The primary aim of our randomised trial is to quantify whether the incidence of metabolic acidosis is decreased by the use of a strategy of fetal monitoring with CTG plus ST-analysis of the fetal ECG, compared to usual care consisting of CTG plus FBS, when indicated.
Secondary aims are to quantify the cost-effectiveness of CTG in combination with non-invasive ST-analysis of the fetal ECG for fetal monitoring during labour as compared to usual care and the differences in incidence of operational deliveries and performance of FBS across both groups.
The trial is a pragmatic randomised diagnostic study. We have choosen a randomised design because ST-analysis of the fetal ECG is an example of a new test that might provide better or other information, potentially leading to other treatment choices than the existing reference (CTG plus FBS). In such a situation there is an indication to do a randomised study rather than a conventional accuracy study in which ST-analysis is compared to CTG plus FBS, to quantify its value on patient outcome [20
]. A design with randomisation at the moment that the CTG group and the ST-analysis group provide discordant test results was not feasible, as this would require randomisation at the moment of medical emergency [21
This study is set in the Dutch Obstetric Consortium, a collaboration of obstetric clinics in the Netherlands. The study will be carried out in nine hospitals, including academic hospitals and non-academic teaching hospitals [23
Women will be eligible if they are in labour with a singleton fetus in vertex position, a gestational age ≥ 36 weeks and a medical indication for electronic fetal monitoring. A medical indication is defined by either a high-risk pregnancy, induction or augmentation of labour, epidural anaesthesia, meconium stained amniotic fluid or non-reassuring fetal heart rate.
Exclusion criteria: breech presentation, twin pregnancies, maternal age < 18 years or absent informed consent.
Procedures, recruitment, randomisation and collection of baseline data
Eligible women will receive the patient study information around 36 weeks of gestation, in the outpatient clinic. Women, having their prenatal controls under supervision of a midwife and being referred to the hospital during labour, will be informed and asked for consent by the attending doctor or midwife at arrival at the hospital.
After consent, women are randomised through a computer-generated randomisation sequence. Stratification will be applied for centre and parity (no previous vaginal delivery versus one or more previous vaginal deliveries). Randomisation will be 1:1 for either monitoring by CTG plus ST-analysis or CTG plus FBS. Both strategies will be performed according to strict protocols (see below).
Women who decide not to participate in the study will be asked for their reasons of refusal. They will be monitored with CTG and, if indicated, FBS.
In each centre an independent gynaecologist will be responsible for the centre specific data collection. Per centre a research nurse or midwife will monitor the protocol also via patient meetings and feedback on potential protocol violations.
At baseline, demographic, past obstetric and medical history data will be recorded for all women.
Control group: CTG
In women randomised to the control group, a scalp electrode will be applied to the fetal head and connected to the conventional CTG-monitor conform routine practice. CTG-interpretation will be guided by the FIGO guidelines (Table ) [24
]. Fetal blood sampling is recommended in case of an intermediate or abnormal CTG.
If the pH of the first FBS measurement is below 7.20 delivery is recommended, unless the cause of fetal distress can be alleviated. If the pH is between 7.20 and 7.25, FBS will be repeated after 30 minutes. If the pH is above 7.25, FBS is repeated according to the consecutive CTG pattern on discretion of the attending doctor or midwife.
Intervention group: CTG and ST-analysis
In women randomised to the intervention group, a scalp electrode will be applied to the fetal head and connected to the STAN®
-monitor. This electrode will allow both standard fetal heart rate monitoring (CTG) as well as ST-analysis of the fetal ECG. The CTG will be classified as normal, intermediate, abnormal or preterminal according to the FIGO-guidelines for fetal heart rate monitoring (Table ) [24
]. Clinical management will be supported by computerised ST waveform assessment and will be guided by the STAN®
-guidelines, indicating when intervention is recommended [25
]. The ST log automatically alerts the attending doctor or midwife if a significant ST-event occurs. Delivery is recommended when there are significant ST-changes unless the cause of fetal distress can be alleviated (Table ).
STAN® clinical guidelines: ST-changes that prompt clinical intervention, such as delivery or solving a cause of fetal distress
Preferably, FBS will not be performed in this group. This, however, does not suffice in all cases. The aim is to keep the performance of FBS as low as possible, limited to three well-defined situations: 1) poor signal quality of the fetal ECG in combination with an abnormal CTG; 2) STAN®-registration starts with an abnormal CTG; 3) 60 to 90 minutes of abnormal CTG recording during first stage of labour without ST-events and the attending doctor decides not to intervene by caesarean section. In the second stage of labour it is advised to perform an instrumental delivery (if possible) in case of the before mentioned situations.
Furthermore, we will also consider the incidence of FBS in both groups as secondary outcome.
Children who were admitted to the neonatal intensive care or high care after delivery because of birth asphyxia or other delivery trauma, will be checked upon at 6 months post-delivery. These check-ups are part of the regular controls of children.
Primary outcome measures
The primary outcome will be the incidence of serious metabolic acidosis defined as a pH < 7.05 and a BDecf
> 12 mmol/l in the umbilical cord artery [26
Secondary outcome measures
Secondary outcomes are:
1. Instrumental delivery rate because of fetal distress, failure to progress or a combination of the two
2. Neonatal outcome defined as Apgar scores < 4 after 1 minute and/or < 7 after 5 minutes
3. Need for admission to neonatal medium or intensive care unit
4. Incidence of performance of FBS in both groups
5. Cost-effectiveness of both monitoring strategies in general and across hospitals.
The sample size calculation is based on the primary endpoint, which is metabolic acidosis in the umbilical cord artery.
Although in the two previous randomised trials the incidence of metabolic acidosis decreased from 1.5 % to 0.5 % in favour of the CTG + ST-analysis group [13
], we assume that the incidence of metabolic acidosis in our higher-risk population (women delivering in the hospital with a medical indication/risk factor) is higher and estimated on 3.5 %, as found in our preliminary study [12
]. Based upon the numbers of the largest clinical trial with a relative risk of 0.5 the required sample size would then yield 2400 cases (1200 per arm), using an alpha of 0.05 (2-sided) and a power of 0.80 [13
]. However, soon after the start of our study a third randomised clinical trial appeared, although much smaller and non-significant, but yielding an opposite effect [16
]. A meta-analysis of the three clinical trials showed the varying relative risks of 0.5, 0.4 and 2.4 [27
]. Hence, to be conservative we used the meta-analysis overall relative risk of 0.6 for our power calculation, implying a reduction of metabolic acidosis, in favour of ST-analysis, from 3.5% to 2.1%. With an alpha of 0.05, a two-sided test – given conflicting results in the literature – and a power of 0.80, about 4638 women should be randomised (2319 per arm). Accounting for 10% loss to follow-up, the study requires inclusion of about 5100 women in order to obtain 4638 analysable cases.
The analysis of the primary endpoint will follow the intention to treat principle. Since this is a randomised trial, we would anticipate minimal differences in baseline characteristics. The relative risk with 95% confidence interval of metabolic acidosis in the CTG plus ST-analysis group compared to the CTG (plus FBS) group will be calculated, accounting for the stratified randomisation by centre and parity. Relative risks and 95% confidence intervals will also be calculated for the dichotomous secondary outcomes, i.e. instrumental delivery rate, neonatal outcome, need for neonatal admission and incidence of performance of FBS across both groups.
The following planned subgroup analyses will be performed: analysis according to risk pregnancies such as women with insulin dependent diabetes mellitus, fever during delivery and start of a STAN®-registration with an abnormal trace.
Missing data rarely occur at random. Simply excluding subjects with missing values thus not only lead to loss of statistical power but also to biased study results. To decrease bias and increase statistical efficiency we will therefore impute missing values rather than perform complete case analysis only. This will be done using single and multiple imputation methods [28
The economic evaluation is primarily a cost-effectiveness analysis (CEA) to find the optimal strategy as the one with the most favourable trade-off between avoided adverse neonatal outcome (fetal distress/metabolic acidosis) and difference in cost.
For this purpose, the process of care is distinguished into two cost stages (delivery/childbirth stage and postnatal stage) and three cost categories (direct medical costs [all costs in the health care sector, such as type of intervention and maternal and fetal monitoring, lab tests, costs associated with intrapartum complications, costs of training, maternal and neonatal care in the postnatal stage], direct non-medical costs [costs outside the health care sector that are affected by health status or health care], and indirect costs [productivity costs, costs of sick leave]).
Valuations of direct medical resources are estimated as cost per unit estimates comprising 'true' economic costs, i.e. including shares of fixed costs and hospital overheads. Cost per unit is estimated for at least one teaching and one non-teaching hospital. Direct medical volumes outside the hospital and direct non-medical volumes are valued using national reference prices [31
]. Indirect costs are quantified according to the friction cost method. Study specific costs are excluded from analysis.
As we anticipate a reduction of metabolic acidosis with the use of CTG plus ST-analysis compared to CTG plus FBS, the economic analysis is planned to be a cost-effectiveness analysis. We will use bootstrap sampling to calculate 95% confidence intervals around the cost-effectiveness ratios. Sensitivity analysis will be used to explore the effect of variation of several key factors.
Baseline obstetric characteristics of first 500 randomised patients
Baseline obstetric characteristics of the first 500 enrolled patients in the study are shown in table .
Baseline obstetric characteristics of the first 500 randomised patients in clinical trial. All values are absolute numbers (%) or mean (SD).
Ethical considerations and Safety Committee
The study protocol has been approved by the Medical Ethical Committee of the University Medical Center Utrecht, Utrecht, The Netherlands (05/157-K). Written informed consent will be obtained from each participating patient.
Serious Adverse Events (SAE) are defined as "metabolic acidosis (arterial pH < 7.00 and BDecf
>12 mmol/L) and admission to a Neonatal Intensive Care Unit (NICU)" or "an Apgar score < 7 after 5 minutes and admission to a NICU". Given the conflicting results in the literature, all SAE's will be reported to the Data Safety Monitoring Committee, consisting of an independent gynaecologist, neonatologist, epidemiologist and a biostatistician. This committee will consider the reported incidence of SAE's at regular intervals to determine whether there are significantly more serious adverse events in the STAN-group and, if so, whether the study should be discontinued. For this purpose the computer program PEST, version 4, will be used [32
This study is the first randomised trial to quantify the cost-effectiveness of ST-analysis of the fetal ECG during labour and the first to achieve data about the use of ST-analysis with fetal blood sampling performed only in well-defined circumstances. This study aims to provide additional evidence on the true clinical value of intrapartum ST-analysis and also the clinical indications in which fetal blood sampling still might be necessary in addition to ST-analysis.