Assessment and Treatment of Post Patent Ductus Arteriosus Ligation Syndrome Assessment and Treatment of Post Patent Ductus Arteriosus Ligation Syndrome Afif F. EL-Khuffash, MD, DCE, FRCPI1, Amish Jain, MD2,3,4, Dany Weisz, MD5, Luc Mertens, MD6, and Patrick J. McNamara, MD, MRCPCH3,4,7,8 Objective To compare differences in tissue Doppler imaging, global longitudinal strain (GLS), and cardiac troponin T (cTnT) between infants with low (<200 mL/kg/min) and high (>200 mL/kg/min) left ventricular (LV) output 1 hour after duct ligation and assess the impact of milrinone treatment on cardiac output and myocardial performance. Study design LV function was assessed preoperatively and 1 and 18 hours postoperatively. Infants were catego- rized into a low-output or a normal-output group based on the echocardiographic assessment of LV output at 1 hour. Results Thirty infants with a mean gestation of 25.3 weeks were enrolled. LV basal lateral S0, basal septal S0, and basal right ventricular S0 were lower in the low-output group (n = 19) at 1 hour postoperatively, with no significant difference in GLS (low-output �10.3% vs high-output �14.4%, P >.05) or cTnT between the groups. Patients in the low-output group were treated with milrinone, and by 18 hours LV performance recovered to levels comparable with the high output group. cTnT values increased at 18 hours in the whole cohort with no significant difference between the groups. Conclusion Tissue Doppler imaging and GLS provide novel insights and further characterization of myocardial performance immediately after patent ductus arteriosus ligation. A reduction in tissue Doppler-derived LV systolic velocity may further help in monitoring cardiac performance after patent ductus arteriosus ligation and for monitoring the effects of treatment. (J Pediatr 2014;-:---). T he decision to undergo patent ductus arteriosus (PDA) ligation is subject to much debate and controversy, in part related to the high morbidity in survivors less than 1000 g.1 Postligation cardiac syndrome (PLCS) complicates the postoperative course in 40%-50% of preterm infants. PLCS is a clinical entity characterized by hypotension requiring inotropic support, and/or ventilation/oxygenation failure, usually occurring 6-12 hours after ligation.2,3 Several studies have investigated the physiologic mechanisms contributing to cardiorespiratory instability after PDA ligation and demonstrated a decrease in shortening fraction and ejection fraction associated with an increase in systemic vascular resistance (SVR) and a sudden reduction in preload.2,4 The sudden increase in SVR is not well tolerated by the preterm myocardium.3,5 Our group recently demonstrated that a left ventricular output (LVO) <200 mL/kg/min on echocardiography evaluation 1 hour after ligation is predictive of developing PLCS over the subsequent few hours.6 Early administration of milrinone to infants with low LVO reduces the incidence of PLCS from 44% to 11%.6 Conventional echocardiographic measures of left ventricular (LV) systolic performance, including fractional shortening and ejection fraction, do not predict PLCS.4 Previously, we demonstrated that PDA ligation resulted in significant changes in myocardial tissue Doppler measures and myocardial strain imaging. The aim of the present study was to characterize LV performance using tissue Doppler imaging (TDI) and speckle tracking echocardiography (STE) in neonates with high (>200 mL/min/kg) vs low (<200 mL/min/kg) LVO 1 hour after PDA ligation. We hypothesized that the tissue Doppler and strain measures at 1 hour postoperatively would be lower in infants with low LVO compared with those with a greater LVO. We also performed a longitudinal evaluation of TDI- and STE-derived estimates of LV systolic performance in the group with low LVO treated with milrinone. BP cTnT FiO2 GLS LV LVEDC LVO MAP Methods From the 1Department of Pediatrics, The Rotunda Hospital, Dublin, Ireland; 2Department of Pediatrics, All preterm neonates with birth weights between 500 and 1500 g and gestation between 24 and 32 weeks who were admitted to the Hospital for Sick Children Mount Sinai Hospital; 3Department of Physiology, University of Toronto; 4Physiology and Experimental Medicine, Hospital for Sick Children; 5Department of Pediatrics, Sunnybrook Health Sciences Centre; 6The Labatt Family Heart Center, The Hospital for Children; and 7Departments of Neonatology and 8Pediatrics, The Hospital for Sick Children, Toronto, Canada Funded by the Physicians’ Services Incorporated Foun- dation, Ontario, Canada. The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.03.048 Blood pressure Cardiac troponin T Fraction of inspired oxygen Global longitudinal strain Left ventricular LV end-diastolic circumference Left ventricular output Mean airway pressure PDA Patent ductus arteriosus PLCS Postligation cardiac syndrome PWD Pulsed-wave Doppler RV Right ventricular STE Speckle tracking echocardiography SVR Systemic vascular resistance TDI Tissue Doppler imaging VTI Velocity time integral 1 Delta:1_given name Delta:1_surname Delta:1_given name Delta:1_surname Delta:1_given name http://dx.doi.org/10.1016/j.jpeds.2014.03.048 THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. - for PDA ligation were considered eligible for inclusion. Infants with major congenital abnormalities, cardiac lesions other than PDA, or a patent foramen ovale were excluded from this study. Referrals for PDA ligation are triaged according to the clinical and echocardiography findings via use of the PDA staging system.7 All infants are anesthetized with fentanyl before the procedure at a dose ranging between 10 and 30 mg. Our institutional approach to the postoperative manage- ment of preterm infants after ligation is as follows. A routine echocardiography assessment is performed 1 hour postoper- atively to assess LVO.8 Infants with LVO less than 200 mL/kg/ min are considered at high risk of developing PLCS and are started on 0.33 mg/kg/min intravenous infusion of milrinone treatment. A bolus of 10 mL/kg of 0.9% saline is coadminis- tered for the first hour to prevent low diastolic blood pressure (BP). Infants who develop systolic hypotension (defined as a value less than the third percentile for any given gestation) in the postoperative period were treated with an intravenous infusion of 5-20 mg/kg/min dobutamine. Infants with diastolic hypotension (defined as a value less than the third percentile for any given gestation) were treated with boluses of 0.9% saline and/or intravenous infusion of 5-15 mg/kg/ min dopamine. Adrenocorticotropic hormone (cosyntropin) testing was performed on all patients before surgical intervention. Intravenous hydrocortisone 0.5 mg/kg every 6 hours was administered to all neonates with intractable hypotension and postadrenocorticotropic hormone cortisol <500 mmol/L despite cardiovascular treatment. The cohort was divided into low (LVO <200 mL/kg/min) and high (LVO >200 mL/kg/min) LVO groups based on the 1 hour postoperative echocardiogram. The study was approved by the institutional research ethics board, and written parental consent was obtained. Neonatal birth and postnatal demographics (antenatal steroid administration, Apgar scores, gestational age and weighs at birth, respiratory distress syndrome, surfactant administration, and ventilation days) were recorded. Details of medical and surgical treatment for the PDA also were collected. Serial evaluation of hemodynamic (eg, heart rate, BP, arterial pH, and base deficit) and respiratory variables (eg, fraction of inspired oxygen [FiO2], mean airway pressure [MAP]) was performed. Oxygenation index was calculated during the 3 time points via the following formula: oxygenation index = (MAP � FiO2 � 100) PaO2. Studies were performed with the use of a Vivid 7 or E9 ultrasound scanner (GE Healthcare, Milwaukee, Wisconsin) with a 10- or 12-MHz neonatal probe, respectively. Each infant underwent 3 echocardiography assessments: 4-6 hours preoperatively (preoperatively), 1 hour postoperatively before the administration of milrinone if needed (1 hour postoperatively), and 16-18 hours postoperatively (18 hours postoperatively). All studies were conducted in accordance with our standardized functional protocol, which follows recently published guidelines 9,10 and included TDI and STE imaging methods. All studies were stored as raw data for offline analysis (EchoPAC; GE, Milwaukee, Wisconsin). 2 Offline measurements were conducted after the infants were discharged from the unit. All infants underwent a preoperative complete echocardiographic assessment by a pediatric cardiologist to exclude congenital heart disease. We obtained the following measurements by using conventional echocardiography methods that have previously been described 10,11 : Assessment of PDA: (1) PDA diameter measured at the pulmonary end (2D imaging); and (2) the direction and peak velocity of flow across the shunt (pulsed-wave Doppler [PWD]). Markers of volume loading: (1) left atrial to aortic root ratio (M-mode); (2) mitral valve E and A velocities and the E wave to A wave ratio (PWD); and (3) LV end-diastolic diameter (M-mode). Measures of myocardial performance: (1) LV and right ventricular (RV) outputs (PWD + 2D measurement); (2) shortening fraction (M-mode); (3) ejection fraction measured by Simpson biplane method (ejection fraction) – (2D imaging); (4) tissue Doppler velocities (S0, E0, and A0) of the mitral, septal, and tricuspid annuli; (5) isovolumic contraction and isovolumic relaxation times measured by TDI; and (6) LV global longitudinal strain (GLS). LVO was calculated by measuring the blood velocity at the aortic valve annulus from an apical5-chamber view via a PWD and the diameter of the aortic annulus from a parasternal long-axis view by measuring the distance between the 2 visible hinges of the aortic valve in early systole. From the PWD tracing, the velocity time integral (VTI) was calculated. The aortic cross-sectional area was calculated using the formula: p � aortic diameter2/4. LVO was then indexed to weight as (mL/kg/min) = (aortic cross-sectional area � VTI � heart rate)/weight. The heart rate was calculated from the R-R interval obtained from electrocardiogram tracing of the aortic Dopplerflowimage.SVRwascalculatedbyusingthefollowing formula: (mean systemic BP � mean tricuspid valve inflow pressure gradient)/LVO.12 We also measured LV mean velocity of circumferential fractional shortening. It is determined by the following method13: velocity of circumfer- ential fractional shortening = (LVEDC � LV end-systolic circumference)/(LVEDC � ejection time corrected for heart rate [ejection time/ORR interval]), where LVEDC = LV end-diastolic circumference. Tissue Doppler velocities were obtained from the apical 4-chamber view. We used a PWD sample gate of 2 mm at the level of the annuli while always maintaining an angle of <20� between the pulsed-wave cursor and the longitudinal plane of motion. The sector width was minimized to improve the frame rate and temporal resolution. On the tissue Doppler traces we measured peak systolic (S0), early diastolic (E0), and late diastolic (A0) velocities. The isovolumic contraction and relaxation times and LV systolic time were measured (Figure 1; available at www.jpeds.com). For strain analysis, 2D grayscale images were recorded from the apical 4-, 3-, and 3-chamber views. GLS was calculated based on the segmental values as previously described by our group. 5 Image acquisition was carried out by 1 of 3 investigators (A.K., A.J., and D.W.). Offline image analysis was carried EL-Khuffash et al http://www.jpeds.com Table I. Patient demographics and clinical and PDA characteristics before ligation Low LVO, n = 19 High LVO, n = 11 P value Gestation, weeks 25.0 [24.4-25.6] 25.2 [24.9-26.6] .39 Birth weight, g 700 [620-825] 750 [640-874] .37 Gestation at ligation, weeks 28.7 [27.8-30.5] 29.9 [28.6-31.1] .42 Weight at ligation, g 948 [820-1071] 990 [874-1102] .66 Postdelivery age at ligation, days 25 [21-36] 30 [23-36] .66 Male 9 (50%) 6 (50%) 1.0 Cesarean delivery 7 (39%) 7 (58%) .46 5-minute Apgar score 7 [5-7] 8 [7-9] .003 Antepartum hemorrhage 1 (6%) 2 (17%) .55 Preeclampsia 2 (11%) 1 (8%) 1.0 Chorioamnionitis 1 (6%) 2 (17%) .55 Antenatal steroids .53 Partial 1 (6%) 2 (17%) Complete 12 (67%) 8 (67%) Days on invasive ventilation 23 [21-35] 24 [20-27] .71 Necrotizing enterocolitis 1 (6%) 3 (25%) .27 Intraventricular hemorrhage (III/IV) 3 (16%) 2 (17%) .48 NSAIDs treatment 16 (89%) 11 (82%) .60 Hemoglobin, g/dL 118 [109-129] 124 [91-129] .44 pH 7.29 [7.28-7.34] 7.31 [7.29-7.39] .79 Oxygen requirement, % 34 [30-40] 35 [26-36] .64 MAP, cm H2O 10 [9-10] 10 [9-10] .71 PDA diameter before ligation, mm 3.1 [2.3-3.4] 3.3 [2.7-3.9] .63 Maximum pressure gradient, mmHg 21 [13-26] 28 [12-39] .26 NSAIDs, nonsteroidal anti-inflammatory drugs. Values are expressed as median [IQR] or absolute value (%). - 2014 ORIGINAL ARTICLES out by 2 investigators (A.K. and D.W.). The intra- and inter- observer reliability of TDI and STE in the preterm population undergoing PDA ligation was previously reported by our group using the same observers.5 The components of the LVO calculation (heart rate, aortic VTI, aortic diameter) were entered into an electronic database, and LVO was calculated post-hoc. As a result, the observers were unaware of the LVO grouping assignment at the time of the analysis. Cardiac troponin T (cTnT) measurements were taken immediately after the preoperative echocardiogram, 1 hour postoperatively, and after the 18-hour postoperative echocardiogram. cTnT increases in the presence of a PDA in premature infants14 and is predictive of adverse neurodevelopmental outcomes associated with a PDA.15 cTnT is a measurement of myocardial damage that also may be associated with the hemodynamic loading.16 The preoperative and postoperative blood samples were performed with routine blood work and an additional 0.15 mL of blood was required per sample. The samples were tested using the Roche bedside cardiac reader (cobas h 232 POC system; Roche Diagnostics Limited, Rotkreuz, Switzerland). For low levels, the troponin reader provides a range in which the true value is: less than 0.03 mg/L or between 0.03 and 0.10 mg/L. A level greater than 0.10 mg/L is shown as an absolute value. We represented a level < 0.03 as “0” and a level between 0.03 and 0.10 as “0.065.” Statistical Analyses We presented data as means � SD for normally distributed variables and as medians and IQR for nonparametric data. Continuous variables between the 2 groups were compared with a Student t test or a Mann-Whitney U test as appropriate. Categorical data were compared using the c2 test or Fisher exact test as appropriate. We compared clinical and echo data between the 2 groups across the 3 time points using 2-way ANOVA with repeated measures. We assessed the correlation between various echocardiography measures using Pearson correlation coefficient for normally distributed data and Spearman correlation coefficient for skewed data. We accepted a P value of less than .05 as significant. The statistical analysis was performed using SPSS version 20 (SPSS Institute, Chicago, Illinois). Results Thirty infants were included in this study. Nineteen infants had LVO less than 200 mL/kg/min (low LVO group) and were started on an intravenous infusion of milrinone. Eleven infants had a LVO greater than 200 mL/kg/min (high LVO group) and did not require milrinone. With the exception of the 5-minute Apgar score, which differed by a single point only between the groups, there were no differences in baseline neonatal demographics, clinical or PDA characteristics between the 2 groups (Table I). None of the infants was on inotropic agents before ligation. In the low-LVO group, there was a greater proportion of infants requiring an escalation of FiO2 >20% (10 [53%] vs 1 [9%], P = .02), Assessment and Treatment of Post Patent Ductus Arteriosus Lig in the postoperative period. Similarly, there was a nonsignificant trend of infants in the low LVO group needing high-frequency oscillation (8 [42%] vs 2 [27%], P = .5) accompanied by an increase in MAP >20% (8 [42%] vs 2 [18%], P = .2). Only 2 (16%) of those infants also required inotropic agents and therefore met the criteria for PLCS. Both developed a systolic BP less than the third percentile for gestation between 6 and 12 hours postoperatively and received intravenous dobutamine treatment. None of the high LVO group developed PLCS. None of the infants in our cohort received hydrocortisone. We identified an increase in median (IQR) preoperative cTnT level from 0.07 mg/L (0.07-0.18) to 0.23 mg/L (0.16-0.28) at 18 hours after surgical intervention (P < .001, 2-way ANOVA). There were no differences in cTnT between groups at any time point. A decrease in LVO (P < .001, 1-way ANOVA) was demonstrated in the entire cohort after PDA ligation (Table II). The magnitude of the reduction from pre-operative values was greater in the low-LVO group (Table III; available at www.jpeds.com). There was, however, recovery in LVO by 18 hours in milrinone-treated patients (low-LVO group) to a level comparable with infants in the high-LVO group. An increase in SVR was seen in both groups 1 hour after PDA ligation (P < .001, 2-way ANOVA), although the magnitude of the increase was greater in the low-LVO group (Table II). By 18 hours, SVR decreased in milrinone-treated patients (low-LVO ation Syndrome 3 http://www.jpeds.com Table II. Conventional echocardiography markers Preoperative 1 hour postoperatively 18 hours postoperatively Group ANOVA Time ANOVALow LVO High LVO Low LVO High LVO Low LVO High LVO Heart rate 155 (12) 156 (18) 142 (13) 150 (18) 155 (17) 147 (14) .19 .79 Mean BP 37 (5) 41 (7) 42 (8) 43 (9) 38 (9) 54 (8.1)* .1 .008 Oxygenation index 10.3 (1.9) 10.0 (1.8) 9.4 (1.5) 9.5 (2.6) 11.0 (4.1) 10.4 (2.7) .23 .66 LVO, mL/kg/min 446 (145) 407 (82) 139 (32)* ,† 223 (44)* 222 (78)* 267 (48)* .19 <.001 RVO, mL/kg/min 256 (89) 263 (105) 202 (72) 301 (92) 267 (117) 344 (139) .04 .07 Shortening fraction (%) 37 (4) 38 (8) 24 (8)* 27 (6)* 30 (9)* 27 (6)* .81 <.001 Ejection fraction (%) 65 (9) 63 (8) 50 (11)* 56 (9)* 54 (10)* 55 (11)* .65 <.001 VcFs, circ/s 3.9 (0.5) 4.1 (0.8) 3.4 (1.2) 3.5 (0.9) 3.8 (1.1) 3.1 (1.0) .06 .52 LVEDD, mm 16.5 (2.6) 17.0 (2.2) 12.5 (2.4)* 13.8 (2.4)* 13.6 (2.7)* 15.1 (1.9)* .34 <.001 LA:Ao 2.5 (0.3) 2.5 (0.5) 1.7 (0.2)* 1.8 (0.4)* 1.8 (0.4)* 1.9 (0.4)* .42 <.001 Mitral valve E:A ratio 0.88 (0.17) 0.89 (0.08) 0.79 (0.24) 0.86 (0.14) 0.80 (0.16) 0.88 (0.12) .25 .27 Mitral valve E wave, cm/s 0.87 (0.25) 0.85 (0.12) 0.40 (0.07)* 0.51 (0.12)* 0.53 (0.16)* 0.57 (0.15)* .38 <.001 Mitral valve VTI 10.8 (2.1) 11.0 (2.7) 5.7 (1.4) 6.5 (1.5) 6.9 (1.9) 7.0 (1.8) .53 <.001 SVR, mmHg/kg/min 92 (49) 102 (27) 329 (137)*,† 198 (57)* 203 (106)* 198 (45)* .07 <.001 E:A, E wave to A wave; LA:Ao, left atrial to aortic root; LVEDD, LV end-diastolic diameter; NS, not significant (P > .05); VcFs, LV mean velocity of circumferential fractional shortening (corrected for heart rate). Values are presented as means (SD). Two-way, repeated-measures ANOVA was used to compare the difference in values across the 3 time points (ANOVA P). Pairwise comparisons were only performed, and the P value displayed, if the 2-way, repeated-measures ANOVA test yielded significant results. *P < .05 vs baseline. †P < .05 vs low-risk infants. THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. - group) to levels comparable with the high-LVO group (Table II). LV systolic function and ejection fraction decreased in the entire cohort after PDA ligation, but there were no intergroup differences. Tissue Doppler velocities were measurable in all infants across the 3 evaluation time points. Isovolumic timing measurements were not possible in 5 (6%) studies because of the difficulty in identifying the start and end of systolic and diastolic waves. GLS measurements were not possible in 8 (9%) scans because of poor image quality. There were early postoperative intergroup differences in tissue Doppler-derived systolic velocities. At 1 hour, LV S0, septal S0, and RV S0 were lower in low-LVO group compared with the high-LVO group (Table IV and Figure 2). The Table IV. GLS and TDI estimates of myocardial performance Preoperative 1 hour post postopera Low LVO High LVO Low LVO Hig LV lateral TDI, cm/s S0 5.1 (1.0) 4.8 (0.7) 3.1 (1.0)*,† 3 E0 8.0 (4.4) 6.3 (2.7) 3.2 (1.5)† 4 A0 9.7 (3.0) 8.1 (2.9) 4.6 (2.0)† 5 Septal TDI, cm/s S0 4.9 (1.0) 4.8 (0.6) 3.0 (0.7)*,† 4 E0 5.9 (2.0) 5.5 (1.6) 3.3 (1.2)† 5 A0 7.0 (1.4) 6.8 (1.3) 4.6 (1.1)† 5 RV TDI, cm/s S0 7.1 (1.7) 8.2 (1.6) 5.6 (1.7)*,† 7 E0 8.5 (3.5) 8.5 (2.1) 5.4 (2.1)† 7 A0 9.8 (3.0) 8.9 (2.0) 7.3 (1.6) 9 LV GLS, % �18.9 (2.6) �20.2 (5.4) �10.6 (3.2)† �14 LV event times LV systolic time, ms 169 (23) 172 (33) 145 (14)† 15 IVCT, ms 38 (14) 35 (6) 54 (16)† 5 IVRT, ms 46 (14) 41 (8) 69 (13)* ,† 5 IVCT, isovolumic contraction time; IVRT, isovolumic relaxation time; LVET, LV ejection time. Values are presented as means (SD). Repeated-measures ANOVA was used to compare the differenc the P value displayed, if the 2-way, repeated-measures ANOVA test yielded significant results. *P < .05 vs low-risk infants. †P < .05 vs baseline. 4 magnitude of change in systolic velocity from preoperative to 1 hour after intervention was only significant for LV lateral S0 and septal S0 (Table III). There was complete recovery by 18 hours in LV systolic velocities in the low-LVO group who received milrinone to levels comparable with the high-LVO group. GLS decreased in both groups after surgical intervention, with a nonsignificant trend towards lower values in the low-LVO group (�10.6 � 3.2 % vs �14.4 � 1.9%, P > .05). By 18 hours, GLS in the low-LVO group who received milrinone returned to a level comparable with the high- LVO group (Tables III, IV, and Figure 2). A negative correlation between SVR and tissue Doppler systolic velocities of the LV and septal walls (r = �0.6, tivley 18 hours postoperativley Group ANOVA Time ANOVAh LVO Low LVO High LVO .8 (0.9)† 4.3 (1.0)† 3.9 (0.8)† .87 <.001 .2 (1.7)† 4.2 (1.2)† 4.7 (1.7)† .85 <.001 .8 (3.0) † 5.5 (2.1) † 5.8 (1.8) † .94 <.001 .2 (0.6) † 4.1 (0.8) † 4.1 (0.3) † .14 <.001 .0 (1.8)† 4.8 (1.9)† 4.8 (1.3)† .22 <.001 .6 (1.5)† 5.8 (1.5)† 5.9 (1.1)† .20 <.001 .8 (1.9) 6.2 (1.5) 6.8 (0.9) .007 .02 .3 (2.9) † 6.2 (2.7) † 7.7 (3.1) † .08 <.001 .1 (2.1) 8.5 (2.3) 9.2 (2.6) .78 .81 .3 (1.9) �15.0 (3.3)† �14.9 (2.5) .6 <.001 3 (18)† 147 (18)† 151 (22)† .48 <.001 2 (9)† 40 (13)† 47 (8)† 1.0 <.001 6 (13) † 54 (11) † 58 (11) † .07 <.001 e in values across the 3 time points (ANOVA P). Pairwise comparisons were only performed, and EL-Khuffash et al Figure 2. LV, septal, and RV S0 velocities and LV GLS in the 2 groups across the 3 time points. Strain values are represented as a positive value for the purposes of graphical representation. The solid line represents the low-LVO group, and the dotted line represents the high-LVO group. The error bars represent the SE. *Depicts a significant difference between the groups at that time point (2-way ANOVA with repeated measures). - 2014 ORIGINAL ARTICLES P < .001 and r = �0.7, P < .001, respectively) was present at the preoperative and 1-hour evaluation. There was a weaker negative correlation between SVR and tissue Doppler diastolic velocities of the LV and septum: LV E0 (r = �0.4, P = .001), LV A0 (r = �0.4, P = .002), and septal E0 (r = �0.6, P < .001). There was no correlation between SVR and septal A0 (r = �0.2, P = .1). There was a negative correlation between SVR and GLS (r = �0.6, P < .001). There was a positive correlation between LVO and LV S0 (r = 0.6, P < .001), septal S0 (r = 0.7, P < .001), and GLS (r = 0.7, P < .001). There was a weak positive correlation between RV S0 and LVO (r = 0.3, P = .02). Similarly, there was a positive correlation between LVO and LV E0 (r = 0.6, P < .001), LV A0 (r = 0.6, P < .001), septal E0 (r = 0.6, P < .001), and septal A0 (r = 0.4, P = .004). Discussion Reference ranges for tissue Doppler velocities in extremely low birth weight infants are emerging.17-19 It is difficult to perform a direct comparison between preoperative TDI Assessment and Treatment of Post Patent Ductus Arteriosus Lig velocities in this cohort and previously published data, because most neonatal studies of TDI were performed in the infant’s first week of life. Nevertheless, the preoperative TDI systolic and diastolic velocities in this cohort were all generally greater than those presented in the literature for similar gestational-age infants. This finding may relate to the impact of a high-volume shunt present in our population increasing preload. It is also worth noting that after PDA ligation, infants in the low-LVO group had TDI velocities that were lower than the previously published values. We also observed significant changes in LV GLS in both groups. Although the magnitude of the change was larger in the lower output group, it did not reach statistical significance. This may relate, at least in part, to the greater variability of the strain measurements compared with the tissue Doppler measurements or to the small study sample size. The effect of the acute changes on tissue Doppler velocities is likely attributable to the load-dependency of the measures rather than a true decrease in LV contractility (intrinsic myocardial function). Evaluation of true myocardial contractility is not feasible with current echocardiography techniques, which ation Syndrome 5 THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. - provide information on myocardial systolic/diastolic performance alone.20 Longitudinal changes in LV function measured by tissue Doppler and strain measurements appear to be influenced by the acute changes in loading conditions. There was a significant negative linear relationship between peak systolic tissue Doppler velocitiesand SVRmeasurements. The same negative correlation was observed between the absolute value of GLS and SVR, suggesting that those markers of myocardial performance are strongly influenced by afterload changes. Although we accepted the limitations of the SVR calculation based on echocardiography, it was interestingto note that thelow-LVO group hadamuch greater SVR than the high-LVO group. In a study in animals, researchers demonstrated that strain was sensitive to acute changes in afterload and that strain rate measurements seem to be less affected.20 Strain rate measurements are, however, technically more difficult and are highly frame rate dependent. In preterm infants with greater heart rates, assessment of strain rate is technically challenging. LV diastolic velocities (E0 and A0) were lower after PDA ligation, but an intergroup difference was not seen. Early tissue Doppler velocities are influenced by different factors, including early relaxation, restoring forces determined by the amount of shortening during systole, and lengthening load which represents preload.21 Only LV isovolumic relaxation time at 1 hour was different with longer interval duration in the low-LVO group, which may relate to the acute volume unloading causing early relaxation abnormalities or may be a consequence of increased afterload.22 Diastolic function may contribute to the evolution of low LVO after ligation. The lack of intergroup differences in most measures of diastolic function suggests that the magnitude of the change in diastolic function (and preload) play a lesser role in the etiology of low LVO. The weak correlation between LVO, SVR, and the diastolic TDI variables further supports this reasoning. RV diastolic velocities were less influenced by ligation. It has been suggested that the reduction in LV systolic function post duct closure may relate to myocardial injury related to ischemia.23 For this reason, we included the use of biomarkers in the current project. cTnT levels were unchanged 1 hour after PDA ligation but increased at 18 hours. The levels were not different between the 2 groups. The lack of intergroup differences at 1 and 18 hours suggests that myocardial ischemia is not a contributory factor in the etiology of low LVO. However, it is possible that the delayed increase in cTnT at 18 hours is a consequence of its sudden exposure to increased SVR, analogous to the increase in plasma troponin that occurs in the transitional period.24,25 The nature of the elevation in cTnT may relate to the release of cytosolic cTnT not bound to the myofibril.26 Administration of milrinone to patients with low LVO was associated with recovery in LVO, LV lateral, and septal S0 velocities to levels comparable with the high-LVO group. Milrinone, a phosphodiesterase III inhibitor, increases tissue bioavailability of cyclic adenosine monophosphate, resulting in vasodilation, positive inotropy, and afterload reduction. 6 We previously demonstrated that the immediate postopera- tive administration of intravenous milrinone to infants with low LVO results in a reduction in the risk of development of PLCS.6 The presumed benefits of milrinone treatment were thought to relate predominantly to its vasodilator and afterload-reducing effects. In this study, we found that intravenous milrinone was associated with longitudinal improvement in TDI- and STE-derived indicators of systolic function by 18 hours. Although milrinone is known to possess a lusitropic effect, with the exception of septal E0, treatment was not associated with recovery in diastolic veloc- ities in the low LVO group by 18 hours of life. This finding suggests that the predominant effect of milrinone is afterload reduction coupled with an improvement in systolic function rather than lusitropy. There were, however, 2 infants in the low-output group who developed PLCS despite treatment with milrinone. It was not possible to further investigate factors contributing to the development of PLCS in these patients because of the low event rate. The physiologic nature of disease and causal factors in neonates with PLCS, despite milrinone treatment, requires further evaluation in larger groups. There were no obvious differences in any echocardiography markers of systolic performance between these 2 infants and the rest of treatment cohort. There are several limitations to the study design. First, because the study is not randomized and lacks a control group of high-risk infants that did not receive milrinone, it is not possible to be certain that the changes seen were related to treatment and not independent temporal changes. Second, the sample size is small, which prohibits making any meaningful statements of treatment effect. Third, as mentioned previously, it is not possible to accurately measure LV afterload in these patients. Also, the calculation of SVR is dependent on LVO, and thus may not represent true vascular resistance in this population. TDI and STE provide novel insights and further character- ization of LV systolic performance immediately after PDA ligation. Treatment with milrinone appears to have a positive influence on LV systolic function, although this needs confirmation in a prospective randomized controlled trial. n Submitted for publication Dec 20, 2013; last revision received Feb 12, 2014; accepted Mar 27, 2014. Reprint requests: Dr Patrick J. McNamara, MD, Hospital for Sick Children, University of Toronto, 555 University Avenue, Toronto, Ontario MG5 1X8. E-mail: patrick.mcnamara@sickkids.ca References 1. Kabra NS, Schmidt B, Roberts RS, Doyle LW, Papile L, Fanaroff A. Neurosensory impairment after surgical closure of patent ductus arteriosus in extremely low birth weight infants: results from the Trial of Indomethacin Prophylaxis in Preterms. J Pediatr 2007;150:229-34. 2. McNamara PJ, Stewart L, Shivananda SP, Stephens D, Sehgal A. Patent ductus arteriosus ligation is associated with impaired left ventricular systolic performance in premature infants weighing less than 1000 g. J Thorac Cardiovasc Surg 2010;140:150-7. 3. Teixeira LS, Shivananda SP, Stephens D, Van AG, McNamara PJ. Postoperative cardiorespiratory instability following ligation of the EL-Khuffash et al mailto:patrick.mcnamara@sickkids.ca http://refhub.elsevier.com/S0022-3476(14)00312-6/sref1 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref1 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref1 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref1 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref2 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref2 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref2 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref2 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref3 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref3 - 2014 ORIGINAL ARTICLES preterm ductus arteriosus is related to early need for intervention. J Perinatol 2008;28:803-10. 4. Kimball TR, Ralston MA, Khoury P, Crump RG, Cho FS, Reuter JH. Effect of ligation of patent ductus arteriosus on left ventricular performance and its determinants in premature neonates. J Am Coll Cardiol 1996;27:193-7. 5. 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. 6. 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. 7. McNamara PJ, Sehgal A. Towards rational management of the patent ductus arteriosus: the need for disease staging. Arch Dis Child Fetal Neonatal Ed 2007;92:424-7. 8. El-Khuffash AF, Jain A, McNamara PJ. Ligation of the patent ductus arteriosus in preterm infants: understanding the physiology. J Pediatr 2013;162:1100-6. 9. Lai WW, Geva T, Shirali GS, Frommelt PC, Humes RA, Brook MM, et al. Guidelines and standards for performance of a pediatric echocardiogram: a report from the Task Force of the Pediatric Council of the American Society of Echocardiography. J Am Soc Echocardiogr 2006;19:1413-30. 10. 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. 11. El-Khuffash AF, McNamara PJ. Neonatologist-performed functional echocardiography in the neonatal intensive care unit. Semin Fetal Neonatal Med 2011;16:50-60. 12. Noori S, Friedlich P, Seri I, Wong P. Changes in myocardial function and hemodynamics after ligation of the ductus arteriosus in preterm infants. J Pediatr 2007;150:597-602. 13. Osborn DA, Evans N, Kluckow M. Left ventricular contractility in extremely premature infants in the first day and response to inotropes. Pediatr Res 2007;61:335-40. 14. El-Khuffash AF, Molloy EJ. Serum troponin in neonatal intensive care. Neonatology 2008;94:1-7. Assessment and Treatment of Post Patent Ductus Arteriosus Lig 15. El-Khuffash AF, Barry D, Walsh K, Davis PG, Molloy EJ. Biochemical markers may identify preterm infants with a patent ductus arteriosus at high risk of death of severe intraventricular haemorrhage. Arch Dis Child Fetal Neonatal Ed 2008;93:407-12. 16. El-Khuffash A, Davis PG, Walsh K, Molloy EJ. Cardiac troponin T and N-terminal-pro-B type natriuretic peptide reflect myocardial function in preterm infants. J Perinatol 2008;28:482-6. 17. Negrine RJ, Chikermane A, Wright JG, Ewer AK. Assessment of myocardial function in neonates using tissue Doppler imaging. Arch Dis Child Fetal Neonatal Ed 2012;97:304-6. 18. 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 Z-score values. Neonatology 2011;100:85-92. 19. 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. 20. Ferferieva V, Van den Bergh A, Claus P, Jasaityte R, Veulemans P, Pellens M, et al. The relative value of strain and strain rate for defining intrinsic myocardial function. Am J Physiol Heart Circ Physiol 2012; 302:188-95. 21. Opdahl A, Remme EW, Helle-Valle T, Lyseggen E, Vartdal T, Pettersen E, et al. Determinants of left ventricular early-diastolic lengthening velocity: independent contributions from left ventricular relaxation, restoring forces, and lengthening load. Circulation 2009;119:2578-86. 22. Gewillig M, Daenen W, Aubert A, Van der Hauwaert L. Abolishment of chronic volume overload. Implications for diastolic function of the systemic ventricle immediately after Fontan repair. Circulation 1992; 86:II93-9. 23. Sehgal A, McNamara PJ. Coronary artery perfusion and myocardial performance after patent ductus arteriosus ligation. J Thorac Cardiovasc Surg 2012;143:1271-8. 24. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction redefined—a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959-69. 25. Noori S, Seri I. Pathophysiology of newborn hypotension outside the transitional period. Early Hum Dev 2005;81:399-404. 26. Katus HA, Remppis A, Scheffold T, Diederich KW, Kuebler W. Intracellular compartmentation of cardiac troponin T and its release kinetics in patients with reperfused and nonreperfused myocardial infarction. Am J Cardiol 1991;67:1360-7. ation Syndrome 7 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref3 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref3 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref4 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref4 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref4 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref4 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref5 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref5 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref5 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref5 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref6 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref6 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref6 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref6 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref7 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref7 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref7 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref8 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref8 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref8 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref9 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref9 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref9 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref9 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref9 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref10 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref11 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref11 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref11 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref12 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref12 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref12 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref13 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref13 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref13 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref14 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref14 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref15 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref15 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref15 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref15 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref16 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref16 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref16 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref17 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref17 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref17 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref18 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref18 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref18 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref18 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref18 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref19 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref19 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref19 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref20 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref20 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref20 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref20 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref21 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref21 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref21 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref21 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref22 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref22 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref22 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref22 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref23 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref23 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref23 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref24 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref24 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref24 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref24 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref25 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref25 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref26 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref26 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref26 http://refhub.elsevier.com/S0022-3476(14)00312-6/sref26 Figure 1. Tissue Doppler velocities of the LV in a preterm in- fant after PDA ligation. The different phases of the cardiac cy- cle are clearly identified. The systolic velocity (S0) and early (E0) and late (A0) diastolic velocities can easily be measured. Iso- volumic relaxation time (IVR and a) time and isovolumic contraction (IVC and b) time in addition to systolic time (c) can also be measured. Table III. Comparison of the change in LVO, SVR, and function parameters between the groups 1 hour postoperatively � preoperatively P value 18 hour postoperatively � 1 hour postoperatively P valueLow LVO High LVO Low LVO High LVO LVO, mL/kg/min �307 (138) �183 (85) .01 +83 (66) + 43 (69) .13 SVR, mmHg/kg/min +237 (130) +96 (62) .003 �125 (140) 0 (80) .01 Shortening fraction, % �12 (7) �11 (8) .4 +5 (8) 0 (5) .1 Ejection fraction, % �15 (8) �6 (9) .009 +4 (13) �1 (14) .3 VcFs, circ/s �0.5 (0.9) �0.6 (1.0) .8 +0.4 (1.1) �0.5 (1.2) .06 LV S0, cm/s �2.1 (1.7) �1.0 (0.7) .03 +1.2 (1.4) 0.1 (0.7) .008 LV E0, cm/s �4.7 (3.6) �2.7 (2.1) .2 1.0 (1.7) 0.5 (0.8) .4 LV A0, cm/s �5.1 (3.5) �2.5 (2.5) .06 1.0 (3.3) �0.1 (3.8) .5 Septal S0, cm/s �1.9 (1.1) �0.6 (1.1) .004 +1.1 (0.7) �0.1 (0.6) <.001 Septal E0, cm/s �2.6 (2.2) �1.1 (1.9) .09 +1.4 (1.7) 0 (1.5) .01 Septal A0, cm/s �2.3 (2.0) �1.6 (1.0) .3 +1.2 (1.8) 0 (1.5) .2 GLS, % �8.6 (3.9) �5.7 (6.4) .2 +4.5 (3.5) 0.5 (3.0) .006 Comparisons are made between preoperative to 1 hour postoperative and between 1 hour postoperative to 18 hours postoperative. A positive sign (+) depicts an increase in the parameter between the compared time points. A negative sign (�) depicts a reduction in the value of the parameter between the compared time points. THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. -, No. - 7.e1 EL-Khuffash et al Assessment and Treatment of Post Patent Ductus Arteriosus Ligation Syndrome Methods Statistical Analyses Results Discussion References