Neonatal Haematology. Irene Roberts

By contrast, expression of the main platelet adhesion receptors appears to be similar in neonatal and adult platelets, with the exception of glycoprotein (Gp) IIb/IIIa, which is expressed at lower levels in neonatal platelets.¹⁵² Importantly, follow‐up studies have shown that previously impaired platelet responses in neonates are similar to adult responses within 2 weeks of birth.^152,153 Another point of note is that most studies of neonatal platelet function have been carried out on cord blood platelet‐rich plasma. However, more recent studies on whole blood have shown that healthy neonates have robust primary haemostasis overall, despite apparently hypofunctional platelets as assessed using standard techniques in platelet‐rich plasma. This may, at least in part, be due to the higher haematocrit and higher levels of von Willebrand factor in neonates. In addition, measures of in vitro platelet activation and aggregation in cord blood do not match those obtained when studying peripheral blood.¹⁵⁴ These observations have led some investigators to conclude that the function of neonatal platelets should not be viewed as impaired but, instead, that neonatal platelets function as a component of a generally well‐balanced neonatal haemostatic system.¹⁴⁹

Practical problems in interpreting neonatal blood counts and films

      Sample quality/artefacts

One of the commonest practical difficulties in interpreting neonatal blood counts and blood films occurs because of the much higher haematocrit in neonates. First, there is a higher frequency of clotted samples, which most likely reflects the challenge of collecting free‐flowing blood samples from neonates, especially those who are very low birthweight and/or preterm. Secondly, the high haematocrit often leads to poor‐quality ‘thick’ or unevenly spread films, which may give the misleading impression of the presence of abnormal red cells, such as spherocytes or target cells, which are not seen when the correctly spread section of the film is reviewed (Fig. 1.18). These effects can often be mitigated by dilution of the sample prior to analysis and preparation of the blood film. Another practical problem is the difficulty of distinguishing the typical changes seen on aged blood samples anticoagulated with ethylene diaminetetra‐acetic acid (EDTA) from the normal red cell features typical of preterm babies. It is therefore particularly important to make blood films on neonatal samples as soon as possible.

Fig. 1.18 Blood film of a healthy term neonate with a normal blood count and no evidence of haemolysis, showing artefactual changes in red cell morphology apparent on different sections of the blood film. (a) Thin end of the blood film showing apparent ‘spherocytes’ that are not present on the well‐spread section of the film (compare with Fig. 1.18c). A normal neutrophil, monocyte and NRBC can also be seen. (b) Thick end of the same blood film showing overlapping red cells, which make it difficult to interpret the red cell morphology, and apparent target cells and hypochromia that are not present on the well‐spread section of the film (compare with Fig. 1.18c). Normal neutrophils and an NRBC are also shown. (c) Good section of the blood film showing normal red cell morphology. Normal white blood cells are also shown. MGG, ×40.

      Site of sampling

The site of sampling also influences blood count analyses in neonates. For example, in the first few hours of life the Hb of venous samples is lower than that of heel‐prick samples collected simultaneously,¹⁵⁵ sometimes by up to 20–40 g/l.³⁵ This difference is greater in preterm neonates and falls with increasing postnatal age, such that by the fifth day of life there is almost no difference in Hb between a well‐taken heel‐prick sample and a venous sample.¹⁵⁵ Similarly, the haematocrit and red blood cell count at birth are lower in venous blood compared with capillary blood samples collected simultaneously, while there is no difference in the other red cell indices by site of sampling.¹⁵⁵ Neonatal heel‐prick samples have also been shown to have white cell, neutrophil and lymphocyte counts about 20% higher than arterial or venous samples; counts are most likely to approximate to those of venous blood if there is a free flow of blood and if early drops, excluding the first, are used for the count.³⁵ In contrast, the platelet count and mean platelet volume (MPV) are lower in capillary samples than in venous samples.^155,156

      Gestational age and postnatal age

Although the Hb, haematocrit and MCV vary little over the first week of life, there is a gradual loss of other distinctive neonatal red cell features after the first week of life. In particular, the numbers of circulating nucleated red cells fall, as mentioned above, and echinocytes are gradually replaced by red cells with the more typical appearance of adult red cells. Consideration of the postnatal age of a baby is most important when interpreting blood films to help in the identification of infection or the cause of anaemia. For example, maternal chorioamnionitis may cause extremely high neutrophil counts, often with associated toxic granulation, in the newborn infant despite the absence of active infection in the neonate (Fig. 1.19). It seems likely that this is due to maternal cytokines crossing the placenta, although this has not been specifically demonstrated. Importantly, virtually identical changes observed on neonatal blood films after the first week of life are highly likely to reflect active infection in the neonate as any maternal cytokine‐driven changes will have resolved by this time.

Fig. 1.19 Blood film in maternal chorioamnionitis showing neutrophilia, left shift (myelocytes and promyelocytes) and toxic granulation. MGG, ×40.

Fig. 1.20 Leucoerythroblastic blood film in a preterm neonate with hypoxic ischaemic encephalopathy showing a myelocyte, an NRBC, an eosinophil and a macropolycyte, likely to be a tetraploid cell. MGG, ×100.

      Pregnancy‐associated complications and mode of delivery

In addition to being influenced by maternal chorioamnionitis, the white cell count (WBC) at birth is also affected by the mode of delivery, being lower after an elective caesarean section than after either vaginal delivery or caesarean section performed after labour has commenced.⁶⁴ Hypoxic ischaemic encephalopathy (HIE) following perinatal hypoxia (e.g. due to the cord being round the neck of the baby) also causes an increased WBC; in this case the blood film is leucoerythroblastic and the neutrophils show varying degrees of toxic granulation, often making it difficult to identify the presence of coexisting infection based on the haematological findings alone (Fig. 1.20).

References

      1 1 Huyhn A, Dommergues M, Izac B, Croisille L, Katz W, Vainchenker W and Coulombel L (1995) Characterization of hematopoietic progenitors from human yolk sacs and embryos. Blood,