  |
|
|
||||
|
||||||
| Journal of the New Zealand Medical Association,
26-November-2004, Vol 117 No 1206
|
||||||
|
Caesarean section in term nulliparous women at
Wellington Hospital in 2001: a regional audit
Michel Sangalli, Alice Guidera
There is an increasing tendency to perform caesarean
sections in most industrialised countries, although there are marked differences
among countries in both the timing and the rate of increase in caesarean section
(CS) rates.1 The reasons for this global
increase in caesarean sections are multiple and complex, and what constitutes a
‘good indication’ for a CS is clearly a matter of controversy.
Greater customer expectations and choice, older maternal age, smaller family
units, inadequate models of care, and fear of litigation represent some aspects
of modern pregnancy care that have been associated with the increase in CS
rate.2
In the USA, where the increase in CS rate was first
observed, the CS rate reached 24% in 1990, then stabilised and even fell
somewhat to 22% in 1999 following initiatives taken to stabilise or reduce the
CS rate.3,4 These initiatives have focuses on
decreasing the rate of first CS and increasing the rate of vaginal deliveries
after CS.2 In New Zealand, the CS rate rose to
22.1% in 2001.5
In the year 2000, at Wellington Hospital, a tertiary
referral centre, the local CS rate has reached 26.6% for all women, one of the
highest in the country.
The aims of this audit were to determine the frequency of CS
in the Wellington region for term nulliparous women, to evaluate the local
demographic and clinical factors associated with CS, and, to assess the quality
of clinical care against agreed standards derived from published literature.
MethodsNulliparous women with a
singleton live pregnancy with a gestational age ≥36 weeks (who delivered
in Wellington Hospital during 1 January 2001 to 30 June 2001) were identified by
Capital and Coast District Health Board’s Perinatal Information Management
System.
Data was extracted from the Perinatal Information
Management System for women who delivered vaginally and directly from the
hospital records of women who had a CS. The data included: maternal age,
gestational age, induction of labour (IOL), epidural anaesthesia for women in
labour, birthweight, neonatal outcome (live/stillborn), and the presence of
severe asphyxia.
For women delivered by CS, ethnicity and indication for
CS was also noted. Severe asphyxia was defined as evidence of metabolic acidosis
in intrapartum fetal scalp blood, umbilical arterial cord, or on very early
neonatal blood sample (pH <7.00 and base deficit >12 mmol/L) with early
onset severe or moderate neonatal
encephalopathy6.
Ethnicity was self-reported as defined by the Ministry
of Health, and grouped as European/other, Maori, Pacific Island, or Asian.
Caesarean section was classified as either an elective or an emergency
intervention. Indications for CS were divided in the following categories:
dystocia (including failure to progress during the latent phase of labour,
failed IOL, failure to progress during the active phase of the first stage and
during the second stage of labour), suspected fetal compromise, malpresentation,
placenta praevia/antepartum haemorrhage, and ‘other’.
An indicated CS was defined as a CS performed in the
presence of at least one of the below-defined indications using a recognised
standard from the literature. All cases where the clinical indication was not
clearly documented (or the criteria for an indicated CS was not clearly met)
were reviewed independently by two obstetricians, and classified either as
either ‘indicated’ (one or both of the obstetricians found that a
criteria for an indicated CS was met) or ‘not indicated’ at the time
of the decision.
Considerations included a careful review of the
partogram and notes describing the progress of labour and the nature and quality
of obstetric intervention, records of clinical assessments, inspection of
cardiotocograph (CTG), and the feasibility of fetal blood sampling (FBS). FBS
was considered feasible if the cervix was ≥4cm
dilated.1 For simplicity of assessment, time
was rounded up to the nearest hour. If documented, the woman’s request for
a CS in the absence of a medical indication was recorded.
Malpresentation was a recognised indication for CS when
the fetus was in breech, transverse, brow, or face presentation at the time of
labour/delivery. Furthermore, when CS was performed for a breech presentation,
it was noted whether external cephalic version (ECV) was offered or attempted
prior to labour.7
Consensus criteria for dystocia (including failed
labour induction, failure to progress during the latent phase, or first stage
and second stage of labour) have not been clearly
established.8–10 However, one of the
recognised auditable standards for failure to progress in labour is that
oxytocin should be used in the management of nulliparous women with suspected
failure to progress in labour prior to CS
(Standard
1).11
To obtain a credible crude estimate of the lowest
possible CS rate for dystocia, a pragmatic approach (taking account of the
current maternity system in New Zealand) was used.
Three factors were taken into account.
Therefore, the least time
stringent recognised standards of management of dystocia were chosen because
they fit best with the current model of labour care. In this study, failed
induction of labour and failure to progress during the latent phase of labour
were defined as an indication for CS when despite ruptured membrane and 6 hours
of oxytocin perfusion the cervix did not reach a dilatation >3cm
(Standard
2).10
The indication for CS for failure to progress during
the active phase of labour was defined as failure to progress when there has
been an arrest in dilatation, despite ruptured membranes and oxytocin infusion
for over 4 hours (Standard
3).9 Failure to progress in the second
stage of labour was an indication for CS if the baby has not been delivered
vaginally after 2 hours with oxytocin augmentation for some time during the
second stage (Standard
4).10
Fetal compromise was a recognised indication for CS if
the CTG was frankly pathological (including any combination of prolonged
bradycardia, repetitive variable/late decelerations, tachycardia with reduced
variability) or when a non-reassuring CTG was associated with either a fetal
blood sample showing a pH <7.20 or a specific clinical situation (eg,
placental abruption).12 It was also established
whether a cord pH result was available in the mother or infant’s
notes.1,12
For statistical analysis, Yates corrected test was used
to evaluate discrete variables, and the Fisher exact test (two-tailed) was used
when any cell contained an expected value of less than five. For normally
distributed continuous variables, the Student t-test was used. P<0.05 was
considered significant.
ResultsBetween 1 January 2001 and 30 June
2001, 743 women at a gestational age ≥36 weeks (with a singleton live
pregnancy) gave birth to their first infant at Wellington Hospital. This total
includes 51 women who gave birth vaginally in two peripherical birthing units
without facilities for CS.
Most women delivering at Wellington Women’s Hospital
were under the private care of their lead maternity carer (LMC), which include
independent midwives, GP obstetricians, and specialist obstetricians. The
majority of LMCs were independent midwives, and specialist team (SHO, registrar
and/or specialist) consultation was required before any significant obstetric
intervention.
Of those 743 women, 209 gave birth by CS, thus giving an
uncorrected CS rate of about 28%. The number of first deliveries outside
hospital settings during that period is unknown but is estimated to be low
(~3%)5, thus giving an estimated minimum
corrected CS rate for the Wellington region of about 27%. The hospital file of 8
women was not available, thus leaving a study population of 201 women for some
of the data.
Ethnicity was reported as either European/other (75%), Maori
(8%), Pacific Island (7%), or Asian (10%). There was no statistically
significant difference in CS rate among women of different ethnicity. The
epidural rate of all nulliparous women (expecting their first infant [elective
CS excluded] with a gestation ≥36 weeks) was 67%. The women delivered by
CS were statistically significantly older by an average of 2 years (30.8 ±
5.8 years versus 28.7 ± 5.6 years) than women who gave birth vaginally.
Women aged ≥40 years were significantly more frequent
in the CS group (10/209 versus 4/534, OR 6.66 [1.90–25.48]). The average
gestational age at delivery was 39.5 ± 1.5 weeks, and did not differ from
women with a vaginal delivery. The proportion of women with a gestational age
>41 completed weeks was not a statistically different amount between the 2
groups. Induction of labour was significantly more frequent in the CS group
(105/201 versus 177/534, OR 2.21 [1.56–3.11]).
Birthweight did not differ statistically between the 2
groups (3410 ± 452 g for the [CS] versus 3477 ± 582 g [vaginal
delivery]). Infants with a birthweight >4000 g were significantly more common
in the CS group (32/209 versus 39/534, OR 2.11, 95% CI 1.35–3.88).
Table 1. Indication for CS (caesarean section)
deliveries
Table 2. CS (caesarean section) for dystocia
Infants with a birthweight <2500 g were not statistically
significantly more common in the CS group (10/209 versus 14/534). In the vaginal
and CS birth groups, there were no stillbirths or infants with severe birth
asphyxia. Table 1 shows the indications for CS in the study
population.
Results for dystocia are shown in Table 2. Overall, 88% of
women who failed to progress in labour (IOL excluded) received oxytocin
(Standard 1). Among the 23 women who
had an emergency CS for failed IOL, 6 had prostaglandins only; 4 had artificial
rupture of the membranes (ARM) and received oxytocin; and 13 had prostaglandins,
ARM, and oxytocin.
Standard 2 compliance for women with a CS for failed IOL
(who received oxytocin) or with failure to progress during the latent phase of
labour was 85%. Standard 3 compliance for women with a CS for failure to
progress during the active phase of the first stage of labour was 48%. Among
women with a CS for failure to progress during the second stage of labour, 78%
(29/37) received oxytocin, 86% (32/37) had a second-stage duration ≥2
hours, and 32% had a second-stage duration ≥ 3 hours (12/37). Compliance
with standard 4 was 65%.
Overall, out of 46 women with a CS for suspected fetal
compromise, 20 had a FBS of which 5 failed. Among the 15 women who had a FBS, 3
had a pH <7.20 before CS. Among the 26 women who did not have a FBS before CS
for suspected fetal compromise, 3 declined to have the test; in 1 case, an
immediate CS was indicated due to a severely pathological CTG; and 9 women were
<4 cm dilated and therefore ineligible for a FBS.
Therefore, an attempt to obtain a FBS was made in 64%
(23/36) of eligible women. After CS for suspected fetal compromise, a cord pH
result was available in the notes of 78% of the women (36/46).
A total of 40 CS were performed for malpresentation,
including 26 elective CS (25 breech, 1 oblique lie; 40/743, 5.3%) and 14
emergency CS (13 breech and 1 brow presentation). Out of the 38 women with a CS
for breech presentation, 5 were thought to have a contraindication to ECV, 7 had
an unsuccessful attempt at ECV, 1 was booked for an ECV, and 2 declined an ECV.
No documentation about a discussion about ECV was found in 23 sets of notes.
Compliance with the (ECV) standard was 33% (11/33).
DiscussionThis audit from Wellington Hospital,
a tertiary referral centre, shows that the CS rate for primiparous women with a
gestational age ≥36 weeks is about 27%. This rate is similar, although
higher, to the rate (24%) published for England, Wales, and Northern Ireland in
2000–2001 in the National Sentinel Caesarean Section Audit Report
(NSCSAR).1
As in other reports from the literature from various
institutions and countries, dystocia (48%), suspected fetal compromise (23%),
and malpresentation (20%) represented the most common indications for
CS.1,2 In line with other studies, an
association with advanced maternal age (≥40
years)13, large babies, and
IOL14 was also shown, but these factors were
only found in a minority of women.
In a consensus statement from 1985, the World Health
Organization (WHO) suggested that there were no additional health benefits
associated with a CS rate above 10%–15%. However, a recent report looking
at obstetric outcome data from three large free-standing hospitals in Dublin
over a 22-year period showed a remarkably consistent association between an
increasing CS rate and a falling mortality rate. Also, the hospital with the
highest CS had consistently the lowest mortality
rate.15 Indeed, this has been the feeling of
many obstetricians over the years.
Despite the WHO statement and CS being one of the most
commonly performed major surgical interventions, there is, currently, no clear
agreement of what constitutes a ‘good indication’ for CS. Instead, a
multitude of objective and subjective factors (including personal values) are
considered when the option of a CS is discussed with the woman in labour and her
partner/support person(s).
Thus, several ‘reasonable’ standards derived
from the literature have been chosen to assess the quality of clinical care
received by Wellington women during the study period. These standards have been
chosen because they are the most compatible with current practice, and are
associated with a high rate of vaginal deliveries without compromising the
health of the mother and infant.
For dystocia, most (88%) women in this study received
oxytocin during established labour, similar to the NSCSAR audit
(81%).1 However, it appears that in about
one-third of cases, this decision could potentially have been safely delayed,
thus allowing opportunity for further progress.
The standards chosen for dystocia reflect either the
treatment received (oxytocin), or both oxytocin and the time allowed for labour
progression, and clearly do no take account of any other factors that may be
pertinent at the time of the decision. Due to the design of this study, it was
not always possible to identify accurately why a CS was performed at the time of
the decision. However, it should be noted that the indication for CS was only
rarely documented as a request from the women against professional advice.
One of the relevant factors in the management of labour
dystocia is the presence and duration of ‘good quality’ uterine
contractions, which is the determining factor to achieve a satisfactory progress
in labour. Indeed, the use of oxytocin is only a surrogate marker for good
contractions, and such assessment requires competent assessment during labour,
and perhaps an intra-uterine pressure catheter.
Clearly, if there is an identified need to
stabilise/decrease the CS rate for dystocia, strategies will need to focus on
public education and on the model of care, acknowledging the fact that the
current maternity system was clearly ineffective in limiting the increase in CS
rates. Efforts at the level of each maternity unit are also required to optimise
the diagnosis and management of slow progress in labour, keeping in mind that
the hospital team has often only a limited input in the women’s global
decision-making process, and in the management of dystocia, which starts many
hours before the decision for CS.
Suspected fetal compromise was the indication for 23% of CS
operations, and about two-thirds of eligible women were offered a FBS, which is
higher than the NSCSAR result (44%),1 probably
because Wellington Hospital is a tertiary referral centre. Seventy-eight percent
of women who had a CS carried out for presumed fetal compromise had a documented
cord pH result in the hospital file, which is similar to the NSCSAR result
(82%).1
In many cases, women were not eligible for FBS because of
lack of cervical dilatation, and in a significant proportion of eligible women,
it was not offered. An alternative in selected cases may be the use of fetal
scalp or vibroacoustic stimulation to check for fetal
wellbeing.16,17 Again, patient and LMC
education as well as staff training (in technical aspects of FBS collection and
handling) may increase the acceptability and success rate of FBS.
However, it is noteworthy that many CS were performed
despite a reassuring pH result. Again, due to the study design, it was not
possible to clearly identify the reasons for the decision to proceed with the
CS. An important fact remains that the clinical decision is usually based on the
likelihood of fetal acidosis and insult, which, in presence of a persistently
non-reassuring CTG increases proportionally to the time taken to deliver the
fetus.
It is currently not a New Zealand standard, even in large
units, to have a dedicated fully staffed anaesthetic and obstetric team on site
24/24hours. This situation is likely to explain, at least partly, why FBS were
not offered in all cases and why it was decided to proceed with a CS despite a
reassuring FBS in the presence of a non-reassuring CTG rather than careful
observation or performing a second FBS.
Almost all CS (95%) for malpresentation were due to breech
presentation. With caesarean section being the currently favoured mode of
delivery,18 external cephalic version has
become a recommended option and an auditable standard for the management of term
breech.1,7 This was offered in about one-third
of eligible women, which is similar to the results of the NSCSAR
audit.1
ECV was also commonly declined. Again, women and LMC
education allowing timely identification of breech fetuses with referral to
skilled ECV teams may have an impact on the number of CS performed for breech
presentation.
In conclusion, this audit looking at the CS rate among women
having their first baby in the Wellington region in 2001 shows that the CS rate,
the indications for CS, and the compliance with recognised standards from the
literature are very similar to the situation in other countries.
Dystocia, suspected fetal compromise, and malpresentation
represent the most common indications for CS. Based on the result of this audit,
it is estimated that better compliance with recognised standards could
potentially allow a decrease in the CS rate from 27% to about 18%–22%.
However, this process is likely (as overseas) to be long, complex and difficult.
Therefore, significant changes in the current maternity system in New Zealand
are probably required to efficiently address these difficulties.
Author information:
Michel Sangalli, Consultant Obstetrician and Gynaecologist; Alice Guidera, House
Surgeon, Women’s Health Service, Capital and Coast District Health Board,
Wellington South
Acknowledgements:
This summer student project was sponsored by The Wellington Medical Research
Foundation. We thank Dr J Tuohy for reviewing relevant hospital files and for
his expert opinion.
Correspondence:
Michel Sangalli, Consultant Obstetrician and Gynaecologist, Women’s Health
Service, Wellington Hospital, Private Bag 7902, Wellington South. Fax: (04) 385
5862; email: Michel.Sangalli@ccdhb.org.nz
References:
|
|
||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|