Over the past 30 years, one of the most significant developments in obstetric practice has been an increase in the number of deliveries by caesarean section (CS). The current World Health Organization (WHO) recommendations state that the rate of CSs should not exceed 15%.1 One of the main fetal indications for CS is fetal distress, especially during labour, diagnosed by either fresh meconium staining of the amniotic fluid or an abnormal cardiotocograph (CTG).
Although cardiotocography is the most widely used tool for antepartum and intrapartum surveillance, suspicious fetal heart rate (FHR) patterns lack specificity, and false-positive FHR tracings often result in unnecessary CSs. Fetal blood sampling (FBS) is a particularly helpful tool for diagnosis of fetal acidosis and therefore hypoxia.
This study was conducted to examine the value of cardiotocography during labour when used as the sole indicator for CSs in fetal distress within our department in Dubai Hospital, United Arab Emirates.
Reasons for clinical audit project
To identify the indicators of fetal distress used in women undergoing a CS for fetal distress.
To ensure a decrease in CS rate for fetal distress once these parameters have been identified.
To update the current protocol used in our department.
To introduce new tools in our department for assessing fetal distress, for example:
Materials and methods
This is a retrospective and descriptive study of cases conducted in Dubai Hospital for 6 months from 1 June 2011 to 31 December 2011. All pregnant women at or beyond 34 weeks’ gestation who underwent an emergency CS primarily for fetal distress were included. The exclusion criteria covered any severe maternal or fetal pathology which may lead to fetal distress before labour or at its onset. During this period, 101 out of 517 CSs were performed for fetal distress, but only 90 CSs were kept for review and analysis, with 11 excluded as not fitting the criteria. Data were collected from antenatal files, delivery book and electronic medical records.
An abnormal CTG was taken as the primary indicator of fetal distress. These CTGs were reviewed by two consultants. In 20 out of 90 files (22.2%), CTGs were not available while reviewing the cases; however, they were available at the time of delivery and the decision to carry out a CS was made based on abnormal FHR tracings. Only those files with CTG tracings available were reviewed and classified according to international guidelines.2 As regards other data (e.g. age, type of caesarean, neonatal outcome), the indication was considered according to the antenatal interpretation of the specialist registrar documented in the files. The other data considered were amniotic fluid colour; time from the decision to carry out a CS until delivery; and neonatal outcome by measurement of an Apgar score at 1 and 5 minutes, cord pH (where possible) and admission to neonatal intensive care. The immediate factors leading to CS were identified. Rates of CSs and CTG abnormalities were compared with the literature.
During the 6-month period, 1890 mothers delivered in our unit, of whom 517 were delivered by CS for different indications. Among the CSs, 101 were due to fetal distress, but only 90 of the CSs were kept for review, representing 17.4% of total CSs and 4.76% of total deliveries. The 11 files excluded did not fit the criteria owing to other complications as follows: severe eclampsia (n = 4), severe intrauterine growth restriction (n = 2), fetal anomalies (n = 2), non-immune hydrops (n = 1) and failure to progress without fetal distress (n = 2).
The average age of the women in our study was 27.8 years (standard deviation ± 5.3 years). More than 50% of the women were primigravidae (n = 52) and 81% delivered at term (n = 73).
Fetal distress indicators
When CTGs were reviewed, 20 were found to be absent for unknown reasons. The 70 CTGs reviewed were classified according to international guidelines as normal, suspicious or pathological. Surprisingly, only 21 CTGs (30%) fitted the criteria of pathological CTG, whereas 6% were considered normal. The most common indicator for emergency CSs for fetal distress was fetal heart deceleration (two out of three cases), as shown in Figure 1.
The relationship between liquor colour and acute fetal distress was also assessed. The findings showed that only 37 out of 90 CSs (41%) had meconiumstained liquor. Analysis of the meconium-stained liquor showed that it was mostly grade 1 (Figure 2). No woman underwent an emergency CS based on thick meconium alone, as this was always associated with suspicious or pathological CTGs.
Time between decision to carry out a caesarean section and birth
It is desirable that the time taken between the decision to carry out a CS and delivery should not exceed 30 minutes. In our study, the average was 48.6 minutes (Figure 3), with a time of 30 minutes from decision to delivery achieved in only 26.6% of patients. Possible explanations include understaffing, anaesthesia check-up prior to CS, time spent transferring patients to the operating theatre and use of regional anaesthesia.
An analysis of the time of day when CSs were carried out clearly showed that the rate increased after normal working hours. Of the 35 CSs that took place at night, 17 were performed after midnight (Figure 4). This may be because consultants are available within the hospital only during normal working hours, and, despite obstetric unit guidelines requiring the on-call consultant to be informed of all proposed CSs, the decision in most cases is taken by the specialist registrar on call. Staff fatigue, the fear of perinatal asphyxia and medicolegal issues may have influenced the decision.
Average Apgar scores were good, being 8 ± 1.6 (range 3–9) at 1 minute and 9.7 ± 0.66 (range 7–10) at 5 minutes, despite the time taken between decision and delivery. Few newborns had low Apgar scores. Seven newborns (7.5%) had a 1 minute Apgar score < 5, but none had a low score (< 7) at 5 minutes.
Umbilical cord pH was tested in 47 out of 90 newborns; however, two samples were clotted. The results from the remaining 45 newborns are shown in Figure 5.
When CTG findings were correlated with pH, it was observed that, out of 21 infants with pathological CTGs, five had a cord pH > 7.25, two had a pH between 7.25 and 7.20, and four had a pH < 7.20. These results should be interpreted with caution as only 45 out of 90 newborns underwent cord blood sampling.
A comparison of cord pH results with Apgar scores showed that, out of 29 newborns with normal cord pH, only four had a 1-minute Apgar score < 5. Also, our study showed that only one newborn out of nine with a borderline cord pH measurement (i.e. pH 7.24–7.20) had a 1-minute Apgar score < 5 and, of the seven newborns with pathological cord pH (i.e.< 7.20), only two of them had a 1-minute Apgar score < 5, and none at 5 minutes. Once again, these results should be interpreted with caution, with a larger study being required. Nevertheless, the results indicate the importance of obtaining additional information, such as fetal scalp blood pH, as soon as possible when deciding whether or not to carry out an emergency CS for fetal distress.
Ten newborns were admitted to the neonatal intensive care unit, of whom two were intubated. Six had pathological, and four had suspicious, CTGs.
Cardiotocography was introduced 30 years ago with the aim of identifying fetuses affected by hypoxia during labour. Despite this, there has been no real long-term improvement in neonatal outcome, with the rate of CSs showing that it has increased fourfold.3
Roy et al.3 reported a very high (6.8%) incidence of CS delivery at ≥ 36 weeks, primarily for non-reassuring FHR during labour. In our study, the incidence of CS was 4.7%. This could be due to multiple factors, such as misinterpreting CTGs during labour and patient and family anxiety. The proportion of emergency CSs performed for non-reassuring CTGs is highest outside normal working hours (working hours: 0730–1430) at about 72% (65 out of 90), with 26% of these (17 out of 65) taking place after midnight in our study. This suggests that a decreased staff concentration may also play a part. In a study by Hueston et al.,4 a significant increase in rates of CS for fetal distress was noted between the hours of 0900 h and 0300 h. The observation that CS deliveries for fetal distress peak during the night raises the possibility that the interpretation of fetal monitor tracing is influenced by physician and patient fatigue or other clinical factors, such as medicolegal issues or the absence of senior consultants.4
As shown in Table 1, the most common fetal heart abnormality was found to be deceleration of fetal heart rate (type I, II or variable, observed in 65% of cases), whereas in the study by Roy et al. it was persistent bradycardia (48.8% of cases).3
|Dubai Hospital||Roy et al.3|
|Other abnormal CTG||34.5%||–|
The mean decision to delivery interval reported by Roy et al.3 in CSs for fetal distress was 38.2 ± 12.5 minutes, whereas in this study the mean was 48.6 ± 25.2 minutes. However, the longer decision to delivery interval time did not lead to poorer neonatal outcomes as expected since CTG is not a real indicator for fetal distress, confirming our belief that other tools are required for accurate diagnosis. In our opinion, neither amniotic fluid nor CTG abnormalities can confirm fetal hypoxia. Only the introduction of fetal scalp blood sampling and/or fetal ECG can improve the diagnosis and hence reduce the incidence of CSs for fetal distress.
The aim in obstetrics is to deliver infants in good health, but relying on traditional cardiotocography alone to detect fetal distress results in a large number of unnecessary CSs. There is an urgent need to use techniques that both are safe and offer more objective evidence of fetal distress.
The Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG)5 recommends that all clinicians using and interpreting CTGs should have:
current knowledge of fetal physiological responses to hypoxia
good pattern recognition skills
the ability to apply this knowledge in every clinical situation.
Regular education and credentialling should be completed by all clinicians using and interpreting CTGs.
How to decrease the incidence of caesarean sections
The integrated CTG and fetal ECG (STAN®-monitor; Neoventa Medical AB, Molndal, Sweden) has been proven to reduce the rates of metabolic acidosis and instrumental delivery. The ST log automatically alerts the attending doctor or midwife if a significant ST event occurs. CS delivery is recommended when there are significant ST changes unless the cause of fetal distress can be alleviated.
It is undeniable that CSs for fetal distress have contributed to rising CS rates in the obstetric gynecology department of Dubai Hospital. In order to decrease CS rates, other tools should be provided which may influence decisions made during labour, especially outside normal working hours.