Dr M Holzmann, Department of Obstetrics and Gynaecology, Karolinska University Hospital, 171 76 Stockholm, Sweden. Email email@example.com
Please cite this paper as: Holzmann M, Cnattingius S, Nordström L. Lactate production as a response to intrapartum hypoxia in the growth-restricted fetus. BJOG 2012;119:1265–1269.
Objective To analyse whether the increase in lactate in response to intrapartum hypoxia differs between small- (SGA), appropriate- (AGA) and large-for-gestational-age (LGA) fetuses.
Design Observational cohort study.
Setting Ten obstetric units in Sweden.
Population A cohort of 1496 women.
Methods A secondary analysis of a randomised controlled trial, in which 1496 women with fetal heart rate abnormalities, indicating fetal scalp blood sampling, were randomised to lactate analyses. After delivery, the neonates were divided according to birthweight for gestational age into SGA, AGA and LGA groups.
Main outcome measure Lactate concentration in fetal scalp blood.
Secondary outcome measures Acid–base balance in cord artery blood and Apgar score <7 at 5 minutes.
Results Median lactate concentrations in the SGA, AGA and LGA groups were 3.8, 3.0 and 2.2 mmol/l, respectively (SGA versus AGA, P = 0.017; LGA versus AGA, P = 0.009). In the subgroups with scalp lactate >4.8 mmol/l (lactacidaemia), the corresponding median (range) values were 6.2 (4.9–14.6), 5.9 (4.9–15.9) and 5.7 mmol/l (5.0–7.9 mmol/l), respectively (no significant differences between the groups). The proportions of neonates with cord artery pH < 7.00, metabolic acidaemia or Apgar score <7 at 5 minutes were similar in all weight groups.
Conclusion SGA fetuses with fetal heart rate abnormalities have the same ability to produce lactate as a response to intrapartum hypoxia as AGA and LGA fetuses. The risk of a poor outcome associated with high lactate concentration is the same in SGA, AGA and LGA fetuses. Scalp blood lactate analysis is therefore a reliable method for intrapartum fetal surveillance of suspected growth-restricted fetuses scheduled for vaginal delivery at ≥34 weeks of gestation.
For more than 30 years, fetal scalp blood sampling (FBS) with lactate analysis has been evaluated for intrapartum fetal surveillance. With the introduction of handheld microvolume devices, it has become a clinically attractive alternative to pH analysis. Studies have suggested that lactate concentration in FBS is better than pH in predicting neonatal morbidity.1,2 Two randomised controlled trials (RCTs), comparing the analysis of scalp blood lactate with pH, found lactate to be just as reliable as pH in the prevention of birth acidaemia and more favourable in clinical practice because of less sampling/analysis failure and reduced time from sampling to the results of the analysis,3,4 At present all obstetric units in Sweden use FBS in intrapartum fetal surveillance and the vast majority (41 of 46 units) use lactate analysis.5 However, the method is still not widely spread internationally.
In normal glycolysis, glucose is metabolised to pyruvate and, through the ‘citric acid cycle’, broken down further to CO2 and H2O as waste products. In a normally oxygenated situation, there is a steady state between pyruvate and lactate (1:10). In cases with glucose supplementation, with a concomitant increase in glucose concentration, both pyruvate and lactate will increase because of the steady state in spite of aerobic glycolysis. Lactate is also produced in the placenta and this lactate is probably used as energy for the fetus in an aerobic environment.6 In an anaerobic environment, glucose metabolism produces lactate and hydrogen ions as end products.7 Experimental animal data have shown that lactate is an earlier marker than pH in the hypoxic process and increases in subcutaneous tissue before pH decreases.8
Growth-restricted fetuses have poor glycogen reserves, and it is speculated that, in pregnancies with growth-restricted fetuses, there may be a lack of substrate to produce lactate, and thereby less increase in lactate concentrations under anaerobic circumstances. If so, the degree of lactacidaemia in growth-restricted fetuses might not reflect the seriousness of the hypoxia. To our knowledge, it has not yet been studied whether there is a difference between appropriately grown and growth-restricted fetuses in terms of fetal lactate production as a response to intrapartum hypoxia. The aim of this study was to compare scalp blood lactate concentrations between small- (SGA), appropriate- (AGA) and large-for-gestational-age (LGA) fetuses in cases in which all had non-reassuring fetal heart rate tracings indicating FBS during the course of labour.
This is a secondary analysis of an RCT comparing pH analysis with lactate analysis of fetal scalp blood during labour. A total of 2992 women with a cardiotocography (CTG) trace indicating FBS were randomised to receive either lactate or pH analysis. All women had given informed consent before randomisation. The inclusion criteria were as follows: singleton pregnancy, cephalic presentation, gestational age ≥34 weeks and a non-reassuring fetal heart rate trace considered as an indication for FBS by the clinician in charge. Details of the trial (ISRTN 1606064) have been published elsewhere.4
In this study, we used data from this RCT to perform an observational cohort study, in which intrapartum lactate determinations on fetal scalp blood were performed with prospectively collected data. We divided the fetuses into three subgroups according to birthweight for gestational age: SGA, AGA and LGA. SGA and LGA were defined as birthweights corrected for gestational age beyond ±2 standard deviations (SDs) (more than ±22% weight deviation) from the mean, according to the sex-specific Swedish fetal growth curve.9
Lactate concentrations were analysed with a microvolume test strip device (Lactate Pro; Arkray, Kyoto, Japan). Interpretation of the scalp blood analyses followed clinical guidelines: normal, <4.2 mmol/l; preacidaemia, 4.2–4.8 mmol/l; acidaemia, >4.8 mmol/l.1 In fetuses with preacidaemia, repeat sampling within 20–30 minutes is recommended and, in those with acidaemia, the clinician should consider delivery.1
For statistical analysis, we used statistica for Windows™, version 10 (StatSoft, Tulsa, OK, USA). We used nonparametric methods considering the study population (fetuses in which blood sampling during labour was considered to be necessary) to be non-normally distributed. Data are presented as medians, ranges and percentiles. Differences in continuous variables between the birthweight groups were tested with the Kruskal–Wallis test and the Mann–Whitney U-test. The chi-square test and Fisher’s exact test were used for proportion comparisons, and correlation between two variables was tested with the Spearman rank test. P < 0.05 was considered to be significant.
There were 1496 women randomised to lactate determinations. With a failure rate of 1.2% and several other reasons for no collection of blood, such as rapid delivery, reassuring CTG and withdrawn consent, there were 1354 women with available data on scalp blood lactate concentration.4 The clinical details of the women are given in Table 1. In this population of pregnancies with CTG abnormalities indicating FBS, the proportions of SGA and LGA fetuses were 5.3% and 3.3%, respectively. Compared with mothers of AGA and SGA infants, mothers of LGA infants were more often multiparous, had a shorter gestational age and had more often a caesarean delivery.
Table 1. Clinical data of the study population (numbers [%], medians [range])
SGA (n = 72)
AGA (n = 1237)*
LGA (n = 45)
AGA, appropriate for gestational age, LGA, large for gestational age; SGA, small for gestational age.
Maternal age (years)
Gestational age (days)
Mode of delivery
Birth weight (g)
Median scalp lactate concentration in the SGA (n = 72), AGA (n = 1237) and LGA (n = 45) groups were 3.8, 3.0 and 2.2 mmol/l, respectively. The median lactate concentration in the SGA group was significantly higher than that in the AGA group (P = 0.017), and the median lactate concentration in the LGA group was significantly lower than that in the AGA group (P = 0.009). The distributions of lactate concentrations in the three groups are shown in Figure 1. In the subgroups with scalp lactate >4.8 mmol/l (SGA, n = 19; AGA, n = 247; LGA, n = 8), the median (range) values were 6.2 (4.9–14.6) for SGA, 5.9 (4.9–15.9) for AGA and 5.7 mmol/l (5.0–7.9 mmol/l) for LGA. In cases with lactacidaemia, there were no significant differences between the SGA and AGA groups, the LGA and AGA groups or the SGA and LGA groups (P = 0.56, P = 0.89 and P = 0.81, respectively). These distributions are shown in Figure 2.
The proportions of neonates with cord artery blood pH (CA-pH) < 7.00, metabolic acidosis (CA-pH < 7.05 and base deficit in blood [BDblood] >12) or Apgar score < 7 at 5 minutes did not differ between the birthweight groups, either when comparing all cases or in the subanalysis of those with intrapartum acidaemia (scalp lactate > 4.8 mmol/l) (Table 2).
Table 2. Neonatal outcome in the birthweight groups
SGA 72 n (%)
AGA* 1237 n (%)
LGA 45 n (%)
SGA vs AGA
LGA vs AGA
CA-pH < 7.10
CA-pH < 7.0
Apgar 5 minutes <7
Number with scalp lactate >4.8****
19 n (%)
247 n (%)
8 n (%)
SGA vs AGA
LGA vs AGA
AGA, appropriate for gestational age; CA-pH, cord artery pH; LGA, large for gestational age; SGA, small for gestational age.
**Numbers with full acid–base data in the weight groups were 65, 1149 and 43, respectively.
****Numbers with full acid–base data in the weight groups were 19, 231 and 8, respectively.
*****Fisher’s exact test.
******pH < 7.05 and base deficit >12 in cord artery.
CA-pH < 7.10
CA-pH < 7.0
Apgar 5 minutes <7
The correlation between the degree of birthweight deviation (in per cent) and lactate concentration in FBS was specifically tested in the SGA group. There was a weak, but significant, negative correlation (ρ = −0.24, P < 0.05) (Figure 3).
The main finding of this study was that the growth-restricted fetuses showing fetal heart rate abnormalities were, in spite of probably smaller energy reserves, capable of increasing their lactate concentration adequately as a response to intrapartum fetal distress. The proportion of depressed or acidaemic neonates was not larger in the SGA group than in the AGA or LGA groups.
When comparing all cases, the median lactate concentration was significantly higher in the SGA group than in the AGA group. This is in accordance with earlier studies of fetal lactate concentrations in the antepartum period and at delivery.10,11 In animal experiments, intrauterine growth-restricted (IUGR) fetuses have a relative hypoglycaemia and diminished insulin production. Insulin should ideally be the facilitator for hepatic glucose uptake.12 Human IUGR fetuses have significantly lower concentrations of glucose and insulin in cord blood compared with fetuses with normal growth.13 This is a probable explanation for the depleted glycogen depots in the myocardium and liver of IUGR fetuses, which are important sources for glucose in fetal metabolism. Lactate concentration at cordocentesis and in cord blood at delivery has been shown to be higher in SGA relative to AGA fetuses, and these findings are interpreted as being caused by mild chronic tissue hypoxia and polycythaemia.10,11 However, the lactate concentrations in the SGA fetuses were within the normal range in both quoted studies, and these findings cannot be transcribed to an acute hypoxic situation during labour, where lactate might reach four to five times higher concentrations than in the antenatal period.14
It is not clear why LGA fetuses had significantly lower concentrations of lactate than AGA fetuses, a finding opposite to that observed previously in routinely collected cord blood at delivery on a larger dataset.11 One possible explanation for the findings in the present series could be the selection of cases. If LGA fetuses more often had prolonged labour, this might have influenced the clinician to perform FBS with less severe CTG abnormalities. Unfortunately, we do not have any data on the duration of labour and CTG traces to explore this hypothesis further. Alternatively, LGA fetuses might be more resistant to CTG abnormalities before anaerobic metabolism begins or may start lactate production at a later stage in the hypoxic process. Future research should explore these hypotheses.
We did not find any difference in lactate concentrations between SGA, AGA and LGA fetuses when looking into the subgroups with intrapartum acidaemia, i.e. scalp blood lactate >4.8 mmol/l. The SGA fetuses reached lactate concentrations as high as the AGA and LGA fetuses. We were concerned whether it was only the mildly growth-restricted fetuses that could respond to hypoxia with a marked increase in lactate concentration. However, when analysing the SGA group specifically, we found a significant, although weak, negative correlation between fetal weight deviation and lactate concentration (ρ = −0.24; P < 0.05), with some of the most growth-restricted fetuses having the highest lactate concentrations.
The weakness of this study is that it is a secondary analysis of data from an RCT, which was designed to compare fetal scalp blood lactate and pH analyses in the prevention of birth acidaemia. Thus, the study was not specifically designed to address the issue of lactate production in IUGR fetuses in response to intrapartum hypoxia. We did not have full data on acid–base balance in cord artery blood at delivery in all cases. However, this did not influence our conclusions, as the primary outcome was scalp blood lactate. As we lacked information about the duration of labour, it was not possible to investigate whether fetal growth influenced the time at which signs of fetal distress were noticed or when lactate concentration was increased.
In addition, we used information about birthweight, gestational age and sex to classify groups as SGA, AGA or LGA. It is necessary to emphasise that SGA and IUGR are not synonymous concepts. IUGR is the functional definition referring to a fetus that, for placental/maternal or fetal reasons, is unable to achieve its predestined weight, whereas SGA is a statistical definition based on birthweight, gestational age and sex. We used the Swedish definition of SGA, which is >2 SDs below the mean of estimated fetal growth, according to the sex-specific ultrasonically estimated fetal weight curves, routinely used in Sweden. As the proportion of acidaemic neonates was not larger among these SGA infants (the lowest 2.5th percentile) than among AGA infants, our findings should also be valid using a more common definition of SGA, i.e. less than the 10th percentile.
Women admitted in labour often have a vigorous fetus demonstrated in a reactive CTG admission test. Some of these cases might deteriorate during the course of labour and develop lactacidaemia in fetal scalp blood, which indicates intervention, usually operative delivery. We have shown previously that fetuses with severe intrapartum acidaemia (fetal scalp lactate > 6.6 mmol/l) still have a window of opportunity for preventive measures.14 In the present study, the SGA fetuses were not more often depressed or acidaemic at birth than fetuses in the other weight groups. Thus, lactate analysis in FBS is a well-functioning surveillance method in the labour of growth-restricted fetuses. If the ability to increase the lactate concentration adequately holds true for the severely growth-restricted fetus that is never allowed to enter a vaginal delivery or in preterm growth-restricted fetuses (<34 weeks of gestation) cannot be concluded from this study.
In summary, we have shown that growth-restricted fetuses entering a vaginal delivery can respond adequately to hypoxia with an increase in lactate concentration in a similar manner to normally grown fetuses. We conclude that the measurement of fetal scalp blood lactate is a reliable diagnostic test for intrapartum hypoxia, regardless of fetal growth status.
Disclosure of interests
The authors have no conflict of interest with regard to the material presented in this article.
Contribution to authorship
LN initiated the study. MH and LN performed the data analyses. MH drafted the manuscript. All three authors contributed to the revision of the manuscript and approved the final version.
Details of ethics approval
This study was approved by the regional ethics committee of Karolinska Institutet, Stockholm, Sweden, reference number 02-109.