Pediatr Cardiol DOI 10.1007/s00246-014-0947-x
ORIGINAL ARTICLE
Cyclooxygenase Inhibitors in Preterm Infants With Patent Ductus Arteriosus: Effects on Cardiac and Vascular Indices Arvind Sehgal • Tejas Doctor • Samuel Menahem
Received: 23 January 2014 / Accepted: 25 April 2014 Ó Springer Science+Business Media New York 2014
Abstract Existing data suggest subendocardial ischemia in preterm infants with patent ductus arteriosus (PDA) and alterations in cardiac function after indomethacin administration. This study aimed to explore the evolution of left ventricular function by conventional echocardiography and speckle-tracking echocardiography (STE) and to ascertain the interrelationship with coronary flow indices in response to indomethacin. A prospective observational study was performed with preterm infants receiving indomethacin for medical closure of PDA. Serial echocardiography was performed, and the results were analyzed using analysis of variance. Intra- and interobserver variability was assessed using the intraclass correlation coefficient. Indomethacin was administered to 18 infants born at a median gestational age of 25.8 weeks (interquartile range [IQR], 24.2–28.1 weeks) with a birth weight of 773 g (IQR, 704–1,002 g). The median age of the infants was 7.5 days (IQR, 4–17). Global longitudinal strain (GLS) values significantly decreased immediately after indomethacin infusion (preindomethacin GLS, -19.1 ± 2.4 % vs. -15.9 ± 1.7 %; p \ 0.0001) but had improved at reassessment after 1 h (-17.4 ± 1.8 %). Conventional echocardiographic indices did not show significant alterations. A significant increase in arterial resistance in the coronary vasculature from 1.7 to 2.4 mmHg/cm/s was
A. Sehgal (&) T. Doctor Monash Newborn, Monash Children’s Hospital, Melbourne, Australia e-mail: [email protected] A. Sehgal Department of Pediatrics, Monash University, Melbourne, Australia S. Menahem Pediatric Cardiology, Monash Health, Melbourne, Australia
demonstrated. A significant correlation was noted between peak systolic GLS and flow resistance in the coronary vasculature. Significant changes in myocardial indices were observed immediately after indomethacin infusion. Compared with conventional methods, STE is a more sensitive tool to facilitate understanding of hemodynamics in preterm infants. Keywords Coronary
Strain Deformation Indomethacin
Introduction Speckle-tracking echocardiography (STE) is a well-established concept in adult and pediatric cardiology, and upcoming data show its clinical usefulness in neonates [9, 13, 14]. It measures myocardial deformation (wall motion strain), expressed as a percentage change from the original dimension. Previous studies have indicated STE to be more sensitive than conventional echocardiography for detecting early myocardial dysfunction in adults and children [9, 13, 14]. El-Khuffash et al. [11] demonstrated the feasibility and reliability of STE and noted significant changes in myocardial function immediately after surgical duct ligation. Impairment in left ventricular (LV) global longitudinal strain (GLS) was temporally associated with elevated vascular resistance. Association between two-dimensional (2D) STE-derived myocardial function and perfusion index (resistance) after exposure to cyclooxygenase inhibitors in human infants has not been investigated previously. The presence of a hemodynamically significant ductus arteriosus is seen in approximately 40 % of extremely low birthweight infants (\1,000 g), and indomethacin is
123
Pediatr Cardiol
commonly used for its medical closure [12]. Indomethacininduced cyclooxygenase inhibition results in decreased secretion of vasodilating prostaglandins and increased production of vasoconstrictor leukotrienes, leading to severe vasoconstriction. Prostaglandins play an important role in the regulation of coronary blood flow [1]. Interference with this autoregulation could alter coronary perfusion, with potential to affect myocardial function. Appleton et al. [2] previously described impairment in diastolic cardiac function 1 h after indomethacin (predominant coronary flow occurs in diastole), although no change in systolic function occurred. In addition, indomethacin also affects organ blood flow and end-organ resistance. An inverse correlation of splanchnic and cerebral blood flow with respective resistance after indomethacin administration has been demonstrated previously in preterm infants [4, 30], although a similar impact on the coronary vasculature is not known. This study aimed to investigate the effects of indomethacin infusion serially using conventional Doppler and 2D STE methods to understand better the effect of acute changes in resistance on the preterm myocardium.
Methods A prospective nonblinded observational echocardiography study performed with premature infants of less than 34 weeks gestation administered indomethacin for medical closure of a symptomatic patent ductus arteriosus (PDA). The diagnosis was initially suspected on the basis of clinical features (murmur, wide pulse pressure, bounding pulses, and/or hyperdynamic precordium) or early-onset hypotension. A significant duct was defined by a transductal diameter greater than 1.5 mm with unrestrictive (\1.5 m/s) left-to-right transductal flow on pulse-wave Doppler and clinical signs of pulmonary overcirculation or systemic hypoperfusion. Normative anatomy was confirmed by a preceding echocardiogram from the pediatric cardiology department. Approval of the study research ethics board and informed parental consent were obtained. At the time of the study, the department used indomethacin at a dose of 0.1 mg/kg as an intravenous infusion 1 h daily for 6 days. The assessments were made at the first dose of the first course only. The decision to treat was determined by the physician, with no influence from the researchers. The unit does not treat infants with necrotizing enterocolitis, renal failure, or right-to-left shunting. 2D echocardiography was performed within 2 h before indomethacin treatment and then again 10 min and 1 h after completion of infusion. All echocardiography evaluations were performed by a single operator (A.S.) with GE Vivid 7 equipment (GE Vingmed Ultrasound, Horten,
123
Norway) using a 7.5- to 10-MHz, high-frequency, phasedarray transducer probe. The study protocol was adapted from a previous study assessing the effects of indomethacin on splanchnic circulation [4]. Data analysis was performed offline. LV output was measured from an apical five-chamber view [26] and calculated as follows: Velocity time integral heart rate cross-sectional area indexed to body weight: LV dimensions for fractional shortening were measured using M-mode from a parasternal long-axis view at the level just distal to the mitral valve leaflet tips at end diastole. Transductal dimensions were calculated as described previously [16]. For STE in the apical fourchamber view, gain settings were optimized to obtain clear differentiation between cavity and wall. The endocardium was traced manually over a single frame, followed by automatic tracking of endocardial borders throughout the cardiac cycle. The left ventricle was divided into six segments: basal septal, middle septal, apical septal, apical lateral, middle lateral, and basal lateral (Fig. 1). The negative strain rate values in the apical view depict longitudinal shortening in systole. The timing of aortic valve closure was defined using the pulse Doppler recording from the apical four-chamber view. The Echopac algorithm generates seven curves that represent strain in the longitudinal direction (along the heart wall) for six specific myocardial segments and one global value that represents the combined strain from all the segments. A frame rate of at least 80 (range 80–112) was used for analysis. The left anterior descending coronary artery was interrogated for flow measurements as described previously [22, 24]. The left coronary artery was best identified from the standard parasternal short-axis view using color Doppler, and the left anterior descending branch could be seen by moving the transducer down one or two intercostal spaces, rotating it clockwise, and angling it superiorly. For color Doppler flow analysis, the scale was adjusted to an upper limit of 15–30 cm/s, allowing low-velocity signals to be identified. The pulse-wave Doppler sample gate was placed distal to the bifurcation, and the blood flow signal was obtained, keeping the angle of the Doppler beam as parallel as possible to the direction of flow (\15° in all recordings). Once a trace was obtained, the diastolic velocity time integral was calculated by tracing the area under the curve from an average of five consecutive beats. The internal dimensions were measured at end diastole with callipers applied to the endothelial border. No change in internal dimensions during the study time was observed in the initial six cases. Hence, to avoid errors
Pediatr Cardiol
Fig. 1 Top left Mapping of the region of interest in the left ventricle myocardium. Bottom left Peak segmental longitudinal strain values depicting the basal-to-apical gradient. Top right Segmental peak
systolic strain curves. Bottom right Curved anatomic M-mode of longitudinal deformation
and abbreviate the scan, internal dimensions measured in the preindomethacin echocardiogram were used subsequently. Flow measurements using Doppler methods correlate well with those obtained using simultaneous recording with intracoronary Doppler guidewire [15, 31]. The previously validated ratio of mean arterial blood pressure to mean flow velocity was used as an independent estimator of relative vascular resistance in the perfusion region [4, 18, 30].
intraobserver variability was assessed by one investigator (AS) performing offline analysis of the same patients 2 weeks apart to reduce recall bias. The interobserver variability was assessed by a second investigator (TD), who was unaware of the previous results. These were assessed using the intraclass correlation coefficient, version 2.1. Significance was set at a P value lower than 0.05.
Statistical Analysis
Results
A sample size of convenience was chosen, and STE analysis was performed offline on the archived images. Descriptive statistics were used to characterize baseline clinical and echocardiographic measurements. Analysis of variance testing was used to analyze serial changes over time using a statistical program (SigmaPlot, version 11.0; Jandel Scientific, San Jose, CA). The Pearson correlation coefficient was used to assess correlation between echocardiography indices. Multiple comparisons versus control group (preindomethacin readings) were performed using the Holm-Sidak method. Every patient served as his or her own control. The
The study enrolled 18 infants born at a median gestational age of 25.8 weeks (interquartile range [IQR] 24.2–28.1) and weight of 773 g (IQR 704–1,002 g). The median age at commencement of the indomethacin treatment was 7.5 days [5, 11]. Volume boluses or inotropes were not administered to any of the subjects in the preceding 12 h, and both fluid administration and ventilator settings (median airway pressure: 10.5 [IQR 9–11], 10.2 [IQR 8.8–11.1], and 10.4 [IQR 8.9–11.1] cm of H2O) remained constant during the study period. All the infants were mechanically ventilated using conventional modes. No significant alteration in median
123
Pediatr Cardiol Table 1 Conventional two-dimensional Doppler parameters during three periods Variable Mean transductal diameter (mm) LVO (ml/kg/min)
b
Mean fractional shortening
Preindomethacin
Postindomethacin (10 min)
Postindomethacin (60 min)
P value
2.8 ± 0.5
2.5 ± 0.6
2.34 ± 0.4
0.02a
423.5 (133.7)
392.1 (127.7)
376.8 (139.3)
NS
43 ± 6.7
38.1 ± 7.2
37.9 ± 7.6
NS
Mean coronary velocity (cm/s)
20.6 ± 5
15.2 ± 4
18.1 ± 5
0.012
Mean coronary diastolic VTI (cm)
3.19 ± 1.2
2.01 ± 0.9
2.4 ± 0.9
0.004
Mean coronary vascular resistance (mmHg/cm/s)
1.68 ± 0.4
2.43 ± 0.6
1.89 ± 0.4
0.0004
Median systolic blood pressure (mmHg)
45 (41–50)
50 (45–51)
46 (41–50)
NS
Median diastolic blood pressure (mmHg)
22 (18–27)
24 (20–29)
25 (20–28)
NS
Mean blood pressure (mmHg)
32 (29–35)
34 (30–37)
33 (31–37)
NS
LVO left ventricular output, VTI velocity time integral Values are expressed are mean ± standard deviation or median (range) a
Between preindomethacin and 60 min after indomethacin
b
Median (interquartile)
Table 2 Intervariable correlations before and after indomethacin
Variable
Preindomethacin
Postindomethacin (10 min)
Postindomethacin (60 min)
Mean velocity and coronary vascular resistance
R2 = 0.6
R2 = 0.55
R2 = 0.44
P \ 0.0001
P = 0.0003
P = 0.002
Coronary diastolic VTI and coronary vascular resistance
R2 = 0.24
R2 = 0.34
R = 0.17
P = 0.04
P = 0.01
P = 0.08
R2 = 0.2
R2 = 0.63
R2 = 0.57
P = 0.06
P \ 0.0001
P = 0.0002
Intervariable correlations
VTI velocity time integral, LV GLS left ventricular global longitudinal strain
Fig. 2 Correlation between coronary vascular resistance and peak systolic global longitudinal strain immediately after indomethacin
123
Coronary vascular resistance and LV GLS
Pediatr Cardiol Table 3 Global and segmental strain in the three periods Variable
Preindomethacin (%)
Postindomethacin (10 min) (%)
Postindomethacin (60 min) (%)
Global longitudinal strain
–19.1 ± 2.4a
–15.9 ± 1.7a
–17.3 ± 1.8a
Basal septal strain
–17.9 ± 1.1
–17.2 ± 1
–17.5 ± 0.9
Middle septal strain
–18.6 ± 1.2a
–17.6 ± 1a
Apical septal strain Basal lateral strain Middle lateral strain Apical lateral strain
–18.1 ± 1.1
–19.6 ± 1.9
ab
–16.7 ± 1.6
a
–19.6 ± 1.6
a
–17 ± 1.3
a
–17.4 ± 1.6b
–21.3 ± 1.2
a
–17.8 ± 1.3
a
–18 ± 1.3a
–21.2 ± 2.4
b
ab
–19.9 ± 1.1a
–18.2 ± 2
–20 ± 2.2a
a
Significant pairwise comparisons; all global, septal wall, and inferior wall strain pairwise comparisons were significant
b
Significant pairwise comparisons; all global, septal wall, and inferior wall strain pairwise comparisons were significant
Table 4 Intraclass correlation and coefficient of variation for vascular and strain indices Echocardiographic parameter
Interobserver Intraclass correlation coefficient and coefficient of variation (%)
Intraobserver Intraclass correlation coefficient and coefficient of variation (%)
Global longitudinal strain
0.76 (8.9)
0.82 (6.5)
Coronary flow velocity Coronary diastolic VTI
0.85 (5.8)
0.87 (5.6)
0.79 (7.9)
0.86 (6.8)
response to indomethacin. Overall, the segmental strain was lower in the basal segments, and the basal septal segments did not show alteration after intervention. Table 4 depicts the intra- and interobserver correlation and the coefficient of variation for the vascular and strain indices. Six of the eight infants with very low diastolic coronary flows immediately after the indomethacin infusion became hypotensive during the subsequent duration of the 6-day indomethacin course and needed inotropic support.
Discussion
VTI velocity time integral
oxygen saturations (90 % [IQR 86–93 %]; 89 % [IQR 85–93 %]; 90 % [IQR 85–92 %]) or median transcutaneous carbon dioxide levels (49 mmHg [IQR 45–52 mmHg]; 51 mmHg [IQR 46–52 mmHg]; 52 mmHg [IQR 49–53 mmHg]) occurred. The left anterior descending coronary artery was clearly visualized in all cases. Table 1 depicts the conventional echocardiographic parameters at the three time points. The changes in LV output, although statistically nonsignificant, temporally coincided with the ductal response. Fractional shortening did not change significantly during the study period, whereas ductal constriction was noted. Coronary flow declined, and resistance increased in the absence of significant alterations in the systemic indices. Table 2 depicts the interrelationships between the coronary indices and GLS. Raised resistance coincided with impaired GLS, most prominently immediately after indomethacin administration (Fig. 2). Table 3 shows the LV longitudinal peak systolic global and segmental strain values during the study. A significant decrease in GLS was noted immediately after indomethacin infusion. The subsequent measurements taken 1 h later showed recovery but did not reach pre-infusion levels. The changes in LV deformation did not coincide with ductal
In the current study, we used serial measurements to assess the LV function in preterm infants administered indomethacin for closure of the PDA. Our study showed a significant impairment in myocardial deformation immediately after the infusion, with a partial recovery subsequently. This temporally coincided with an increase in coronary resistance. These changes occurred in the absence of major alterations in the systemic Doppler indices. Speckle-Tracking Echocardiography: Usefulness and Interactions With Vascular Indices STE is a non-Doppler method allowing quantitative assessment of myocardial function. Additional advantages include nondependence on the angle of insonation and assessment in multiple segments (regional) simultaneously [5]. Data derived by STE have been validated against tissue Doppler imaging and magnetic resonance imaging studies [7]. Its utility for understanding cardiac adaptation in clinical neonatal settings had been demonstrated previously [8, 11]. Czernik et al. [8] studied eight severely asphyxiated neonates undergoing whole-body hypothermia (33–34 °C), assessing LV longitudinal strain and fractional shortening at the start and end of hypothermia, immediately after
123
Pediatr Cardiol
rewarming, and at the age of 1 week. LV strain showed initial impairment followed by recovery, whereas fractional shortening remained unchanged. El-Khuffash et al. [11] studied GLS using 2D STE in 19 preterm infants (median gestational age, 25 weeks) before, 1 h after, and 18 h after PDA ligation. The GLS values were decreased significantly 1 h after the procedure (-19.7 ± -3.8 % vs. -11.5 ± -3.5 %; P = 0.001) but improved significantly at 18 h (-15.1 ± -2.9 %; P = 0.01). The decline in LV GLS values coincided with an increase in vascular resistance. The increase in vascular resistance after surgical duct ligation has been recognized previously [20]. We noted relative regional asynchrony. Basal segments showed reduced deformation, and the basal septal segment had the least serial deformation. Similar asynchrony (significantly lower values) in the basal segments has been noted in children with aortic stenosis [9]. This heterogeneous pattern could be determined by the orientation of the muscle fibers and its interaction with local wall stress and LV pressure. The segments with the highest wall stress were the basal part (predominantly the septum). Because the wall stress is higher, the contractile force is smaller, and systolic deformation is diminished. The presence of reduced systolic deformation in the basal septal segment is considered to be a sensitive indicator of increased pressure loading of the ventricle [6]. The sensitivity of STE compared with conventional echocardiography for detecting early myocardial dysfunction has been tested in various pediatric and adult clinical settings. Ng et al. [21] used 2D STE in 47 asymptomatic adults with early diabetic cardiomyopathy and noted impaired longitudinal LV cardiac function. Conventional markers between subjects and control subjects were comparable. Similar results were noted also in other studies showing that despite normal LV ejection fractions, significant impairments in longitudinal systolic strain were noted [9, 13, 14]. We noted impairment on STE in the setting of no major alterations in conventional parameters. Effects of COX Inhibition on End-Organ Vascular Indices In our study, the relationship between end-organ blood flow and resistance was most pronounced immediately after indomethacin infusion. The hemodynamic effects of indomethacin are due to formation of inhibiting prostaglandins or vasoconstrictive actions, and its stimulatory effect results in an increased vascular resistance [29]. Studies with pregnant ewes and newborn lambs have demonstrated increased umbilical and coronary vascular resistance as well as reduced flow after indomethacin administration [17, 19]. In rabbit hearts perfused by the
123
Langendorf technique, a combination of nitro arginine and indomethacin abolished desflurane-induced coronary vasodilation and significantly reduced coronary blood flow [3]. The effects of indomethacin on systemic vascular indices in preterm infants have been studied previously. In a study on preterm infants administered prophylactic indomethacin, Yanowitz et al. [30] noted decreased cerebral and mesenteric blood flow velocities accompanied by a significant increase in cerebral (38 %) and mesenteric (18 %) relative vascular resistance. During the study, the conventional echocardiographic measures such as cardiac output, stroke volume, and fractional shortening remained unchanged. Other investigators have shown a similar correlation in splanchnic and cerebral vasculatures and its evolution in response to indomethacin when administered at a postnatal age of 3–35 days [4]. In summary, evidence from animal and human studies indicates that indomethacin has transient but significant effects on various systemic circulations via prostaglandin/ leukotriene pathways, direct arteriolar vasoconstriction, or both. Findings have shown that the increase in vascular resistance after indomethacin therapy is similar to the raised systemic resistance after duct ligation. This physiologic relationship was first noted in preterm baboons, in which the temporal relationship between impaired LV performance and increased vascular resistance was identified [27], Subsequent prospective observational studies in human neonates have replicated the findings [20, 23]. Clinical Relevance The aforementioned finding has important clinical implications. Infants with a significant compromise in diastolic flow have needed subsequent inotropic support. Although echocardiographic evidence of cumulative impairment in coronary perfusion or cardiac function after completion of the 6-day course was not obtained, it is possible that alterations in coronary flow/resistance and interactions with cardiac function could underlie the need for inotropic support. Appleton et al. [2] previously proposed that the impaired cardiac function seen in premature infants after indomethacin administration may be explained by coronary vasoconstriction and myocardial ischemia. The relevance of these findings is even greater when taken in the context of previous data demonstrating ST segment depression on the electrocardiogram and elevated plasma troponin (biochemical marker of myocardial perfusion) in premature infants with a significant PDA [10, 28]. Way et al. [28] studied preterm infants (mean gestational age, 31.6 weeks; birthweight, 1.5 kg) and noted features of subendocardial ischemia. The authors suggested that this could be another mechanism contributing to
Pediatr Cardiol
refractory congestive heart failure in infants with a significant PDA. Compared with matched controls, the presence of a PDA also is associated with higher troponin levels. This may relate to diastolic steal and lower coronary blood flow leading to ischemia. Our study brings into focus the role of vascular resistance in elevating effects of cyclooxygenase inhibition in the setting of preexisting myocardial underperfusion. The importance of adequate myocardial perfusion toward maintaining cardiac output in both healthy and critically sick infants has been demonstrated previously [22, 25]. A direct linear relationship between coronary artery flow and cardiac systolic performance was demonstrated. We propose that the impairment in LV deformation in our study could be multifactorial, with reduction in flow and increased resistance in the coronary circulations playing important roles. Cyclooxygenase inhibitors affect the preterm myocardium, which is undergoing rapid postnatal adaptation related to gestational maturation and alterations of systemic and pulmonary resistances. These findings also give fresh impetus to the need to rationalize treatment strategies related to the medical closure of PDA.
Study Limitations This study had many limitations, including the small sample. The diameter of a coronary artery in this patient population ranged from 0.08 to 1 mm. Therefore, small errors may have caused significant errors in flow calculations. Images of adequate quality for STE analysis were available for apical four-chamber views only. Although longitudinal assessments from the apical four-chamber approach enable assessment of the interaction between inferolateral and anteroseptal segments simultaneously, radial and circumferential measurements could give additional information. Currently, it is not possible to measure true organ resistance, and we acknowledge this discrepancy. The measurement relies on the previously validated relationship of blood pressure and flow velocity.
Conclusions Our study demonstrated the usefulness of cardiac function assessment using the speckle-tracking method. LV strain was impaired, although conventional markers showed no significant alterations. Impairment in strain temporally coincided with elevation in flow resistance and was most pronounced immediately after indomethacin infusion.
References 1. Afonso S, Bandow GT, Rowe GG (1974) Indomethacin and the prostaglandin hypothesis of coronary blood flow regulation. J Physiol 241:299–308 2. Appleton RS, Graham TP, Cotton RB, Moreau GA, Boucek RJ (1988) Decreased early diastolic function after indomethacin administration in premature infants. J Pediatr 112:447–451 3. Beaussier M, Mouren S, Souktani R, Arthaud M, Massias L, Vicaut E, Lienhart A, Coriat P (2002) Role of nitric oxide and cyclooxygenase pathways in the coronary vascular effects of halothane, isoflurane, and desflurane in red blood cell-perfused isolated rabbit hearts. Br J Anaesth 88:399–407 4. Bel FV, Van Zoeren D, Schipper J, Guit GL, Baan J (1990) Effect of indomethacin on superior mesenteric artery blood flow velocity in preterm infants. J Pediatr 116:965–970 5. Belghitia H, Brette S, Lafitte S, Reant P, Picard F, Serri K, Lafitte M, Courregelongue M, Dos Santos P, Douard H, Roudaut R, DeMaria A (2008) Automated function imaging: a new operatorindependent strain method for assessing left ventricular function. Arch Cardiovasc Dis 3:163–169 6. Bijnens B, Cikes M, Butakoff C, Sitges M, Crispi F (2012) Myocardial motion and deformation: what does it tell us and how does it relate to function? Fetal Diagn Ther 32:5–16 7. Cho GY, Chan J, Leano R et al (2006) Comparison of twodimensional speckle and tissue velocity based strain and validation with harmonic phase magnetic resonance imaging. Am J Cardiol 47:789 8. Czernik C, Rhode S, Helfer S, Schmalisch G, Beuhrer C (2013) Left ventricular longitudinal strain and strain rate measured by 2D speckle-tracking echocardiography in neonates during whole body hypothermia. Ultrasound Med Biol 8:1343–1349 ¨ cal B, Orun UA, Ozgur S, Yılmaz O, Keskin M, 9. Dogan V, O Ceylan O, Karademir S, S¸ enocak F (2013) Strain and strain rate echocardiography findings in children with asymptomatic congenital aortic stenosis. Pediatr Cardiol 34:1152–1158 10. El-Khuffash AF, Molloy EJ (2008) Influence of a patent ductus arteriosus on cardiac troponin T levels in preterm infants. J Pediatr 153:350–353 11. El-Khuffash A, Jain A, Dragulescu A et al (2012) 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 25:1058 12. Ellison RC, Peckham GJ, Lang GJP, Talner NS, Lerer TJ, Lin L et al (1983) Evaluation of the preterm infant for patent ductus arteriosus. Pediatrics 71:364–372 13. Friedberg M, Sloarch C (2008) Relation between left ventricular regional radial function and radial wall motion abnormalities using two-dimensional speckle tracking in children with idiopathic dilated cardiomyopathy. Am J Cardiol 102:335–339 14. Ho E, Brown A, Barrett P, Morgan RB, King G, Kennedy MJ, Murphy RT (2010) Subclinical anthracycline- and trastuzumabinduced cardiotoxicity in the long-term follow-up of asymptomatic breast cancer survivors: a speckle-tracking echocardiographic study. Heart 96:701–707 15. Hozumi T, Yoshida K, Akasaka T, Asami Y, Ogata Y, Takagi T et al (1998) Noninvasive assessment of coronary flow velocity and coronary flow velocity reserve in the left anterior descending coronary artery by Doppler echocardiography: comparison with invasive technique. J Am Coll Cardiol 32:1251–1259 16. Jarvisalo MJ, Jartti L, Nanto-Salonen K, Irjala K, Ronnemaa T, Hartiala JJ et al (2001) Increased aortic intima-media thickness: a marker of preclinical atherosclerosis in high-risk children. Circulation 104:2943–2947
123
Pediatr Cardiol 17. Klautz RJM, van Bel F, Teitel DF, Steenduk P, Baan J (1993) Myocardial perfusion and performance after indomethacin administration in newborn lambs. Pediatr Res 33:295–301 18. Martinussen M, Brubakk AM, Vik T, Yao AC (1996) Mesenteric blood flow velocity and its relation to transitional circulatory adaptation in appropriate for gestational age preterm infants. Pediatr Res 39:275–280 19. McLaughlin MK, Brennan SC, Chez RA (1978) Effects of indomethacin on sheep uteroplacental circulations and sensitivity to angiotensin II. Am J Obstet Gynecol 132:430–435 20. McNamara PJ, Stewart L, Shivananda SP, Stephens D, Sehgal A (2010) Patent ductus arteriosus ligation is associated with impaired left ventricular systolic performance in premature infants weighing less than 1,000 g. J Thorac Cardiovasc Surg 140:150–157 21. Ng AC, Delgado V, Bertini M, van der Meer RW, Rijzewijk LJ, Shanks M, Nucifora G, Smit JW, Diamant M, Romijn JA, de Roos A, Leung DY, Lamb HJ, Bax JJ (2009) Findings from left ventricular strain and strain rate imaging in asymptomatic patients with type 2 diabetes mellitus. Am J Cardiol 104:1398–1401 22. Oskarsson G, Pesonen E (2000) Coronary blood flow abnormalities in neonates with aortic stenosis. J Pediatr 137:875–877 23. Sehgal A, McNamara PJ (2012) Coronary artery perfusion and myocardial performance after patent ductus arteriosus ligation. J Thorac Cardiovasc Surg 143:1271–1278 24. Sehgal A, Ramsden CA, McNamara PJ (2012) Indomethacin impairs coronary perfusion in infants with haemodynamically significant ductus arteriosus. Neonatology 101:20–27
123
25. Sehgal A, Wong F, Mehta S (2012) Reduced cardiac output and its correlation with coronary blood flow and troponin in asphyxiated infants treated with therapeutic hypothermia. Eur J Pediatr 171:1511–1517 26. Silverman NH, Schiller NB (1984) Cross-sectional echocardiographic assessment of cardiac chamber size and ejection fraction in children. Ultrasound Med Biol 10:757–769 27. Taylor AF, Morrow WR, Lally KP, Kinsella JP, Gerstmann DR, de Lemos RA (1990) Left ventricular dysfunction following ligation of the ductus arteriosus in the preterm baboon. J Surg Res 48:590–596 28. Way GL, Pierce JR, Wolfe RR, McGrath R, Wiggins J, Merenstein GB (1979) ST depression suggesting subendocardial ischemia in neonates with respiratory distress syndrome and patent ductus arteriosus. J Pediatr 95:609 29. Wennmalm A, Carlsson I, Edlund A, Eriksson S, Kaijser L, Nowak J (1984) Central and peripheral haemodynamic effects of nonsteroidal antiinflammatory drugs in man. Arch Toxicol Suppl 7:350–359 30. Yanowitz TD, Yao AC, Werner JC, Pettigrew KD, Oh W, Stonestreet BS (1998) Effects of prophylactic low-dose indomethacin on haemodynamics in very-low-birth-weight infants. J Pediatr 132:28–34 31. Yasuoka K, Harada K, Orino T, Tamura M, Takada G (2002) Left anterior descending coronary artery flow and its relation to age in children. J Am Echocardiogr 15:69–75
ORIGINAL ARTICLE
Cyclooxygenase Inhibitors in Preterm Infants With Patent Ductus Arteriosus: Effects on Cardiac and Vascular Indices Arvind Sehgal • Tejas Doctor • Samuel Menahem
Received: 23 January 2014 / Accepted: 25 April 2014 Ó Springer Science+Business Media New York 2014
Abstract Existing data suggest subendocardial ischemia in preterm infants with patent ductus arteriosus (PDA) and alterations in cardiac function after indomethacin administration. This study aimed to explore the evolution of left ventricular function by conventional echocardiography and speckle-tracking echocardiography (STE) and to ascertain the interrelationship with coronary flow indices in response to indomethacin. A prospective observational study was performed with preterm infants receiving indomethacin for medical closure of PDA. Serial echocardiography was performed, and the results were analyzed using analysis of variance. Intra- and interobserver variability was assessed using the intraclass correlation coefficient. Indomethacin was administered to 18 infants born at a median gestational age of 25.8 weeks (interquartile range [IQR], 24.2–28.1 weeks) with a birth weight of 773 g (IQR, 704–1,002 g). The median age of the infants was 7.5 days (IQR, 4–17). Global longitudinal strain (GLS) values significantly decreased immediately after indomethacin infusion (preindomethacin GLS, -19.1 ± 2.4 % vs. -15.9 ± 1.7 %; p \ 0.0001) but had improved at reassessment after 1 h (-17.4 ± 1.8 %). Conventional echocardiographic indices did not show significant alterations. A significant increase in arterial resistance in the coronary vasculature from 1.7 to 2.4 mmHg/cm/s was
A. Sehgal (&) T. Doctor Monash Newborn, Monash Children’s Hospital, Melbourne, Australia e-mail: [email protected] A. Sehgal Department of Pediatrics, Monash University, Melbourne, Australia S. Menahem Pediatric Cardiology, Monash Health, Melbourne, Australia
demonstrated. A significant correlation was noted between peak systolic GLS and flow resistance in the coronary vasculature. Significant changes in myocardial indices were observed immediately after indomethacin infusion. Compared with conventional methods, STE is a more sensitive tool to facilitate understanding of hemodynamics in preterm infants. Keywords Coronary
Strain Deformation Indomethacin
Introduction Speckle-tracking echocardiography (STE) is a well-established concept in adult and pediatric cardiology, and upcoming data show its clinical usefulness in neonates [9, 13, 14]. It measures myocardial deformation (wall motion strain), expressed as a percentage change from the original dimension. Previous studies have indicated STE to be more sensitive than conventional echocardiography for detecting early myocardial dysfunction in adults and children [9, 13, 14]. El-Khuffash et al. [11] demonstrated the feasibility and reliability of STE and noted significant changes in myocardial function immediately after surgical duct ligation. Impairment in left ventricular (LV) global longitudinal strain (GLS) was temporally associated with elevated vascular resistance. Association between two-dimensional (2D) STE-derived myocardial function and perfusion index (resistance) after exposure to cyclooxygenase inhibitors in human infants has not been investigated previously. The presence of a hemodynamically significant ductus arteriosus is seen in approximately 40 % of extremely low birthweight infants (\1,000 g), and indomethacin is
123
Pediatr Cardiol
commonly used for its medical closure [12]. Indomethacininduced cyclooxygenase inhibition results in decreased secretion of vasodilating prostaglandins and increased production of vasoconstrictor leukotrienes, leading to severe vasoconstriction. Prostaglandins play an important role in the regulation of coronary blood flow [1]. Interference with this autoregulation could alter coronary perfusion, with potential to affect myocardial function. Appleton et al. [2] previously described impairment in diastolic cardiac function 1 h after indomethacin (predominant coronary flow occurs in diastole), although no change in systolic function occurred. In addition, indomethacin also affects organ blood flow and end-organ resistance. An inverse correlation of splanchnic and cerebral blood flow with respective resistance after indomethacin administration has been demonstrated previously in preterm infants [4, 30], although a similar impact on the coronary vasculature is not known. This study aimed to investigate the effects of indomethacin infusion serially using conventional Doppler and 2D STE methods to understand better the effect of acute changes in resistance on the preterm myocardium.
Methods A prospective nonblinded observational echocardiography study performed with premature infants of less than 34 weeks gestation administered indomethacin for medical closure of a symptomatic patent ductus arteriosus (PDA). The diagnosis was initially suspected on the basis of clinical features (murmur, wide pulse pressure, bounding pulses, and/or hyperdynamic precordium) or early-onset hypotension. A significant duct was defined by a transductal diameter greater than 1.5 mm with unrestrictive (\1.5 m/s) left-to-right transductal flow on pulse-wave Doppler and clinical signs of pulmonary overcirculation or systemic hypoperfusion. Normative anatomy was confirmed by a preceding echocardiogram from the pediatric cardiology department. Approval of the study research ethics board and informed parental consent were obtained. At the time of the study, the department used indomethacin at a dose of 0.1 mg/kg as an intravenous infusion 1 h daily for 6 days. The assessments were made at the first dose of the first course only. The decision to treat was determined by the physician, with no influence from the researchers. The unit does not treat infants with necrotizing enterocolitis, renal failure, or right-to-left shunting. 2D echocardiography was performed within 2 h before indomethacin treatment and then again 10 min and 1 h after completion of infusion. All echocardiography evaluations were performed by a single operator (A.S.) with GE Vivid 7 equipment (GE Vingmed Ultrasound, Horten,
123
Norway) using a 7.5- to 10-MHz, high-frequency, phasedarray transducer probe. The study protocol was adapted from a previous study assessing the effects of indomethacin on splanchnic circulation [4]. Data analysis was performed offline. LV output was measured from an apical five-chamber view [26] and calculated as follows: Velocity time integral heart rate cross-sectional area indexed to body weight: LV dimensions for fractional shortening were measured using M-mode from a parasternal long-axis view at the level just distal to the mitral valve leaflet tips at end diastole. Transductal dimensions were calculated as described previously [16]. For STE in the apical fourchamber view, gain settings were optimized to obtain clear differentiation between cavity and wall. The endocardium was traced manually over a single frame, followed by automatic tracking of endocardial borders throughout the cardiac cycle. The left ventricle was divided into six segments: basal septal, middle septal, apical septal, apical lateral, middle lateral, and basal lateral (Fig. 1). The negative strain rate values in the apical view depict longitudinal shortening in systole. The timing of aortic valve closure was defined using the pulse Doppler recording from the apical four-chamber view. The Echopac algorithm generates seven curves that represent strain in the longitudinal direction (along the heart wall) for six specific myocardial segments and one global value that represents the combined strain from all the segments. A frame rate of at least 80 (range 80–112) was used for analysis. The left anterior descending coronary artery was interrogated for flow measurements as described previously [22, 24]. The left coronary artery was best identified from the standard parasternal short-axis view using color Doppler, and the left anterior descending branch could be seen by moving the transducer down one or two intercostal spaces, rotating it clockwise, and angling it superiorly. For color Doppler flow analysis, the scale was adjusted to an upper limit of 15–30 cm/s, allowing low-velocity signals to be identified. The pulse-wave Doppler sample gate was placed distal to the bifurcation, and the blood flow signal was obtained, keeping the angle of the Doppler beam as parallel as possible to the direction of flow (\15° in all recordings). Once a trace was obtained, the diastolic velocity time integral was calculated by tracing the area under the curve from an average of five consecutive beats. The internal dimensions were measured at end diastole with callipers applied to the endothelial border. No change in internal dimensions during the study time was observed in the initial six cases. Hence, to avoid errors
Pediatr Cardiol
Fig. 1 Top left Mapping of the region of interest in the left ventricle myocardium. Bottom left Peak segmental longitudinal strain values depicting the basal-to-apical gradient. Top right Segmental peak
systolic strain curves. Bottom right Curved anatomic M-mode of longitudinal deformation
and abbreviate the scan, internal dimensions measured in the preindomethacin echocardiogram were used subsequently. Flow measurements using Doppler methods correlate well with those obtained using simultaneous recording with intracoronary Doppler guidewire [15, 31]. The previously validated ratio of mean arterial blood pressure to mean flow velocity was used as an independent estimator of relative vascular resistance in the perfusion region [4, 18, 30].
intraobserver variability was assessed by one investigator (AS) performing offline analysis of the same patients 2 weeks apart to reduce recall bias. The interobserver variability was assessed by a second investigator (TD), who was unaware of the previous results. These were assessed using the intraclass correlation coefficient, version 2.1. Significance was set at a P value lower than 0.05.
Statistical Analysis
Results
A sample size of convenience was chosen, and STE analysis was performed offline on the archived images. Descriptive statistics were used to characterize baseline clinical and echocardiographic measurements. Analysis of variance testing was used to analyze serial changes over time using a statistical program (SigmaPlot, version 11.0; Jandel Scientific, San Jose, CA). The Pearson correlation coefficient was used to assess correlation between echocardiography indices. Multiple comparisons versus control group (preindomethacin readings) were performed using the Holm-Sidak method. Every patient served as his or her own control. The
The study enrolled 18 infants born at a median gestational age of 25.8 weeks (interquartile range [IQR] 24.2–28.1) and weight of 773 g (IQR 704–1,002 g). The median age at commencement of the indomethacin treatment was 7.5 days [5, 11]. Volume boluses or inotropes were not administered to any of the subjects in the preceding 12 h, and both fluid administration and ventilator settings (median airway pressure: 10.5 [IQR 9–11], 10.2 [IQR 8.8–11.1], and 10.4 [IQR 8.9–11.1] cm of H2O) remained constant during the study period. All the infants were mechanically ventilated using conventional modes. No significant alteration in median
123
Pediatr Cardiol Table 1 Conventional two-dimensional Doppler parameters during three periods Variable Mean transductal diameter (mm) LVO (ml/kg/min)
b
Mean fractional shortening
Preindomethacin
Postindomethacin (10 min)
Postindomethacin (60 min)
P value
2.8 ± 0.5
2.5 ± 0.6
2.34 ± 0.4
0.02a
423.5 (133.7)
392.1 (127.7)
376.8 (139.3)
NS
43 ± 6.7
38.1 ± 7.2
37.9 ± 7.6
NS
Mean coronary velocity (cm/s)
20.6 ± 5
15.2 ± 4
18.1 ± 5
0.012
Mean coronary diastolic VTI (cm)
3.19 ± 1.2
2.01 ± 0.9
2.4 ± 0.9
0.004
Mean coronary vascular resistance (mmHg/cm/s)
1.68 ± 0.4
2.43 ± 0.6
1.89 ± 0.4
0.0004
Median systolic blood pressure (mmHg)
45 (41–50)
50 (45–51)
46 (41–50)
NS
Median diastolic blood pressure (mmHg)
22 (18–27)
24 (20–29)
25 (20–28)
NS
Mean blood pressure (mmHg)
32 (29–35)
34 (30–37)
33 (31–37)
NS
LVO left ventricular output, VTI velocity time integral Values are expressed are mean ± standard deviation or median (range) a
Between preindomethacin and 60 min after indomethacin
b
Median (interquartile)
Table 2 Intervariable correlations before and after indomethacin
Variable
Preindomethacin
Postindomethacin (10 min)
Postindomethacin (60 min)
Mean velocity and coronary vascular resistance
R2 = 0.6
R2 = 0.55
R2 = 0.44
P \ 0.0001
P = 0.0003
P = 0.002
Coronary diastolic VTI and coronary vascular resistance
R2 = 0.24
R2 = 0.34
R = 0.17
P = 0.04
P = 0.01
P = 0.08
R2 = 0.2
R2 = 0.63
R2 = 0.57
P = 0.06
P \ 0.0001
P = 0.0002
Intervariable correlations
VTI velocity time integral, LV GLS left ventricular global longitudinal strain
Fig. 2 Correlation between coronary vascular resistance and peak systolic global longitudinal strain immediately after indomethacin
123
Coronary vascular resistance and LV GLS
Pediatr Cardiol Table 3 Global and segmental strain in the three periods Variable
Preindomethacin (%)
Postindomethacin (10 min) (%)
Postindomethacin (60 min) (%)
Global longitudinal strain
–19.1 ± 2.4a
–15.9 ± 1.7a
–17.3 ± 1.8a
Basal septal strain
–17.9 ± 1.1
–17.2 ± 1
–17.5 ± 0.9
Middle septal strain
–18.6 ± 1.2a
–17.6 ± 1a
Apical septal strain Basal lateral strain Middle lateral strain Apical lateral strain
–18.1 ± 1.1
–19.6 ± 1.9
ab
–16.7 ± 1.6
a
–19.6 ± 1.6
a
–17 ± 1.3
a
–17.4 ± 1.6b
–21.3 ± 1.2
a
–17.8 ± 1.3
a
–18 ± 1.3a
–21.2 ± 2.4
b
ab
–19.9 ± 1.1a
–18.2 ± 2
–20 ± 2.2a
a
Significant pairwise comparisons; all global, septal wall, and inferior wall strain pairwise comparisons were significant
b
Significant pairwise comparisons; all global, septal wall, and inferior wall strain pairwise comparisons were significant
Table 4 Intraclass correlation and coefficient of variation for vascular and strain indices Echocardiographic parameter
Interobserver Intraclass correlation coefficient and coefficient of variation (%)
Intraobserver Intraclass correlation coefficient and coefficient of variation (%)
Global longitudinal strain
0.76 (8.9)
0.82 (6.5)
Coronary flow velocity Coronary diastolic VTI
0.85 (5.8)
0.87 (5.6)
0.79 (7.9)
0.86 (6.8)
response to indomethacin. Overall, the segmental strain was lower in the basal segments, and the basal septal segments did not show alteration after intervention. Table 4 depicts the intra- and interobserver correlation and the coefficient of variation for the vascular and strain indices. Six of the eight infants with very low diastolic coronary flows immediately after the indomethacin infusion became hypotensive during the subsequent duration of the 6-day indomethacin course and needed inotropic support.
Discussion
VTI velocity time integral
oxygen saturations (90 % [IQR 86–93 %]; 89 % [IQR 85–93 %]; 90 % [IQR 85–92 %]) or median transcutaneous carbon dioxide levels (49 mmHg [IQR 45–52 mmHg]; 51 mmHg [IQR 46–52 mmHg]; 52 mmHg [IQR 49–53 mmHg]) occurred. The left anterior descending coronary artery was clearly visualized in all cases. Table 1 depicts the conventional echocardiographic parameters at the three time points. The changes in LV output, although statistically nonsignificant, temporally coincided with the ductal response. Fractional shortening did not change significantly during the study period, whereas ductal constriction was noted. Coronary flow declined, and resistance increased in the absence of significant alterations in the systemic indices. Table 2 depicts the interrelationships between the coronary indices and GLS. Raised resistance coincided with impaired GLS, most prominently immediately after indomethacin administration (Fig. 2). Table 3 shows the LV longitudinal peak systolic global and segmental strain values during the study. A significant decrease in GLS was noted immediately after indomethacin infusion. The subsequent measurements taken 1 h later showed recovery but did not reach pre-infusion levels. The changes in LV deformation did not coincide with ductal
In the current study, we used serial measurements to assess the LV function in preterm infants administered indomethacin for closure of the PDA. Our study showed a significant impairment in myocardial deformation immediately after the infusion, with a partial recovery subsequently. This temporally coincided with an increase in coronary resistance. These changes occurred in the absence of major alterations in the systemic Doppler indices. Speckle-Tracking Echocardiography: Usefulness and Interactions With Vascular Indices STE is a non-Doppler method allowing quantitative assessment of myocardial function. Additional advantages include nondependence on the angle of insonation and assessment in multiple segments (regional) simultaneously [5]. Data derived by STE have been validated against tissue Doppler imaging and magnetic resonance imaging studies [7]. Its utility for understanding cardiac adaptation in clinical neonatal settings had been demonstrated previously [8, 11]. Czernik et al. [8] studied eight severely asphyxiated neonates undergoing whole-body hypothermia (33–34 °C), assessing LV longitudinal strain and fractional shortening at the start and end of hypothermia, immediately after
123
Pediatr Cardiol
rewarming, and at the age of 1 week. LV strain showed initial impairment followed by recovery, whereas fractional shortening remained unchanged. El-Khuffash et al. [11] studied GLS using 2D STE in 19 preterm infants (median gestational age, 25 weeks) before, 1 h after, and 18 h after PDA ligation. The GLS values were decreased significantly 1 h after the procedure (-19.7 ± -3.8 % vs. -11.5 ± -3.5 %; P = 0.001) but improved significantly at 18 h (-15.1 ± -2.9 %; P = 0.01). The decline in LV GLS values coincided with an increase in vascular resistance. The increase in vascular resistance after surgical duct ligation has been recognized previously [20]. We noted relative regional asynchrony. Basal segments showed reduced deformation, and the basal septal segment had the least serial deformation. Similar asynchrony (significantly lower values) in the basal segments has been noted in children with aortic stenosis [9]. This heterogeneous pattern could be determined by the orientation of the muscle fibers and its interaction with local wall stress and LV pressure. The segments with the highest wall stress were the basal part (predominantly the septum). Because the wall stress is higher, the contractile force is smaller, and systolic deformation is diminished. The presence of reduced systolic deformation in the basal septal segment is considered to be a sensitive indicator of increased pressure loading of the ventricle [6]. The sensitivity of STE compared with conventional echocardiography for detecting early myocardial dysfunction has been tested in various pediatric and adult clinical settings. Ng et al. [21] used 2D STE in 47 asymptomatic adults with early diabetic cardiomyopathy and noted impaired longitudinal LV cardiac function. Conventional markers between subjects and control subjects were comparable. Similar results were noted also in other studies showing that despite normal LV ejection fractions, significant impairments in longitudinal systolic strain were noted [9, 13, 14]. We noted impairment on STE in the setting of no major alterations in conventional parameters. Effects of COX Inhibition on End-Organ Vascular Indices In our study, the relationship between end-organ blood flow and resistance was most pronounced immediately after indomethacin infusion. The hemodynamic effects of indomethacin are due to formation of inhibiting prostaglandins or vasoconstrictive actions, and its stimulatory effect results in an increased vascular resistance [29]. Studies with pregnant ewes and newborn lambs have demonstrated increased umbilical and coronary vascular resistance as well as reduced flow after indomethacin administration [17, 19]. In rabbit hearts perfused by the
123
Langendorf technique, a combination of nitro arginine and indomethacin abolished desflurane-induced coronary vasodilation and significantly reduced coronary blood flow [3]. The effects of indomethacin on systemic vascular indices in preterm infants have been studied previously. In a study on preterm infants administered prophylactic indomethacin, Yanowitz et al. [30] noted decreased cerebral and mesenteric blood flow velocities accompanied by a significant increase in cerebral (38 %) and mesenteric (18 %) relative vascular resistance. During the study, the conventional echocardiographic measures such as cardiac output, stroke volume, and fractional shortening remained unchanged. Other investigators have shown a similar correlation in splanchnic and cerebral vasculatures and its evolution in response to indomethacin when administered at a postnatal age of 3–35 days [4]. In summary, evidence from animal and human studies indicates that indomethacin has transient but significant effects on various systemic circulations via prostaglandin/ leukotriene pathways, direct arteriolar vasoconstriction, or both. Findings have shown that the increase in vascular resistance after indomethacin therapy is similar to the raised systemic resistance after duct ligation. This physiologic relationship was first noted in preterm baboons, in which the temporal relationship between impaired LV performance and increased vascular resistance was identified [27], Subsequent prospective observational studies in human neonates have replicated the findings [20, 23]. Clinical Relevance The aforementioned finding has important clinical implications. Infants with a significant compromise in diastolic flow have needed subsequent inotropic support. Although echocardiographic evidence of cumulative impairment in coronary perfusion or cardiac function after completion of the 6-day course was not obtained, it is possible that alterations in coronary flow/resistance and interactions with cardiac function could underlie the need for inotropic support. Appleton et al. [2] previously proposed that the impaired cardiac function seen in premature infants after indomethacin administration may be explained by coronary vasoconstriction and myocardial ischemia. The relevance of these findings is even greater when taken in the context of previous data demonstrating ST segment depression on the electrocardiogram and elevated plasma troponin (biochemical marker of myocardial perfusion) in premature infants with a significant PDA [10, 28]. Way et al. [28] studied preterm infants (mean gestational age, 31.6 weeks; birthweight, 1.5 kg) and noted features of subendocardial ischemia. The authors suggested that this could be another mechanism contributing to
Pediatr Cardiol
refractory congestive heart failure in infants with a significant PDA. Compared with matched controls, the presence of a PDA also is associated with higher troponin levels. This may relate to diastolic steal and lower coronary blood flow leading to ischemia. Our study brings into focus the role of vascular resistance in elevating effects of cyclooxygenase inhibition in the setting of preexisting myocardial underperfusion. The importance of adequate myocardial perfusion toward maintaining cardiac output in both healthy and critically sick infants has been demonstrated previously [22, 25]. A direct linear relationship between coronary artery flow and cardiac systolic performance was demonstrated. We propose that the impairment in LV deformation in our study could be multifactorial, with reduction in flow and increased resistance in the coronary circulations playing important roles. Cyclooxygenase inhibitors affect the preterm myocardium, which is undergoing rapid postnatal adaptation related to gestational maturation and alterations of systemic and pulmonary resistances. These findings also give fresh impetus to the need to rationalize treatment strategies related to the medical closure of PDA.
Study Limitations This study had many limitations, including the small sample. The diameter of a coronary artery in this patient population ranged from 0.08 to 1 mm. Therefore, small errors may have caused significant errors in flow calculations. Images of adequate quality for STE analysis were available for apical four-chamber views only. Although longitudinal assessments from the apical four-chamber approach enable assessment of the interaction between inferolateral and anteroseptal segments simultaneously, radial and circumferential measurements could give additional information. Currently, it is not possible to measure true organ resistance, and we acknowledge this discrepancy. The measurement relies on the previously validated relationship of blood pressure and flow velocity.
Conclusions Our study demonstrated the usefulness of cardiac function assessment using the speckle-tracking method. LV strain was impaired, although conventional markers showed no significant alterations. Impairment in strain temporally coincided with elevation in flow resistance and was most pronounced immediately after indomethacin infusion.
References 1. Afonso S, Bandow GT, Rowe GG (1974) Indomethacin and the prostaglandin hypothesis of coronary blood flow regulation. J Physiol 241:299–308 2. Appleton RS, Graham TP, Cotton RB, Moreau GA, Boucek RJ (1988) Decreased early diastolic function after indomethacin administration in premature infants. J Pediatr 112:447–451 3. Beaussier M, Mouren S, Souktani R, Arthaud M, Massias L, Vicaut E, Lienhart A, Coriat P (2002) Role of nitric oxide and cyclooxygenase pathways in the coronary vascular effects of halothane, isoflurane, and desflurane in red blood cell-perfused isolated rabbit hearts. Br J Anaesth 88:399–407 4. Bel FV, Van Zoeren D, Schipper J, Guit GL, Baan J (1990) Effect of indomethacin on superior mesenteric artery blood flow velocity in preterm infants. J Pediatr 116:965–970 5. Belghitia H, Brette S, Lafitte S, Reant P, Picard F, Serri K, Lafitte M, Courregelongue M, Dos Santos P, Douard H, Roudaut R, DeMaria A (2008) Automated function imaging: a new operatorindependent strain method for assessing left ventricular function. Arch Cardiovasc Dis 3:163–169 6. Bijnens B, Cikes M, Butakoff C, Sitges M, Crispi F (2012) Myocardial motion and deformation: what does it tell us and how does it relate to function? Fetal Diagn Ther 32:5–16 7. Cho GY, Chan J, Leano R et al (2006) Comparison of twodimensional speckle and tissue velocity based strain and validation with harmonic phase magnetic resonance imaging. Am J Cardiol 47:789 8. Czernik C, Rhode S, Helfer S, Schmalisch G, Beuhrer C (2013) Left ventricular longitudinal strain and strain rate measured by 2D speckle-tracking echocardiography in neonates during whole body hypothermia. Ultrasound Med Biol 8:1343–1349 ¨ cal B, Orun UA, Ozgur S, Yılmaz O, Keskin M, 9. Dogan V, O Ceylan O, Karademir S, S¸ enocak F (2013) Strain and strain rate echocardiography findings in children with asymptomatic congenital aortic stenosis. Pediatr Cardiol 34:1152–1158 10. El-Khuffash AF, Molloy EJ (2008) Influence of a patent ductus arteriosus on cardiac troponin T levels in preterm infants. J Pediatr 153:350–353 11. El-Khuffash A, Jain A, Dragulescu A et al (2012) 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 25:1058 12. Ellison RC, Peckham GJ, Lang GJP, Talner NS, Lerer TJ, Lin L et al (1983) Evaluation of the preterm infant for patent ductus arteriosus. Pediatrics 71:364–372 13. Friedberg M, Sloarch C (2008) Relation between left ventricular regional radial function and radial wall motion abnormalities using two-dimensional speckle tracking in children with idiopathic dilated cardiomyopathy. Am J Cardiol 102:335–339 14. Ho E, Brown A, Barrett P, Morgan RB, King G, Kennedy MJ, Murphy RT (2010) Subclinical anthracycline- and trastuzumabinduced cardiotoxicity in the long-term follow-up of asymptomatic breast cancer survivors: a speckle-tracking echocardiographic study. Heart 96:701–707 15. Hozumi T, Yoshida K, Akasaka T, Asami Y, Ogata Y, Takagi T et al (1998) Noninvasive assessment of coronary flow velocity and coronary flow velocity reserve in the left anterior descending coronary artery by Doppler echocardiography: comparison with invasive technique. J Am Coll Cardiol 32:1251–1259 16. Jarvisalo MJ, Jartti L, Nanto-Salonen K, Irjala K, Ronnemaa T, Hartiala JJ et al (2001) Increased aortic intima-media thickness: a marker of preclinical atherosclerosis in high-risk children. Circulation 104:2943–2947
123
Pediatr Cardiol 17. Klautz RJM, van Bel F, Teitel DF, Steenduk P, Baan J (1993) Myocardial perfusion and performance after indomethacin administration in newborn lambs. Pediatr Res 33:295–301 18. Martinussen M, Brubakk AM, Vik T, Yao AC (1996) Mesenteric blood flow velocity and its relation to transitional circulatory adaptation in appropriate for gestational age preterm infants. Pediatr Res 39:275–280 19. McLaughlin MK, Brennan SC, Chez RA (1978) Effects of indomethacin on sheep uteroplacental circulations and sensitivity to angiotensin II. Am J Obstet Gynecol 132:430–435 20. McNamara PJ, Stewart L, Shivananda SP, Stephens D, Sehgal A (2010) Patent ductus arteriosus ligation is associated with impaired left ventricular systolic performance in premature infants weighing less than 1,000 g. J Thorac Cardiovasc Surg 140:150–157 21. Ng AC, Delgado V, Bertini M, van der Meer RW, Rijzewijk LJ, Shanks M, Nucifora G, Smit JW, Diamant M, Romijn JA, de Roos A, Leung DY, Lamb HJ, Bax JJ (2009) Findings from left ventricular strain and strain rate imaging in asymptomatic patients with type 2 diabetes mellitus. Am J Cardiol 104:1398–1401 22. Oskarsson G, Pesonen E (2000) Coronary blood flow abnormalities in neonates with aortic stenosis. J Pediatr 137:875–877 23. Sehgal A, McNamara PJ (2012) Coronary artery perfusion and myocardial performance after patent ductus arteriosus ligation. J Thorac Cardiovasc Surg 143:1271–1278 24. Sehgal A, Ramsden CA, McNamara PJ (2012) Indomethacin impairs coronary perfusion in infants with haemodynamically significant ductus arteriosus. Neonatology 101:20–27
123
25. Sehgal A, Wong F, Mehta S (2012) Reduced cardiac output and its correlation with coronary blood flow and troponin in asphyxiated infants treated with therapeutic hypothermia. Eur J Pediatr 171:1511–1517 26. Silverman NH, Schiller NB (1984) Cross-sectional echocardiographic assessment of cardiac chamber size and ejection fraction in children. Ultrasound Med Biol 10:757–769 27. Taylor AF, Morrow WR, Lally KP, Kinsella JP, Gerstmann DR, de Lemos RA (1990) Left ventricular dysfunction following ligation of the ductus arteriosus in the preterm baboon. J Surg Res 48:590–596 28. Way GL, Pierce JR, Wolfe RR, McGrath R, Wiggins J, Merenstein GB (1979) ST depression suggesting subendocardial ischemia in neonates with respiratory distress syndrome and patent ductus arteriosus. J Pediatr 95:609 29. Wennmalm A, Carlsson I, Edlund A, Eriksson S, Kaijser L, Nowak J (1984) Central and peripheral haemodynamic effects of nonsteroidal antiinflammatory drugs in man. Arch Toxicol Suppl 7:350–359 30. Yanowitz TD, Yao AC, Werner JC, Pettigrew KD, Oh W, Stonestreet BS (1998) Effects of prophylactic low-dose indomethacin on haemodynamics in very-low-birth-weight infants. J Pediatr 132:28–34 31. Yasuoka K, Harada K, Orino T, Tamura M, Takada G (2002) Left anterior descending coronary artery flow and its relation to age in children. J Am Echocardiogr 15:69–75
