Prediction of neonatal acidemia at birth with total deceleration area on intrapartum cardiotocogram

Introduction

Intrapartum foetal distress and neonatal acidemia are closely linked conditions that lead to neonatal mortality. A WHO report states that 1.9 million stillbirths happen every year and intrapartum stillbirths contribute 40% to it.1 Hypoxia is believed to contribute to 90% of foetal deaths occurring during labour.2 It can have significant implications on the health of the newborn. Complications of foetal distress also include neonatal morbidity in the form of neonatal intensive care unit (NICU) admissions, birth asphyxia, respiratory distress, neurological damage, hypoxic-ischaemic encephalopathy, seizure disorder, developmental delays and cerebral palsy.3 This also leads to maternal morbidity in the form of caesarean section, as most of the intrapartum foetal distress cases are managed by emergency caesarean sections for immediate delivery and a better neonatal outcome.4 Recently, there has been a marked rise in the rate of caesarean deliveries across the globe. Foetal distress is the chief indication for primary emergency caesarean sections and the second most common indication for overall caesarean sections.5 In 2017, a report carried out at our tertiary care hospital revealed a caesarean section rate of 44.61%.5 Foetal distress was the second most common indication for overall caesarean deliveries, accounting for 16.24% after previous lower segment caesarean section (LSCS).5

Foetal distress is a vague term. In clinical practice, it is used to describe a wide range of foetal heart rate abnormalities that, if not corrected, will result in decompensation of physiological responses and cause permanent central nervous system damage or death.2 3 Electronic foetal monitoring is commonly used to detect foetal distress. In clinical practice, cardiotocography (CTG) based National Institute of Child Health and Human Development (NICHD) classification is widely used for diagnosis of foetal distress such as category II and category III traces which include tachycardia, repetitive variable deceleration, late deceleration, and marked and decreased variability.6 It has been observed in previous research that irrespective of the type of decelerations, acidic pH and NICU admissions are seen.7 Acidic pH is often observed with variable and late deceleration.7 CTG is criticised for increased rates of caesarean deliveries for foetal distress as the false-positive rate of diagnosis of foetal distress with CTG is high (60%).8 Nevertheless, CTG is preferred as it is documented proof for the basis of intervention performed for managing cases of foetal distress. CTG interpretation needs technical training. Emphasis is given to avoid foetal acidemia, for late and persistent variable deceleration. It is observed that there is a knowledge gap in identifying a better CTG parameter that can be interpreted easily and aid in the early diagnosis of foetal distress so that a timely decision is taken for the prevention of neonatal acidemia. Total deceleration area (TDA), a simple formulated calculation, was first described by Cahill et al, which has shown promising effects for predicting acidemia.9 In recent research by Geva et al, increased TDA was positively associated with neonatal encephalopathy and with MRI changes in affected neonates.10 11 We intended to study TDA on intrapartum cardiotocogram which will predict neonatal acidemia at birth in the Indian population.

Material and methods

This case-control study was conducted at KAHER’S Dr. Prabhakar Kore charitable hospital, Belagavi, India, for 1 year (1 February 2023–31 January 2024). The study population included pregnant women at or above 37 weeks of gestation, who were admitted to the labour room for delivery. Ethical clearance was obtained from the JNMC Institutional Ethics Committee (MDC/JNMC/IEC/69) (CTRI registration number: CTRI/2023/02/049506). The study population was screened as per the inclusion & exclusion criteria. The eligible study participants were enrolled based on the following criteria.

Operational definitions

TDA: the deceleration area is the width of the widest aspect of deceleration (below baseline) measured in minutes, multiplied by the maximum depth below the baseline (beats) divided by two. TDA is a summation of all the deceleration areas in the last 30 minutes before delivery. Its value is expressed as missed beats12 (figure 1).

Figure 1

Request permission

Illustration of deceleration area on foetal cardiotocogram.

Neonatal acidemia at birth: umbilical cord artery pH <7.2 at birth.13

Late deceleration: these are visually apparent, gradual decreases in FHR and return to baseline (onset to nadir is equal to or greater than 30 s) associated with uterine contractions. The onset, nadir and resolution of the deceleration usually occur after the peak of the contraction.9

Variable deceleration: these are visually apparent abrupt decreases in FHR (onset to nadir < 30 s). Onset, depth, duration and return to baseline vary with uterine contraction.9

Beat-to-beat variability: the visual quantification of fluctuation in FHR at baseline (decreased variability <5 bpm, marked variability is >25 bpm).9

Bradycardia: FHR<110 bpm. Tachycardia: FHR >160 bpm9

Repetitive variable deceleration: FHR changes that occur when variable decelerations happen with 50% or more of uterine contractions.14

High-risk pregnancy: it is a pregnancy affected by a condition that is harmful to the health and well-being of the foetus or the mother.15

The following high-risk factors are included in the study.

Postdatism: it is defined in the study as pregnancy which crosses the expected date of delivery.16

Foetal growth restriction: foetal growth restriction is defined as expected foetal weight (EFW) or abdominal circumference (AC) <10th percentile for gestational age (GA).17

Gestational diabetes mellitus: it is diagnosed based on a 2-hour, 75 gm oral glucose challenge test value of >140 mg/dl in pregnancy.(Diabetes in Pregnancy Society of India).18

Pre-eclampsia: acute onset hypertension in a previously normotensive pregnant woman involving symptoms related to one maternal organ dysfunction of the liver, kidney, brain and placenta.19

Macrosomia: foetus with EFW or AC above the 90th percentile.20

Oligohydramnios: it is defined as the amniotic fluid index <5.21

Anaemia: anaemia in pregnancy is a condition where maternal haemoglobin (Hb) concentration is less than 11 g/dl (WHO).22

Hypothyroidism: pregnancy with elevated maternal thyroid-stimulating hormone (TSH) of more than 4 m IU/ml.23

Premature rupture of membrane (PROM): it is a prelabour rupture of the amniotic membrane before the onset of labour pain.24

Prolonged latent phase: > 20 hours in a nullipara and >14 hours in a multipara.25

Results

Among a total of 4194 deliveries, 402 women were screened for eligibility during the study period. The number of excluded participants was 234 (preterm (88), insufficient trace (112), unavailability of umbilical cord blood pH (34)). A total of 168 participants were enrolled in the study. The newborns with neonatal cord blood pH <7.2 were 42 (25%), and they were categorised as cases, while 126 (75%) newborns with pH >7.2 were grouped as controls. The additional findings noted were two foetal tachycardia, five foetal bradycardia, one marked variability and one decreased variability among the case groups (figure 2).

Table 1 gives the clinical details of study participants among case and control groups. The median age of the study participants in both case and control groups was 24 years. There were 35 (83.33%) primigravidae in the case group and 103 (81.75%) in the control group, respectively. The high-risk gravidas were 31 (73.80%) and 94 (74.60%) in the case and control groups. Postdatism accounted for 11 (35.48%) and 32 (34.04%) gravidas in the case and control groups, respectively. There were 7 (22.58%)and 17 (18.08%) participants with foetal growth restriction (FGR) in case and control groups. The incidence of induced labour was 24 (57.14%) and 73 (57.94%) in the case and control groups. A total of 23 (54.76%) cases had foetal distress in the latent phase, five (11.9%) in the active phase and 14 (33.33%) in the second stage of labour (p value 0.002). There were 12 (28.57%) instrumental deliveries in the case group and seven (5.56%) in the control group (P value <0.001).

Table 2 highlights the distribution of various CTG variables over case and control groups. There is a significant difference in the distribution of foetal bradycardia over groups with five (11.9%) in case and three (2.38%) in the control group (P value 0.029). No significant difference is noted in the distribution of foetal tachycardia, beat-to-beat variability <5 or >25 beats, and repetitive variable deceleration over groups. Late decelerations are associated significantly with cases (P value 0.003). From the Mann-Whitney U test, it is observed that there is a significant (P value 0.008) difference in the distribution of mean TDA over groups. The mean TDA of 254.62±170.67 is in the case group, and the mean TDA of 165.1 ± 94.17 is in the control group.

Table 3 describes the birth characteristics of newborns in the case and control groups. Among cases, seven (16.67%) neonates did not cry at birth (P value 0.008), 31 (73.81%) were normal-weight neonates and 11 (26.19%) neonates had low birth weight. A total of 9 (21.43%) newborns in the cases group were admitted to the NICU for further management (P value 0.003).

Table 4 highlights the findings of the logistic regression analysis of various CTG variables for predicting neonatal acidemia. Foetal bradycardia and late deceleration AUC were 0.5476 (95% CI 0.4963 to 0.5990): P value 0.023 and 0.6597(95% CI 0.5552 to 0.7643) and P value 0.015 times more likely to be associated with neonatal acidemia, respectively. TDA cut-off of ≥195 was significantly associated with neonatal acidemia with an AUC of 0.6576 (95% CI 0.5305, 0.7847) (P value 0.001).

Table 5 demonstrates various diagnostic characteristics of study variables, that is, area under curve (AUC) was 0.5476 (95% CI 0.4963 to 0.5990), 0.6597 (95% CI 0.5552 to 0.7643) and 0.6576 (95% CI 0.5305 to 0.7847) for foetal bradycardia, late deceleration and TDA cutoff >195, respectively. PPV of TDA cutoff >195 was 83.78% (95% CI 70.64% to 89.35).

Discussion

The results of this case-control study highlighted the predictive ability of TDA (≥195 missed beats) observed on foetal cardiotocogram, in the last 30 min of labour, as statistically significant (P value 0.001) with neonatal acidemia at birth and an AUC of 0.6576 (95% CI 0.5305 to 0.7847).

These results are similar to those of other studies conducted in various other population settings. Furukawa et al computed TDA based on missed beats, whereas Cahill et al used seconds, and Gamboa et al’s study employed square centimetres on CTG trace for calculation.9 12 26 Furukawa et al found that TDA of 250 missed beats had the highest AUC value of 0.702 (95% CI 0.655 to 0.749), indicating its superior predictive ability for metabolic acidosis.12 The cut-off obtained in our study is noted to be lesser when compared with other studies. This may be explained by the proportion of high-risk populations in our study (75% in both case and control) compared with the different studies. Additionally, the study was conducted among the South East Asian ethnic population. Gamboa et al found that the optimal threshold for predicting neonatal acidemia based on TDA was 8.37 cm² in the last half an hour of labour, with a sensitivity of 71.6%, a specificity of 71% and an AUC of 0.83.26 For every square centimetre increase in the area during the final 30 min of labour, there was a corresponding decrease of 0.08 units in pH levels.26 Cahill et al also observed that TDA is the most predictive electronic foetal monitoring pattern for acidemia and found the respective Youden cut-off point of deceleration area for optimal prediction to be 42,152 for neonatal acidemia with an AUC of 0.76. Additionally, they found when combined with tachycardia, it poses a significant risk for neonatal morbidity.9

Though we found foetal tachycardia in 8 cases (19.05%) and foetal bradycardia in 5 cases (11.9%) in our study, only foetal bradycardia along with TDA was significantly associated with neonatal acidemia (P value 0.029) and AUC of 0.5476 (0.5084–0.6265). Cavoretto et al studied metabolic acidemia at birth with a duration of the second stage of labour. They found that the risk of metabolic acidemia increases (> 3.4–15%) at 60 min of second-stage duration in the presence of foetal bradycardia.27 Among cases with foetal bradycardia in our study, 83.33% experienced an increase in TDA from baseline. This trend suggests that bradycardia may be more associated with an increase in TDA, but the difference is not statistically significant (P value 0.242). Similarly, foetal tachycardia is related to a decrease in TDA from baseline, though not statistically significant (p value 0.061).

A decreased variability was noted in 11 cases and marked variability in nine cases. Repetitive variable decelerations were observed in three cases (7.14%). None of these variables show any significance in predicting neonatal acidemia at birth. Different studies have shown varying results concerning these variables. The findings reported by Gamboa et al and colleagues noted a significant association between acidemic babies and decreased beat-to-beat variability.26 The study by Cahill et al showed no significance in the predictive ability of decreased or marked variability.9 In the presence of repetitive persistent late deceleration, the risk of acidemia increases above 3.4% at the 50-min duration of the second stage of labour.28 In clinical practice, more than a certain decelerative pattern, the depth, duration and frequency of deceleration should be considered important while managing cases of intrapartum foetal distress.12 Similar to previous literature, our study shows a higher incidence of neonates experiencing acidemia in association with late decelerations, which was noted in 11 cases (26.9%) (P value 0.003 and AUC of 0.659).29–31 Cahill et al in another study found that combining all significant features (tachycardia, prolonged decelerations, repetitive late and variable decelerations) did not enhance the NICHD model’s predictive ability for acidemia.29 Though the early deceleration pattern was not significantly associated with neonatal acidemia in this study, acidic pH with early decelerations was noted in two cases (4.67%) with an average TDA of 145 missed beats. Hence, neonatal acidemia was consistently noted irrespective of the deceleration type observed. Among our study population, most participants had antenatal risk factors (73.80% of the case and 74.60% of the control groups). Postdatism emerged as the most prevalent risk factor in both groups (11 participants (35.48%) in the case group), followed by FGR (seven participants (22.58%) in the case group), oligohydramnios and GDM (two participants (6.45%)) each. The association of TDA with umbilical artery pH in high-risk pregnancies like postdatism, FGR and pre-eclampsia has been studied recently, and it was found to have a significant association in these high-risk pregnancies.32–34 Our study shows that induced labour appears to be the predominant type of labour compared with spontaneous labour in both the case and control groups, comprising 57.14% and 57.94%, respectively. The labour induction rate by Cahill et al was 44.8%.9 Our study group consisted of a greater number of high-risk pregnancies like postdates, FGR and oligohydramnios which are induced for better maternal and neonatal outcomes. In this study, 54.76% of study participants in the case group had foetal distress in the latent phase. This could be attributed to the increased rate of caesarean sections in women who undergo admission in the latent phase. Augmentation with oxytocin and caesarean deliveries were also significantly noted more in women admitted into latent labour.35 Foetal distress detected during the second stage of labour was observed in 33.3% of the cases. It shows a significant P value of 0.002. Cavoretto et al concluded in their research that the duration of the second stage of labour is directly proportional to metabolic acidemia (escalating 12-fold (1.2–15.3%) from 30 min to 180 min) at birth irrespective of the type of deceleration on CTG.27 We observed that 42.8% of cases had intrapartum meconium-stained liquor. It was not found to be significant in determining foetal acidemia. These findings align with results reported in previous research.36 In this study, 64.29% of the case group underwent caesarean section, and instrumental delivery was performed in 28.5% of cases. This might happen because most of the foetal distress was diagnosed during the latent phase of labour where the pregnancy was terminated by caesarean section for emergent delivery. It was noted that 92.86% of the control group underwent caesarean delivery. This shows that an increased number of LSCS are being performed for foetal distress, but the neonates are found not to have developed acidemia. This gives power to TDA as a more relevant predictor of foetal acidemia. Among the babies who did not cry at birth, seven (16.67%) belonged to cases, and four (3.17%) belonged to the control group (P value 0.008). Notably, babies who did not cry at birth had significantly worse outcomes than those who cried immediately, regarding NICU admissions (21.43%; P value −0.003)belonged to case group. Previous research conducted in other South Asian populations shows similar results.37 A proportion of 73.81% of babies in the case group had a normal birth weight. These findings closely resemble those reported in the study conducted by Cahill et al.9 It is noted that the majority of the study group consists of nulliparous individuals, comprising up to 80% in both the case and control groups. Comparable results were observed in a study conducted by Furukawa and colleagues, where nulliparity was evident in 68.2% of the cases.12

Conclusion

An intrapartum TDA of ≥195 missed beats was significantly associated with neonatal acidemia at birth in this study. TDA represents a simple calculation in the management of intrapartum foetal distress. This will help clinicians to make a timely decision in the management of foetal distress cases in order to prevent foetal acidemia without classifying decelerations into different categories that are subject to interpersonal variations.