The Journal of Pediatrics
Volume 148, Issue 3 , Pages 289-291, March 2006

Cardiovascular support in the preterm: Treatments in search of indications

  • K.J. Barrington, MB, ChB

      Affiliations

    • Corresponding Author InformationReprint requests: Keith J. Barrington, Paediatrics and Obstetrics & Gynaecology, McGill University, Director NICU, Royal Victoria Hospital, Room C7.80, 687 Pine Ave W, Montreal, Quebec H3A 1A1, Canada
    • ,
  • E.M. Dempsey, MB, BCh

McGill University, Royal Victoria Hospital, Montreal, Quebec H3A 1A1, Canada

Article Outline

 

As many as 50% of very preterm infants in intensive care units receive fluid boluses, pressor/inotrope medications, or both, for cardiovascular support.1 The most frequent indication for intervention is a low systemic blood pressure in the first few days of life, which, although statistically associated with worse outcome in some studies, poorly discriminates between infants with a good outcome and those who will have complications or die.

One common approach is to intervene solely because blood pressure falls below an arbitrary cut-off, most commonly when the mean arterial pressure is less than the gestational age in weeks. This approach is difficult to justify, it neglects the considerable variation between published blood pressure standards, the normal spontaneous increase in blood pressure over the first few hours and days of life, and the very poor relationship between blood pressure and systemic blood flow.2 Blood pressure is, of course, determined both by cardiac output and vascular resistance, thus a low blood pressure may be associated with high, normal, or low cardiac output,2 and treatments for hypotension may elevate blood pressure but decrease perfusion.3 It would be physiologically more rational to assess the adequacy of the perfusion of vital organs to decide when to intervene. This requires measurements of flow, which are difficult in the preterm. Left ventricular output is equivalent to pulmonary venous return (plus or minus any net flow across the foramen ovale), the sum of pulmonary artery flow and the left to right component of patent ductus arteriosus flow. Right ventricular output is an estimate of systemic blood flow (inferior vena cava [IVC] and superior vena cava [SVC] flow) plus or minus any net flow across a patent foramen ovale. Because of these complications, the group at the University of Sydney has focused on SVC flow, a measurement of perfusion of the upper body. They have demonstrated that low SVC blood flow is statistically associated with an increased risk of late intraventricular hemorrhage,4 death, and adverse neurodevelopmental outcome.5 What is not yet clear is whether low SVC blood flow causes poorer outcomes, whether we can reliably increase low SVC flow, and whether doing so will improve outcomes without adverse effects. Although physiologically rational, measurement of SVC flows may not be adequately predictive. Osborn et al4 showed that 70 of 217 preterm infants without late intraventricular hemorrhage (IVH) had low SVC flow, and 31 of 37 with late IVH had low flow. Although highly significant, the positive predictive value is only 30%. This may be a high enough positive predictive value to intervene if the intervention is shown to be safe, but a further complicating consideration is that, if low SVC flow causes cerebral injury, it probably occurs when the perfusion is at its lowest, thus it may be important to prevent low SVC flow.

In this issue of The Journal of Pediatrics, Paradisis et al6 have taken the next step in their series of investigations, attempting to prevent the low-flow state. They enrolled infants with an elevated risk for the development of low SVC flow. Sixty-one percent of the group enrolled was predicted to have development of low flow, which leads by multiplication to an 18% PPV for late IVH. Their criteria enroll, for example, all infants less than 27 weeks gestation. Most of such infants will have good outcomes; it is very important therefore to demonstrate that the intervention is safe. It must be said that 55% of preterm infants less than 27 weeks already receive pressors,7 without evidence of benefit. As well as the new information describing milrinone pharmacokinetics, this uncontrolled pilot study suggests that milrinone, using the new dosage regime, could prevent low SVC flow in small preterm infants; however, without randomized controls, this must remain a suggestion, to be confirmed by randomized controlled trials (RCTs), of which safety must be a primary concern. This report appears to be the first report of the use of milrinone in the preterm infant; the use of a new agent to support the circulation of the preterm infant raises many questions.

Back to Article Outline

Whom should we treat, and what should we treat with? 

The current poor state of knowledge is that we do not know which infants, if any, may benefit from cardiovascular support, and, if we could identify them, we do not know what is clinically effective. Furthermore, without knowing what is clinically effective, it is difficult to design studies to determine who may benefit.

Interventionscommonly used and recommended include fluid boluses and catecholamines. Fluid boluses are given despite evidence that hypotension is not commonly associated with hypovolemia and despite observational data suggesting an association with intraventricular hemorrhage.8 A search of PubMed and the Cochrane database of systematic reviews, performed for this editorial, could find not one RCT comparing fluid boluses to no boluses in infants with circulatory compromise.

The most commonly used catecholamine, dopamine, appears to elevate blood pressure predominantly by vasoconstriction at the expense of systemic blood flow9 and also causes suppression of pituitary and thus thyroid function.10 Analyses of large databases suggest that dopamine use is a better predictor of poor outcome than is hypotension,7 which suggests the possibility that dopamine treatment of hypotension could be as harmful, or more harmful, than hypotension itself. Other agents such as dobutamine, epinephrine, or milrinone have different hemodynamic effects and side effects but have also not been shown to be safe or effective. Corticosteroids are now being frequently prescribed as second-line therapy for hypotension, with large variation in doses, of fourfold or even greater, despite a complete lack of studies showing clinical benefit and despite recent demonstrations that steroids cause serious short- and long-term adverse effects in the preterm infant.

The only way to determine which infants may benefit from cardiovascular support will be by performing RCTs of an intervention with different criteria. Such criteria may include low right ventricular output, low SVC flow, or risk for low flow, if other investigators can replicate the important work of the group from the University of Sydney. Treatment at differing blood pressure thresholds, or only when defined clinical signs of poor perfusion are present, also should be compared. These trials will be required as a prerequisite to being able to determine which forms of cardiovascular support may be beneficial.

Back to Article Outline

Should we consider using phosphodiesterase III inhibitors in the newborn? 

Milrinone increases intracellular cyclic AMP by inhibiting class III phosphodiesterase (PDE), causing both inotropic effects and vasodilation. However, in neonatal mammalian models class III PDE inhibitors have minimal,11 no, or even negative12, 13 inotropic effects, probably because of a developmental imbalance between class III and class IV PDE14 in neonatal sarcoplasmic reticulum. The negative inotropic effects seen in neonatal puppies rapidly become positive in the first few days after birth.13 The effects on preterm human myocardium are unknown and cannot be predicted from the animal studies. Increases in blood flow and systemic perfusion, if confirmed in RCTs, could be due solely to vasodilation, an effect shown to be intact in limited studies in newborn mammals.15 Milrinone may have other adverse effects: in dogs treatment with milrinone at 1 mg/kg produced lesions in the left ventricle and the right atrium.16 Similar lesions have been noted with other inotropes or pressors and may occur whenever the myocardial workload is increased, because this may exceed any increases in myocardial oxygen and substrate supply. Milrinone must be subjected to rigorous evaluation both against placebo and against other common therapies, with evaluation of myocardial and vascular responses, and, more importantly, clinical outcomes.

Back to Article Outline

What outcomes should be studied? 

Dopamine has been shown to increase blood pressure more than dobutamine;17 given that dopamine is a vasoconstrictor and dobutamine a vasodilator, this is hardly surprising! However, we do not know whether dopamine improves clinically important outcomes compared with dobutamine,17 only 3 of the 5 comparative trials even reported death (n = 103), and only 2 reported the incidence of severe IVH (total n = 83). As for steroids, the only proven effect of their use for hypotension is that they decrease catecholamine use. Showing that a drug improves pressures or flows, or that one potentially toxic drug can replace another, are not appropriate end points for clinical trials. Effects on death, brain, lung, and intestinal injury in the short term and on long-term neurodevelopment and health must be investigated.

Back to Article Outline

Can we perform RCTs to determine the value of cardiovascular support? 

Some very preterm babies are desperately sick, with poor cardiac function, very poor perfusion, and predictably poor outcomes, and withholding cardiovascular support from such babies in the context of an RCT would be ethically questionable. However, even in such babies, comparisons of different interventions are surely possible. On the other hand many infants who currently receive boluses or pressors/inotropes are relatively stable and could be enrolled in RCTs.

We have recently presented a cohort study of extremely low birth weight infants with a very restricted treatment protocol (“permissive hypotension”).18 Only 11% of 118 extremely low birth weight babies received catecholamine infusions for blood pressure support, and only 6% received fluid boluses—despite this our overall results are quite comparable to other centers. Although this is only level 2 evidence with a small sample and short-term outcomes, it is, sadly, a better level of evidence than is available for the routine treatment of hypotension with fluid boluses and dopamine! Despite the widespread use of dopamine for 30 years, there is no published evidence to demonstrate that babies benefit. There is clearly sufficient equipoise to perform the required RCTs. The total number of preterm infants in published studies comparing mortality rates with dopamine compared with dobutamine is only 103. The total number of preterm infants in reported RCTs comparing outcomes of treating cardiovascular compromise or hypotension using dopamine, or fluid boluses, to no therapy is zero. In contrast, with 1% of babies being of very low birth weight, and up to 50% receiving dopamine, the total number of preterm infants treated with dopamine annually may exceed 1000 in Canada, and 10,000 in the United States. It is essential that we now design and perform the right trials to determine whether the infusion of inotropic agents, pressor agents, or the use of fluid boluses is helping these patients rather than harming them.

Back to Article Outline

References 

  1. Al-Aweel I , Pursley DM , Rubin LP , Shah B , Weisberger S , Richardson DK . Variations in prevalence of hypotension, hypertension, and vasopressor use in NICUs . J Perinatol . 2001;21:272–278
  2. Kluckow M , Evans N . Relationship between blood pressure and cardiac output in preterm infants requiring mechanical ventilation . J Pediatr . 1996;129:506–512
  3. Zhang J , Penny DJ , Kim NS , Yu VH , Smolich JJ . Mechanisms of blood pressure increase induced by dopamine in hypotensive preterm neonates . Arch Dis Child Fetal Neonatal Ed . 1999;81:F99–F104
  4. Osborn DA , Evans N , Kluckow M . Hemodynamic and antecedent risk factors of early and late periventricular/intraventricular hemorrhage in premature infants . Pediatrics . 2003;112:33–39
  5. Hunt RW , Evans N , Rieger I , Kluckow M . Low superior vena cava flow and neurodevelopment at 3 years in very preterm infants . J Pediatr . 2004;145:588–592
  6. Paradisis M , Evans N , Kluckow M , Osborn D , McLachlan AJ . Pilot study of milrinone for low systemic blood flow in very preterm infants . J Pediatr . 2006;148:306–313
  7. Synnes AR , Chien LY , Peliowski A , Baboolal R , Lee SK . Variations in intraventricular hemorrhage incidence rates among Canadian neonatal intensive care units . J Pediatr . 2001;138:525–531
  8. Goldberg RN , Chung D , Goldman SL , Bancalari E . The association of rapid volume expansion and intraventricular hemorrhage in the preterm infant . J Pediatr . 1980;96:1060–1063
  9. Osborn D , Evans N , Kluckow M . Randomized trial of dobutamine versus dopamine in preterm infants with low systemic blood flow . J Pediatr . 2002;140:183–191
  10. Van den Berghe G , de Zegher F , Lauwers P . Dopamine suppresses pituitary function in infants and children . Crit Care Med . 1994;22:1747–1753
  11. Artman M , Kithas PA , Wike JS , Strada SJ . Inotropic responses change during postnatal maturation in rabbit . Am J Physiol . 1988;255(Pt 2):H335–H342
  12. Klitzner TS , Shapir Y , Ravin R , Friedman WF . The biphasic effect of amrinone on tension development in newborn mammalian myocardium . Pediatr Res . 1990;27:144
  13. Binah O , Legato MJ , et al.   Developmental changes in the cardiac effects of amrinone in the dog . Circ Res . 1983;52:747–752
  14. Akita T , Joyner RW , Lu C , Kumar R , Hartzell HC . Developmental changes in modulation of calcium currents of rabbit ventricular cells by phosphodiesterase inhibitors . Circulation . 1994;90:469–478
  15. Hassan S , Mori M , Obara H . The vasorelaxant effects of amrinone on neonatal rabbit thoracic aortic strips . Kobe J Med Sci . 1996;42:73–91
  16. Hanton G , Gautier M , Bonnet P , Herbet A . Effect of milrinone on echocardiographic parameters after single dose in beagle dogs and relationship with drug-induced cardiotoxicity . Toxicol Lett . 2005;155:307–317
  17. Subhedar NV, Shaw NJ. Dopamine versus dobutamine for hypotensive preterm infants. The Cochrane Database of Systematic Reviews (2003 Issue 4).
  18. Dempsey EM, Alhazzani F, Barrington K. Permissive hypotension in the ELBW. Pediatric Academic Societies Annual Meeting 2005;Abstract 560.

PII: S0022-3476(05)01242-4

doi:10.1016/j.jpeds.2005.12.056

Refers to article:

  • Pilot study of milrinone for low systemic blood flow in very preterm infants

    Mary Paradisis, Nick Evans, Martin Kluckow, David Osborn, Andrew J. McLachlan
    The Journal of Pediatrics March 2006 (Vol. 148, Issue 3, Pages 306-313)

The Journal of Pediatrics
Volume 148, Issue 3 , Pages 289-291, March 2006

Article Tools