Risk factors for hypoxic–ischemic encephalopathy in cases of severe acidosis: A case–control study

Abstract

Introduction

The aim of the study was to identify the obstetric risk factors for hypoxic–ischemic encephalopathy (HIE) in infants with asphyxia at birth.

Material and methods

This multicenter case–control study covered the 5‐year period from 2014 through 2018 and included newborns ≥36 weeks of gestation with an umbilical pH at birth ≤7.0. Cases were newborns who developed moderate or severe HIE; they were matched with controls with pH ≤7.0 at birth over the same period without moderate or severe HIE. The factors studied were maternal, gestational, intrapartum, delivery‐related, and neonatal characteristics. A multivariable analysis was performed to study the maternal, obstetric, and neonatal factors independently associated with moderate or severe HIE.

Results

Our review of the records identified 41 cases and 98 controls. Compared with controls, children with moderate or severe HIE had a lower 5‐min Apgar score, lower umbilical artery pH, and higher cord lactate levels at birth and at 1 h of life. Obstetric factors associated with moderate or severe HIE were the occurrence of an acute event (adjusted odds ratio [aOR] 6.4; 95% confidence interval [CI] 1.8–22.5), maternal fever (aOR 3.5; 95% CI 1.0–11.9), and thick meconium during labor (aOR 2.9; 95% CI 1.0–8.6).

Conclusions

HIE is associated with a lower 5‐min Apgar score and with the severity of acidosis at birth and at 1 h of life. In newborns with a pH <7.0 at birth, the occurrence of an acute obstetric event, maternal fever, and thick meconium are independent factors associated with moderate or severe HIE.

Abbreviations

FHR

Fetal heart rate

HIE

Hypoxic–ischemic encephalopathy

Key message.

Acute obstetric events may be risk factors for hypoxic–ischemic encephalopathy, independent of the umbilical pH value.

1. INTRODUCTION

Intrapartum asphyxia results from the impairment of uteroplacental gas exchanges leading to dysfunctional cell metabolism and then metabolic acidosis. It is defined by an umbilical artery pH ≤7.0 and a base deficit ≥12 mmol/L at birth. ¹ , ² Intrapartum asphyxia is associated with severe short‐term neonatal outcomes: multiorgan failure, hypoxic–ischemic encephalopathy (HIE), and death. ¹ , ³ , ⁴

HIE stages have been classified by clinical criteria, in accordance with Sarnat and Sarnat. ⁵ Its prevalence is estimated at between 1.5 and 2.5 per 1000 livebirths. ⁴ , ⁶ Among infants with moderate and severe HIE (stages 2 and 3), 30% die and 30% of the survivors have major neurodevelopmental disabilities. ⁷

Several studies in the general population have identified risk factors for intrapartum asphyxia and HIE, including in particular intrapartum discharge of thick meconium, fetal heart rate (FHR) anomalies, chorioamnionitis, and the occurrence of an acute obstetric event. ³ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ Nonetheless, only about 10–20% of children born with intrapartum asphyxia develop HIE or die, and why some children with severe acidosis develop HIE whereas others do not remains unknown. ¹⁹ From a physiological perspective, it is possible that some contributing factors increase the risk of HIE in cases of neonatal asphyxia. For example, an acute obstetric event leading to asphyxia without the possibility of fetal adaptation may be associated with the occurrence of HIE. Moreover, some fetuses that are known to be more vulnerable (post‐term pregnancy, small‐for‐gestational‐age fetus) may be more likely to develop HIE. Similarly, other maternal and obstetric factors such as age, smoking, diabetes, and fever during labor might affect fetal brain adaptation to acidosis. The objective of our study was to identify risk factors for HIE in newborns with an umbilical artery pH ≤7.0.

2. MATERIAL AND METHODS

This retrospective case–control study took place in Trousseau hospital in Paris, which is a tertiary referral center. In addition to the cases of HIE that occurred in infants born at Trousseau, all cases of HIE referred there for controlled hypothermia from four other Paris hospitals during calendar years 2014–2018 were included.

Inclusion criteria were an umbilical artery pH ≤7.0 at birth, a gestational age ≥36 weeks, and a moderate or severe HIE (stage 2 or 3 according to Sarnat and Sarnat ⁵ ) managed by controlled hypothermia. Exclusion criteria were missing umbilical artery pH data, chromosomal abnormality, and neurologic trauma (cerebral hemorrhage, spinal cord injuries). Control children were those with an umbilical artery pH ≤7.0 who did not have moderate or severe HIE born in Trousseau hospital during calendar years 2014–2017.

The data were extracted from medical records in the maternity wards and neonatal units and included four categories of factors. The first comprised maternal variables: age, ethnicity, health insurance, body mass index, medical and surgical history (diabetes, hypertension, chronic diseases), parity, and obstetric history (cesarean delivery, in utero fetal death, gestational hypertension, and history of small‐for‐gestational‐age fetus). The gestational factors were gestational diabetes, gestational hypertension, preeclampsia, intrahepatic cholestasis of pregnancy, and small‐for‐gestational‐age fetus (estimated weight <10th percentile for gestational age). The intrapartum factors were induction of labor, maternal fever during labor (defined by a temperature ≥38 °C, measured in the ear and systematically recorded every 2 h during labor), thick meconium (defined by dark green viscous fluid with “pea soup” characteristics), use of oxytocin, and occurrence of an acute event (placental abruption, uterine rupture, cord prolapse, bleeding from vasa previa, shoulder dystocia, fetomaternal hemorrhage, and amniotic fluid embolism). The last 2 h before birth of all FHR traces were reviewed and classified as either normal, suspicious, or pathological according to the International Federation of Gynecology and Obstetrics (FIGO) 2015 classification. ²⁰ Finally, the delivery factors considered included cesarean delivery (emergency or planned), decision‐to‐delivery interval for emergency cesarean deliveries, spontaneous or operative vaginal delivery, duration of pushing (expulsive efforts), and FHR during pushing. We also recorded neonatal characteristics: umbilical artery pH and lactate level at birth, 5‐min Apgar score, birthweight, sex, and pH and lactate level measured at 1 h of life.

2.1. Statistical analyses

Characteristics of the two groups were first compared using a univariate analysis. Chi‐squared and Fisher’s exact tests were used to compare categorical variables, and Student’s t test was used to compare the continuous variables. The statistical significance threshold was set as a p value <0.05.

A multivariable analysis was used to study the independent factors associated with HIE. The first model investigated the risk factors for HIE in the overall population. A second model studied the association between labor characteristics and HIE in a subpopulation after exclusion of the women with cesarean deliveries before labor. The variables included in the models were those associated with HIE in the univariate analysis with a p < 0.2, maternal age and parity. Mode of delivery was not included in the multivariable analysis models because birth by cesarean or operative vaginal delivery is more frequent in severe obstetric situations; it is therefore an intermediate factor.

The population source of the cases was Trousseau hospital and four other centers referring cases to it for controlled hypothermia. All control patients were born in Trousseau hospital. Because that could have biased the comparison of women’s characteristics between cases and controls, a sensitivity analysis was performed on a population restricted to infants born at Trousseau hospital to analyze the maternal characteristics associated with HIE.

Statistical analyses were performed with Stata 13 software (TX, USA).

2.2. Ethical approval

The study received a favorable opinion from the Committee for Ethics and Research in Obstetrics and Gynecology on March 9, 2021 (n°CEROG 2020‐OBS‐1203).

3. RESULTS

During the study period, 73 infants admitted to Trousseau hospital’s neonatal intensive care unit were treated with therapeutic hypothermia for moderate or severe HIE. After exclusion of 15 with missing data for umbilical artery pH at birth and 17 with umbilical pH >7.0, 41 neonates with an umbilical pH ≤7.0 made up the case group (HIE group). Among them, 28 (68.3%, 95% confidence interval [CI] 54.1–82.5) had moderate HIE and ten (24.4%, 95% CI 11.3–37.5) had severe HIE.

Among the 14 001 children born in Trousseau hospital between 2014 and 2017, 115 had an umbilical pH artery ≤7.0 (0.82%; 95% CI 0.68–0.99). Among them, 98 did not have moderate or severe HIE and constituted the control group (Figure 1).

FIGURE 1.

Study flow chart. HIE, hypoxic–ischemic encephalopathy; NICU, neonatal intensive care unit

The two groups did not differ for the women’s individual characteristics or obstetric history (Table 1). The sensitivity analysis restricted to infants born in Trousseau hospital yielded similar results (Table 2).

TABLE 1.

Comparison of maternal and obstetric characteristics between the case (stage 2 and 3 hypoxic–ischemic encephalopathy) and control groups

Characteristics	Cases n = 41 (%)	Controls n = 98 (%)	p
Age (years, mean ± SD)	31.4 ± 5.3	32.3 ± 9.1	0.43
Age ≥35 years	15 (36.6)	41 (41.8)	0.57
BMI (kg/m2) ≥30	13 (31.7)	23 (23.5)	0.31
Origin
European	20 (50.0)	46 (63.0)	0.56
North Africa	7 (17.5)	10 (13.7)
Sub‐Saharan Africa	11 (27.5)	15 (20.6)
Other	2 (5.0)	2 (2.7)
Missing data	1	25
Health insurance
Has health insurance	32 (78.0)	82 (83.7)	0.36
Has no health insurance	9 (22.0)	15(15.3)
Missing data	0	1
Previous cesarean delivery	9 (22.0)	24 (24.5)	0.75
Hypertension before pregnancy	0 (0.0)	4 (4.1)	0.32
Diabetes before pregnancy	1 (0.6)	1 (1.0)	0.50
Obstetric history
Previous SGA, preeclampsia	0 (0.0)	5 (5.1)	0.32
Previous in utero fetal death	1 (2.4)	2 (2.0)	1
Nulliparity	26 (63.4)	56 (57.1)	0.49
Multiple pregnancy	4 (9.8)	6 (6.1)	0.48
Gestational hypertension or preeclampsia	3 (7.3)	18 (18.4)	0.10
SGA	3 (7.3)	15 (15.3)	0.20
Gestational diabetes	9 (22)	12 (12.2)	0.15
Intrahepatic cholestasis of pregnancy	3 (7.3)	0 (0.0)	0.02
Gestational age of birth
37–41 weeks	30 (73.2)	81 (82.7)	0.46
<37 weeks	2 (4.9)	3 (3.1)
>41 weeks	9 (22.0)	14 (14.3)
Vaginal delivery	14 (34.2)	42 (42.9)	0.33
Instrumental vaginal delivery	13 (31.7)	23 (23.5)	0.31
Cesarean section	27 (65.9)	56 (57.1)	0.34
Planned cesarean section	2 (4.9)	13 (13.3)	0.23
Emergency cesarean section before labor	7 (17.1)	4 (4.1)	0.02
Emergency cesarean section during labor	18 (43.9)	39 (39.8)	0.65
Temperature ≥38 °C at the time of delivery	10 (24.4)	11 (11.2)	0.05
Breech presentation	4 (9.8)	11 (11.2)	1
Thick meconium	15 (37.5)	15 (15.6)	0.005
Identified acute event
Total	14 (34.1)	9 (9.2)	0.001
Placenta abruption	1 (2.4)	1 (1.0)
Cord prolapse	1 (2.4)	4 (4.1)
Uterine rupture	8 (19.5)	4 (4.1)
Shoulder dystocia	3 (7.3)	0 (0.0)
Amniotic fluid embolism	1 (2.4)	0 (0.0)

Abbreviations: BMI, body mass index; SD, standard deviation; SGA, small for gestational age.

TABLE 2.

Comparison of the maternal characteristics of women in the case (stage 2 and 3 hypoxic–ischemic encephalopathy) and control groups who gave birth at Trousseau hospital (sensitivity analysis)

Characteristics	HIE n = 17 (%)	No HIE n = 98 (%)	p
Age (years, mean ± SD)	31.3 ± 18.1	32.3 ± 9.1	0.46
Age ≥35 years	6 (35.3)	41 (41.8)	0.79
BMI (kg/m2) ≥30	3 (17.7)	23 (23.5)	0.76
Origin
European	9 (52.9)	46 (63.0)	0.76
North Africa	2 (11.8)	10 (13.7)
Sub‐Saharan Africa	5 (29.4)	15 (20.6)
Other	1 (5.9)	2 (2.7)
Missing data	0	25
Health insurance
Has health insurance	16 (94.1)	93 (94.9)	0.74
Does not have health insurance	1 (5.9)	4 (4.1)
Missing data	0	1
Previous cesarean delivery	4 (23.5)	24 (24.5)	0.93
Hypertension before pregnancy	0 (0.0)	4 (4.1)	1
Diabetes before pregnancy	0 (0.0)	1 (1.0)	1
Obstetric history
Previous SGA, preeclampsia	0 (0.0)	5 (5.1)	1
Previous in utero fetal death	1 (5.9)	2 (2.0)	0.38

Abbreviations: BMI, body mass index; HIE, hypoxic–ischemic encephalopathy; SD, standard deviation; SGA, small for gestational age.

Rates of preexisting or gestational diabetes, preexisting or gestational hypertension, or preeclampsia during pregnancy were also similar between the two groups (Table 1). There were significantly more diagnoses of intrahepatic cholestasis of pregnancy in the HIE group (7.3 vs 0%, p = 0.02) as well as significantly fewer spontaneous vaginal deliveries and more emergency cesareans before labor.

Compared with the control group, the case group had a higher rate of maternal fever ≥38 °C during labor and of thick meconium discharge during labor. An acute event occurred in 14 (32.1%) women in the HIE group vs nine (9.2%) in the control group (p < 0.001) (Table 1).

Among the 23 acute events, 17 (73.9%) had pathological FHR according to the FIGO classification, ²⁰ and four (17.4%) had suspicious FHR. The one (4.3%) event with normal FHR was shoulder dystocia.

After exclusion of the women with a cesarean delivery before labor (emergency or planned), 32 (78.0%) and 81 (82.7%) women in the case and control groups, respectively, gave birth after labor. In this population, factors associated with HIE were the use of oxytocin and the mode of delivery (Table 3).

TABLE 3.

Comparison of labor characteristics between the case (stage 2 and 3 hypoxic–ischemic encephalopathy) and control groups

Characteristics	Cases n = 32 (%)	Controls n = 81 (%)	p
Onset of labor
Spontaneous	19 (59.4)	55 (67.9)	0.39
Induction	13 (40.6)	26 (32.1)
Oxytocin use	20 (62.5)	31 (38.3)	0.02
FHR 2 h before birth (FIGO)
Normal	3 (9.4)	6 (7.4)	0.48
Suspicious	6 (18.8)	9 (11.1)
Pathological	23 (71.9)	66 (81.5)
Mode of delivery
Vaginal delivery	1 (3.1)	23 (28.4)	0.01
Operative vaginal delivery	13 (40.6)	23 (28.4) a
Cesarean	18 (56.3)	39 (48.2) a

Abbreviations: FHR, fetal heart rate; FIGO, International Federation of Gynecology and Obstetrics.

^a Four cesarean sections for failure of operative delivery.

The case group also had lower umbilical artery pH levels and higher lactate levels at birth and 1 h after birth as well as a lower 5‐min Apgar score (Table 4).

TABLE 4.

Comparison of neonatal outcomes between the case (stage 2 and 3 hypoxic–ischemic encephalopathy) and control groups

Outcomes	Cases n = 41 (%)	Controls n = 98 (%)	p
Male	27 (65.9)	50 (51.0)	0.11
Weight (g, mean ± SD) SGA	3235.2 ± 510.0 5 (12.2)	3269.6 ± 479.2 6 (6.1)	0.72 0.30
Umbilical pH (mean ± SD)	6.87 ± 0.086	6.93 ± 0.053	<0.001
[6.90–7.00]	17 (41.5)	80 (81.6)	<0.001
[6.85–6.90]	9 (22.0)	8 (8.2)
[6.80–6.85]	12 (29.3)	10 (10.2)
<6.80	3 (7.3)	0 (0.0)
Umbilical lactate level (in mmol/L, mean ± SD)	11.2 ± 3.3	9.3 ± 1.7	<0.001
[5–10]	14 (34.2)	64 (68.8)	<0.001
[10–12]	15 (36.6)	22 (23.7)
≥12	12 (29.3)	7 (7.5)
Missing data	0	5
5‐min Apgar score (mean ± SD)	4.1 ± 2.4	8.6 ± 2.0	<0.001
7–10	7 (17.1)	83 (84.7)	<0.001
4–6	17 (41.5)	12 (12.2)
0–3	17 (41.5)	3 (3.1)
pH at 1 h (mean ± SD)	7.10 ± 0.16	7.22 ± 0.09	<0.001
≥7.20	8 (30.8)	60 (65.9)	<0.001
[7.10–7.20]	6 (23.1)	23 (25.3)
[7.00–7.10]	5 (19.2)	7 (7.7)
[6.00–7.00]	7 (26.9)	1 (1.1)
Missing data	15	7
Lactate level at 1 h (in mmol/L, mean ± SD)	13.5 ± 11.4	8.5 ± 7.9	<0.001
<5	1 (4.0)	12 (13.3)	<0.001
[5–10]	5 (20.0)	48 (53.3)
[10–15]	9 (36.0)	28 (31.1)
≥15	10 (40.0)	2 (2.2)
Missing data	16	8

Abbreviations: SD, standard deviation; SGA, small for gestational age.

Factors independently associated with HIE in the overall population were the occurrence of an acute event (adjusted odds ratio [aOR] 6.4; 95% CI 1.8–22.5), a maternal temperature ≥38 °C at delivery (aOR 3.5; 95% CI 1.0–11.9), thick meconium at delivery (aOR 2.9; 95% CI 1.0–8.6), and a low umbilical artery pH (Table 5). In the subpopulation of women who underwent labor, only a low umbilical artery pH and an acute event were still associated with HIE.

TABLE 5.

Predictive factors for hypoxic–ischemic encephalopathy: Multivariable analysis of all women and women in labor

Factors	Overall population		Women in labor
	Crude OR[95% CI]	Adjusted OR[95% CI]	Crude OR[95% CI]	Adjusted OR[95% CI]
Age	1.0 [0.9–1.0]	1.0 [0.9–1.1]	1.0 [0.9–1.0]	1.0 [0.9–1.1]
Nulliparity	1.3 [0.6–2.8]	2.3 [0.8–6.4]	0.9 [0.4–2.1]	1.7 [0.5–5.2]
Hypertension, preeclampsia, SGA	0.5 [0.2–1.3]	0.3 [0.1–1.4]	0.4 [0.1–1.2]	0.3 [0.1–1.8]
Gestational diabetes	2.0 [0.8–5.2]	3.3 [0.9–12.1]	1.8 [0.6–5.7]	1.8 [0.4–8.3]
Acute event	5.1 [2.0–13.1]	6.4 [1.8–22.5]	5.5 [2.0–15.2]	6.6 [1.6–27.2]
Thick meconium at delivery	3.2 [1.4–7.5]	2.9 [1.0–8.6]	2.7 [1.1–6.7]	2.9 [0.9–9.3]
Umbilical pH at birth
[6.90–7.00]	1	1		1
[6.85–6.90]	5.3 [1.8–15.7]	10.5 [2.9–38.3]		4.8 [1.1–20.3]
<6.85	7.1 [2.7–18.4]	8.6 [2.5–29.9]		6.2 [1.7–22.8]
Maternal temperature ≥38 °C at delivery	2.6 [1.0–6.7]	3.5 [1.0–11.9]	2.4 [0.9–6.6]	2.7 [0.7–10.1]
Oxytocin use			2.6 [1.2–6.5]	1.6 [0.5–4.7]

Abbreviations: CI, confidence interval; OR, odds ratio; SGA, small for gestational age.

4. DISCUSSION

In this case–control study, the risk factors for stage 2 and 3 HIE in newborns with an umbilical artery pH ≤7.0 were the occurrence of an acute event, a maternal temperature ≥38 °C at delivery, thick meconium, and the degree of acidosis at birth.

The selection of patients with umbilical artery pH ≤7.0 is one of the strengths of our study. The pH threshold at birth of 7.0 is frequently used to define infants with asphyxia at birth because the risk of short‐ and long‐term complications increases sharply below this cutoff. ⁴ , ¹⁹ Specifically, the risk of severe neonatal complications, including encephalopathy, convulsions, or death, is six times higher in newborns with a pH ≤7.0 than in those with a pH between 7.01 and 7.05. ¹⁹ A systematic review including ten studies and 386 newborns with a cord pH <7.0 found neurological morbidity among 5–30% of them. ⁴ Therefore, an umbilical artery pH <7.0 is one of the criteria set by the American College of Obstetricians and Gynecologists and the American Academy of Pediatrics to establish a causal relation between neonatal encephalopathy and intrapartum asphyxia. ² We did not include HIE stage 1 in this study as we considered moderate or severe HIE—the condition requiring management with controlled hypothermia—to be more clinically relevant.

Our study had some limitations. Cases and controls did not come from exactly the same population, which might have biased the comparison of women’s individual characteristics. However, the sensitivity analysis restricted to women who gave birth in the same hospital provided similar results. A second weakness was the power of the study. Although quite large for such a rare event, the sample size in our study provided a power of 90% to detect a difference of 2.5‐fold for a risk factor present in 20% in the control group and a power of only 60% for a 2‐fold difference. It was therefore possible that we failed to identify some factors associated less strongly with HIE.

After searching the MEDLINE and Embase databases with adequate medical subject heading terms, we found only one published study that examined the risk factors for HIE stage 2 and 3 in cases of pH ≤7.0. The design was similar to ours but included only half the number of cases with HIE stages 2 and 3; it reported an association between decreased FHR variability and HIE. ²¹ However, the authors did not include the umbilical artery pH value in the regression model used to study the risk factors for neonatal encephalopathy. As this factor is the main one associated with a worse outcome, the findings of that study may be only those associated with the severity of acidosis at birth.

Higher rates of metabolic acidosis (with lactate ≥12 mmol/L) were observed in the HIE group. This is consistent with the existing literature, as metabolic acidosis is a well‐known risk factor for HIE and neonatal complications. ¹ , ³ , ²² , ²³ Moreover, we found that the degree of acidosis is independently associated with HIE. In a retrospective cohort of 504 newborns with a pH <7.0, Kelly et al. ²² found a dose‐dependent association between acidosis and adverse neonatal outcomes, including HIE and later neurodevelopmental outcome. Goodwin et al. ²³ showed a correlation between the degree of umbilical artery acidemia and HIE and renal, cardiac, and pulmonary dysfunction in 129 term newborns with pH <7.0.

We also found that the newborns with moderate or severe HIE had lower arterial pH and higher lactate levels at 1 h of life. The severity of acidosis at 1 h of life is poorly documented. One previous retrospective study of 1690 infants showed worse prognosis for infants born with acidemia that persisted at 1 h of life. ²⁴ Furthermore, a prospective study including 50 infants with HIE found that a longer time to serum lactate normalization was related to the severity of encephalopathy on electroencephalogram and with seizure burden. ²⁵ Blood lactate, a hallmark of anaerobic metabolism, has been traditionally employed as a surrogate marker for tissue hypoxia and/or ischemia. Monitoring lactate could be useful for assessing HIE severity. ²⁶ A study seeking to correlate serum and brain lactate concentration during HIE reported a correlation between serum and cerebral lactate and severity of insult during perinatal asphyxia. ²⁷ In our study, a secondary increase in lactate or a delay in its decrease was associated with HIE severity.

An acute event was the main risk factor for moderate to severe HIE. This finding has also been observed among the general population of women giving birth, with ORs ranging from 16 to 75, but not among the subgroup of neonates with an umbilical artery pH <7.0 at birth. ¹¹ , ¹² , ¹³ , ¹⁵ This suggests that, regardless of the degree of intrapartum asphyxia they cause, acute events are risk factors for moderate to severe HIE. One hypothesis is that acute and intense hypoxia is more deleterious to the brain than more moderate but prolonged hypoxia, possibly because compensation mechanisms do not have time to occur, consistent with findings in experimental animal studies. ²⁸ , ²⁹ In hypoxic situations, the first response is an increase in blood pressure through increased vascular resistance. ³⁰ Redistribution of blood flow through the ductus venosus follows to maintain the brain’s oxygen supply. In addition, a decrease in electrocortical activity reduces the brain’s need for oxygen. If hypoxia persists, these mechanisms fail and brain injury occurs. ²⁸ , ³¹ Cerebral blood flow does not increase in severe and sudden hypoxia, because there is no cerebral vascular vasodilation. ²⁸ Therefore, a severe hypoxic event, even if brief, can lead to severe brain damage because of this organ’s high metabolic rate. Moreover, the mechanism of neuronal death differs according to the initial hypoxic event: necrotic after a sudden, severe event and apoptotic following a more moderate and longer event. ³² , ³³ Moreover, these two situations differ for the anatomical location of brain lesions. ³⁴ , ³⁵ Some authors suggest that the short‐term outcomes of children with HIE treated with hypothermia after an identifiable intrapartum acute event are worse than for the same population with no such identifiable event. ³⁶ Further studies are needed to determine whether children exposed to acute events respond differently to hypothermia and other neuroprotective strategies.

Even after adjustment for umbilical artery pH, a maternal temperature ≥38 °C at delivery tends to be associated with HIE. In some of these cases, this fever may have resulted from chorioamnionitis, which is a known risk factor for HIE, probably mediated by an inflammatory process. ¹² , ¹⁴ , ³⁷ However, it is possible that maternal fever has a direct negative impact on the fetal brain. According to Blume et al., ³⁸ chorioamnionitis and isolated maternal fever are independently associated with neonatal encephalopathy. Fever during or after an anoxo‐ischemic episode is known to enhance the extension of brain lesions, and fetal temperature rises as maternal temperature rises. ³⁹

The univariate analysis showed significantly more diagnoses of intrahepatic cholestasis of pregnancy in the HIE group. We could not include this variable in the multivariable analysis because the control group had no diagnoses of intrahepatic cholestasis. Gestational cholestasis is associated with both a worse neonatal prognosis and intrapartum asphyxia, especially for bile acid levels above 40 μmol/L. ⁴⁰ , ⁴¹ It is therefore possible that gestational cholestasis worsens the prognosis of neonates with asphyxia. However, because we cannot rule out the possibility that this finding was observed by chance, it should be confirmed by other studies.

We found no difference between the two groups according to FHR. It must be noted that most of the FHR tracings were pathological, which appears logical given that all children had a pH ≤7.0 at birth. Moreover, FHR may not be a good predictor of HIE. In a case–control study comparing FHR 1 h before the birth of neonates with or without HIE treated with hypothermia, Graham et al. ⁴² found no difference between the groups in the tracing categories according to the American College of Obstetricians and Gynecologists classification. Another study found that experienced clinicians assessing FHR blinded to perinatal outcome detected 75% of neonates who were subsequently diagnosed with HIE and recommended expedited birth in nearly half of those cases. ⁴³ Nonetheless, FHR has very low specificity for predicting HIE because of the high frequency of FHR abnormalities in neurologically normal newborns and the low prevalence of HIE.

5. CONCLUSION

This study shows that the degree of acidosis at birth is associated with an increased risk of stage 2 and 3 HIE. Moreover, for a given pH at birth, the presence of an acute event, maternal fever at delivery, and thick meconium is associated with a higher risk of HIE. It provides new insight about the pathophysiology of HIE. After birth, for a given pH, these obstetric factors may help pediatricians to identify neonates who need specific care to prevent long‐term neurological sequelae.

CONFLICT OF INTEREST

None.

AUTHOR CONTRIBUTIONS

All authors contributed to the conception and design, processing of data, interpretation of data, drafting of the article, critical revision, and final approval. PL and EG contributed to the data collection, and PL and GK to the statistical analysis of the data.

Lorain PP, Bower A, Gottardi E, et al. Risk factors for hypoxic–ischemic encephalopathy in cases of severe acidosis: A case–control study. Acta Obstet Gynecol Scand. 2022;101:471–478. doi: 10.1111/aogs.14326