Perinatal/Neonatal Case Presentation and Right Ventricular Dysfunction in Growth-Restricted, Extremely Low Birth Weight Neonates

Olivier Danhaive, MD hypotensive episode. The echocardiograms, performed during the Rene´e Margossian, MD acute episode as part of the work-up, showed severe pulmonary Tal Geva, MD hypertension and right ventricle (RV) dysfunction. The goals of this Stella Kourembanas, MD report are to describe the clinical and hemodynamic features of these patients, and to discuss the pathophysiology and the therapeutic implications of this condition.

This case series reports an acute episode of hypoxemia and systemic PATIENTS AND METHODS hypotension in seven infants under 1 kg, taking place several hours or days after birth, after a period of stability and in the absence of significant lung These cases were observed between December 1998 and December disease. These patients were growth-restricted at birth and had a history of 2000 in two Harvard University-affiliated teaching centers in chronic fetal and oligohydramnios. Pulmonary hypertension and Boston, MA, USA. During that period, a total of 153 newborns right ventricular dysfunction were found by echocadiography. Right <1 kg had been hospitalized, 49 of which had a birth weight 2 ventricular ejection fraction was significantly depressed during the acute <10th percentile for gestational age (32%). Echocardiograms episode, compared to baseline values measured after recovery. The timing of were performed with a Hewlett Packard Sonos 5500 cardiac scanner symptoms seemed related to ductus arteriosus closure or constriction. using transducers appropriate for patient size. The diagnosis of Oxygenation and right ventricular function improvement occurred within a pulmonary hypertension was established on the basis of tricuspid few days under ventilatory and inotropic support, while milrinone was regurgitant jet velocity or, when impossible, on the systolic administered in five cases. In conclusion, pulmonary hypertension is a rare configuration of the ventricular septum. RV pressure greater than but significant cause of hypoxemia in preterm infants, and pulmonary half-systemic or a flattened septum were reported as ‘‘moderate’’, vasodilator therapy should be considered in the presence of right ventricular whereas RV pressure equal or superior to systemic or a septum 3 dysfunction. bulging into the left ventricle were reported as ‘‘severe’’. RV Journal of Perinatology (2005) 25, 495–499. doi:10.1038/sj.jp.7211299 dysfunction was assessed qualitatively as ‘‘mild’’, ‘‘moderate’’ or Published online 14 April 2005 ‘‘severe’’. RV ejection fraction was determined using a previously described technique based on standard two-dimensional echocardiograms, in which RV systolic and diastolic volumes are calculated using the formula: volume ¼ 2/3 area  maximum length.4 All ultrasound scans were reviewed by a single cardiologist INTRODUCTION (R.M.), who performed the morphological analyses and Idiopathic pulmonary hypertension in the absence of significant measurements. Each patient was used as his/her own control, by lung disease is a rare finding in very low birth weight infants, in comparing data obtained during the acute episode with those after which hypoxemia typically results of surfactant deficiency, clinical improvement. Medical chart review was approved by the increased alveolar–arterial gradient, ventilation–perfusion institutional review boards of the two hospitals. mismatch and intrapulmonary shunting. 1 We report a series of seven premature infants with an extremely low birth weight who presented initially with a mild form of respiratory distress syndrome RESULTS or no lung disease visible on chest radiograms, then, following a Pre- and Perinatal History (Table 1) period of cardio-respiratory stability, had an acute hypoxemic and Three of the were multiple. Routine fetal infection screenings were unremarkable, and there was no evidence of fetal malformations on morphological ultrasonograms. Signs of fetal distress, such as intrauterine growth restriction, decreased Department of Neonatology (O.D.), Bambino Gesu` Children’s Hospital, Rome, Italy; Division of movements, nonreassuring heart rhythm or low 1-minute Apgar Newborn Medicine (O.D., S.K.), Children’s Hospital, Boston, MA, USA; and Department of Cardiology (R.M., T.G.), Children’s Hospital, Boston, MA, USA. score were present in all cases. -induced hypertension,

Address correspondence and reprint requests to Olivier Danhaive, MD, Department of pre-eclampsia, oligohydramnios or reversed placental end-diastolic Neonatology, Bambino Gesu` Children’s Hospital, 4 Piazza Sant’Onofrio, Rome 00165, Italy. flow were reported in most, suggesting placental insufficiency as an

Journal of Perinatology 2005; 25:495–499 r 2005 Nature Publishing Group All rights reserved. 0743-8346/05 $30 www.nature.com/jp 495 Danhaive et al. Pulmonary Hypertension in Growth-Restricted ELBW Infants

Table 1 Pre- and Perinatal Data

Case Sex GA BW Apgar Pre- and perinatal history

1 M 29 weeks 650 g 2–6 Pre-eclampsia, moderate oligohydramnios since 28 140 g (<10th p), hypermature placenta, decidual <10th p weeks, IUGR, reversed EDPF, absent fetal movements vasculopathy, two infacts (3 and 1.5 cm) 2 F 26 weeks 450 g 1–5 PIH, vaginal , IUGR since 24 weeks, severe 235 g (50th p), hypermature placenta, <5th p oligohydramnios, reversed EDPF, absent fetal circumvallate insertion, , movements chronic abruptio and retroplacental hematoma 3 F 32 weeks 490 g 8–9 Quadruplet reduced to twin, pre-eclampsia, extreme 390 g fused twin placenta, marginal cord insertion <3rd p IUGR and absent EDPF since 24w, oligohydramnios; decreased fetal movements 4 F 25 weeks 750 g 5–7 Placental hematoma since 12 weeks, fetal growth 230 g (50th p) circumvallate insertion, chronic 25th p arrest and absent EDPF since 23 weeks abruptio and subchorionic hematoma 5 M 29 weeks 655 g 6–8 Dizygotic twin, discordant IUGR Not available <3rd p 6 M 29 weeks 790 g 7–7 Vaginal bleeding, placenta previa, prolonged 186 g (<10th p) hypermature placenta, <10th p premature rupture of membranes since 28 weeks, velamentous cord insertion, multiple infarcts oligohydramnios 7 F 29 weeks 450 g 5–7 Dizygotic twin 2nd born, PIH, discordant IUGR, severe Single umbilical artery <3rd p oligohydramnios, absent fetal movements

M: male; F: female; p: percentile; Apgar: scores at 1 and 5 minutes; PIH: pregnancy-induced hypertension; IUGR: intrauterine growth restriction; EDPF: end-diastolic placental flow; percentiles of BW are expressed for gestational age; percentiles of placental weight are expressed for BW.9 etiology. All the infants were delivered by cesarean section. Five therapy for 4 and 6 months, respectively. Two babies died: patient 6 mothers (patients 3 to 7) had been treated with betamethasone at 5 months of age, out of respiratory syncytial virus infection, and prior to delivery. The median gestational age was 29±2.1 weeks, patient 7 at 10 days after a brain hemorrhage. and the birth weight 655±132 g (median±SD). In the five cases where the placenta was examined, structural anomalies were Echocardiography Results (Table 2) noted. Prior to the hypoxemic event, between day 1 and 6 (median: 3), five infants had been examined by echocardiography because of a Postnatal Clinical Course (Table 2) heart murmur, and had shown a PDA with either a bidirectional or Six of the patients were intubated at birth and received surfactant left-to-right (L-R) shunt. At the onset of hypoxemia, all patients within 6 hours. At 48 hours, four were extubated, while three were had evidence of pulmonary hypertension and global RV still receiving mild mechanical ventilation; all had a FiO2r30%. dysfunction. The RV ejection fraction was 13 to 40%, median: 28 At this stage, despite echocardiographic signs of mild pulmonary (normal: 58 to 715). The ductus was closed in three of the five hypertension, patients 1, 3 and 5 received indomethacin given patients with a previous PDA, and had become restrictive with an the large size of the patent ductus arteriosus (PDA) and a inverted, right-to-left (R-L) shunt in two. In the two cases not predominantly left-to-right shunt. After a variable interval of checked before (2 and 6), the ductus was also closed. On follow-up stability (2 to 12 days from birth), all patients became abruptly echocardiograms performed 2 to 8 days after the onset of hypoxemic and hypotensive. The PaO2/FiO2 ratio decreased symptoms, signs of RV hypertension had disappeared in all cases significantly (minus 40 to 76%). Chest radiograms at that time except patient 3, and the RV ejection fraction had increased 1.2- to showed only minimal or no lung disease. Echocardiography was 3.6-fold. In patient 3, after 7 days the ductus was still patent, but performed then (see below). Mechanical ventilation needed to be with a reversed, L-R shunt, and was ultimately ligated at the age of resumed or intensified, and vasopressors (dopamine, dobutamine) 3 months; mild pulmonary hypertension persisted until 6 months. were used for a duration of 3 to 10 days (median: 4). Milrinone, a RV hypertrophy developed on patients 3 and 6, who were also found phosphodiesterase-III inhibitor, was used in five of the infants at a to have restricted pulmonary artery branches. dosage of 0.25 to 0.75 mg/kg/min for a duration of 3 to 10 days, based on the presence of RV dysfunction. All patients improved DISCUSSION substantially within 2 to 5 days. The PaO2/FiO2 index increased gradually, and mechanical ventilation was weaned off in patients Premature infants with surfactant deficiency and severe respiratory 1 to 6 within 5 to 44 days (median: 33). Patients 2 and 6 distress syndrome may show persistently elevated RV pressures subsequently developed chronic lung disease and required as a consequence of ventilation–perfusion mismatch and

496 Journal of Perinatology 2005; 25:495–499 Pulmonary Hypertension in Growth-Restricted ELBW Infants Danhaive et al.

Table 2 Cardiac Findings at the Onset of Hypoxemia and After Recovery

Case Onset Previous treatments PaO2/FiO2 (OI): Echocardiography findings (days) RV ejection fraction: Prior Acute Acute (%) Post (%) Ratio

1 day 4 Surfactant: two doses 266 130 day 3: PFO with L-R flow; large PDA 22 46 2.1 Indomethacin: two doses days 3–4 (13.8) with bi-di flow; mild RV hypertension day 4: PFO with L-R flow; DA closed; severe RV hypertension; severe RV dysfunction 2 day 2 Dopamine, dobutamine: days 1–2 257 152 day 2: PFO with bi-di flow; DA closed; 40 50 1.3 Hydrocortisone: day 1 (7.5) moderate RV hypertension 3 day 12 Surfactant: four doses 300 72 day 6: PFO with L-R flow; large PDA 34 47 1.4 Dopamine, dobutamine: days 1–4 (11.1) with bi-di flow Indomethacin: 5 doses days 8–12 day 12: PFO with bi-di flow; small PDA with R-L flow; severe RV hypertension (RV pressure B95 mmHg); severe RV dysfunction; RV hypertrophy

4 day 2 Surfactant: three doses 177 81 day 1: PFO with bi-di flow; moderate 30 54 1.8 (7.7) PDA with L-R flow day 2: PFO with bi-di flow; PDA closed; moderate RV hypertension (RV pressure B40 mmHg); moderate RV dysfunction 5 day 7 Surfactant: one dose 194 55 day 2: large PDA with L-R shunt, 25 42 1.7 Dopamine: days 1–3 (12.0) mild RV hypertension Indomethacin: three doses days 2–4 day 7: PFO with R-L flow; DA closed; severe RV hypertension; (RV pressure B70 mmHg); moderate RV dysfunction; moderate RV dilatation

6 day 8 Surfactant: two doses 357 124 day 8: PFO with R-L shunt; DA closed; 13 47 3.6 (5.6) severe RV hypertension (RV pressure B75 mmHg); severe RV dysfunction 7 day 6 Surfactant: one dose 285 40 day 4: PFO with L-R flow; moderate 28 48 1.7 Dopamine, dobutamine: days 1–6 (10.1) PDA with L-R flow Hydrocortisone: day 4 day 6: PFO with bi-di shunt; restrictive PDA with R-L shunt (20 mm gradient); severe RV hypertension (55 mmHg)

Prior: events prior to the onset of symptoms; acute: data acquired at the onset of symptoms; post: data acquired after recovery; ratio: ratio between post and acute data, expressed as the percentage of improvement; PaO2/FiO2: arterial partial oxygen pressure to inspired fraction of oxygen ratio; OI: oxygenation index (FiO2 Â mean airway pressure/PaO2) ; PFO: patent foramen ovale; PDA: patent ductus arteriosus; R-L: right-to-left; L-R: left-to-right; bi-di: bi-directional; RV: right ventricle; LV: left ventricle. hypoxia-related lack of relaxation of the small pulmonary arteries.6 from intrapulmonary shunting or decreased cardiac output. However, clear, well-expanded lungs on chest radiograms and the Systemic hypotension was likely the consequence of reduced left delayed onset of symptoms after a period of stability make it ventricle preload.7 unlikely in these cases. Conditions commonly associated with All except patient 4 had a birth weight <10th percentile, persistent pulmonary hypertension such as sepsis, perinatal indicating intrauterine growth restriction (IUGR). In patient 4, asphyxia or meconium aspiration were not found. The acute despite a birth weight within normal range, fetal growth had symptoms observed in these infants, hypoxemia and hypotension, decreased since 2 weeks prior to delivery. In the three cases born are suggestive of R-L shunting and cardiac output restriction.7 from multiple pregnancy, the birth weight was well <10th None of the babies had ductal or interatrial shunts large enough to percentile even for twins standards.8 Whereas there was no evidence account for the degree of hypoxemia: five of the ducti were closed, of genetic disorders or fetal infections, several risk factors for and the other two were small. Therefore, hypoxemia resulted either uteroplacental insufficiency were present, such as placental

Journal of Perinatology 2005; 25:495–499 497 Danhaive et al. Pulmonary Hypertension in Growth-Restricted ELBW Infants

anomalies,9 maternal hypertension and pre-eclampsia,10 decreased reverts, the ductus may constitute a significant pressure relief to the fetal/placental weight ratio and hypermature villi.11 Uteroplacental RV afterload.6 In our patients, the closure of this physiologic shunt insufficiency may lead to postnatal pulmonary hypertension by may have triggered pulmonary vasoconstriction and restricted two mechanisms: oligohydramnios12 and chronic fetal hypoxia. ventricular outputs, causing hypoxemia and hypotension. Indeed, five of our cases had oligohydramnios, only one of which Although several studies support the safety and efficacy of related to P-PROM. Hypoxia increases endothelial synthesis of inhaled nitric oxide in preterm infants with respiratory distress vasoconstrictors and smooth muscle mitogens such as endothelin-1, syndrome.19–22 Inhaled nitric oxide was not approved for use in platelet-derived growth factor-b and vascular endothelial growth newborns under 34 weeks of gestation in our institutions at the factor; it also inhibits endothelial nitric oxide synthase (eNOS).13 time of these observations, mostly because of concerns regarding In an animal model, hypoxia exposure in the perinatal period lead platelet inhibition. Milrinone, a phosphodiesterase-III inhibitor to increased RV pressure and abnormal pulmonary vascular shown to improve right ventricular output in term neonates after reactivity in adulthood, related to permanent alterations of the cardiac surgery,23 was used instead, given its inotrope and eNOS pathway.14 IUGR may be considered a hallmark of chronic vasodilator properties. With respiratory and hemodynamic support fetal hypoxia: the direct relationship between umbilical venous and plus milrinone in five cases, our patients gradually restored arterial oxygen saturation and birth weight has been well adequate RV function and pulmonary circulation. established in term infants.10 Myocardial performance has been Although growth restiction is frequently associated with shown to deteriorate during the third trimester in IUGR fetuses,15,16 prematurity F 30% of infants born <32 weeks have a birth probably because of increased pulmonary resistance. All these weight <10th percentile of age-matched fetuses24 F pulmonary considerations support the hypothesis that our patients, exposed to hypertension is a rare event in these patients: we report seven cases chronic intrauterine hypoxia, developed structural and functional out of a cohort of 153 infants, and three out of 135 are reported in alterations of the RV and the pulmonary vasculature, and were the literature.17 prone to develop pulmonary hypertension and right ventricular This retrospective case series clearly has inherent limitations. dysfunction. Besides the qualitative evaluation of RV function, we quantified RV The delayed deterioration after a period of stability seems to be volumes derived from two-dimensional measurements using a triggered by a precipitating factor at some distance from birth. The mathematical model. The appropriateness of this method, validated five patients who had been examined by echocardiography prior to in children, was never established on preterm infants, but the the deterioration all had a moderate or large PDA with L-R or percentage of improvement we calculated is likely to be accurate, bidirectional shunting. The repeat echocardiography after they had since it was derived from comparison of each patient’s data during become symptomatic showed either a closed or a restrictive ductus the symptomatic episode and after recovery. Moreover, the RV with a R-L shunt. The short gap between ductal closure and the ejection fraction values we report after stabilization (42 to 54%; onset of pulmonary hypertension, as observed in patients 1, 4 and median: 47) are close to those recently published in preterm 7, suggests a cause–effect relationship between ductal closure and infants.5 More sophisticated methods for evaluating RV function the hypoxemic event. In the two patients not examined before and volume would be necessary. Although this study lacks the (2 and 6), the ductus was closed when hypoxemia occurred. In power for changing current clinical practices regarding the use of patient 2, the clinical deterioration occurred at 2 days, an expected COX-inhibitors in growth-restricted babies, it supports a prompt timing for natural ductal closure in preterm infants. For patient 6, and thorough evaluation of RV pressure and contractility in case of in whom hypoxemia occurred at 8 days, ductal closure might have unexplained hypoxemia and/or hypotension. Our case study been delayed by persistently elevated pulmonary resistances, as supports the use of pulmonary vasodilators in such conditions, that suggested by the RV hypertrophy observed for several weeks and may permit to minimize ventilatory support escalation, but a severe pulmonary hypertension when he died at 5 months. Three clinical trial would be necessary to establish the efficacy and safety of the patients had received indomethacin few hours prior to of milrinone in hypoxemic preterm infants with RV dysfunction. deteriorating. A similar observation was recently reported in three preterm infants treated with ibuprofen.17 Cyclooxygenase (COX) inhibitors decrease the synthesis not only of prostaglandin E2, a References major mediator of ductal patency, but also of prostacyclin, a potent 1. Walther FJ, Benders MJ, Leighton JO. Persistent pulmonary hypertension in premature neonates with severe respiratory distress syndrome. pulmonary vasodilator. The ductus arteriosus is a dynamic 1992;90:899–909. structure during transitional circulation. It has been shown that, 2. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States when pulmonary vascular resistance is still high, a L-R shunt national reference for fetal growth. Obstet Gynecol 1996;87:163–8. through the ductus may increase significantly the pulmonary 3. Sholler GF, Colan SD, Sanders SP. Effect of isolated right ventricular 18 blood flow and contribute to oxygenation. Conversely, when outflow obstruction on left ventricular function in infants. Am J Cardiol pulmonary arterial pressure is greater than systemic and the shunt 1988;62:778–84.

498 Journal of Perinatology 2005; 25:495–499 Pulmonary Hypertension in Growth-Restricted ELBW Infants Danhaive et al.

4. Helbing WA, Bosch HG, Maliepaard C, et al. Comparison of echo- 14. Tolsa JF, Muchlethaler V, Peyter AC. Late effects of perinatal hypoxia on the cardiographic methods with magnetic resonance imaging for assessment pulmonary circulation. Pediatr Res 2003;54:607A. of right ventricular function in children. Am J Cardiol 1995;76: 15. Tsutsumi T, Ishii M, Eto G, Hota M, Kato H. Serial evaluation for 589–94. myocardial performance in fetuses and neonates using a new doppler index. 5. Clark SJ, Yoxall CW, Subdehar NV. Right ventricular volume measurements Pediatr Int 1999;41:722–7. in ventilated preterm neonates. Pediatr Cardiol 2004;25:149–53. 16. Rizzo G, Capponi A, Rinaldo D, Arduini D, Romanini C. Ventricular ejection 6. Skinner JR, Boys RJ, Hunter S, Hey EN. Pulmonary and systemic force in growth-retarded fetuses. Ultrasound Obstet Gynecol 1995;5:247–55. arterial pressure in hyaline membrane disease. Arch Dis Child 1992;67: 17. Gournay V, Savagnier C, Thiriez G, Kuster A, Roze JC. Pulmonary 366–73. hypertension after ibuprofen prophylaxis in very preterm infants. Lancet 7. Evans N, Kluckow M, Currie A. Range of echocardiographic findings in 2002;359:1486–8. term neonates with high oxygen requirements. Arch Dis Child Fetal 18. Clyman RI, Mauray F, Heymann MA, Roman C. Cardiovascular effects of Neonatal Ed 1998;708:F105–11. patent ductus arteriosus in preterm lambs with respiratory distress. J Pediatr 8. Naeye RL, Bernischke K, Hagstrom JWC, Marcus CC. Intrauterine growth of 1987;111:579–87. twins, as estimated from liveborn birth weight data. Pediatrics 19. Van Meurs KP, Rhine WD, Asselin JM, Durand DJ, the Preemie NO 1996;37:409–16. Collaborative Group. Response of premature infants with severe respiratory 9. Lin CC, Santolaya-Forgas J. Current concepts of fetal growth restriction failure to inhaled nitric oxide. Pediatr Pulmonol 1997;24:319–23. Part I. Causes, classification, and pathophysiology. Obstet Gynecol 1998;92: 20. Cheung P-Y, Peliowski A, Robertson C. The outcome of very low birth 1044–55. weight neonates (<1500 g) rescued by inhaled nitric oxide: neurodevelop- 10. Yagel S, Anteby Y, Shen O, Cohen SM, Friedman ZM, Achiron R. Placental ment in early childhood. J Pediatr 1998;133:735–9. blood flow measured by simultaneous multigate spectral doppler imaging in 21. Kinsella JP, Walsh WF, Bose CL, et al. Inhaled nitric oxide in premature pregnancies complicated by placental vascular anomalities. Ultrasound neonates with severe hypoxaemic respiratory failure: a randomised obstet gynecol 1999;14:262–6. controlled trial. Lancet 1999;354:1061–5. 11. Lackman F, Capewell V, Gagnon R, Richardson B. Fetal 22. Schreiber MD, Gin-Mestan K, Marks JD, Huo D, Lee G, Srisuparp P. Inhaled oxygen values and birth to placental weight ratio in relation to size at birth. nitric oxide in premature infants with the respiratory distress syndrome. New Am J Obstet Gynecol 2001;185:674–82. Engl J Med 2003;349:2099–107. 12. Peliowski A, Finer NN, Etches PC, Tierney AJ, Ryan CA. Inhaled nitric oxide 23. Chang AC, Atz AM, Wernowsky G, Redmond P, Wessel DL. Milrinone: for premature infants after prolonged rupture of the membranes. J Pediatr systemic and pulmonary hemodynamic effects in neonates after cardiac 1995;126:450–3. surgery. Crit Care Med 1995;23:1907–14. 13. Kourembanas S, Morita T, Christou H, et al. Hypoxic responses of vascular 24. Bukowski R, Gahn D, Denning J, Saade G. Impairment of growth in fetuses cells. Chest 1998;114:25S–8S. destined to deliver preterm. Am J Obstet Gynecol 2001;185:463–7.

Journal of Perinatology 2005; 25:495–499 499