The incidence of renal injury in term newborns that experience perinatal asphyxia is substantial (50%–72%), and this complication is associated with an increased risk of mortality. It should be noted that there are no fixed criteria to determine its presence. The usual approach relies on the elevation of urea and creatinine levels associated with a decreased urine output,8 although it is important to remember that creatinine levels do not usually increase until 25%–50% of renal function is lost, and that renal failure is not always accompanied by oliguria.5,8

Despite the significant incidence of renal injury, few studies have assessed the changes undergone by the kidney. Before birth, the high renal vascular resistance, high activity of plasma renin and low glomerular filtration rate, among other factors, make this organ particularly sensitive to hypoperfusion. The main type of renal injury is acute tubular necrosis and, in more extreme cases, diffuse renal cortical necrosis.8 Prolonged hypoperfusion may not only reduce glomerular filtration, but also damage cells with a higher energy consumption (tubular cells), causing necrosis of the proximal tubule and the ascending loop of Henle. In addition, the infiltration of neutrophils circulating in the ischaemic renal tissue and the release of cytotoxic substances may promote damage not only in endothelial cells but throughout the kidney.9

In an in vitro study of human proximal renal tubule cells subjected to hypoxia followed by reoxygenation, Du et al10 found a lower number of viable cells and an increased lactate dehydrogenase activity. The authors concluded that the integrity of the membrane is compromised by hypoxic episodes and that tubular cells may suffer irreversible damage. In a newborn pig model, Satas et al11 detected the presence of renal infarction secondary to hypoxia-ischaemia.

Based on the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines,12 the AWAKEN study defined acute kidney injury in newborns as an increase in serum creatinine by 0.3mg/dL or greater (26.5μmol/L) or to 1.5 times the baseline value or more, and/or a urine volume of less than 1mL/kg/hour in days 2–7 post birth.13 In this study, the serum creatinine threshold used to define stage 3 kidney injury was 2.5mg/dL (221μmol/L) rather than the 4.0mg/dL (353.6μmol/L) applied in adults.12,13 Predictive tools have also been developed, such as the renal angina index, which aims to identify children at risk of acute kidney injury lasting more than 72h beyond functional impairment. The clinical signs of injury are based on changes in the estimated creatinine clearance or the percentage of fluid overload.14 The neonatal STARZ score, developed recently, is used for quick quantitative assessment of the risk of acute kidney injury in newborns admitted to the neonatal intensive care unit (NICU) by means of variables such as age at admission, gestational age, sepsis, significant cardiac disease, urine output, serum creatinine and the use of nephrotoxic drugs, furosemide or inotropes.15

When it comes to newborns with neonatal encephalopathy treated with therapeutic hypothermia, there is substantial heterogeneity in renal function and injury. Some authors have asserted that serum creatinine values relative to postnatal age may provide relevant information for the management of these infants.16 Other studies have assessed the predictive value of several urine markers of acute kidney injury: while the levels of neutrophil gelatinase-associated lipocalin, kidney injury molecule-1 and interleukin-18 were found to be elevated in newborns with HIE managed with therapeutic hypothermia who developed acute kidney injury, due to the substantial variability in their levels and poorly defined thresholds their clinical usefulness is unclear.17

Asphyxiated newborns are at risk of developing hyponatraemia secondary to oliguric acute kidney injury due to water retention and a decrease in sodium reabsorption at the tubular level.8 They may also develop syndrome of inappropriate antidiuretic hormone, which can contribute to the development of hyponatraemia, considerable increase in weight, decreased serum osmolality considerable, oliguria and/or increased urine osmolality.2,8

An important aspect to consider in these newborns at higher risk of hyperkalaemia due to renal failure is the increase in cardiac excitability.8 This cardiac abnormality can be exacerbated by hypocalcaemia.2 Magnesium levels may also be abnormal, either reduced or elevated,8 with evidence of low levels in the polyuric phase of acute tubular necrosis.2

There is evidence of an association between perinatal asphyxia and liver damage, although the reported incidence varies widely (22%–80%) (Table 1). Hankins et al20 reported the highest incidence (80%), although this figure could be related to the severity of the cases in the study, as patients had to have developed encephalopathy after the asphyxia episode to be included. In a later study,21 nearly 50% of newborns had not developed HIE. However, both studies applied less stringent criteria to define liver damage compared to other studies, like the one by Tarcan et al.,22 in which the levels of alanine aminotransferase had to exceed 100 U/L, or the one by Barnett et al.,23 in which the peak values of alanine aminotransferase or aspartate aminotransferase had to be 200 U/L or greater at 48h. Possibly as a result of this, the incidence reported in these studies was of 39% and 22%, respectively. Furthermore, the latter one included neonates that had died of perinatal asphyxia, which is indicative of the severity of the asphyxia event in the neonates included in the sample.

This variation in the incidence could be explained not only by discrepancies in the definition of liver damage, but by the inclusion criteria applied in each study. It is also worth noting that liver damage may be underdiagnosed if liver markers are used for its identification, for, while there is a correlation between liver enzyme levels and liver damage,21 normal levels of transaminases or lactate dehydrogenase do not rule out damage.

In a study of 19 newborn pigs, Karlsson et al.24 found foci of necrosis in 5 animals, infarction in 6, neutrophil infiltration in 6, thrombi in the intrahepatic portal veins in 3 and subcapsular haematoma in 1. Ikeda et al.18 reported detection of cytoplasmic eosinophilia and canalicular cholestasis in samples obtained from asphyxiated lambs and fatty changes in tissue samples from lambs that suffered mild asphyxia. Two brain-damaged lamb foetuses had fatty changes and congestion in the liver, in addition to inflammatory cell infiltration and centrilobular necrosis.

In the pathological examination of infants that died of post-asphyxial HIE, Barnett et al.23 detected congestion in 22, fatty changes in 13, haemorrhage in 4, centrilobular necrosis in 7 and other necrotic-ischaemic changes in 2.

Although there is evidence that progression to complete hepatic failure is rare,21 there is also evidence that hepatic involvement is associated with a high mortality.22

In infants that have experienced perinatal asphyxia, the prothrombin time and the international normalized ratio are significantly elevated,19,21 and there is a risk of haemorrhage if there are additional alternations, such as a lower platelet count or decreased fibrinogen level.5

The liver synthesises a large part of the proteins involved in coagulation, but the mechanism by which hypoxia leads to bleeding is not well understood. Thus, it may be reasonable to expect coagulation disturbances if liver function is impaired.11 However, Chadd et al.,25 in a case-control study that included 47 term newborns, found that neonates that had experienced hypoxia exhibited coagulation abnormalities compared to controls, and that the source of the disturbance was the consumption of clotting factors, that is, disseminated intravascular coagulation. Similar results have been described by Chessells and Wigglesworth26 in 9 infants that suffered severe hypoxia, suggesting that direct liver damage would play a secondary role.

Perinatal asphyxia may result in decreased perfusion of the gastrointestinal tract, manifesting with vomiting, diarrhoea, gastrointestinal haemorrhage and even necrotising enterocolitis (NEC).4 The incidence of these events is 29%.27

The intestine is one of the organs most sensitive to ischaemia,27 as it has areas located between two major arteries (mesenteric arteries) that are prone to injury secondary to perinatal asphyxia.5 In addition, intestinal injury is not caused only by ischaemia, but also by reperfusion.27

In addition to an underdeveloped regulation of vascular resistance, the immature intestine has increased metabolic demands. This can cause problems in episodes of cardiovascular stress, as the immature body may not be able to increase intestinal blood flow and metabolic demands may overwhelm the infant’s ability to increase oxygen consumption. This results in defective pressure flow autoregulation in response to hypotension, leading to tissue hypoxia.28 Thus, gastrointestinal complications are among the chief disturbances present in the context of perinatal asphyxia, with a pattern of intestinal injury similar to that found in NEC.29

Necrotising enterocolitis is a potentially devastating acquired intestinal disorder in preterm infants. Its incidence in very low birth weight infants is 7%, and it is associated with a high mortality (15%–30%) and substantial morbididy.29–32 It is characterised by intestinal inflammation and ischaemia and changes in intestinal microcirculation.35,36

Hypoxic-ischaemic myocardial injury is increasingly contemplated as a potential cause of cardiovascular dysfunction in newborns, as it is present in up to one third of affected infants and could contribute to an increase in neonatal morbidity and mortality.7,37 Hankins et al.20 conducted a cohort study on acute perinatal asphyxia followed by encephalopathy and found that 78% of cases were associated with cardiac injury, defined as inotrope requirement after 2h post birth and/or elevation of creatinine kinase-MB. Meanwhile, Shah et al.6 found an incidence of 62%, in contrast to the 29% established by Martín-Ancel et al.27

When the heart is affected by the asphyxia event, a transitory myocardial ischaemia may develop that is known as hypoxic cardiomyopathy. A cohort study was conducted to determine the frequency of this disorder in newborns with cardiovascular symptoms, finding an incidence of 70% in newborns with perinatal asphyxia.38

One of the markers used to assess for the presence of myocardial injury is creatine kinase-MB.14,37,39 However, other authors consider that this biomarker offers a low specificity for in newborns, as enzyme levels decrease considerably from 24h after the asphyxia event.38 Therefore, normal values of this isoenzyme do not reliably rule out myocardial injury.27 Other sensitive indicators are troponin-T, troponin-I and brain natriuretic peptide.39,40

Using these markers and other diagnostic methods, such as electrocardiography chest radiography, makes early detection of myocardial injury possible. This is of utmost importance, as evidence has emerged that transitory myocardial ischaemia secondary to perinatal asphyxia is more frequent than previously expected.38 Early diagnosis of myocardial injury can help guide management, thus improving neonatal outcomes.38,40

In an observational study in neonates with severe asphyxia, Barberi et al.37 found that, in addition to elevation of cardiac enzymes and relevant ischaemic electrocardiographic changes, patients also had depressed ventricular function.

Hochwald et al.41 compared left ventricular cardiac output in neonates with HIE and healthy neonates, and found a decrease of 42.34% in the former. Gurgul et al.34 found that in newborn rat pups, contractile force was only affected in the ventricles.

The incidence of myocardial injury established based on histological findings after perinatal asphyxia appears to be higher compared to estimates based on clinical manifestations. Gurgul et al.34 examined cardiac cells with electron microscopy and found degenerative changes in some myocardial cells that exhibited thinning and rupture of myofibrils, as well as degenerative mitochondrial changes. Yang et al.42 demonstrated that these alterations in mitochondria were reflected in their function, as the mitochondrial respiratory chain in myocardial cells was affected, probably due to an intracellular calcium overload.