SHOCK, Vol. 44, No. 2, pp. 115Y120, 2015

THE EFFECT OF MILRINONE ON SPLANCHNIC AND CEREBRAL PERFUSION IN INFANTS WITH CONGENITAL HEART DISEASE PRIOR TO SURGERY: AN OBSERVATIONAL STUDY Maria Otilia Bianchi,*† Po-Yin Cheung,* Ernest Phillipos,* Abimael Aranha-Netto,† and Chloe Joynt* *Division of Neonatalogy, Department of Pediatrics, University of Alberta, Edmonton, Alberta, Canada; and † Department of Pediatrics, UNICAMP, Universidade Estadual de Campinas, Sa˜o Paulo, Brazil Received 3 Feb 2015; first review completed 23 Feb 2015; accepted in final form 7 Apr 2015 ABSTRACT—Despite the advancement in the postoperative care of neonates with congenital heart disease (CHD), there is little information on preoperative management of systemic and regional hemodynamics, which may be related to outcomes. We aimed to determine the preoperative effect of milrinone, a phosphodiesterase III inhibitor, on cardiac output and splanchnic and cerebral perfusion in neonates with CHD. Neonates with CHD requiring cardiac surgery were enrolled in a prospective, single-blinded study once a clinical decision of starting milrinone (0.75 2g/kg per minute intravenously) using institutional criteria was made. Demographic and clinical variables and outcomes were recorded. Combined cardiac output and measures of splanchnic (superior mesenteric and celiac arteries) and cerebral (anterior and middle cerebral arteries) perfusion were determined by Doppler studies at 0, 6, 24, and 48 h after milrinone infusion. Investigators were unaware of intervention time points and patients in analyzing blood flow measurements. Seventeen term (39.2 T 1.3 weeks) neonates were included with hypoplastic left-sided heart syndrome (78.5%) as the most common diagnosis. Combined cardiac output increased by 28% within 48 h (613 T 154 vs. 479 T 147 mL/kg per minute at baseline, P G 0.05). Superior mesenteric artery mean velocity increased at 6 h and throughout 48 h of milrinone infusion (P G 0.05). Peak and mean velocities at cerebral arteries increased with milrinone infusion (P G 0.05~0.08), and the corresponding changes at celiac artery were modest. There were no significant changes in splanchnic and cerebral resistive and pulsatility indices during milrinone infusion. Milrinone increases cardiac output with concurrent effects on splanchnic and cerebral blood flows during the short-term preoperative use in neonates with CHD. KEYWORDS—Cerebral, mesenteric, hemodynamics, congenital heart disease, milrinone, newborn, preoperative

INTRODUCTION

impacting their quality of life; and increasing mortality rates (4Y8). Indeed, NEC in patients with CHD occurs before surgery is associated with high morbidity, especially in infants with cyanotic heart disease (9, 10). Despite a growing wealth of literature, significant gaps exist in our knowledge about the timing and effects of preoperative therapeutic strategies to protect the brain and gut. Milrinone, a specific phosphodiesterase III inhibitor, increases cardiac output via inotropic and lusitropic effect (11, 12) and also decreases systemic vascular resistance (11, 13). Because of these qualities, milrinone can be utilized to manipulate the Qp:Qs ratio and increase cardiac output preoperatively in children awaiting cardiac surgery. However, the cerebral and splanchnic hemodynamic effects of milrinone preoperatively in neonates awaiting cardiac surgery for CHD have not been studied. Several authors have reported the use of Doppler sonography as an additional tool to measure the blood flow velocities and estimate the volume blood flow in the superior mesenteric artery (SMA). Harrison et al (14) used the resistive index at SMA as an estimate measure of mesenteric perfusion in neonates with CHD. The objective of this prospective study was to use Doppler ultrasound to assess the effect of milrinone on splanchnic and cerebral perfusion in neonates with CHD requiring inotropic support prior to cardiac surgery.

Congenital heart disease (CHD) refers to a spectrum of cardiac structural abnormalities in which the normal pattern of venous and/or arterial blood flow is interrupted or misdirected and can lead to persistent arterial desaturation, acidosis, and cardiac dysfunction. Establishing blood flow to the lungs or systemic circulation is often dependent on a connection between systemic and pulmonary arteries, usually via a patent ductus arteriosus, or an atrial or ventricular septal defect (1). The ductal-dependent lesions may have a diastolic steal in tissue perfusion through a systemic-to-pulmonary shunt. Elements of decreased cardiac output, reversed intestinal diastolic blood flow, and increased mesenteric vascular resistance could lead to abnormal blood flow patterns and the potential to compromise flow, perfusion, and oxygen delivery to vital organs including the intestines and brain (2, 3). These injuries may manifest as necrotizing enterocolitis (NEC) (4) and neurodevelopmental abnormalities (5), contributing to morbidity, nutritional compromise, and prolonged length of stay;

Address reprint requests to Po-Yin Cheung, MBBS, PhD, NICU, Rm 5027, Royal Alexandra Hospital, 10240 Kingsway Ave, Edmonton, Alberta, Canada T5H 3V9. E-mail: [email protected]. This study was funded by a Grant-in-Aid and a Pediatric Trainee Research Grant from the Women and Children_s Health Research Institute, University of Alberta, and an operating grant from the Canadian Institutes of Health Research (MOP10336). DOI: 10.1097/SHK.0000000000000388 Copyright Ó 2015 by the Shock Society

PATIENTS AND METHODS This was a prospective, single-blinded trial performed after institutional and ethics review board approval and parental informed consent between June 2008 115

Copyright © 2015 by the Shock Society. Unauthorized reproduction of this article is prohibited.

116

SHOCK VOL. 44, NO. 2

and July 2010 at the University of Alberta, Stollery Children_s Hospital Neonatal Intensive Care Unit. Inclusion criteria for the study population included neonates (1) less than 1 month old admitted to the hospital with a diagnosis of CHD requiring cardiac surgery and (2) in whom the attending neonatologist independently decided to start a milrinone infusion based on clinical criteria. Exclusion criteria included the concomitant diagnosis of major noncardiac congenital defects, suspected perinatal asphyxia, previous cardiac surgery, prematurity with gestational age at birth less than 36 weeks, previous suspected NEC, current sepsis, and prior milrinone use in the last 24 h. Standard support prior surgery for these cardiac lesions included the lowest dose of prostaglandins (0.005Y0.02 2g/kg per minute) to maintain a patent ductus arteriosus and placement of venous and arterial umbilical catheters. As per protocol in our unit, the independent neonatologist would consider starting milrinone in this population if at least two of the following were present: oliguria (urine output G2 mL/kg per hour at 924 h of life), increased lactate level in the blood (92 mmol/L), oversaturating (SaO2 990% as in single ventricle physiology), increasing metabolic acidosis, tachycardia (heart rate 9180 beats/min not related to fluid status), tachypnea with bilateral alveolar pulmonary edema on chest x-ray, or clinical assessment with poor perfusion. All enrolled infants were started on a continuous infusion of milrinone (Apotex Inc, Toronto, Ontario, Canada) at 0.75 2g/kg per minute without a loading dose as per institutional guidelines. The nonsedated neonate underwent the ultrasound examinations of the combined cardiac output and SMA, celiac, anterior cerebral, and middle cerebral (ACA and MCA, respectively) arteries in a supine position prior to initiation of milrinone and at preset times of 6, 24, and 48 h of milrinone infusion. Data collected also included gestational age, sex, birth weight, and type of CHD. Before each ultrasound scan, hemodynamic and respiratory variables were recorded and a blood gas was obtained. Data included heart rate; systolic, diastolic, and mean blood pressure; urine output; respiratory rate; SaO2; PaCO2; and lactate level. Each data set at a given time period for a given patient was assigned a randomly generated number that was attached to the variable sheet, blood gas, and Doppler measurements. A master list of numbers and patients was kept in a locked file by a study investigator not involved in data collection or interpretation. Arterial blood pressures were monitored with an indwelling arterial, umbilical, or peripheral arterial line (Philips Intellivue with Pressure Transducer CPJ840J6, Philips Healthcare, Andover, Mass). Hypotension was defined as drop in mean arterial pressure of 10% below baseline. The therapeutic use or weaning of concurrent inotropic therapy was at the discretion of the health care team. Preoperative treatment and concurrent inotropic support were recorded. Each patient acted as his/her own control. Arterial plasma samples were prepared at 6, 24, and 48 h of milrinone infusion and stored at j80 -C for milrinone level determinations. Using a high-performance liquid chromatographyYvalidated method (15, 16), plasma milrinone was determined with standards (0Y2,000 ng/mL) for calibration. The limit of quantification was 5 ng/mL.

BIANCHI

ET AL.

cardiac cycles. Resistive and pulsatility indices were calculated by [PSV j end-diastolic velocity] / PSV (Pourcelot) and [PSV j end-diastolic velocity] / MV (Gossling), respectively. All splanchnic and cerebral blood flow velocities were calculated with the Sigma Plot 11 (Systat Software Inc, Chicago, Ill) operated by a single and blinded operator.

Statistical analysis Demographic data are reported as mean T SD, number, and percentage. An independent statistician performed the statistical analysis with SAS System for Windows version 9.2 (SAS Institute Inc, Cary, NC). Repeated-measures analysis of variance (ANOVA) was used to compare differences in between quantitative variables on different times in each patient of blood flow velocities. Tukey post hoc test was used. Repeated-measures ANOVA was used if the data passed normality or equal variance testing. A significance level of P G 0.05 was used for all tests. A sample size of 17 neonates was determined using repeated-measures ANOVA for four time points with an ! of 0.05 and " of 0.20 to detect a 20% change in SMA resistive index based on our experimental findings in newborn piglets, pilot observations, and previous experience of 17% variance in the measurement of regional blood flows (17, 18).

RESULTS Seventeen singleton term neonates, with a mean gestational age of 39.2 T 1.3 weeks and birth weight of 3,289 T 450 g, were enrolled. Their demographic and clinical characteristics are shown in Table 1. Hypoplastic left-sided heart syndrome was the most common diagnosis (78.5%). One neonate had a chromosomal abnormality with a balanced translocation 46 XY t(10,19) (q11.2; q11.2). All neonates were outborn, and prostaglandin infusion was initiated at birth and continued for duct arteriosus patency until surgery. Six neonates were intubated and ventilated during the study period. One neonate had a stroke with seizures prior to milrinone infusion. None of the neonates had been fed prior to or during the study. The length of hospital stay was 35 T 16 days, and 16 neonates (94.1%) survived to hospital discharge. One infant was withdrawn of life support postoperatively because of multiorgan failure. There was no case of NEC in this population and no abnormal neurological events after milrinone infusion.

Doppler studies

Hemodynamic parameters during milrinone infusion

Two-dimensional echocardiograms and Doppler studies were performed using the SonoSite M-Turbo Ultrasound Machine (SonoSite Inc, Bothell) utilizing a 5- to 8-MHz transducer. The echocardiogram was used to calculate combined cardiac output, and the pulse Doppler flow velocity was recorded in the MCA and ACA, as well as SMA and celiac artery. The cardiac output was calculated from both cardiac outflow tracts with consideration for singleoutflow lesion presentations, and each recording was obtained at the emergence of the aortic and/or main pulmonary artery 2 mm above valve insertion. The combined cardiac output was estimated from the results of both velocity measurements obtained with pulsed Doppler multiplied by vessel cross sectional area and heart rate. The final result was normalized to birth weight. Doppler sagittal subcostal plane images of the SMA were obtained through the epigastrium, in the proximal portion of the SMA where flow is in a posterior to anterior direction, with a 5- to 8-MHz phased array transducer using a less than 15-degree angle correction. The celiac artery was interrogated near its origin from the aorta. To study the ACA flow, the transducer will be placed on the anterior fontanel, and the ACA was insonated at its curve around the corpus callosum. The MCA was interrogated via the anterior fontanel using a sagittal section. Three sets of measurements were taken at each time point to account for variability in the angle of insonation, and the mean value was used for data calculation. All recordings were digitally stored for further analysis under the random number assigned. All ultrasound measurements were obtained by a single sonographer experienced in the measurements. The analysis of the Doppler studies was done by two independent investigators (M.O.B. and E.P.) who were blinded to the time point of the intervention, identity and history of the patient, and specific cardiac defect. Peak systolic (PSV), end-diastolic, and mean (MV) velocities were averaged from measurements in three consecutive cardiac cycles per time point. Mean velocity was obtained by integrating the area under the velocity curve over three

Milrinone was commenced, on average at 12 h (range, 1Y168 h) of age as per previously outlined institutional guidelines. The association of decreased urine output and lactate of more than 2.0 mmol/L was the most common indication for milrinone infusion (n = 11, 64.7%). All studied participants received milrinone infusion for at least 48 h in the studied period of hemodynamic changes. Combined cardiac output increased from 479 T 147 to 587 T 150 mL/kg per minute (P G 0.05) after 6 h and was maintained at this significantly increased output (~25% over baseline) at 24 and 48 h (Table 2). A statistically significant 14% increase in the heart rate was noted after 6 h of milrinone infusion (142 T 14 to 161 T 10 beats/min), but this was transient, and the heart rate returned back to only 6% above baseline by 24 h (Table 2). There was no episode of hypotension or any significant difference in the serial measurements of mean arterial pressure from baseline during milrinone infusion (Table 2). The urine output increased and plasma lactate concentrations decreased significantly 24 h after milrinone infusion (Table 2). Plasma milrinone levels were recovered from 12 subjects (70.6%). Ten (83.3%), seven (58.3%), and 10 patients (83.3%) had the levels at 6, 24, and 48 h of infusion within the therapeutic level of 100 to 300 ng/mL (Table 2) (16).

Copyright © 2015 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK AUGUST 2015

HEMODYNAMICS AT MILRINONE INFUSION

TABLE 1. Characteristics of the studied neonates who received milrinone therapy prior to cardiac surgery Variable

Neonates (n = 17)

Vaginal delivery

11 (64.7)

Male sex

11 (64.7)

Gestational age, wk

39.2 T 1.3 3,289 T 450

Mean weight, g Patients on other concurrent inotropes

5 (29.4)

Patients on prostaglandin

17 (100)

Hypoplastic left-sided heart syndrome*

11 (64.7)

Chromosomal abnormality†

1 (5.9)

Neurological abnormalities‡

3 (17.6)

Necrotizing enterocolitis

0 (0)

Hospital stay, d

35 T 16

Death

1 (5.9)

Data are presented as n (percentage) or mean T SD. *Other heart defects include transposition of the great arteries (n = 2), unbalanced atrioventricular septal defect (n = 2), tetralogy of Fallot (n = 1), and double-outlet right ventricle (n = 1). † Balanced translocation 46 XY t(10,19) (q11.2; q11.2). ‡ Neurological findings prior to milrinone infusion include stroke, seizures, and upper-limb hypertonia.

Five received concurrent therapy of other inotropes (dobutamine, dopamine, epinephrine, norepinephrine; median duration of 12 h) during milrinone infusion (Table 1). Of these five patients, three arrived at our unit on dopamine (n = 2) or multiple inotropes (n = 1). The dopamine-treated patients were transitioned to milrinone as they demonstrated the aforementioned criteria, and dopamine was gradually stopped within 12 h of milrinone infusion. The patient on multiple inotropes did so throughout the milrinone infusion and deceased after surgery. One patient had epinephrine, and another had dobutamine added after 24 h of milrinone infusion at the discretion of the health care team, and both additional inotropes were stopped significantly prior to the 48-h study time point. Splanchnic blood flow during milrinone infusion

After 6 h of milrinone infusion, PSV and MV at SMA significantly increased from the respective pretreatment baseline (Table 3 and Fig. 1). The increases in SMA blood flow persisted

117

during the 48 h of milrinone infusion and were greater in percentage change above baseline than concurrent percentage change in cardiac output (Fig. 1). Celiac artery PSV was significantly higher than the baseline values at 24 h only and did not demonstrate the same marked increase as seen in the SMA (Table 3 and Fig. 1). Celiac artery MV did not change significantly during milrinone infusion. Cerebral blood flow during milrinone infusion

The PSV increased significantly in the ACA at 24 h of milrinone infusion and trended in the MCA during the infusion (P = 0.06) (Table 3). The MV at the ACA and MCA increased significantly over 24 h of milrinone infusion and remained significant in the ACA and trended in the MCA (P = 0.08) at 48 h. The temporal changes in ACA and MCA MV from respective baseline were similar to the cardiac output increases (Fig. 1). The PaCO2 value was analyzed, and although there was a significant increase over 24 h, the values remained within normocapnic range (39 T 6 at baseline to 47 T 10 mmHg) with no significant correlation between PaCO2 and cerebral flow detected (Table 2). Resistive and pulsatility indices were not significantly different with milrinone infusion in the SMA and celiac and cerebral arteries measured and remained relatively unchanged (Table 3). End-diastolic velocities were not significantly different from baseline for any of the arteries measured (data not shown). DISCUSSION Preoperative care in neonates with CHD strives to ensure adequate systemic blood flow and oxygen delivery. Milrinone, a type III phosphodiesterase inhibitor, has been a commonly reported agent used to balance the Qp:Qs preoperatively as it may alter vascular smooth muscle resistance as well as provide luisitropic and a positive inotropic support to further augment systemic blood flow with minimal changes in myocardial oxygen (11, 12, 17, 19Y22). A recent survey of 31 countries and 90 hospitals has demonstrated that 26.5% of neonates with hypoplastic left-sided heart syndrome had received milrinone combination therapy preoperatively, intraoperatively, and postoperatively (21). In a study of normal healthy term neonates, the cardiac output gradually increased by 27% over 5 days of life, which

TABLE 2. Systemic hemodynamics, clinical parameters, and milrinone levels in 17 neonates with CHD treated with milrinone infusion at 0.75 2g/kg per minute Time 0

Time after milrinone infusion

Parameter

Baseline

6h

Combined cardiac output, mL/kg per minute

479 T 147

587 T 150*

24 h 583 T 151*

48 h 613 T 154*

44 T 6

45 T 6

43 T 8

47 T 6

Heart rate, beats/min

141 T 14

161 T 10*

151 T 9*

150 T 10

Urine output, mL/kg per hour

1.59 T 1.42

2.61 T 1.21

3.04 T 1.43*

3.63 T 1.34*

Plasma lactate, mmol/L

2.55 T 1.23

2.34 T 0.93

1.77 T 0.89*

1.21 T 0.41*

Mean arterial pressure, mmHg

PaCO2, mmHg Plasma milrinone, ng/mL

39 T 6

41 T 12

47 T 10*

47 T 6

Not done

220 T 76

299 T 103

208 T 73

Data are presented as mean T SD. *P G 0.05 versus respective value at time 0 (repeated-measures ANOVA with Tukey post hoc test).

Copyright © 2015 by the Shock Society. Unauthorized reproduction of this article is prohibited.

118

SHOCK VOL. 44, NO. 2

BIANCHI

ET AL.

TABLE 3. Blood flow parameters for SMA and celiac and cerebral arteries in 17 neonates with CHD treated with milrinone infusion at 0.75 2g/kg per minute Time 0 Parameter

Time after milrinone infusion

Baseline

6h

24 h

48 h

50.2 T 23.3

86.4 T 35.5*

78.9 T 24.7*

91.1 T 36.3*

SMA PSV, cm/s

19.3 T 5.9*

20.9 T 11.8*

18.6 T 6.6*

Resistive index

1.18 T 0.17

1.14 T 0.13

1.11 T 0.21

1.15 T 0.15

Pulsatility index

5.35 T 2.69

5.60 T 2.59

5.27 T 3.25

6.43 T 4.03

PSV, cm/s

75.6 T 36.2

90.9 T 30.6

94.6 T 27.3*

89.7 T 26.1

MV, cm/s

29.8 T 22.6

31.2 T 14.1

32.9 T 13.8

36.9 T 14.5

Resistive index

0.96 T 0.19

0.97 T 0.15

0.91 T 0.14

0.91 T 0.10

Pulsatility index

3.11 T 2.01

3.08 T 1.31

2.94 T 1.31

2.76 T 1.31

PSV, cm/s

28.6 T 10.1

34.9 T 7.3

38.3 T 13.4*

37.7 T 10.1

MV, cm/s

14.6 T 5.9

17.7 T 5.8

19.6 T 9.2*

20.1 T 5.9*

Resistive index

0.81 T 0.13

0.77 T 0.12

0.75 T 0.13

0.76 T 0.15

Pulsatility index

1.67 T 0.58

1.66 T 0.77

1.66 T 0.76

1.55 T 0.89

MV, cm/s

11.39 T 4.7

Celiac artery

Anterior cerebral artery

Middle cerebral artery PSV, cm/s

41.6 T 10.3

47.7 T 9.2

49.1 T 12.2

49.7 T 12.2

MV, cm/s

18.3 T 3.8

23.0 T 6.9

24.0 T 8.9*

23.7 T 5.7

Resistive index

0.86 T 0.14

0.81 T 0.12

0.82 T 0.16

0.79 T 0.14

Pulsatility index

2.04 T 0.86

1.90 T 0.88

1.90 T 1.03

1.82 T 0.96

Data are presented as mean T SD. *P G 0.05 versus respective value at Time 0 (repeated-measures ANOVA with Tukey post hoc test).

was attributed to an increased stroke volume (23). In our study, neonates with CHD had an increase of 25% in cardiac output over their pretreatment baseline within 24 h of milrinone infusion, which would suggest that the increase in cardiac output was augmented by milrinone and not due to a pure physiological increase with age. The sustained increased cardiac output, even at 48 h, outlasted the transient increase in heart rate found at 6 h and may indicate that inotropy played a significant role in increased cardiac output of these preoperative hearts especially in light of the maintained blood pressure. There was no significant hypotension with the use of milrinone, which is consistent to our previous observation in hypoxic newborn piglets (17). Similar effects in cardiac output have been observed in neonates with milrinone after cardiac surgery (11). Evaluating successive Doppler images of blood flow velocity, we observed significant changes of cerebral, superior mesenteric, and celiac flows that acutely improved at 6 to 24 h after milrinone infusion. Developmental changes to SMA flow occur during the first days and weeks of life, including increasing PSV and MV, and these changes are thought to be caused by a decrease in peripheral vascular resistance, increasing blood pressure and stroke volume in the face of a closing ductus arteriosus (24, 25). We speculate that in our group of neonates with CHD the rapid change in regional flows was a response to milrinone infusion and was unlikely related to a normal adaptive response to postnatal life, as the

duct was kept opened and no measured changes in blood pressure or mesenteric vascular resistance were found. Of interest, SMA resistive indices in this preoperative population at the time of milrinone commencement were higher than those reported for normal-term infants (1.1 vs. 0.8) at similar chronological time points of life (24) and did not significantly

FIG. 1. Temporal effects of milrinone infusion (0.75 2g/kg per minute) on percentage changes in MVs from baseline at SMA, celiac artery, ACA, and MCA and combined cardiac output in 17 neonates with CHD.

Copyright © 2015 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK AUGUST 2015 decline over the first few days of life as would be expected in normal-term infants. These increased preoperative resistive indices are in keeping with others_ findings that neonates with hypoplastic left-sided heart syndrome have increased SMA resistive index (14). Despite increased resistive index, the SMA pulsatility index was preserved on milrinone, if not inflated, compared with normal-term babies (1.8Y2.1) at similar time points of life (26). Interestingly, Miller et al. (27) recently demonstrated that neonates with hypoplastic left-sided heart syndrome and NEC had a decreased preoperative abdominal aorta pulsatility index. Inadequate mesenteric blood flow may contribute to a significantly increased risk (10Y100 folds) of developing NEC in children with CHD (10). Our data of preoperative neonates with CHD indicate that mesenteric blood flow prior to milrinone was compromised with increased resistive index and lower baseline PSV and MV values compared with normative values of healthy term neonates. Thus, this suggests this group of neonates has an increased risk of developing mesenteric ischemia and NEC. Within 24 to 48 h of milrinone infusion, the SMA PSV and MV improved to values that were equivalent to healthy term neonates. The rate of change of SMA blood flow with milrinone infusion (started within 6 h and 180% in 48 h) is much faster than that expected by postnatal adaptation (115%Y130% over days and weeks) (24, 25, 28). A similar increase in SMA flows has been demonstrated in an animal model of hypoxia and reoxygenation (17, 29). While etiologies of diminished intestinal blood flow and NEC are multifactorial, it is encouraging that milrinone improved intestinal blood flows to approach that of normal-term infants in these neonates with CHD. No episode of NEC was recorded in this small studied population. In light of the relative brain-sparing effect of the neonatal circulation, the cerebral artery flows do not change significantly (10%) between 12 and 72 h of life (24). Our study demonstrated that neonates with CHD appear to have the same relative brain-sparing effect in the MCA with PSV and MV values near that of normal-term infants but to a lesser extent in the ACA. The cerebral arteries had increases in MV and PSV over the 48 h of milrinone that were greater (30%Y40% above baseline) than would be predicted by term infant adaptation. The resistive indices of both the ACA and MCA are higher than normal-term-neonate values and do not vary significantly with the milrinone infusion (Table 3) (24). Carbon dioxide, a well-described cerebral vasodilator, had significantly increased during the study evaluation, but its effect on cerebral circulation is correlated to the decrease in the cerebral vascular resistance (30) that was not apparent during the same time. Our study is limited by the absence of a comparable control group of neonates with CHD. However, a placebo-treated control group in these symptomatic neonates would not be ethical. Significant controversies that exist over the choice of inotropes, which is usually based on the discretion of practitioners, also make the comparison of effects with milrinone difficult. The protocol design focused only on the immediate hemodynamic effects of a standard dose of intravenous milrinone. The baseline for each neonate served as his/her own baseline, and only those who met predetermined criteria were

HEMODYNAMICS AT MILRINONE INFUSION

119

started on milrinone. Nevertheless, the uniform response during the short period of investigation makes it unlikely that there were random or otherwise unexpected hemodynamic changes. Furthermore, the possible effect of other inotropes used concurrently with milrinone in this study is possible but unlikely to significantly influence our data analysis and interpretation, given its low percentage and short duration in studied population receiving concurrent inotropic therapy. We acknowledge that as preoperative NEC and stroke are a rare occurrence in this population in our institution, a greater number of patients would be required to show a correlation between improved blood flows and clinical outcomes. In addition, there are no preoperative Bnormative data[ for changes in regional flows over time for neonates with CHD, especially in single ventricle. Thus, we compared the improvements in regional flows during milrinone infusion with the postnatal changes in healthy term neonates. Furthermore, because milrinone levels were measured in 71% of neonates and of whom 58% to 83% had levels within the therapeutic range (16), cautious interpretation of the milrinone effect is required. Because of the low cost, easy access, and clinical availability, Doppler ultrasound has been widely used to measure mesenteric blood flow in pediatric patients including neonates with CHD (2, 4, 7, 14). Lynch et al. (31) compared Doppler ultrasound and perfusion fluorometry regarding the sensitivity and clinical applicability in a model of bowel perfusion and ischemia. Doppler ultrasound compares favorably with perfusion fluorometry (sensitivity of 86% and 95%, respectively), because of low cost and accessibility lending to its use in the clinical and clinical research setting. In summary, our findings suggest that milrinone would increase cardiac output with concurrent effects on splanchnic and cerebral blood flows during the short-term preoperative use in neonates with CHD. The acute increase in SMA blood flow suggests milrinone therapy may have the potential to alleviate mesenteric hypoperfusion in these neonates at risk for NEC. Additional investigation is required to ascertain the dose-dependent effects of milrinone on regional circulation with controlled trials in neonates. ACKNOWLEDGMENTS The authors thank Cleide Aparecida Moreira Silva of Universidade Estadual de Campinas for statistical analysis. They acknowledge the technological support of the Cardiovascular Research Centre, University of Alberta.

REFERENCES 1. Wren C, Reinhardt Z, Khawaja K: Twenty-year trends in diagnosis of lifethreatening neonatal cardiovascular malformations. Arch Dis Child Fetal Neonatal 93:33Y35, 2008. 2. Cheung Y, Ho MHK, Cheng VYW: Mesenteric blood flow response to feeding after systemic-to-pulmonary arterial shunt palliation. Ann Thorac Surg 75:947Y951, 2003. 3. Martin CG, Snider AR, Katz SM, Peabody JL, Brady JP, Chir B: Abnormal cerebral blood flow patterns in preterm infants with a large patent ductus arteriosus. J Pediatr 101:587Y593, 1982. 4. Carlo WF, Kimball TR, Michelfelder EC, Border WL: Persistent diastolic flow reversal in abdominal aortic Doppler-flow profiles is associated with an increased risk of necrotizing enterocolitis in term infants with congenital heart disease. Pediatrics 119:330Y335, 2007. 5. Limperopoulos C, Majnemer A, Shevell M, Rosenblat B, Roblicek C, Tchervenkov C: Neurodevelopmental status of newborns and infants with

Copyright © 2015 by the Shock Society. Unauthorized reproduction of this article is prohibited.

120

6. 7.

8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

SHOCK VOL. 44, NO. 2

congenital heart defects before and after open heart surgery. J Pediatr 137: 638Y645, 2000. Sherlock RL, McQuillen PS, Miller SP, on behalf of ACCENT: Preventing brain injury in newborns with congenital heart disease. Stroke 40:327Y332, 2009. Murdoch EM, Sinha AK, Shanmugalingam ST, Smith GC, Kempley ST: Doppler flow velocimetry in the superior mesenteric artery on the first day of life in preterm infants and the risk of neonatal necrotizing enterocolitis. Pediatrics 118:1999Y2003, 2006. Hintz SR, Benitz WE, Colby CE, Sheehan AM, Rycus P, Van Meurs KP; ELSO Registry: Utilization and outcomes of neonatal cardiac extracorporeal life support: 1996Y2000. Pediatr Crit Care Med 6:33Y38, 2005. Dees E, Lin H, Cotton RB, Graham TP, Dodd DA: Outcome of preterm infants with congenital heart disease. J Pediatr 137:653Y659, 2000. McElhinney DB, Hedrick HL, Bush DM, Pereira GR, Stafford PW, Gaynor JW, Spray TL, Wernovsky G: Necrotizing enterocolitis in neonates with congenital heart disease: risk factors and outcomes. Pediatrics 106:1080Y1087, 2000. Hoffman TM, Wernovsky G, Atz AM, et al.: Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease. Circulation 107:996Y1002, 2003. Hoffman TM, Wernovsky G, Atz AM, et al.: Prophylactic intravenous use of milrinone after cardiac operation in pediatrics (PRIMACORP) study. Am Heart J 143:15Y21, 2002. Chang AC, Atz AM, Wernovsky G, Burke R, Wessel DL: Milrinone: systemic and pulmonary hemodynamic effects in neonates after cardiac surgery. Crit Care Med 23:1907Y1914, 1995. Harrison AM, Davis S, Reid JR, Morrison SC, Arrigain S, Connor JT, Temple ME: Neonates with hypoplastic left heart syndrome have ultrasound evidence of abnormal superior mesenteric artery perfusion before and after modified Norwood procedure. Pediatr Crit Care Med 6:445Y447, 2005. Brocks DR, Spencer TJ, Shayeganpour A: A sensitive and specific high performance liquid chromatographic assay for milrinone in rat and human plasma using a commercially available internal standard and low sample volume. J Pharm Pharm Sci 8:124Y131, 2005. Guerra G, Joffe A, Senthilselvan A, Kutsogiannis D, Parshuram C: Incidence of milrinone blood levels outside the therapeutic range and their relevance in children after cardiac surgery for congenital heart disease. Intensive Care Med 39:951Y957, 2013. Joynt C, Bigam DL, Charrois G, Jewell LD, Korbutt G, Cheung PY: Doseresponse effects of milrinone on hemodynamics of newborn pigs with hypoxiareoxygenation. Intensive Care Med 34:1321Y1329, 2008.

BIANCHI

ET AL.

18. Rigo V, Escoredo S, Robertson M, Campbell ME, Phillipos EZ: Regional flows after balancing circulations in hypoplastic left heart syndrome. [abstract] Can J Cardiol 19(Suppl):A716, 2003. http://hdl.handle.net/2268/41128 19. Krushanasky E, Burbano N, Morell V, Moguillansky D, Kim Y, Orr R, Chrysostomou C, Munoz R: Preoperative management in patients with single ventricle physiology. Congenit Heart Dis 7:96Y102, 2012. 20. Villa CR, Marino BS, Jacobs JP, Cooper DS: Intensive care and perioperative management of neonates with functionally univentricular hearts. World J Pediatr Congenit Heart Surg 3:359Y363, 2012. 21. Vogt W, Laer S: Prevention for pediatric low cardiac output syndrome: results from the European survey EuLoCOS-Paed. Paediatr Anaesth 21:1176Y1184, 2011. 22. Dugal B, Pratap U, Slavik Z, Kaplanova J, Macrae D: Milrinone and low cardiac output following cardiac surgery in infants: is there direct myocardial effect? Pediatr Cardiol 26:642Y645, 2005. 23. Mandelbaum VHA, Alverson DC, Kirchgessner A, Linderkamp O: Postnatal changes in cardiac output and haemorrheology in normal neonates at full term. Arch Dis Child 66:391Y394, 1991. 24. Ilves P, Lintrop M, Talvik I, Muug K, Asser K, Veinia M: Developmental changes in cerebral and visceral blood flow velocity in healthy neonates and infants. J Ultrasound Med 27:199Y207, 2008. 25. Havranek T, Thompson Z, Carver JD: Factors that influence mesenteric artery blood flow velocity in newborn preterm infants. J Perinatol 26:493Y497, 2006. 26. Matasova K, Dokus K, Zubor P, Danko J, Zibolen M: Physiological changes in blood flow in the superior mesenteric and coeliac artery in healthy term fetuses and newborns during perinatal period. J Matern Fetal Neonatal Med 24: 827Y832, 2011. 27. Miller TA, Minich LL, Lambert LM, Joss-Moore L, Puchalski MD: Abnormal abdominal aorta hemodynamics are associated with necrotizing enterocolitis in infants with hypoplastic left heart syndrome. Pediatr Cardiol 35:616Y621, 2014. 28. Papacci P, Giannantonio C, Cota F, Latella C, Semeraro CM, Fioretti M, Tesfagabir MG, Romagnoli C: Neonatal colour Doppler ultrasound study: normal values of abdominal blood flow velocities in the neonate during the first month of life. Pediatr Radiol 39:328Y335, 2009. 29. Joynt C, Bigam DL, Charrois G, Jewell LD, Korbutt G, Cheung PY: Intestinal hemodynamic effects of milrinone in asphyxiated newborn pigs after reoxygenation with 100% oxygen: a dose-response study. Shock 31:292Y299, 2009. 30. Archer LNJ, Evans DH, Paton JY, Levene MI: Controlled hypercapnia and neonatal cerebral artery Doppler ultrasound waveforms. Pediatr Res 20:218Y221, 1986. 31. Lynch TG, Hobson RW, Kerr JC, et al.: Doppler ultrasound, laser Doppler, and perfusion fluorometry in bowel ischemia. Arch Surg 123:483Y486, 1988.

Copyright © 2015 by the Shock Society. Unauthorized reproduction of this article is prohibited.