Cord Blood Erythropoietin and Markers of Fetal Hypoxia

October 10, 2010

Objective: To investigating the relationship between cord blood erythropoietin and clinical markers of fetal hypoxia.


Objective: To investigating the relationship between cord blood erythropoietin and clinical markers of fetal hypoxia.

Methods: Erythropoietin (EPO) concentrations in umbilical venous blood were measured from 80 neonates with low-risk and different risk factors by radioimmunoassay (RIA). The relation between EPO concentration and clinical markers of fetal asphyxia was evaluated.

Results: Median EPO concentrations were 29mU/mL, in infants with no risk factors or complications during pregnancy (n=23). In infants with complicated pregnancy and no complication during labor the median level of EPO was 68mU/ml (n=25) Infants showing abnormal FHR pattern during labor had median EPO levels of 42mU/ml (n=13) while infants showing umbilical blood acidosis showed a median EPO levels of 76mU/ml.

Using multiple regression, we found that the erythropoietin concentration correlated significantly (p<0.01) with fetal growth retardation and umbilical acidosis but not with gestational age, abnormal fetal heart rate (FHRT) pattern or Apgar score at 5 minutes.

Conclusion: Elevated Erythropoietin concentrations in umbilical venous blood may indicate prolonged fetal hypoxia.


More than forty years have passed since Virigina Apgar introduced her score to evaluate the condition of the newborn (1). Although not meant primarily for this purpose, the score has become the most widely used indicator of perinatal asphyxia. Nevertheless, it is still difficult to determine the presence, duration and extent of fetal hypoxia. Certain abnormal fetal heart rate patterns are considered to indicate imminent fetal risk (2). Whether meconium stained amniotic fluid is related to fetal distress is controversial (3). Umbilical arterial pH is widely used as an indicator of fetal hypoxia.

All of these are indirect markers. Moreover, neurologic outcome seems to correlate poorly with acidosis at birth (4,5), low Apgar score or meconium stained amniotic fluid (5). Therefore, additional markers are required to distinguish between acute and chronic fetal hypoxia.

Erythropoietin production is stimulated by hypoxia in adults (6) and fetuses (7,9). Because this glycoprotein does not cross the placenta (10,11) elevated cord blood concentration should be a marker of fetal hypoxia. Increased cord EPO concentrations have been found in maternal diabetes (12,14), maternal hypertension (12,15) and acute fetal hypoxia (15,16). Fetal polycythemia frequently occurs in intrauterine growth retardation and increased levels of EPO have been reported in pathological pregnancies possibly associated with placental insufficiency (17). Severe anemia was accompanied by increased EPO concentration (18). Elevated EPO in these cases suggested that prolonged intrauterine hypoxia is an important factor among these fetuses.

This study was conducted to investigate the relationship between erythropoietin concentration in umbilical venous blood and the indirect markers of fetal hypoxia (Apgar score, Abnormal FHR pattern, Meconium stained amniotic fluid, Umbilical arterial pH and blood gases).

Subjects and Methods

Eighty pregnant women with singleton pregnancy were selected for this study. Thirty women had no complications during pregnancy and fifty women had one or more complications during pregnancy; 19(38%) had pregnancy induced hypertension and/or pre-eclampsia; 11(22%) had Diabetes mellitus; 6(12%) had both Diabetes and hypertension; 8 (16%) had rheumatic heart disease and 6 (12%) had vaginal bleeding during pregnancy.

Forty-six (57.5%) infants were delivered spontaneously; 20 (25%) by Caesarian section before the onset of labor; 10 (12.5%) by Caesarian section after the onset of labor and 4 (5%) vaginally by operative means (forceps or ventous).
The infants' birth weight was nearly normally distributed with a range of 1050 to 4700gm. Seven (8.75%) weighed below the tenth percentile for gestational age (20)

The gestational age ranged from 33 to 42 completed weeks; 22(27.5%) were born before completed 37 weeks and 58 (72.5%) were term infants. Sixteen infants (20%) were transferred to our neonatal intensive care unit after birth, 12 for prematurity; 3 because of hypoglycemia and one because of pulmonary maladaptation.

Based on clinical signs of fetal distress, we defined four groups of infants:
1- Low-risk group (n=23) singleton term newborns (37-42 completed weeks) with no event detected that could lead to fetal hypoxia during pregnancy e.g. maternal diabetes, hypertension, vaginal bleeding, preterm labor, rupture of the membranes more than 24 hours before delivery or heart disease. Also there was no signs of fetal distress during labor e.g. meconium stained amniotic fluid, abnormal FHR pattern; umbilical pH below 7.20 or need for special care after birth.
2- Complicated pregnancy (n=25+ with events detected during pregnancy that can lead to fetal hypoxia and no signs of fetal distress.
3- Abnormal FHR pattern and/or meconium stained amniotic fluid (n=13) during labor but no umbilical acidosis.
4- Umbilical acidosis (n=19); umbilical arterial pH below 7.20 with or without abnormal FHR pattern

Erythropoietin was measured by competitive binding double antibody radioimmuno-assay technique (RIA)(20). Using Commercial Kit of Incstar-EPO trace, 125 I (Incstar Corporation, Stillwater, Minnesota USA).

Fetal heart rate was continuously recorded during labor using IM76 intrapartum fetal monitor (advanced Medical System Inc.) All recordings were evaluated retrospectively during the last 40 minutes of the first stage of labor by the same obstetrician who classified FHR according to that of MacDonald et al (21).

Blood gases and pH were analyzed in umbilical arterial blood drown by needle puncture immediately after birth. An ABL30 blood gas analyzer (Radiometer, Copenhagen, Denmark) was used.

To determine the relationship between EPO concentration in umbilical venous blood and clinical findings we performed multiple regression analysis and reported the results as statistically significant at p<0.05. All calculations were done with the SPSS-statistical package (SPSS Inc. Chicago,.n).


Table 1 shows values of EPO, pH and blood gases in low-risk group and other groups. Significant difference between mean EPO in low-risk group and other groups was found.

Erthyropoietin concentrations did not correlate significantly with umbilical arterial oxygen pressure (PO2) and carbon dioxide pressure (PCO2) and correlate significantly with umbilical arterial pH [r=-0.28;p<0.01].

Table 2 shows the Erthyropoietin concentrations of low-risk group in our study and other studies.

Erthyropoietin concentration in preterm infants [median 55mU/ml; range 29-160] were not different from those in term infants [median 72.5mU/ml; range 29-150]. However, EPO concentrations were markedly higher in small for gestational age infants (p<0.01) below the tenth percentile than in average gestational age infants [median 62.4mU/ml Vs median34mU/ml). Infants who need special care had higher EPO concentrations than well infants [median 53.6mU/ml Vs 29mU/ml] p<0.04.

Table 3 presents results of EPO concentration in low-risk group and the other three groups with clinical signs of fetal distress. The four groups differ significantly with regard to EPO concentration in umbilical venous blood (p<0.01). No infant in the low-risk group had an EPO above 58mU/ml. On the other hand, several infants with complicated pregnancy or umbilical acidosis had EPO levels above 300mU/ml.

The median EPO concentration increased with decreasing umbilical arterial pH (p<0.01).

Using multiple regression analysis we evaluated gestational age and factors related to fetal distress; fetal growth retardation, abnormal FHR pattern, umbilical arterial blood pH and Apgar score at 5 minutes.

Erthyrpoietin concentration correlated significantly with fetal growth retardation (p<0.01) and with umbilical arterial pH. The levels of EPO did not correlate with gestational age, abnormal FHR pattern or Apgar score at 5 minutes.


Levels of EPO in umbilical venous blood observed in our low-risk group is comparable to those measured in other studies, however, inclusion and exclusion criteria for control groups are not comparable in all of these studies (table 2).

Like Widness et al (13, 14, 22), Ruth et al (15,17), Forestier et al (23) and Maier et al (24) we found no difference in EPO concentrations between preterm and term infants. These results contrast to those of Thomas et al (11) and Ekardt et al (25), who reported an increase with progressing gestational age. However, the latter studies involved only infant in an advanced stage of gestation (36-43 weeks).

In contrast to Ekardt et al (25), who observed no differences in EPO levels between average for gestational age (AGA) and small for gestational age (SGA) infants in the absence of fetal distress, we found increased EPO levels in SGA infants, suggesting a prolonged period of hypoxia. This agrees with Ruth et al (15) and Maier et al (24) who found a correlation between cord blood EPO levels and the degree of fetal growth retardation.

Whereas adults produce EPO in the kidney, the main production site during fetal life is thought to be the liver in animals (7), as well as in human beings. Widness et al (26) observed that EPO concentrations in the cord blood of infants with Potter syndrome were similar to those in healthy term infants. Furthermore, they found no difference between a nephric infants and those with severe renal disease. They concluded that nearterm human kidney is not important for fetal erytheropoietin synthesis.

We found no relation between erythropoietin concentration in umbilical venous blood and umbilical arterial PO2. This in agreement with the observations of Teramo et al (12), Widness et al (14) and Maier et al (24). This can be explained in that tissue oxygenation depends not only on arterial PO2 but also on the oxygen dissociation characteristic, which is influenced by acidosis. Furthermore, fluctuations in arterial PO2 occur more rapidly than those in serum erythropoietin concentration.

In healthy adults following exposure to hyboparic hypoxia, elevated serum EPO concentrations decline with an average half-life of 5.1 hours (6). Ruth et al (15) found a shorter half-life in newborn infants, after cessation of a hypoxic stimulus serum EPO declined with a mean half life of 3.7 hours. Assuming that these results (15) obtained immediately after birth can be extrapolated to fetal life, we can detect a proceeding hypoxia by elevated EPO concentrations several hours after cessation, even when the PO2 returned to normal levels.

In our study EPO rose with increasing degree of acidosis. A negative correlation between cord blood EPO concentration and umbilical arterial blood pH has also been observed by Ekardt et al (25), Ruth et al (15), Termo et al (12) in infants of mothers with pre-eclampsia, and Widness et al (14) in infants of diabetic mothers, this was also observed by Maier et al (24) where they measured EPO by ELISA.

In our study EPO concentrations in umbilical venous blood were not related to umbilical arterial PCO2. Widness et al (8) demonstrated that prolonged hypoxia in fetal sheep was followed by metabolic acidosis and increased serum EPO. On the other hand they found that lactic acidosis without hypoxia did not stimulate Epo production.

Some of our infants with acidosis and/or other clinical signs of fetal distress had EPO concentrations within the range of low-risk group, which may reflect a delay between the start of a hypoxic stimulus and the increase of EPO in circulating blood. This delay suggests that there are no stores for rapid EPO release.

In adult rabbits, EPO began to rise 3 hours after exposure to hypobaric hypoxia (27). In fetal sheep, a progression increase in plasma EPO could be measured during the fourth hour of maintained hypoxemia. Ekardt et al (6) found the first significant increase 1.5 hours after exposure of healthy human adults to acute hypobaric hypoxia. If these results can be applied to human fetuses, elevated EPO levels in umbilical venous blood combined with arterial blood pH could help to distinguish between acute and prolonged fetal hypoxia.

Ruth et al (28) reported that a high cord blood EPO levels after normal pregnancy indicates an increased risk of cerebral palsy or death. We found that higher EPO concentration in infants who needed special care compared to those who did not. Preterm infants did not differ from term infants with regard to EPO concentrations, this finding can not be explained by the fact that preterm infants are more likely to need special care than term infants.

There is no standard diagnosis of fetal hypoxia. Sykes et al (29) noted that Apgar score does not usually reflect the degree of acidosis at birth. We found that abnormal FHR pattern during the last 40 minutes before delivery and Apgar score at 5 minutes were not closely correlated to EPO concentration in umbilical venous blood. Therefore, we concluded that abnormal FHR pattern may signal imminent fetal risk but does not indicate fetal hypoxia. We can identify infants with acidosis at birth, but further investigations are needed to determine the exact relation between EPO and the duration of fetal hypoxia in human fetuses.

Table 1 - Comparison between EPO, pH, and blood gases in low-risk group and other groups


 Mean ±SD









1 2 t p7.30(± 0.06) 7.23(± 0.04) 4.2 <0.0000128.04( 13.7) 73.84( 29.6) 6.7 <0.0000121.17(± 2.9) 13.17(± 3.1) 9.12 <0.00001  33.78(± 7.5) 51.38( ±5.7) 9.17 <0.00001  
1 3 t p7.30( ±0.06) 7.30(± 0.05) 0.15 NS    28.04(± 13.7) 43.92(± 26.1) 2.4 <0.0221.17(± 2.9) 13.8(± 4.9) 5.5 <0.0003    37.78( ±7.5) 41.91(± 9.6) 2.8 <0.008
  1 4 t p7.30(± 0.06) 7.13(± 0.05) 8.8 <0.00001  28.04( ±13.7) 109.57(± 86.4) 4.4 <0.0006  21.17(± 2.9) 10.3(± 0.5) 15.8 <0.00001    33.78(± 5.6) 53.5( ±5.6) 9.4 <0.00001

NS=not significant


Table 2 - Cord blood EPO concentrations in current study and previous reports


Erythropoietin concentration mU/mlNo.ValueExplanation
Eckardt et al (26)7533.6/17-56Median/95%CI
Ruth et al (16)17 3723/18-27 40/33-47Median/95%CI-no labor Median/95%CI-labor
Ruth et al (17)920/6-39Median/range
Teramo et al (13)1719.7/9.8-50.3Median/range
Widness et al (14)1632/7-64Median/range
Widness et al (15)2323/9.8-50.3Median/range
Widness et al (23)18 2326±10 46±34Mean±SD No labor Mean±SD-Labor
Maier et al (25)1925.1/11.9-55.3Median/range
Abdel Aziz et al(present study)2329/5-58Median/range

CI= Confidence Interval

Table 3 - Erythropoietin Concentration in umbilical venous blood in the 4 studied groups


GroupNo.Median mU/mlRange  mU/mlMean mU/ml 
Complicated pregnancy256830-16073.84 
Abnormal FHR134211-10043.92
Umbilical acidosis197625-300109.5



1- Apgar V. A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Anag 1953;32:260-6.

2- Hammacher K. Die Semiquantitative. Auswertung von Kardiotokogrammen anhand eines CTG-Scores. Stuttgart: George Thieme Verlage, 1975;240-4 (Dudenhausen JW, Saling E.Schmidt E. Perinatale Medizin; vol6)

3- Heltzman RB, Banzhaf WC, Silver RK, Hageman J.R. Perinatal management of meconium staining of the amniotic fluid. Clin. Perinatol 1989; 16:825-38.

4- Dennis J., Johnson A, Mutch L, Yudkin P, Johnson P. Acid base status at birth and neurodevelopmental outcome at four and one-half years. Am J Obstet Gynecol 1989; 161:213-20.

5- Dijxhoorn MJ, Visser GHA, Fidler VJ, Touwen BCL, Huisjes HJ. Apgar score, meconium and acidaemia at birth in relation to neonatal neurological morbidity in term infants. Br. J. Obstet Gynecol 1986; 93:217-22

6- Ekardt KU, Boutellier U., Kurtz A, Schopen M, Koller EA, Bauer C. Rate of erythropoietin in humans in response to acute hypobaric hypoxia. J.Appl Physiol 1989;66:1785-8.

7- Clemons GK, Fitzsimmons SL, DeManincor D. Immuno-reactive erythropoietin concentrations in fetal and neonatal rats and the effect of hypoxia. Blood 1986; 68:892-8.

8- Widness JA, Teramo KA, Clemons GK, et al. Temporal response of immunoreactive erythropoietin to acute hypoxemia in fetal sheep. Pediatr Res 1986; 20:15-19.

9- Zanjani ED, Mann LI, Burlington H, Gordon AS, Wasserman LR. Evidence for a physiologic role of erythropoietin in fetal erythropoiesis. Blood 1974; 44:285-90.

10- Cotes PM, Canning CE, Lind T. Changes in serum immunoreactive erythropoietin during the menstrual cycle and normal pregnancy. Br.J Obstet Gynaecol 1983;90:30-31.

11- Thomas RM, Canning CE, Cotes PM et al. Erythropoietin and cord blood haemoglobin in the regulation of human fetal erythropoiesis. Br.J Obstet Gynaecol 1983;90:795-800.

12- Teramo KA, Widness JA, Clemons GK, Voutilainen P, McKinaly S, Schwartz R. Amnoitic fluid erythropoietin correlates with umbilical plasma erhthropoietin in normal and abnormal pregnancy. Obstet Gynecol 1987;69:710-16.

13- Widness JA, Susa JB, Garcia JF, et al. Increased erythropoiesis and elevated erythropoietin in infants born to diabetic mothers and in hyperinsulinemic rhesus fetuses. J. Clin Invest 1981; 67:637-42.

14- Widness JA, Teramo KA, Clemons GK, et al. Direct relationship of antepartum glucose control and fetal erythropoietin in human type I (insulin-dependent) diabetic pregnancy. Diabetologia 1990; 33:378-83.

15- Ruth V; Fyhrquist F, Clemons G, Raivio Ko. Cord plasma vasopressin, erythropoietin, and hypoxanthine as indices of asphyxia at birth. Pediatr Res 1988;24:490-4.

16- Ruth V, Widness JA, Clemons G, Raivio KO. Postnatal changes in serum immunoreactive erythropoietin in relation to hypoxia before and after birth. J. Pediatr 1990;116:950-4.

17- Ruth V, Clemons G and Raivio KO. Cord plasma vasopressin, EPO and hypoxia as indices of Asphyxia at birth. Pediatr Res. 1988;24:490-4.

18- Roslyn M, Cella ECP, Mary and David C. Erythropoietin ,cord blood Hb in the regulation of human fetal erythropoiesis. Br.J Obstet Gynaecol 1983;90:745-50.

19- Keen DV, Pears RG. Weight , length and head circumference curves for boys and girls of between 20 and 42 weeks'gestation. Arch. Dis Child 1988;63:1170-2

20- Gregory JB. Principles of immunochemical techniques in Fundamental of Clinical Chemistry 1987, 3rd Ed. P.144

21- MacDonald D, Grant A, Sheridan-Pereira M, Boylan P, Chalmer I. The Dublin randomized controlled trial of intrapartum fetal heart rate monitoring. Am J Obstet Gynecol. 1985;152:425-539.

22- Widness JA, Clemons GK, Garcia JF, Shwartz R. Increased immunoreactive erythropoietin in cord serum after labor. Am J Obstet Gynecol 1984;148:194-7.

23- Forestier F, Daffos F, Catherine N, Renard M, Andreux JP. Developmental hematopoiesis in normal human fetal blood. Blood 1991;77:2360-3

24- Maier RF, Bohme K, Dudenhavsen JW, Obladen M. Cord blood erythropoietin in relation to different markers of fetal hypoxia. Obstet Gynecol 1993;81(4):575-580.

25- Ekardt KU, Hartmann W, Vetter U, Pohlandt F, Burghardt R, Kurtz A. Serum immunoreactive erythropoietin of childern in health and disease. Eur.J Pediatr 1990; 149:459-64.

26- Widness JA, Philipps AF, Clemons GK. Erythropoietin levels and erythropoiesis at birth with Potter syndrome. J Pediatr 1990;117:155-8.

27- McMullin MF, Lappin TRJ, Elder GE, Savage GA, Bridges JM. Erythropoietic response to hypobaric hypoxia in rabbits. Biomed Biochem Acta 1988;47:523-27.

28- Ruth V, Autti-Ramo I, Granstrom ML, Korkman M, Raivio KO. Prediction of perinatal brain damage by cord plasma vasopressin, erythropoietin, and hypoxanthine values. J Pediatr 1988;13:880-5.

29- Sykes GS, Molloy Pm, Johnson P, et al. Do Apgar scores indicate asphyxia? Lancet 1982;I:494-6.