Pathogenesis of myocardial hypoxic injury

Oxygen homoeostasis is critical for survival and function of all cells in particular the myocardium.8 Hypoxia is characterised by inadequate oxygen delivery to the myocardium and elicits both indirect and direct effects on the heart which are mediated by neurohormonal mechanisms.9 In the acute or chronic situation, it is likely that a reduction in arterial blood oxygen partial pressure is responsible for the mismatch between oxygen supply and delivery, and this in turn may result in a reduction in myocardial contractility.8

At the cellular level, anoxia ablates adenosine triphosphatase synthesis leading to cell death by induction of apoptosis.10 Mammalian cells respond to hypoxia by activating transcription factors and hypoxia-inducible factors which bind to hypoxia-responsive elements and consensus sequences in the promoter region of more than 100 genes. The transcription of these genes allows the cell to adapt to and survive the hypoxic environment.8

Although hypoxia is thought to exhibit a negative effect on myocardial contractility, several neural and humoral changes act conjointly to increase myocardial contractility in hypoxia. The increase in sympathetic activity and the release of apelin, which has a potent positive inotropic effect, were mediated in part by an enhanced myofilament sensitivity to calcium which increases contractility. In contrast, adenosine produced from the cardiac myocyte and endothelium in response to the hypoxic injury reduces the contractile responsiveness of the myocardium to adrenergic stimulation. However, the production of nitric oxide may reduce or increase contractility depending on the micro-environmental circumstances.8

Normally, the pulmonary vasculature relaxes after birth. However, hypoxia prevents this normal physiological process resulting in increased pulmonary versus systemic vascular resistance and causes deoxygenated blood to be shunted to the systemic vasculature.11 Persistent pulmonary hypertension of the newborn (PPHN) precipitated by hypoxia is characterised by impairment of nitric oxide synthase and is treated with exogenous inhaled nitric oxide therapy.12

Clinical features of cardiovascular dysfunction in hypoxia–ischaemia

Antenatally, even in infants with non-academia-related neurological impairments, 74% had shown intrapartum non-reassuring fetal heart rate patterns.13 This is also mirrored in the persistent bradycardia often seen in infants with neonatal encephalopathy. Changes in baseline heart rate variability was found to be insensitive for detecting seizures in asphyxiated neonates,14 but diminished heart rate variability was considered to be a poor prognostic sign.15

Martin-Ancel et al found cardiac involvement in 29% of neonates with birth asphyxia (n=72). ECG changes consistent with myocardial ischaemia were found in 19%, and 21% had a transient systolic murmur.2 Barberi et al graded ECG changes in asphyxiated newborns from 1 to 4 based on T, ST and Q wave measurements taken from a standard 12-lead ECG.16 Infants with the most severe hypoxic damage had grades 3–4 ECG changes.5 Similarly, Kanik et al found that 38% of perinatally asphyxiated neonates had ECG changes consistent with myocardial ischaemia.17

Echocardiographic measures of cardiovascular dysfunction function in hypoxic injury

Traditionally, left ventricular systolic function is evaluated using M Mode to measure ejection and shortening fraction (SF).18 19 Barberi et al found that SF decreased in severely versus mildly asphyxiated newborns.5 Wei et al also demonstrated lower left ventricular ejection (LVEF) and SF in infants with severe asphyxia compared to the mildly asphyxiated and control groups.20

Tissue Doppler imaging (TDI) is evolving as a technique to allow measurement of myocardial contraction and relaxation velocity from the myocardium and could allow recognition of myocardial dysfunction related to hypoxic injury.21 22 Wei et al used TDI to serially evaluate left ventricular systolic function in newborns with mild to severe asphyxia. The peak systolic velocity of the anterior mitral valve leaflet (Sm wave) was measured with TDI. The LVEF and SF of the severe asphyxia group were significantly lower than in the mild and control groups. The Sm wave of the asphyxia group was significantly lower than that of the control group (p<0.001). In the severe asphyxia group, the Sm wave at 24 h was significantly lower than that at 48 or 72 h (p<0.001). The findings of this study suggest that Sm by TDI is a more sensitive indicator of left ventricular systolic function than LVEF or SF measured by M-mode echocardiography.23

A Doppler index of myocardial performance combining systolic and diastolic time indexes (Tei index) has been used to characterise ventricular function in normal infants.24 25 Ichihashi et al found that the Tei index was higher in infants with mild asphyxia compared to controls.25 Similarly, Matter et al found both left and right Tei indexes to be higher in asphyxiated infants.26 Myocardial performance index is a measure of both systolic and diastolic dysfunction; it may be a better overall determinant of cardiac dysfunction.

Pulmonary arterial hypertension is an important complication of perinatal asphyxia and treatment using therapeutic hypothermia.27,–,29 PPHN can be assessed indirectly by measuring the velocity of the tricuspid regurgitant jet using Doppler echocardiography and the modified Bernoulli equation.

Biomarkers of cardiac dysfunction in hypoxic injury

Cardiac troponins I and T are well-established markers of myocardial ischaemia and cardiac failure in adults and children.30 31 Troponins are the calcium binding site of the myofibrillary thin filament of the cardiac sarcomere. There are three distinct proteins: troponin T, troponin C and troponin I. Troponins T and I are cardiac specific and can be markers of cardiac injury in adults with ischaemic or haemorrhagic stroke in the absence of myocardial cell injury.32 Troponins T and I have been used to detect myocardial injury in the neonatal period.33 34

Gunes et al found that troponin T levels were significantly higher in infants with severe asphyxia (n=15) compared with the mildly affected group (n=15). Troponin T reached a peak concentration on day 1 but remained high for 1 week.35 Szymankiewicz et al also found that troponin T was significantly higher in asphyxiated premature infants versus controls.36 Troponin I is elevated in asphyxiated neonates, and it has been suggested that these levels are maternal in origin. However, Trevisanuto et al concluded that elevated troponin I is of neonatal origin and may be attributed to the earlier switching from skeletal troponin I to cardiac troponin I in the fetal–neonatal myocardium, an increased supply of cardiac troponin I in the free cytosol of the cardiac myocytes.33 Kanik et al also suggested that troponin I measured on days 1 and 3 of life following perinatal asphyxia was a valuable marker of myocardial injury.17

Creatine kinase (CK) exists as a dimer of two subunits: B and M, and three isoenzymes: CK-MM, CK-BB and CK-MB. Myocardium contains 15–20% CK-MB, and when myocardial injury occurs, blood levels of total CK and CK-MB begin to rise within 4–8 h.37 38 CK-MB levels are elevated in infants with moderate or severe hypoxic ischaemic encephalopathy.5 35

Brain-type natriuretic peptide (BNP) is a 32-amino acid ring structure which has its sequence present on chromosome 1. It is found in high concentration in the ventricles of the heart and is released in response to volume and pressure loading and ventricular stress. Pro-BNP is the inactive precursor and is cleaved into BNP, the active component and N-terminal pro-BNP (NTpBNP), an inactive by-product.39 BNP and NTpBNP levels are high at birth and fall slowly over the first 2 weeks of life. Normal ranges of BNP and NTpBNP have been established in neonates but vary depending on age of neonate and the testing kit used.38,–,41

NTpBNP and troponin predict poor neonatal outcome (grade III/IV intraventricular haemorrhage or death) in preterm infants with a patent ductus arteriosus (PDA).40 Troponin, NTpBNP and a PDA scoring system at 48 h may facilitate the identification of those infants with a PDA, who are at greatest risk of poor neurodevelopmental outcome at 2 years of age.40 The use of serial echocardiography to determine cardiovascular dysfunction in infants with hypoxic injury may help to predict neurodevelopmental outcome; however, at present, there is a paucity of literature in this field and warrants further research.

Management of cardiac dysfunction in hypoxic injury

Following fluid resuscitation, as recommended by the American College of Critical Care guidelines42 and the 2010 International Liason Committee Recommendations,43 caution with fluid is widely recommended to prevent cerebral oedema and subsequent brain injury.44 Hypotension is observed in up to 62% of patients following perinatal asphyxia,3 supportive therapies to improve hypotension include dopamine, epinephrine and dobutamine. These agents may improve cardiac function via β1 adrenoceptor stimulation (table 1). However, adverse effects such as tachycardia, increased oxygen consumption and altered tissue perfusion complicate their use.4 Dopamine is an endogenous cathecholamine which improves blood pressure, cardiac output and stroke volume on starting doses of 10 micrograms/kg/min. However, there is insufficient evidence to suggest that the use of dopamine in term infants with perinatal asphyxia either improves mortality or long-term neurodevelopmental outcome.45 Epinephrine has both α and β receptor agonist effects. At low dose, it is a potent inotrope, chronotrope, systemic and pulmonary vasodilator.46 Dobutamine is an inotrope with predominantly β receptor effects, and up to 20 micrograms/kg/min is the dose range used to increase cardiac output.46

Milrinone is a phosphodiesterase III inhibitor and has been used postcardiac surgery for neonates with low cardiac output syndrome and also in the treatment of pulmonary hypertension.47 In animal models, the vasodilatory effects of milrinone have been found to alleviate pulmonary hypertension in newborn piglets. Epinephrine, dobutamine and milrinone increase cardiac output, stroke volume and systemic oxygen delivery without aggravating pulmonary hypertension in asphyxiated newborn piglets48 (table 1). Despite there being no clear evidence in the literature advocating the use of one inotrope over the other in infants, they are all used is clinical practice.4 Pulmonary hypertension is a recognised complication of hypoxic injury and may be treated with nitric oxide which improves oxygenation by selective pulmonary vasodilatation.12 Milrinone may be a useful therapy in PPHN as a result of hypoxic injury.48

Consequences of hypothermia on cardiac function

Therapeutic hypothermia is now standard practice as a neuroprotective therapy following perinatal asphyxia, for infants who meet the inclusion criteria.28 49,–,51

Cardiovascular side effects have been consistently noted in these cooled infants, in particular, an increase in sinus bradycardia and hypotension with an increased need for inotropic support during the hypothermia therapy.51 During mild hypothermia, cardiac output is reduced to approximately 67% of the posthypothermic level with a concurrent decrease in heart rate and stroke volume.49 50

Meta analysis of six clinical trials of hypothermia did not demonstrate a significant effect of hypothermia on cardiac arrhythmia requiring medical treatment.51

Novel therapies for cardiovascular dysfunction secondary to hypoxia ischaemia

Acute myocardial infarction (MI) in adults leads to cardiac remodelling, which includes thinning of the infarct wall, and cardiac dilatation. At the cellular level, this is associated with apoptosis, necrosis of cardiac myocytes, hypertrophy, fibrosis and infiltration of inflammatory cells.52 A similar process may occur in ischaemia secondary to hypoxic injury in newborns. No single treatment allows complete regeneration of the injured myocytes whose abnormal state may ultimately lead to cardiac failure. However, stem cell research is under investigation in a variety of animal models to treat heart failure due to myocardial hypoxic injury.52 A cardiac progenitor cell can differentiate into any of the cardiac cell lineages, including endothelial cells and cardiomyocytes.53 Transplanted mouse embryonic stem cells in the infarcted heart also show significantly improved cardiac function up to 3 months following transplantation.54 Similar results were noted in an engrafted sheep heart, suggesting that embryonic stem cells have the potential to engraft into cardiac myocytes in small and large animal models.53

Transplantation of bone marrow mononuclear cells in adults following acute MI increases myocardial contractility.55

In adults, intracoronary infusions of autologous bone marrow cells (BMCs) are safe and feasible in patients with acute MI and demonstrated up to 9% improvement in global LVEF (BMC).56 57 The REPAIR-AMI trial demonstrated a significant improvement of global and regional ejection fraction in the BMC group (+5.5%) compared to placebo (+3%) during 4months of follow-up.55 58

Human umbilical cord blood stem cells (hUCBSCs) may represent an alternative source of stem cells for myocardial-cell replacement and can be highly expanded. They spontaneously express proteins of paramount importance for cardiovascular regeneration, such as Cx-43, SERCA-2 and SDF-1α.59 hUCBSC were injected into a neonatal sheep heart and detected in the myocardium up to 6 weeks after transplantation following pulmonary artery banding. Transplantation was accompanied by right ventricular functional improvement.60 Apparent neonatal myocardial recovery following severe hypoxia–ischaemia may be related to the large number of hUCBSC. However, there is a paucity of studies on long-term cardiovascular function in this group of patients. Therefore, future advance in stem cell therapy in animal models and evaluation of long-term outcomes are required.

Conclusions

In neonates following perinatal asphyxia, cardiovascular dysfunction may contribute to increased mortality and morbidity. Early assessment with echocardiography (tissue Doppler and myocardial velocity imaging) combined with serum biomarkers (eg, BNP, troponin) will allow more accurate evaluation of cardiac dysfunction. This may allow rapid diagnosis of cardiac dysfunction and allow early initiation of inotropic support. Future studies of long-term cardiovascular outcomes in relation to neurodevelopmental status are required.