Introduction

Neonatal resuscitation is unique in its frequency, immediacy, and favourable outcome when compared to adult resuscitation. It is usually carried out by Senior House Officers (residents) with the assistance of a midwife, and needs high levels of training and ability.

However, this is in a setting where a “natural” process, childbirth, is being seen to be over-medicalised, and where the demands on the paediatricians are such, especially in the wake of the a need to curtail doctor's hours of work, as to make attendance at all births an impossibility. Any medical input should therefore be directed to the episodes where it is needed, and be as efficient as possible₁.

There are nine possible types of births as detailed below (table 1). In our hospital the paediatrician was asked to attend every birth type excepting normal, meconium-free vaginal deliveries, and elective and semi-elective caesarean sections. Each hospital has its own protocols for attendance of the paediatrician to the delivery room, aiming to have a doctor standing by in case intervention is needed.

Most hospital protocols, and the American Academy of Paediatrics₂, now do not require routine endotracheal intubation of thin meconium staining of the liquor, and some hospitals have now stopped asking the paediatrician to attend thin meconium stained deliveries₃. This has been in response to studies showing that there is no increased risk of meconium aspiration syndrome with solely thin meconium₄. We wished to study whether paediatric attendance at these births was appropriate.

Figure 1

Table 1: Types of births possible and paediatrician attendance

Previous studies have investigated attendances at caesarean sections₅ and found no difference between resuscitation rates for babies born by caesarean when compared with normal cephalic vaginal deliveries. A study in Anchorage likewise found that local anaesthesia caesarean sections had no difference in resuscitation rates to their matched controls₆. A large cohort study in Tasmania performed between 1980-1989 concluded that there was no need for a paediatrician for epidural sections. However they noted a significant, unexplained drop in intubation rates in the second half of the decade, and they based their recommendations on that₇. Finally, a group from Chicago determined that there was no need for a paediatrician at local anaesthetic caesarean sections₈.

We felt it necessary to formally identify those deliveries where a paediatrician in attendance was of worth, so as to make childbirth more comfortable, natural, and private for mothers and to lessen the paediatrician's workload. With the current pressure for the reduction of working hours for doctors in the UK, efficiency of input is paramount if the same amount of work is to be carried out by a decreased workforce. Rationalisation of the paediatric junior doctor workforce will impact smaller hospitals most significantly. The setting of the District General Hospital has not been extensively researched in the past, and there are often more demands on the paediatrician as one doctor may have to be responsible for the paediatric ward, paediatric admissions to Accident and Emergency, as well as the neonatal intensive care, whereas a larger hospital may have separate teams of doctors for different areas. Constant interruptions are a significant burden.

Method

This was a questionnaire-based study with the documentation included in the standard paperwork pack with each admission to labour ward. This was done in conjunction with the midwives' full co-operation. The sheets were printed on to yellow paper to aid recognition. There were 15 questions in total. The midwife or paediatrician attending each delivery was asked to fill the sheet in. The study was anonymous and the Huntingdonshire Local Research and Ethics Committee granted ethical approval. The obstetric consultants were all informed and supportive.

It was concluded that at least 1000 deliveries would be needed to achieve statistical significance. The data was collected during 2000. The completed sheets were data entered by hand in to a Microsoft Excel spreadsheet. Statistics were performed within Microsoft Excel.

Our main outcome measure was based on the Apgar score at one minute. This was considered an “intervention independent” score, which best reflects the need for intervention. Although the Apgar score at one minute is not associated with long-term outcome, this may well be because of early intervention which has been indicated by the low score. For instance, an apnoeic baby who is intubated early can have an Apgar of 8 at five minutes without having any spontaneous ventilation, but without intervention would probably have a poor outcome. Apgar scores at 5 and 10 minutes, although recorded, were not processed because they may have been skewed by (beneficial) intervention by the medical team.

The mean Apgar scores were calculated with 95% confidence intervals. The odds ratios of having an Apgar of 4-6 and of 0-3 were also calculated for every type of birth. Finally, the odds ratios for having facemask ventilation or being intubated were calculated with 95% confidence intervals. The term ‘intubation' was used to refer only to those cases where intubation was performed for ease of ventilation. Suctioning of the trachea after a meconium-stained delivery was not counted as intubation.

All results were compared to the “normal” births, ie the normal vaginal deliveries without meconium, for which a paediatrician would not be called.

Results

Our sample was of 1255 babies. 45 forms were excluded due to inadequate data entry, giving 1210 valid data points.

The demographics were as follows:

Figure 2

The breakdown of modes of birth were as follows:

Figure 3

Figure 1 presents the data for mean Apgar score at one minute, including 95% confidence interval error bars, with the baseline of the graph at the mean Apgar for normal vaginal deliveries without meconium (ie at 8.40).

Figure 4

Figure 1: Mean Apgars at 1 minute with 95% Confidence Intervals Baseline at NVD without meconium

Figure 2 shows the odds ratios of having an Apgar at 1 minute of either between 0-3, or 4-6, when compared to normal vaginal deliveries without meconium, for each type of birth. 95% confidence interval error bars are presented.

Figure 5

Figure 2: Odds ratios of Apgar scores of 0-3 or 4-6 at 1 Minute by Type of Birth with 95% Confidence Intervals Compared With Normal Vaginal Deliveries without Meconium

Figure 3 shows the odds ratios of a baby needing intubation or facemask ventilation, again by type of birth and compared to normal vaginal deliveries without meconium.

Figure 6

Figure 3: Odds Ratios of Face Mask Ventilation/Intubation by type of birth with 95% Confidence Intervals Compared to NVD without Meconium

Discussion

The Apgar score was developed by Virginia Apgar to determine which babies require intervention to resuscitate after delivery. Any high-risk birth should be defined as one which has a significantly depressed Apgar score at the intervention-independent stage (ie at one minute), or which needs significantly more interventions, when compared to the normal vaginal delivery without meconium.

Our data of frequency of different types of births is comparable to other studies. Previous incidences of meconium staining have ranged between 8.5%₉ and 18%₁₀, compared to 15% in this study.

There is a risk when examining data based on a numerical mean of a discrepancy between clinical and statistical significance. In the case of deliveries with thin meconium, depression in the mean Apgar scores, although being significant, was only 0.36 points. Therefore the grouped data (Apgars 0-3, 4-6) and the pragmatic data (intubation and face mask ventilation rates) were examined as well.

We looked at five ways of classifying high-risk births, each of which would be valid in their own right. Any significant depression in mean Apgar scores, a significant increase in odds ratio of having an Apgar of between 0-3 or 4-6, and a significantly increased odds ratio of being intubated or having face mask ventilation. The summary table below shows which modes of birth showed significant differences.

Assisted deliveries

Figure 7

Ventouse deliveries had a significant decrease in Apgar score at one minute, and a significant increase in the odds ratio of having face mask ventilation. Forceps similarly had poorer Apgars and increased face mask ventilation,and also a higher chance of having an Apgar between 4-6. Therefore we recommend that a paediatrician attends these births as both modes of delivery are at higher risk of intervention and poorer Apgars.

Caesarean Sections

Meconium-stained liquor deliveries

Acknowledgement

Correspondence to

Dr Patrick Davies Email: daviespatrick@hotmail.com