The decision-to-delivery interval for emergency caesarean section: is 30 minutes a realistic target?

Authors


*Mr M. H. Jones, Darent Valley Hospital, Darenth Wood Road, Dartford, Kent DA2 8DA, UK.

Abstract

Objective To evaluate whether a 30-minute decision-to-delivery interval is a realistic target for emergency caesarean section.

Design An audit of all emergency caesarean sections over five separate periods.

Setting A district general hospital.

Participants Five groups of women (343 women) with an indication for emergency caesarean section.

Methods Following an initial survey, a structured time sheet was introduced, followed by four other surveys to complete the audit cycle.

Main outcome measure The proportion of caesarean sections where the decision-to-delivery interval measures was achieved within 30 minutes. The reasons for delay.

Results In the first survey of 73 emergency caesarean sections, the time to deliver the infant exceeded 30 minutes in 47 women (64%). The main sources of delay were transferring the women to the operating theatre and starting the anaesthetic. After the introduction of a structured time sheet, there was an improvement with each survey, but the target of 30 minutes was reached in only 71% of caesarean sections in the final survey.

Conclusions The introduction of a time sheet can improve the decision-to-delivery interval for emergency caesarean section; however, a universal standard of 100% in 30 minutes is unrealistic.

Introduction

The 30-minute decision-to-delivery interval for emergency caesarean section, despite being a pragmatic rather than an evidence-based rule, is widely accepted1. It is also a requirement by the Clinical Negligence Scheme for Trusts2. It has therefore become important for obstetric units to carry out surveys of their decision-to-delivery times, as recommended by the Confidential Enquiry into Stillbirths and Deaths in Infancy3. The aim of this study was to assess how close our obstetric unit was to achieving this target of 30 minutes for emergency caesarean section and to identify ways of improving our performance.

Methods

The maternity unit is in a district general hospital. It has 2300 deliveries per year and a caesarean section rate of approximately 24% (elective 16%). A survey of all emergency caesarean sections occurring over three months was carried out. The delivery suite was fully staffed with trained midwives, obstetricians and anaesthetists, and in the operating theatre there were trained operating department assistants. The operating theatre was a few yards from the midwives' station. An emergency caesarean section was defined as one which required prompt delivery to reduce the risk to the pregnant woman or her infant. The decision-to-delivery interval was the time between the decision to perform the caesarean section and the delivery of the infant. The decision to perform the caesarean section was always made by the consultant obstetrician, who may or may not be in the hospital. During these three months, information was extracted from the case notes of the women whose decision-to-delivery interval was longer than the standard of 30 minutes, in order to identify reasons for the delay. A time sheet was then devised (Fig. 1). The survey was then repeated four times, at three-to-six-monthly intervals, in order to measure the improvement in the decision-to-delivery associated with the introduction of the time sheet.

Figure 1.

The time sheet for recording decision and actions.

Results

In the initial survey there were 73 emergency caesarean sections. 26 caesareans (36%) were performed within the standard of 30 minutes (Table 1). Of the 47 caesarean sections not achieving the 30-minute standard, 21 were found not to be emergency caesarean sections after review of the case notes. Table 2 summarises some of the clinical features, as well as the theatre transit time and the quality of note keeping, in the 26 true emergency caesarean sections. The standard of note keeping was generally poor, making it sometimes difficult to account for the delay. The main sources of delay were in transferring women to the operating theatre and in starting the anaesthetic.

Table 1.  Initial survey of decision-to-delivery interval in 73 emergency caesarean sections.
Decision-to-delivery interval (min)n(%)
30 or less26(36)
31 to 3513(68)
36 to 406(8)
41 to 509(12)
51 to 609(12)
>6010(14)
Mean time46 
Table 2.  Reasons for the delay in delivery in 26 women undergoing caesarean section. DDI = decision delivery interval; GA = general anaesthetic; TTT = theatre transit time.
 GestationAnaestheticDDI (min)ApgarsComment
140 + 5GA318,9TTT not recorded, poor note keeping
240 + 2GA324,9TTT 20 minutes
340 + 1GA324,7TTT 20 minutes, poor note keeping
439 + 3GA338,9Very poor note keeping. TTT not recorded
541GA337,9Very poor note keeping
641 + 4GA339,9Slight delay in TTT
737 + 3Spinal349,9Poor management i.e. 13 hour labour
840 + 1GA347,9TTT 20 minutes, 10 minutes to start GA
938GA366,8TTT 15 minutes, 9 minutes to start GA
1031GA365,7TTT 20 minutes, poor note keeping
1140Spinal389,9TTT 25 minutes, poor note keeping
1234 + 2GA388,10TTT 25 minutes, poor note keeping
1339 + 1Epidural399,9TTT 30 minutes, poor note keeping
1441 + 6Spinal409,9TTT 25 minutes, poor note keeping
1541 + 1GA409,9TTT 30 minutes, poor note keeping
1633 + 1GA417,715 minutes for anaesthetist to arrive, no explanation given
1741 + 1GA425,9TTT 10 minutes, BUT 20 minutes to start GA, no explanation given
1841GA449,9TTT 28 minutes, poor note keeping
1941Spinal457,9TTT 29 minutes, delay in siting spinal
2036GA506,9TTT to start GA was 41 minutes, Poor note keeping
2141GA558,9TTT 40 minutes, very poor note keeping
2241GA569,9TTT 35 minutes, poor note keeping
2340 + 2Spinal588,9TTT 40 minutes, very poor note keeping
2440 + 1Epidural609,10TTT not recorded, extremely poor note keeping
2539 + 6GA779,9TTT 30 minutes, no explanation given for starting GA so late
2638 + 1Spinal979,9Substandard management of labour, no reason given for delay with either TTT or calling anaesthetist

Table 3 shows the decision-to-delivery interval recorded in the four surveys following the introduction of the time sheet into the case notes. The results show an improvement in the number of caesarean sections achieving a decision-to-delivery interval of 30 minutes. This improvement continued with each survey, reaching a maximum of 71%. Simple linear regression shows an improvement of 10.3% per survey (95% confidence limits +5.3 to +15.2%). The χ2 for trend is 16.7, P < 0.001.

Table 3.  Decision-to-delivery interval from four surveys after the labour ward performance time sheet was introduced.
Time (min)Survey 2n= 85Survey 3n= 29Survey 4n= 86Survey 5n= 70
n(%)n(%)n(%)n(%)
≤3037(44)15(52)61(71)50(71)
31 to 357(8)6(21)12(14)8(11)
36 to 4010(12)3(10)6(7)4(6)
41 to 5012(14)2(7)4(5)6(9)
51 to 608(9)2(7)2(2)
>6011(13)1(3)1(1)2(3)
Mean time37.9 minutes33 minutes28.3 minutes28.7 minutes    

If 40 minutes is made the standard, simple linear regression shows that the improvement is non-linear, F (df 2266) = 3.0, P= 0.05. An asymptotic regression is a significantly better model F (df 1268) = 5.4, P < 0.025. This indicates that repeated audit cycles show a dose–response relationship with an asymptote of 90.7% achieving the 40-minute standard.

Discussion

There is a generally agreed recommendation that emergency caesarean sections should be achieved within 30 minutes of the decision to operate4–6. However, this does not appear to be an evidence-based rule. The Confidential Enquiries of Stillbirths and Deaths in Infancy in the UK recommend that all hospital trusts will be required to audit this standard. Our first survey showed that in the majority of emergency caesarean sections the decision-to-delivery interval was more than 30 minutes. This arbitrary 30-minute limit is not based on scientific studies, but has the approval of respected authorities, including medico-legal bodies7,8.

We have then shown that significant improvement in the decision-to-delivery interval can be achieved by the introduction of time sheets. We hoped that the fifth survey would show continued improvement, but the results were no better than in the fourth survey.

The problem of failed epidural or spinal anaesthesia, or inadequate ‘top-ups’, will always cause delays that cannot be compensated. In our unit, failed regional anaesthesia occurs in less than 3% of caesarean sections and we have a firm policy of resorting to general anaesthesia if 15 minutes after the decision for caesarean section regional anaesthesia is inadequate.

Two main issues arise from our study. The first is the definition of emergency caesarean section. One could argue that varying degrees of emergency would have resulted in different degrees of urgency in the response of the midwifery and obstetric staff. In our first survey, about half of the caesarean sections where the decision-to-delivery interval was greater than 30 minutes were reclassified as non-emergency caesarean sections after review of the case notes. The Standing Joint Committee of the Royal College of Obstetricians and Gynaecologists and Royal College of Anaesthetists, classifies ‘emergency’ as ‘urgent’, defined as maternal or fetal compromise which is not immediately life-threatening. This definition agrees with that in a recent report by obstetric anaesthetists and encompasses all the caesarean sections in our audit, which were described as ‘emergency’9.

The next issue is whether a more lenient standard should be adopted. There is no scientific evidence that failure to achieve the 30-minute target is associated with an adverse neonatal outcome10–12. The evidence suggests that achieving the 30-minute standard in fetal distress does not benefit the infant7. Chauhan et al.13 suggest that an interval between decision and incision in some instances may allow intrauterine resuscitation of the infant. An investigation into assisted vaginal delivery for fetal distress did not show that a decision-to-delivery interval of longer than 30 minutes had adverse consequences for the infant14.

In our opinion a standard of 40 minutes for the decision-to-delivery interval is more realistic and can be achieved in more than 90% of emergency caesarean sections (Table 3).

Acknowledgements

The authors would like to thank Ms M. Van Dyke and all her midwife colleagues who helped collect data. We would also like to thank Dr J. J. Jones for his assistance with the statistical analysis.

Ancillary