In the absence of outcome-based evidence to guide circulatory support in the newborn, a logical approach is one that uses individual- or precision-based therapy. That is, one should try to define the individual’s hemodynamic as accurately as possible and apply treatment logically to that hemodynamic based on what we know about the effects of the pharmacologic interventions on physiology. As in the case of all approaches to circulatory support in the newborn, there is no evidence that this will improve outcomes.
Keywordscardiac ultrasound, circulatory support, infant newborn, infant premature, inotropes
There is little outcome based evidence to guide circulatory support in the newborn.
In clinical situations of high risk for circulatory compromise, there are a range of hemodynamics.
A ‘one size fits all’ approach to neonatal circulatory support is not logical or likely to be successful.
It is logical to define the individual hemodynamic in a baby and apply therapy logically on the basis of those individual findings.
Shock refers to a circulatory state where the delivery of oxygen to the organs and tissues of the body is inadequate to meet demand. In neonatology, the terms hypotension and shock have tended to be used synonymously. This is erroneous, and although many shocked babies will be hypotensive, not all hypotensive babies are shocked and not all shocked babies are hypotensive. True shock is uncommon in neonatology, but borderline states of circulatory incompetence are not uncommon.
Circulatory competence depends on adequate systemic blood flow, which is dependent on cardiac output. Cardiac output is the product of stroke volume and heart rate. Stroke volume is determined by preload (positively), myocardial performance (positively), and afterload (negatively, above an individually variable threshold). Thus cardiac output can be increased by increasing stroke volume, heart rate, or both. Blood pressure (BP) is the product of cardiac output and systemic vascular resistance (SVR), so BP can be improved by increasing cardiac output, SVR, or both. Thus neonatal circulatory compromise can result from four basic mechanisms: reduced intravascular volume (low preload), failure of myocardial performance, obstruction in the circulation (high afterload), and loss of vascular tone or distributive disorders of the peripheral circulation. In the first three processes, the main hemodynamic feature is low systemic blood flow, which results in tissue hypoperfusion and hypoxia. In contrast, vasodilatory shock, if myocardial performance keeps up with the increased workload, exhibits a normal or high systemic blood flow. Abnormal distribution of the circulation and microcirculatory alterations play a primary role in the development of tissue hypoxia. Microcirculatory shock is the condition in which the microcirculation fails to support tissue oxygenation in the face of normal systemic hemodynamics. This complexity is magnified many times in the newborn because we are often dealing with a circulation in transition. The fetal channels may not close, leading to blood either bypassing the lungs or recirculating through them, depending on the pressure differentials in the pulmonary and systemic circulation. If the fetal channels do close (as they often do in near-term or term babies) and pulmonary vascular resistance remains high, the circulation may be compromised because the blood cannot get through the lungs to reach the systemic circulation.
These hemodynamics vary between different clinical situations and even among babies in the same clinical situation. As a result, to determine the logical approach to managing shock in an individual baby, there is a need to understand the mechanism of circulatory compromise in that baby and to apply therapy logically on the basis of our understanding of the actions of the pharmacologic agents that are available. This is not an area where a “one size fits all” evidence-based treatment can be applied. In this chapter, I will initially detail what is understood about the pharmacology of the commonly used circulatory support interventions and what evidence of neonatal effect there is about each of them. I will then suggest a logical approach to utilizing these interventions based on hemodynamic scenarios that are commonly found in the newborn in clinical situations of high circulatory risk.
Circulatory Support Interventions
In a recent European-based survey of neonatal circulatory support, the most common first-line interventions were volume expansion (85%), dopamine (62%), dobutamine (18%), both dopamine and dobutamine (18%), epinephrine (2%), and norepinephrine (1%). The same interventions were used with varying frequency as second-line interventions; however, steroids (10%) and milrinone (1%) were also used as second-line interventions. How do these interventions work and what is the evidence that they provide any benefit?
Volume expansion is probably one of the most widely used circulatory support interventions despite evidence that true hypovolemia is rare and the fact that we have very little understanding of the effects of volume expansion in a normovolemic infant. Volume expansion will increase preload on the heart. This will be lifesaving in a truly hypovolemic baby; in a normovolemic infant, however, there may well be an immediate increase in cardiac output, the maintenance of this will depend on how long the extra volume stays in the circulation. If volume expansion keeps being pushed, the distribution of extra volume out of the intravascular compartment may well create interstitial edema in the lungs and other organs. Excessive volume expansion in preterm babies may be associated with higher mortality.
Studies have shown that volume expansion in hypotensive preterm babies has little effect on BP. One study in preterm babies with low systemic blood flow showed an immediate increase in systemic blood flow but did not study how long this increase was maintained. Cerebral blood flow, whether studied by xenon extraction or near infrared spectroscopy (NIRS), does not seem to be consistently increased by volume in hypotensive preterm infants. Clinical trials of routine early volume expansion in preterm babies with fresh frozen plasma, plasma substitutes, or isotonic saline have shown no improvement in outcomes as compared with no intervention. However, delayed umbilical cord clamping probably provides benefit, in part by facilitating a routine volume expansion with an extra 10 to 20 mL/kg of blood; there is evidence that this intervention does improve outcomes. The benefit of delayed cord clamping in very preterm infants has been confirmed with the results of the large Australian Placental Transfusion Study (APTS). This study showed a reduction in hospital mortality that did not quite reach statistical significance after correction for multiple comparisons. However, when the APTS results are combined with the other randomized clinical trials, a significant reduction in mortality was confirmed, together with less need for blood transfusions.
The evidence to guide what type of fluid to use in volume expansion is not consistent. Two small trials in preterm babies showed no difference between isotonic saline and 5% albumin in improving BP. One slightly larger trial showed that hypotensive preterm babies given 5% albumin were more likely to achieve normal BP and less likely to be given vasopressors than those randomized to saline.
Because relative hypovolemia in the context of a mechanically ventilated baby is difficult to exclude, there is merit in including some volume expansion in a circulatory support protocol. However, volumes in excess of 20 mL/kg should be used with caution. Delayed cord clamping should now be a standard of care in preterm infants.
Dopamine is the naturally occurring catecholamine precursor to noradrenaline. In older children and adults, dopamine has dose-dependent effects. At low doses (2 to 4 μg/kg/min), it stimulates dopaminergic receptors in the coronary, renal, and mesenteric systems, causing vasodilation. At moderate doses (5 to 10 μg/kg/min), it increases myocardial contractility and heart rate by stimulating β 1 , β 2 , and α receptors. At high doses (10 to 20 μg/kg/min), vascular α-adrenergic stimulation causes an increase in systemic and probably pulmonary vascular resistance. What evidence there is suggests a similar dose-dependent effect in the newborn, particularly the high-dose α-adrenergic effects.
The clinical trial evidence around dopamine is largely related to its effect on BP, and there are consistent data that it is more effective than volume and dobutamine at increasing BP and similarly efficacious to epinephrine, steroids, and vasopressin. Much of this effect on BP seems to be due to the vasopressor effect; the evidence around effects on systemic and cerebral blood flow is less consistent. Dopamine at doses between 4 and 10 μg/kg/min increased cardiac output in a cohort of late preterm and term infants, most of whom were asphyxiated. In hypotensive preterm babies, Zhang et al. showed no significant effect on cardiac output. Roze et al. showed a reduction in left ventricular (LV) output with a dose of dopamine sufficient to normalize BP, and Osborn et al. showed no change in superior vena cava (SVC) flow with dopamine despite achieving a significant increase in BP. This study showed a rise in LV wall stress (a marker of afterload) on increasing the dose of dopamine from 10 to 20 μg/kg/min, consistent with an α-adrenergic effect at high doses. There is no evidence that this vasoconstriction significantly limits cerebral blood flow. Studies of changes in cerebral blood flow after dopamine, mainly in hypotensive preterm babies and using NIRS surrogate markers of cerebral blood flow, tend to show either increases or no change.
With reference to effects of dopamine on the pulmonary vasculature, the study is limited to preterm babies with patent ductus arteriosus (PDA). Liet et al. showed variable effects on the ratio of systemic-to-pulmonary pressure, but overall there was no change in that ratio. Bouissou et al. showed an increase in pulmonary artery pressure in response to 8 μg/kg/min of dopamine. They also showed an increase in SVC flow, which they suggest may relate to a reduction in PDA shunt.
In summary, dopamine will reliably increase BP, and it probably does this by increasing SVR more than systemic blood flow. The weight of evidence is that dopamine likely increases cerebral blood flow but that higher doses should be used with caution, as high afterload in either the systemic or pulmonary circulation can result in compromise. In addition, dopamine has a broader spectrum of effects on other organs, including the inhibition of thyroid stimulating hormone production in the pituitary gland. This has special clinical relevance for the newborn.
Dobutamine is a synthetic catecholamine that was modified from isoprenaline to reduce some of the chronotropic effects. Dobutamine has a half-life of about 2 minutes in children and adults and has β 1 and β 2 effects with some α 1 effects; it therefore causes an increase in myocardial contractility and heart rate and some reduction in peripheral vascular resistance. This vasodilatory effect is likely why clinical trials have consistently shown that dobutamine is not as good as dopamine at improving BP in hypotensive preterm babies. Nevertheless, in the few studies that have measured it, dobutamine seems better than dopamine at improving systemic blood flow. There was no difference in neurodevelopment at 3 years in the only trial to report this, although the trial was not powered for this outcome. One small randomized trial has compared dobutamine with placebo in babies with low systemic blood flow and showed that systemic blood flow improved in both groups; however, other markers of perfusion were significantly better in babies treated with dobutamine. This study was unable to show any differences or changes in NIRS markers of cerebral blood flow with either dobutamine or placebo. Other observational studies have shown an impact on Doppler markers of organ blood flow, including cerebral blood flow, but the placebo effect in the study by Bravo et al. reminds us that most blood flow parameters will improve spontaneously after about 12 hours of age in very preterm babies. Dobutamine also has dose-related effects, with low doses (<5 μg/kg/min) probably having little effect but with increasing effects from 5 to 20 μg/kg/min. Osborn et al. showed a continuing drop in LV wall stress (afterload) with increases from 10 to 20 μg/kg/min.
In summary, dobutamine has dose-related central inotropic and chronotropic effects with peripheral vasodilation. It is likely to be most effective where there is myocardial dysfunction and/or increased afterload leading to low systemic blood flow. It is unlikely to be effective when hypotension is due to vasodilatation.
Epinephrine has very similar dose-related effects to dopamine with low doses (0.01–0.1 μg/kg/min) primarily stimulating the cardiac and vascular β 1 – and β 2 -adrenoreceptors, leading to increased inotropy, chronotropy, and conduction velocity as well as peripheral vasodilation. Its inotropic effect is superior to that of dopamine. At doses greater than 0.1 μg/kg/min, epinephrine stimulates the vascular and cardiac α 1 receptors, causing vasoconstriction and increased inotropy, respectively. There has been limited study of the cardiovascular effects of epinephrine in the newborn. The previously cited randomized trial of dopamine and epinephrine in hypotensive preterm babies showed similar effects in improving BP and cerebral oxygenation index, although the epinephrine group had significantly higher lactate and more hyperglycemia than those randomized to dopamine. This is a recognized metabolic effect of epinephrine, which is most likely explained by the drug-induced stimulation of β 2 -adrenoreceptors in the liver and skeletal muscle, resulting in decreased insulin release and increases in glycogenolysis, leading to increases in lactate production, respectively. This does have the clinical side effect of rendering the lactate level an unreliable marker of tissue perfusion. This was one of the few neonatal cardiovascular support trials that assessed neurodevelopmental outcome, and it showed no difference between the two groups at 2 to 3 years of age.
In summary, the vasoconstrictive effects of epinephrine make it best suited for the management of vasodilatory shock. As in the case of dopamine, caution should be exercised in using higher doses because of the risk of afterload compromise.
Norepinephrine is a naturally occurring sympathomimetic amine that, while having less β 2 receptor stimulatory effects, has mainly strong α agonist effects; it is therefore a potent vasoconstrictor. There is no clinical trial evidence on the use of norepinephrine in the newborn and not much observational data. A retrospective study of norepinephrine use in 48 babies born before 33 weeks’ gestation showed that normotension could be achieved in all but one baby at a median dose of 0.5 μg/kg/min. Apart from tachycardia (in 31%), no immediate side effects were observed. One small cohort study showed improved BP in term babies with septic shock refractory to dopamine and dobutamine. An observational study in term babies with pulmonary hypertension suggested that norepinephrine has a beneficial effect on both systemic and pulmonary hemodynamics by causing systemic vasoconstriction and pulmonary vasodilation. There are no long-term safety data on the use of norepinephrine, but it may have a role in refractory vasodilatory shock and in babies with persistent pulmonary hypertension of the newborn (PPHN) and low BP.
Vasopressin is a naturally occurring hormone that is mainly involved in the regulation of extracellular osmolarity, although it does play a role in regulating the function of the cardiovascular system as well. Indeed, vasopressin also has potent vasoconstrictive effects, and it is this effect that is used in its role in circulatory support. There has been one small randomized trial ( n = 20) comparing dopamine with vasopressin in hypotensive preterm infants. Outcomes were limited to physiologic parameters and both agents had similar effects in increasing BP whereas vasopressin had less tachycardic effect. Otherwise there are small case series of improvement in BP in preterm babies with inotrope refractory hypotension. There is some evidence from animal studies that vasopressin has a vasodilatory effect on the pulmonary vasculature. This would be consistent with the observational report of the use of vasopressin in 10 babies with PPHN refractory to inhaled nitric oxide (iNO). A vasopressin infusion improved BP, reduced oxygenation index, and improved urine output. There are not enough safety data to recommend the routine use of vasopressin, but it may have a role in treating vasodilatory shock refractory to vasopressor-inotropes.
Milrinone is a phosphodiesterase-3 inhibitor that inhibits the degradation of cyclic adenosine monophosphate (cAMP). By increasing the concentration of cAMP, milrinone enhances myocardial contractility, promotes myocardial relaxation, and decreases vascular tone in the systemic and pulmonary circulation. The role of milrinone is best established in preventing and treating low-cardiac-output syndrome (LCOS) after cardiac bypass surgery. A large randomized controlled trial (RCT) has shown that milrinone significantly reduced the incidence of LCOS in a neonatal and pediatric population. The similarities between LCOS and the postnatal drop in systemic blood flow seen in some preterm babies led to a trial of milrinone to prevent low systemic blood flow in preterm babies. This trial showed no difference in the incidence of low systemic blood flow between milrinone and placebo, although milrinone did increase the need for vasopressor-inotropes to support BP and seemed to slow constriction of the ductus arteriosus. There was a pharmacokinetic arm to this study which showed a considerably longer half-life of milrinone in preterm babies than in term babies (10 hours vs. 4 hours). The afterload-reducing properties of milrinone led to an observational study suggesting that milrinone may prevent the afterload compromise which can follow PDA ligation in preterm babies. The pulmonary vasodilatory properties may also explain the observation of improvement in oxygenation and hemodynamics in babies with PPHN refractory to iNO. There is also a small case series of milrinone being used in conjunction with iNO in seven preterm babies with pulmonary hypertension, with improvement in oxygenation index, right ventricle (RV) myocardial function, and pulmonary pressure. Milrinone may have a role in the management of hemodynamics associated with systemic or pulmonary vasoconstriction, but there are not enough safety data to recommend its routine use. It should be used with caution in very preterm babies because of its long half-life and the risk of hypotension.
In a recent review of circulatory support measures in a U.S. pediatric health information system, hydrocortisone was second to dopamine as the most commonly used drug in extremely low birth weight infants (dopamine 83% vs. hydrocortisone 33%). This is surprising as we do not know much about the hemodynamic impact of hydrocortisone apart from the effect on BP, and we know little about the long-term effects when used early to treat BP. Bourchier et al. randomized hypotensive preterm babies to hydrocortisone or dopamine and showed that both had similar effects on BP. Ng et al. randomized preterm babies needing more than 10 μg/kg/min of dopamine to hydrocortisone or placebo and showed faster weaning of inotropes in those treated with hydrocortisone. Efird et al. randomized preterm babies to hydrocortisone or placebo and confirmed that vasopressor use was reduced. To the best of my knowledge only one study has looked at the hemodynamic effects of hydrocortisone. In a cohort of 15 preterm and 5 term babies with hypotension requiring high-dose dopamine, Noori et al. showed that the immediate effect of hydrocortisone on BP was mediated via an increase in SVR without any change in cardiac output or stroke volume. However, later, with weaning of dopamine, there were increases in stroke volume and cardiac output.
None of these studies looked at longer-term outcomes, and the current Cochrane review on corticosteroids to treat hypotension in preterm babies urges caution, concluding: “With long-term benefit or safety data lacking, steroids cannot be recommended routinely for the treatment of hypotension in preterm infants.” Although not addressing the circulatory support use of hydrocortisone, there is some reassurance about the short-term safety of hydrocortisone from the large PREMILOC trial ( n = 523), where the effect of prophylactic low-dose hydrocortisone on chronic lung disease was assessed. A follow-up to this study is in progress, which should provide data on longer-term safety.
There is some evidence of an increased risk of gastrointestinal perforation, particularly when hydrocortisone is used in conjunction with indomethacin. It is prudent to avoid using these two drugs together.
Applying the Evidence in Clinical Practice
The bottom line is that we know a lot about how our interventions affect BP, less about how they affect hemodynamics, and almost nothing about how they affect outcomes. In light of this, it is the author’s view that the logical way to approach circulatory support in the newborn is to use an individual or “precision medicine-based” approach. That is, one should try to define the individual hemodynamic as accurately as possible and apply logical treatment based on what we know about the effects of the interventions on physiology.
How to best define individual hemodynamic is controversial. Many neonatologists would still use vital signs, mainly BP, and this will reflect reality for many with limited access to technologies such as Doppler ultrasound and/or NIRS. But to define a hemodynamic one really needs a measure of pressure and flow so that resistance can be estimated and indicate whether it is flow or resistance or both that must be increased to support pressure. How best to define flow is controversial, particularly whether that should be a measure of global systemic flow or a marker of organ (usually brain) blood flow. Both are clearly important; the ideal situation would be to measure systemic and organ blood flow as well BP. Indeed, in the future there may be comprehensive, real-time hemodynamic monitoring as described in Chapter 21 . One would also need a measure of preload or volume status, but such a marker remains elusive except in extreme cases of hypovolemia. The author would argue that it is difficult to define an individualized approach to the circulation without any information about what is happening in the organ that drives the circulation and, indeed, the organ one is trying to manipulate with various interventions, the heart. Without this knowledge, one is flying blind.
Cardiac ultrasound assessment of an individual’s hemodynamic is complex; the measurements are covered in more detail in Chapter 10 , Chapter 11 , Chapter 12 , Chapter 13 . It is clearly important to document structural normality, fetal channel shunts, volume status, myocardial function, and pulmonary pressure, but the essentials for circulatory support are a measure of BP, a measure of systemic blood flow, and, in some babies, an estimate of pulmonary blood flow. The author uses SVC flow and/or RV output as markers of systemic blood flow and left pulmonary artery velocity as an estimate of pulmonary blood flow. In addition, ideally, information on blood flow distribution to the organs must be available to fully understand the cardiovascular status of the neonate.
Neonatal hemodynamics exist on a continuum, not in discrete categories. Notwithstanding this, there are several patterns of abnormal hemodynamic seen in the newborn. Although these are discussed separately in the following paragraphs, it must be highlighted that there will be overlap between these hemodynamic patterns in individual babies. One should also emphasize the importance of integrating the cardiac ultrasound findings with the clinical history and vital signs to define the likely diagnosis and individual therapeutic needs.
There are essentially five patterns of abnormal hemodynamic found in the newborn. These are summarized in Table 29.1 and are as follows:
Hypovolemic hemodynamic (see Chapter 27 )
Absolute hypovolemia is rare in the newborn but is seen in the immediate postnatal period in babies who have suffered intrapartum fetal blood loss and, postnatally, with subgaleal bleeds and postsurgical blood loss. Such babies may have a perinatal or clinical history consistent with blood loss and will be pale, invariably tachycardic, and have low BP if they have progressed beyond the compensatory phase. The main cardiac ultrasound finding is dramatically poor biventricular filling, so that the chambers look small and often have an appearance of poor contractility, reflecting low preload. The systemic veins will also be poorly filled and measures of systemic blood flow will be low or low normal depending on the degree of hypovolemia.
Management: Such babies need immediate volume replacement with isotonic saline and blood as soon as available. The latter may have to be un–cross-matched O-negative blood if the urgency of the situation demands it. Babies with ongoing blood loss of more than 40 mL/kg are vulnerable to develop transfusion-related coagulopathy and should be managed according to the principles of a massive transfusion protocol, where platelets and clotting factors are replaced proactively according to a defined schedule (see Chapter 27 ).
This pattern of loss of vascular resistance is invariable in hypotensive preterm babies who are more than 24 hours old; it can be seen during the first 24 hours, sometimes in combination with low systemic blood flow (see further on). It is also seen on recovery from shock or severe asphyxia and is the invariable hemodynamic in late late-onset sepsis (see Chapter 27 ). These babies will either be very preterm or have a clinical history consistent with sepsis or asphyxia. They will be hypotensive and often tachycardic; the impact on other clinical signs and markers of shock such as lactate will depend on severity. The cardiac ultrasound findings will show well-filled ventricles with a good, sometimes hyperdynamic appearance of the myocardial function. Measures of systemic blood flow will be high normal or high, reflecting the loss of resistance. If treatment is delayed, myocardial function will deteriorate as, after a certain period, myocardial oxygen delivery starts failing to meet tissue oxygen demand.
Management: The suggested management of septic shock is covered in detail in Chapter 27 ; there are overlaps between the management of septic shock and that of postasphyxial vasodilatory shock.
In preterm neonates with vasodilatory hypotension, the first question is whether it needs to be treated. If the hypotension is borderline and there are no changes in the markers of perfusion, such as lactate and/or urine output, then an expectant approach to management may be reasonable. The decision in these cases also depends on the etiology of the vasodilatory shock. For instance, in preterm neonates with suspected sepsis, there should be a lower threshold to initiate cardiovascular support. In babies with BP well below the normal range and/or with other markers of poor tissue perfusion, intervention is indicated. This hemodynamic needs a vasopressor effect. Some volume expansion may help to fill the additional vascular space created by the vasodilation. I would suggest starting some volume expansion (up to 20 mL/kg) as well as dopamine (because there is the most experience with this vasopressor) at 5 μg/kg/min and titrate the infusion rate up in small increments of perhaps 2 μg/kg/min to achieve a minimal acceptable BP. As long as the drug has reached and is being infused into the patient, one does not have to wait for a response longer than approximately 5 minutes to find out if the given dose must be increased again. It is unusual to need more than 10 μg/kg/min but if the BP remains low at higher doses of dopamine (20 μg/kg/min), then consideration can be given to adding another vasopressor because the hypotension may well be resistant to dopamine. Different neonatologists will recommend different vasopressors in this situation, including epinephrine, norepinephrine, and vasopressin as well as hydrocortisone. There is no evidence to guide the choice. My preference would be to add hydrocortisone at 1 mg/kg every 8 hours in that we know a bit more about its hemodynamic effects (e.g., probably vasopressor by potentiating catecholamine receptors), and it covers the risk of relative or absolute adrenocortical insufficiency. Because of hydrocortisone’s prolonged half-life in preterm neonates of less than 34 weeks’ gestation, dosing every 12 hours has recently been recommended for this patient population as a starting dosage interval with the option of shortening the dosage interval if the increase in BP is not sustained for the 12 hours. After that, I would consider norepinephrine or vasopressin with a preference for the former because there is more reported experience.
Low systemic blood flow hemodynamic
This pattern usually reflects absolute or relative myocardial dysfunction. Relative myocardial dysfunction means that a myocardium is struggling against increased or, in the case of the preterm myocardium, unfamiliar afterloads. This is the most common pattern of abnormal hemodynamic seen in the very preterm baby in the first 12 hours after delivery (see Chapter 26 ). It probably reflects maladaptation of the immature myocardium to the higher afterloads of extrauterine life, possibly compounded by the negative circulatory effects of positive-pressure ventilation as well as shunts out of the systemic circulation through the fetal channels and early cord clamping. It is unusual to find this pattern in preterm babies after the first 24 hours, where the abnormal hemodynamic is invariably vasodilatory. Low systemic blood flow due to afterload compromise has also been proposed as the pathophysiology of the cardiorespiratory compromise seen after PDA ligation. Low systemic blood flow is found in more mature babies with high ventilator and oxygen requirements, where primary myocardial dysfunction merges with the PPHN hemodynamic discussed below. Low systemic blood flow is also found in clinical situations associated with an injured myocardium, most commonly in the asphyxiated newborn but also in viral myocarditis or congenital cardiomyopathies. The impact of low systemic blood flow on vital signs is variable, particularly in the preterm baby, where BP may be normal. Because of this, recognition of low systemic blood flow often depends on proactive use of cardiac ultrasound in the high-risk clinical situations previously mentioned. The findings on cardiac ultrasound will be low SVC flow (<50 mL/kg/min) and/or low RV output (<150 mL/kg/min) and, particularly in the more mature baby, poor myocardial function, which may well be subjectively apparent as well as assessed by myocardial function measures.
Management: Most of these babies are not hypovolemic but borderline volume status is difficult to exclude and there is some evidence of short-term improvement in systemic blood flow in preterm babies with low systemic blood flow in response to volume. I would therefore suggest some volume expansion in these babies with 10 mL/kg of isotonic saline. The pathophysiology here would indicate the need for augmentation of myocardial function and, particularly if the BP is normal, with an agent that will tend to reduce afterload. At the same time as the volume, I would suggest starting dobutamine at 10 μg/kg/min. One practical difficulty in this situation is how to monitor the effect. Dobutamine may or may not increase BP even though systemic blood flow has improved. Ideally, the cardiac ultrasound should be repeated about an hour after starting the infusion but, experientially, if that option is not available, the dose can be titrated up until a chronotropic effect is seen in the heart rate, as this will make it likely that inotropy has been achieved. Dobutamine has a good dose-response relationship and, unless there is severe tachycardia, doses of up to 20 μg/kg/min are unlikely to do harm. The more mature baby with low systemic blood flow is often quite responsive to dobutamine, but low systemic blood flow in the preterm baby may be refractory to the intervention. If the BP is low, that suggests combined low flow and vasodilation; I would then suggest adding in dopamine at 5 μg/kg/min and titrating up to a minimally acceptable BP. The concept of titrating to a minimally acceptable BP becomes more important during this early postnatal period because of the risk of afterload compromise. There tends to be an assumption that more is better when it comes to BP, so people are quick to start vasopressor-inotropes if the BP is low but slow to reduce them if the BP gets pushed well into the normal range or high.
It is also important to consider the optimal time for weaning inotropes. The natural history of systemic blood flow in preterm and term babies is for flow to improve between 12 and 24 hours of age. There is often a reluctance to wean the vasopressor-inotropes or inotropes and the theoretical risk this creates is that continuing treatment may drive reperfusion in a system that is in spontaneous recovery. I would suggest repeat cardiac ultrasound at 24 hours of age to confirm normal systemic blood flow and then an aggressive wean of the cardiovascular medications over about 4 to 6 hours. If one does not have access to cardiac ultrasound, it is reasonable to wean the inotropes anyway as long as vital signs and other perfusion markers are normal.
The situation in the very preterm baby may be confused by the presence of large left-to-right ductal shunting even at this early time. This moves blood out of the systemic circulation back to the pulmonary circulation, which may be a factor contributing to low systemic blood flow. Management of the PDA is outside the scope of this chapter (see Chapter 22 , Chapter 23 , Chapter 24 ), but early medical closure of the poorly constricted ductus arteriosus with a large left-to-right shunt could be considered.
Persistent pulmonary hypertension hemodynamics (see Chapters 8 and 9 ). The plural is pointedly used in the title of this subsection because the hemodynamics are as varied as the causes of the high ventilation and oxygen requirements that lead neonatologists to consider PPHN as a diagnosis. It is important to remember that pressure in the pulmonary arterial system is determined by the same factor as in the systemic circulation, which is flow and resistance. Therefore, as in the systemic circulation, pulmonary arterial pressure can be high because flow is high, or resistance is high, or because both are high.
In fact, the hemodynamics here represent a continuum with, at one end, the classic concept of PPHN with high resistance, low pulmonary blood flow, and right-to-left shunts across the fetal channels, while, at the other end, there is normal pulmonary blood flow with varying degrees of increased pulmonary vascular resistance and pressure as well as variable fetal channel shunts. There are some parallels in these hemodynamics with the classic categorization of PPHN into primary (with relatively normal lungs) and secondary (with parenchymal lung disease); primary PPHN is more likely to involve low pulmonary blood flow hemodynamic and secondary PPHN is more likely to involve normal pulmonary blood flow hemodynamic.
PPHN with low pulmonary blood flow
This hemodynamic is commonly seen in term or late preterm babies with idiopathic primary PPHN. Clinically these babies will have increased oxygen requirements that are out of keeping with any parenchymal lung disease; they will often have relatively normal chest x-rays (CXRs) and may have relatively mildly increased work of breathing. This hemodynamic is also found in babies with hypoplastic lungs from conditions such as diaphragmatic hernia, renal anomalies associated with decreased fetal urine production, or premature prolonged rupture of the membranes. It is uncommon in very preterm babies except those born after prolonged oligohydramnios, where this is the invariable hemodynamic. This was traditionally explained as due to pulmonary hypoplasia; however, the responsiveness to iNO suggests a significant reversible component to the pulmonary hypertension in these babies (see case history 4 in Chapter 10 ).
These babies must have congenital heart disease excluded as soon as possible. Assuming that the heart is structurally normal, cardiac ultrasound usually shows evidence of suprasystemic pulmonary pressure with predominantly right-to-left shunting through the ductus (if patent) and bidirectional shunting through the foramen ovale. There will be low-velocity flow in the left pulmonary artery (mean velocity <0.2 m/s), a finding that predicts responsiveness to iNO. Measures of systemic blood flow may be normal, particularly if the flow through the fetal channels is not restricted. If the flow through the fetal channels is restricted, systemic blood flow may be low, with the likely pathophysiology being low pulmonary blood flow restricting systemic blood flow. The LV can only pump what it receives from the RV.
Management (see Chapter 9 for details): The primary problem in this hemodynamic is high pulmonary vascular resistance due to pulmonary vasoconstriction; therefore therapy must be aimed at dilating the pulmonary arterioles. iNO would be the first choice for this and, when this hemodynamic is associated with otherwise normal lungs, there is usually a brisk response to iNO and rapid normalization of the hemodynamic. In babies who are not responsive to iNO, consideration should be given to adding milrinone if the systemic BP is normal or, if the baby has hypotension, norepinephrine. In iNO-unresponsive babies, where the ductus arteriosus is constricting, there is some logic to considering opening the ductus arteriosus with prostaglandin E2. This will serve to decompress the right heart and improve systemic blood flow, albeit with relatively poorly oxygenated blood. In neonates not responding or not having a sustained response to iNO, the addition of sildenafil can be considered as the next step in the management.
PPHN with normal pulmonary blood flow
This hemodynamic is most commonly seen in babies with severe acquired lung disease such as meconium aspiration, pneumonia, or hyaline membrane disease. The main clinical feature is a history and CXR changes consistent with one of these diagnoses. The babies will usually have presented with severe respiratory distress and, when ventilated, will have high oxygen and pressure requirements. These babies must also have congenital heart disease excluded as soon as possible. If the heart is structurally normal, then pulmonary pressures in these babies will be variable and many of these babies will have subsystemic pulmonary artery pressures. Within this group there is a positive relationship between oxygenation index and pulmonary artery pressure, with the most severely affected babies usually having peri- or suprasystemic pulmonary artery pressures. As in the case of pulmonary artery pressures, shunts will be variable but will often have more left-to-right than right-to-left patterns. The ductus arteriosus often constricts early and closes by the second postnatal day. There is a high incidence of low systemic blood flow on the first day with improvement over time, much as seen in very preterm babies.
Management: The primary problem in these babies is pulmonary; the cardiac and pulmonary vascular effects are secondary to the pulmonary disease. With respect to high-pressure ventilation, it is important to remember that inappropriately high positive intrathoracic pressures will have a negative effect on cardiac output and pulmonary artery pressure will have to be higher to drive the blood through the lungs. The primary focus in these babies should be on optimizing respiratory and conventional ventilator management with surfactant, optimizing ventilator settings, or using other modes of ventilation (as indicated). If oxygenation is still poor despite optimized respiratory management, then iNO can be considered depending on the estimated pulmonary artery pressure and the patient’s oxygenation status.