), 8 septal (7%) and 8 RV (7%) where images were deemed to be poor quality. Strain rate measurement was not possible in 16 LV (15%), 9 septal (8%) and 8 of RV (7%) basal segments. The mean (SD) frames rates used for image acquisition were 266 (43), 281 (43) and 279 (40) frames per second for the LV, septum and RV respectively. E/A wave fusion occurred in 28 LV (26%), 27 septal (25%) and 46 RV (43%) images. The intra- and inter-observer variability of strain and strain rate measurements are described in Table 3. LV free wall strain and LV SR parameters demonstrated the highest degree of intra and inter-observer variability. Septal and RV free wall strain and SR parameters were more reliable measurements. RV dimensions, TAPSE and global FAC were feasible in all studies. Overall most RV measurements are highly reliable with the highest variability in FAC (Table 3). 3.4. LV functional parameters and dimensions There was an increase in systolic and diastolic tissue Doppler velocities of the LV free wall and the septum across the two time points. LV systolic time significantly increased without a change in LV diastolic time resulting in an increase in SD ratio. The MPI demonstrated a reduction in value (signifying increased function) across the two time points. We observed a small but significant increase in basal septal SR and SRE between the two time points (Table 4). 3.5. RV functional parameters and dimensions

Echo 1

Echo 2

p

10 [6–13] 156 (13) 45 (8) 29 (8) 35 (8) 8 (2) 21 [21–50] 33 (61%) 83 (8) 7.34 (0.06)

45 [41–48] 164 (12) 53 (9) 32 (8) 40 (8) 8 (2) 21 [21–35] 21 (39%) 119 (27) 7.29 (0.07)

b0.001 0.001 b0.001 0.09 0.005 0.5 0.03 0.02 b0.001 b0.001

Hours of life are presented as medians [inter-quartile range] and oxygen as median [range] . The remainder of data is presented as means (standard deviation) or absolute values and (percentages). BP: blood pressure. Invasive ventilation refers to intermittent positive pressure ventilation. None of the infants were on high frequency oscillation. All non-invasively ventilated infants were on continuous positive pressure ventilation during the study period.

There was an increase in systolic and diastolic tissue Doppler velocities of the RV across the two time points. In addition, RV SR and SRA demonstrated a small increase. While RV dimensions did not significantly change between the two time points, TAPSE and FAC significantly increased (Table 5). We found a positive linear correlation between birth weight and RV four-chamber EDA (r = 0.60, p b 0.001), RVBD (r = 0.44, p = 0.001), RVMC (r = 0.52, p b 0.001) and RVL (r = 0.65, p b 0.001). There was no correlation between birth weight and TVAD (0.24, p = 0.08). 3.6. The effect of ventilation and PDA on function During echo 1, 16 (48%) ventilated infants had a bidirectional shunt across the PDA compared to 3 (9%) infants on CPAP (p = 0.018). The only function parameter different between ventilated and CPAP infants

A.T. James et al. / Early Human Development 90 (2014) 829–835

833

Table 3 Reliability of parameters of left, septal and right ventricular function and dimensions. Intra-observer variability

Inter-observer variability

Bias mean (SD)

ICC (95% CI, p)

Bias mean (SD)

ICC (95% CI, p)

Left ventricle free wall Strain (%) Systolic strain rate (1/s) Diastolic E′ strain rate (1/s) Diastolic A′ strain rate (1/s)

0.05 (3.02) 0.05 (0.24) −0.03 (0.50) 0.15 (0.85)

0.43 (−0.23–0.73) 0.93 (0.85–0.97) 0.80 (0.49–0.92) 0.62 (0.18–0.82)

−1.87 (3.03) −0.18 (0.49) −0.17 (0.41) −0.25 (0.59)

0.69 (0.23–0.86) 0.81 (0.59–0.91) 0.78 (0.36–0.93) 0.85 (0.68–0.93)

Intraventricular septum Strain (%) Systolic strain rate (1/s) Diastolic E′ strain rate (1/s) Diastolic A′ strain rate (1/s)

−0.12 (1.59) 0.00 (0.12) 0.01 (0.20) −0.01 (0.31)

0.95 (0.90–0.98) 0.92 (0.84–0.96) 0.97 (0.93–0.99) 0.95 (0.90–0.98)

−0.42 (1.53) 0.02 (0.20) −0.05 (0.20) −0.02 (0.40)

0.97 (0.94–0.99) 0.89 (0.76–0.94) 0.98 (0.97–1.00) 0.92 (0.84–0.96)

RV function parameters Strain (%) Systolic strain rate (1/s) Diastolic E′ strain rate (1/s) Diastolic A′ strain rate (1/s) M-mode TAPSE (mm) TD TAPSE (mm) 3-Chamber FAC (%) 4-Chamber FAC (%) Global FAC (%)

−1.58 (2.26) −0.13 (0.23) −0.06 (0.45) 0.26 (0.59) 0.08 (0.28) 0.14 (0.41) 0.9 (6) 0.09 (6) 0.5 (5.0)

0.94 (0.82–0.98) 0.93 (0.82–0.97) 0.87 (0.60–0.96) 0.92 (0.81–0.97) 0.98 (0.96–0.99) 0.97 (0.93–0.99) 0.60 (0.11–0.81) 0.62 (0.26–0.81) 0.77 (0.51–0.89)

1.60 (2.82) −0.02 (0.36) 0.02 (0.27) 0.40 (0.58) −0.19 (0.38) −0.27 (0.34) 7.0 (6) 0.02 (6) 5.0 (4.3)

0.92 (0.82–0.97) 0.91 (0.81–0.96) 0.96 (0.89–0.99) 0.92 (0.76–0.97) 0.97 (0.93–0.99) 0.97 (0.84–0.99) 0.62 (−0.14–0.86) 0.69 (0.35–0.85) 0.78 (−0.47–0.93)

RV dimension parameters 4C diastolic fractional area (cm2) 3C diastolic fractional area (cm2) Annulus (mm) Base (mm) Mid cavity (mm) RV length (mm)

0.01 (0.19) −0.07 (0.22) −0.11 (0.47) 0.21 (0.60) 0.25 (0.61) −0.07 (0.97)

0.90 (0.78–0.95) 0.90 (0.80–0.95) 0.78 (0.53–0.89) 0.92 (0.83–0.93) 0.90 (0.78–0.96) 0.94 (0.87–0.97)

0.11 (0.19) −0.03 (0.21) 0.04 (0.54) 0.69 (0.79) −0.90 (0.74) 1.06 (1.06)

0.83 (0.57–0.93) 0.92 (0.83–0.96) 0.71 (0.38–0.86) 0.81 (0.33–0.93) 0.89 (0.77–0.95) 0.79 (0.16–0.93)

SD: standard deviation; ICC; infraclass correlation coefficient; 95% CI: 95% confidence intervals; p: p values; TAPSE: tricuspid annulus plane systolic excursion; FAC: fractional area change. 3C: three-chamber; 4C; four-chamber.

on day 1 was basal septal longitudinal strain [Ventilated −14.8% (2.6) vs. CPAP −17.1% (2.7), p = 0.005]. On day 2, the only parameter different between ventilated and CPAP infants was RV basal longitudinal strain [Ventilated −24.9% (3.9) vs. CPAP −22.1% (4.8), p = 0.04]. During echo 2, there was no difference in any of the functional parameters between infants with a PDA N 1.5 mm vs. those with a PDA b 1.5. 4. Discussion 4.1. Feasibility and reproducibility of the measurements Tissue Doppler-derived strain and strain rate are newer echocardiography techniques that measure the degree of myocardial deformation (strain) and speed at which this deformation occurs (strain rate) [16, 17]. We chose to use the tissue Doppler-derived method for assessment of strain and strain rate instead of speckle tracking due to the higher frame rates attainable with this method. The higher temporal resolution achievable with the tissue-Doppler derived method is of particular importance for the measurement of peak systolic and diastolic strain rates in this population. We demonstrated that measurement of longitudinal strain and strain rate along with RV specific function and dimension parameters is feasible in this population with the majority of the images suitable for analysis. The reproducibility of strain and SR in our study was comparable to those by Helfer et al. [12]. Their group demonstrated the lowest variability to be present in the septum, and the highest variability in the left ventricular wall. We found a similar pattern in our cohort. Assessment of left ventricle free wall was less reliable with ICC values for LV strain, SRS, SRE and SRA ranging from 0.43 to 0.93. The poor reliability of LV free wall measurements is thought to relate to artifact produced by the left lung obstructing a clear view of the LV free wall. In addition, segmental FAC demonstrated moderate reliability with ICC values ranging from 0.60 to 0.69. However, global FAC showed stronger agreement.

4.2. Changes in LV function Shortening fraction and EF did not change over the two time points in our cohort. However, there was an increase in LV and septal systolic and diastolic tissue Doppler velocities. We found that LV strain and strain rate parameters did not significantly change. This may be due to the low signal to noise ratio of the LV lateral wall strain parameters and relatively poor reliability achieved in our study resulting in a failure to detect true changes. Septal wall systolic strain did not significantly change either but septal systolic and diastolic strain rate parameters demonstrated a significant increase. Tissue Doppler and strain measurements are more sensitive to changes in myocardial performance when compared to conventional echocardiography measurements such as SF and EF [18–21]. Myocardial velocities measured by tissue Doppler are influenced by loading conditions [2]. Recent animal data reveal that systolic strain is highly influenced by afterload and as a result is not a good surrogate measure of intrinsic contractility unless the confounding influence of afterload is considered. Strain rate on the other hand is less influenced by changes in cardiac loading conditions and is therefore a more reliable measure of contractility [22]. The changes in those parameters in our cohort may be multifactorial. There was a possible fall in PVR over the two time points as suggested by a fall in the number of bidirectional shunts, and an increase in left to right flow across the PDA without a change in diameter coupled with an increase in pulmonary artery acceleration time. During the second scan, fewer infants were ventilated and there was an increased volume of fluid intake reflecting the progression of care of those infants over the first 2 days of life. Those clinical changes were also coupled with an increase in LVO and an increase in shunt velocity across the PFO. Therefore, many of those changes may have contributed to the increase in LV and septal function observed. Recently, the impact of transitional changes on myocardial performance in the first 24 h was explored by Lee et al. on a group of infants

834

A.T. James et al. / Early Human Development 90 (2014) 829–835

Table 4 Left ventricle and septal functional parameters and dimensions. Echo 1

Echo 2

p

LV TDI (cm/s) s′ Normalized s′ e′ a′

2.8 (0.8) 1.6 (0.5) 3.8 (1.5) 4.6 (1.8)

3.3 (0.8) 1.8 (0.5) 4.7 (1.5) 5.2 (1.8)

b0.001 b0.001 0.004 0.01

Septal TDI (cm/s) s′ Normalized s′ e′ a′

2.6 (0.7) 1.5 (0.4) 2.9 (0.9) 4.2 (1.3)

3.1 (0.8) 1.7 (0.4) 3.9 (1.4) 5.3 (1.7)

b0.001 0.001 b0.001 b0.001

LV free wall strain and SR Strain (%) Systolic Strain Rate (1/s) Diastolic E′ strain rate (1/s) Diastolic A′ strain rate (1/s)

−12.8 (3.3) −1.5 (0.6) 1.8 (0.8) 2.3 (0.7)

−13.1 (3.6) −1.6 (0.5) 2.0 (0.8) 2.7 (1.2)

0.6 0.2 0.5 0.07

Septal strain and SR Strain (%) Systolic strain rate (1/s) Diastolic E′ strain rate (1/s) Diastolic A′ strain rate (1/s)

−15.8 (2.8) −1.6 (0.3) 1.7 (0.6) 2.3 (0.7)

−16.8 (3.6) −1.8 (0.5) 2.2 (0.6) 2.6 (1.2)

0.07 0.002 0.002 0.1

LV event times and dimensions IVCT (ms) IVRT (ms) Systolic time (ms) Diastolic time (ms) SD ratio Myocardial performance index Septal wall diameter (mm) LV internal diameter (mm) LV posterior wall diameter (mm) LV length

60 (12) 60 (13) 139 (15) 122 (18) 1.16 (0.17) 0.87 (0.20) 2.7 (0.6) 11.0 (2.2) 2.5 (0.6) 17.7 (1.6)

51 (10) 53 (13) 145 (16) 116 (24) 1.29 (0.23) 0.72 (0.17) 2.6 (0.6) 11.7 (2.1) 2.4 (0.6) 18.4 (1.8)

b0.001 0.003 0.025 0.08 b0.001 b0.001 0.3 0.002 0.7 0.014

Data are presented as means (SD). TDI: tissue Doppler indices; LV: left ventricle; IVCT: isovolumic contraction time; IVRT: isovolumic relaxation time; SD ratio: systolic time to diastolic time ratio.

at 5 h (n = 32), 12 h (n = 18) and 24 h (n = 22) with a mean gestation of 27 weeks using color TDI [23]. They demonstrated a significant Table 5 Right ventricle functional parameters and dimensions. Echo 1 mean (SD)

Echo 2 mean (SD)

p

RV TDI (cm/s) s′ Normalized s′ e′ a′

3.8 (0.9) 2.0 (0.5) 4.1 (1.2) 7.5 (1.9)

4.5 (1.1) 2.5 (0.6) 5.1 (1.5) 8.7 (2.5)

b0.001 b0.001 0.001 0.003

RV strain and strain rate Strain (%) Systolic strain rate (1/s) Diastolic E′ strain rate (1/s) Diastolic A′ strain rate (1/s)

−22.1 (5.1) −2.0 (0.6) 2.4 (0.9) 3.4 (1.0)

−23.1 (4.7) −2.4 (0.6) 2.5 (0.6) 4.4 (1.4)

0.3 0.001 0.9 b0.001

RV TAPSE and FAC M-mode TAPSE (mm) TD TAPSE (mm) Normalized TD TAPSE (mm) 3-Chamber FAC (%) 4-Chamber FAC (%) Global FAC (%)

5.0 (1.0) 5.1 (1.0) 2.7 (0.5) 43 (8) 37 (8) 40 (7)

5.8 (1.1) 5.9 (1.1) 3.1 (0.6) 48 (7) 43 (8) 46 (6)

b0.001 b0.001 b0.001 0.002 b0.001 b0.001

RV dimensions parameters 4C diastolic fractional area (cm2) Tricuspid valve annular diameter (mm) Right ventricular basal diameter (mm) Right ventricular mid cavity diameter (mm) Right ventricular length (mm)

1.5 (0.3) 6.4 (1.0) 11.1 (1.3) 9.9 (1.6) 18.7 (2.2)

1.5 (0.3) 6.4 (0.9) 10.9 (1.4) 9.6 (1.6) 18.7 (2.4)

0.9 0.9 0.1 0.3 0.9

TDI: tissue Doppler indices; TAPSE: tricuspid annulus plane systolic excursion; FAC: fractional area change. 3C: three-chamber; 4C; four-chamber. TVAD: tricuspid valve annular diameter; RVBD: RV basal diameter; RVMC: RV mid cavity; RVL: RV length.

reduction in LV and RV systolic and diastolic velocities from 5 to 12 h of life with a non-significant increase in those velocities by 24 h life. The group did not assess changes in function beyond 24 h. The authors attributed this fall in velocities to the postnatal increase in afterload occurring after birth. Similarly, Murase et al. demonstrated a similar reduction in function using TDI in preterm infants less than 1500 g between 3 and 12 h of life [24]. However, they did not demonstrate a recovery until day 5 of life. There is emerging evidence that peak systolic strain and strain rate can detect differences in myocardial performance in preterm infants with and without a significant PDA and chronic lung disease beyond the second week of life [13]. In a recent study, Eriksen et al. reported LV and septal s′ velocities normalized to cardiac length in a cohort of term and moderately preterm infants on day 3 and 4–10 weeks post natal age. Interestingly, our normalized LV and Septal s′ values are very similar to those reported by this group. This highlights the potential usefulness of normalization by heart size to allow comparison of function between infants of differing gestations and across studies [25]. 4.3. Changes in RV function and dimensions The objective assessment of RV systolic and diastolic function in the preterm population using various echocardiography techniques such as TAPSE, tissue Doppler velocities and strain is gaining momentum [6,13, 23,26]. However, the serial assessment of RV function during the early preterm neonatal phase warrants further study. In this cohort, we included tissue Doppler velocities, and tissue Doppler-derived strain and strain rate, in addition to TAPSE and FAC to provide a comprehensive appraisal of RV function. We demonstrated that systolic and diastolic tissue Doppler velocities increased over the two time points. RV strain did not change but there was an increase in RV systolic SR and diastolic SRA. Our RV strain and systolic strain rate values were similar to those obtained by Helfer et al. [12]. Direct comparisons however are not possible as their infant population included a wider range of weights and gestations in addition to the fact that their “Day 1” values were obtained within the first 4 days of life. TAPSE and FAC also increased over the same time points. TAPSE is a reliable marker of RV function and correlates well with peak systolic tricuspid annular velocity (S′) in term and preterm infants as well as in the paediatric population [5]. Koestenberger et al. established reference values for TAPSE in preterm infants (≥26 weeks) up to term. However, the group did not present serial values during the first 48 h of life and only presented one value for that time period [7]. Global FAC is another marker of RV systolic function. Given that RV dimensions were measured at end-diastole and unchanged between studies, an increased preload is unlikely to explain the increase in those values. Instead, it is more likely that both FAC and TAPSE increased due to smaller RV endsystolic dimensions, which likely reflect the decreased afterload imposed on the right ventricle by the decreased pulmonary vascular resistance. Developing a more objective approach to RV functional assessment will help us to better understand the impact of haemodynamic changes on function and be useful in defining disease states and monitoring the response to treatment. Similarly, defining RV dimensions in the early neonatal period will aid in detecting changes in RV size and cavity secondary to preterm-associated conditions such as chronic lung disease. Similarly our normalized RV s′ and TAPSE values are very comparable to those by Eriksen et al. further highlighting the importance of normalization of function parameters [25]. 4.4. Effect of ventilation and PDA The effect of ventilation on the function parameters is interesting. Infants ventilated on day 1 had lower septal basal strain compared to those on CPAP. Whether this was a consequence of ventilation, or the fact that most ventilated infants had a higher PVR warrants further

A.T. James et al. / Early Human Development 90 (2014) 829–835

exploration. In addition, on day 2, ventilated infants had higher RV basal strain compared to controls. Again, the difference in RV strain between those two groups needs further investigation. The effect of the PDA on those markers was not possible to study as almost all of the infants in this cohort had similar PDA characteristics. 5. Limitations Although we attempted to assess myocardial performance in a group of stable preterm infants, they are a very heterogeneous group with a variety of pathologies, antenatal and postnatal factors that are likely to have a significant impact on the studied function parameters. In addition, in an attempt to limit the amount of time spent performing scans on this population, we did not assess the reliability of obtaining those functional measurements from two independent scans performed in close succession. The period of peak PVR changes was missed in this study as it occurs in the first hours following delivery. However, administering a comprehensive assessment of myocardial performance in extremely preterm infants during the first few hours of life may interfere with the intensive care requirements during this period. We used surrogate echocardiography markers to determine the change in PVR across the two time points in this study, and this may have overestimated the influence of PVR changes on the study parameter. The changes observed over the study period may have been a result of a variety of factors, including mechanical ventilation, infant maturity and medical management. In addition, as this unit adopts a conservative approach to PDA treatment, the majority of infants in our study had a patent duct and the influence of the PDA per se was not possible over this time period. 6. Conclusion The use of tissue Doppler velocities, strain, strain rate in addition to a more comprehensive assessment of RV function and dimension is feasible in extremely preterm infants. Myocardial performance in preterm infants increases in the first 48 h of life and this change is probably related to changes in loading conditions. Newer echocardiography markers can identify differences in myocardial performance over the first 48 h of life in this population. Studying those parameters in stable preterm infants during the transitional period can pave the way for the assessment of those parameters in more unstable conditions. Funding This research has received funding from the EU FP7/2007–2013 under grant agreement no. 260777 (The HIP Trial) Conflict of Interest None of the authors have any conflict of interest to declare. References [1] Noori S, Seri I. Pathophysiology of newborn hypotension outside the transitional period. Early Hum Dev 2005;81:399–404. [2] El-Khuffash AF, Jain A, Dragulescu A, McNamara PJ, Mertens L. Acute changes in myocardial systolic function in preterm infants undergoing patent ductus arteriosus ligation: a tissue Doppler and myocardial deformation study. J Am Soc Echocardiogr 2012;25:1058–67. [3] Alagarsamy S, Chhabra M, Gudavalli M, Nadroo AM, Sutija VG, Yugrakh D. Comparison of clinical criteria with echocardiographic findings in diagnosing PDA in preterm infants. J Perinat Med 2005;33:161–4.

835

[4] Colan SD, Borow KM, Neumann A. Left ventricular end-systolic wall stress-velocity of fiber shortening relation: a load-independent index of myocardial contractility. J Am Coll Cardiol 1984;4:715–24. [5] Eriksen BH, Nestaas E, Hole T, Liestol K, Stoylen A, Fugelseth D. Longitudinal assessment of atrioventricular annulus excursion by grey-scale m-mode and colour tissue Doppler imaging in premature infants. Early Hum Dev 2013;89:977–82. [6] Koestenberger M, Nagel B, Ravekes W, Gamillscheg A, Pichler G, Avian A, et al. Right ventricular performance in preterm and term neonates: reference values of the tricuspid annular peak systolic velocity measured by tissue Doppler imaging. Neonatology 2013;103:281–6. [7] Koestenberger M, Nagel B, Ravekes W, Urlesberger B, Raith W, Avian A, et al. Systolic right ventricular function in preterm and term neonates: reference values of the tricuspid annular plane systolic excursion (TAPSE) in 258 patients and calculation of Zscore values. Neonatology 2011;100:85–92. [8] Mor-Avi V, Lang RM, Badano LP, Belohlavek M, Cardim NM, Derumeaux G, et al. Current and evolving echocardiographic techniques for the quantitative evaluation of cardiac mechanics: ASE/EAE consensus statement on methodology and indications endorsed by the Japanese Society of Echocardiography. J Am Soc Echocardiogr 2011;24:277–313. [9] Mertens L, Seri I, Marek J, Arlettaz R, Barker P, McNamara P, et al. Targeted neonatal echocardiography in the neonatal intensive care unit: practice guidelines and recommendations for training writing group of the American Society of Echocardiography (ASE) in collaboration with the European Association of Echocardiography (EAE) and the Association for European Pediatric Cardiologists (AEPC). J Am Soc Echocardiogr 2011;24:1057–78. [10] El-Khuffash AF, McNamara PJ. Neonatologist-performed functional echocardiography in the neonatal intensive care unit. Semin Fetal Neonatal Med 2011;16:50–60. [11] Rowland DG, Gutgesell HP. Use of mean arterial pressure for noninvasive determination of left ventricular end-systolic wall stress in infants and children. Am J Cardiol 1994;74:98–9. [12] Helfer S, Schmitz L, Bührer C, Czernik C. Reproducibility and optimization of analysis parameters of tissue Doppler-derived strain and strain rate measurements for very low birth weight infants. Echocardiography 2013;30:1219–26. [13] Helfer S, Schmitz L, Buhrer C, Czernik C. Tissue Doppler-derived strain and strain rate during the first 28 days of life in very low birth weight infants. Echocardiography Dec. 23 2013. http://dx.doi.org/10.1111/echo.12463. [14] Poon CY, Edwards JM, Joshi S, Kotecha S, Fraser AG. Optimization of myocardial deformation imaging in term and preterm infants. Eur J Echocardiogr 2011;12:247–54. [15] Saleemi MS, Bruton K, El-Khuffash A, Kirkham C, Franklin O, Corcoran JD. Myocardial assessment using tissue doppler imaging in preterm very low-birth weight infants before and after red blood cell transfusion. J Perinatol 2013;33:681–6. [16] Teske AJ, De Boeck BW, Melman PG, Sieswerda GT, Doevendans PA, Cramer MJ. Echocardiographic quantification of myocardial function using tissue deformation imaging, a guide to image acquisition and analysis using tissue Doppler and speckle tracking. Cardiovasc Ultrasound 2007;5:27. [17] Bijnens BH, Cikes M, Claus P, Sutherland GR. Velocity and deformation imaging for the assessment of myocardial dysfunction. Eur J Echocardiogr 2009;10:216–26. [18] Jain A, Sahni M, El-Khuffash A, Khadawardi E, Sehgal A, McNamara PJ. Use of targeted neonatal echocardiography to prevent postoperative cardiorespiratory instability after patent ductus arteriosus ligation. J Pediatr 2012;160:584–9. [19] Di Salvo G, Pacileo G, Del Giudice EM, et al. Abnormal myocardial deformation properties in obese, non-hypertensive children: an ambulatory blood pressure monitoring, standard echocardiographic, and strain rate imaging study. Eur Heart J 2006; 27:2689–95. [20] Marcus KA, Barends M, Morava-Kozicz E, Feuth T, de Korte CL, Kapusta L. Early detection of myocardial dysfunction in children with mitochondrial disease: An ultrasound and two-dimensional strain echocardiography study. Mitochondrion 2011; 11:405–12. [21] Khoo NS, Smallhorn JF, Kaneko S, Myers K, Kutty S, Tham EB. Novel insights into RV adaptation and function in hypoplastic left heart syndrome between the first 2 stages of surgical palliation. JACC Cardiovasc Imaging 2011;4:128–37. [22] Ferferieva V, Van den Bergh A, Claus P, et al. The relative value of strain and strain rate for defining intrinsic myocardial function. Am J Physiol Heart Circ Physiol 2012;302:H188–95. [23] Lee A, Nestaas E, Liestol K, Brunvand L, Lindemann R, Fugelseth D. Tissue Doppler imaging in very preterm infants during the first 24 h of life: an observational study. Arch Dis Child Fetal Neonatal Ed 2014;99:F64–9. [24] Murase M, Morisawa T, Ishida A. Serial assessment of left-ventricular function using tissue Doppler imaging in premature infants within 7 days of life. Pediatr Cardiol 2013;34:1491–8. [25] Eriksen BH, Nestaas E, Hole T, Liestøl K, Støylen A, Fugelseth D. Myocardial function in term and preterm infants. Influence of heart size, gestational age and postnatal maturation. Early Hum Dev 2014;90:359–64. [26] Levy PT, Holland MR, Sekarski TJ, Hamvas A, Singh GK. Feasibility and reproducibility of systolic right ventricular strain measurement by speckle-tracking echocardiography in premature infants. J Am Soc Echocardiogr 2013;26:1201–13.