Subsequent obstetric performance related to primary mode of delivery

British Journal of Obstetrics and Gynaecology March 1999, Vol 106, pp. 227-232

Subsequent obstetric performance related to primary mode of delivery John Jolly Specialist Registrar;James Walker Professor;*Kalvinder Bhabra Consultant St Juinesk University Hospital, b e d s

arid *HuddersfieldRoyal In/irmary

Objective To relate subsequent obstetric performance with primary mode of delivery. Design Postal questionnaire survey of women who delivered their first child five years ago. Setting Huddersfield Royal Infirmary. Population Women who were delivered of their first baby in 1991:250 by normal vaginal delivery; 250 by instrumental vaginal delivery; and 250 by caesarean section. Main outcome measures Answers to fixed choice questions on fear of future childbirth, number of subsequent children and reasons for no further children. Results The response rate was 64%.Overall, 222 (4606%)women were initially frightened about future childbirth. According to mode of delivery: 93 (57.1%) after instrumental vaginal; 79 (47.9%) after caesarean section; and 50 (33.8%)after normal vaginal delivery. Five years after the primary delivery, 99 women (20.8%) were still frightened about future childbirth: instrumental vaginal group 41 (25.2%); caesarean section group 43 (26.1%);and normal vaginal group 15 (10.1%). In the group of women who were delivcred by caesarean section 13% more women had not had a second child after five years compared with the normal vaginal delivery group ((P c 0-03, relative risk 1.46 (1.07-1.99)). In the group of women who had a vaginal instrumental delivery 6%more had not had a second child after five years compared with normal vaginal delivery group. Of the women who had no further children, 30% who had caesarean section and 28% vaginal instrumental delivery had involuntary infertility. Conclusions Caesarean section or vaginal instrumental delivery leaves many mothers frightened about future childbirth. Primary caesarean section and to some extent vaginal instrumental delivery is associated with an increased risk of voluntary and involuntary infertility.

INTRODUCTION Over the last 25 years caesarean section has become an increasingly common method of delivery. In 1970 the caesarean section rate in the United Kingdom was reported to be 4.8%‘.The Audit Commission Report of Maternity Services in 1997? found this rate increased to 11%-18%. The most commonly quoted reason for performing caesarean section, particularly in the USA7, is a ‘fear of litigation’. Also, the operation is now safer than in the past because of improvements in anaesthesia, antibiotics and blood transfusion‘. Hence doctors often perform a caesarean section, even when the evidence for improved outcome is uncertain, such as breech presentation. However, a caesarean section still cames significant risk to the mother compared with a normal vaginal delivery’. The increase in the Correspondence: Mr J. F. Jolly, Department of Obstetrics and Gynaecology, Clinical Sciences Centre, Northern General Hospital, Henies Road, Sheffield S5 7AU, UK. 0 RCOG 1999 British Journal of Obstetrics and Gynaecology

caesarean section rate has not had a significant effect on the perinatal mortality rate, which has improved due to other factors, such as better perinatal screening and social economic changes. Studies have shown that primiparous women who have caesarean sections have fewer children and more difficulty conceiving than control^^.^. This subfertility is greatest if there has been a post-operative pelvis abscessx.Although the past teaching of ‘once a caesarean always a caesarean’ no longer applies, there is still a sizeable ‘repeat section’ rate. This has been reported as 80%-95% in the USA’.’” and 56% in the UK”. There have been few studies of long term psychosocia1 consequences following caesarean section. One study reported, after four years follow up, increased fatigue and an increased need for psychological counselling in caesarean section women compared with controls7.We were interested to study how delivery by caesarean section may affect psychological morbidity. With increasing concern about litigation, any research that will elicit areas that will lead to better maternal 221

228

J . J O L L Y ET A L

satisfaction with the birthing process will certain help professionals caring for pregnant women. The aims of the study were to assess fertility and psychological morbidity of women according to mode of delivery of their first child. To exclude the immediate effects of post-operative problems and postnatal depression, we questioned women five years after their first delivery. We were also interested in the fertility rates in these mothers and the mode of delivery of any subsequent infants. We developed a questionnaire to examine women’s experiences following caesarean section and current feelings about future childbirth. To control for other social and psychological factors the questionnaire was sent to a group of women who had normal vaginal deliveries and a second group who had had vaginal instrumental deliveries. The null hypothesis was that women who had been delivered of their first child by caesarean section felt no different than women who had had normal vaginal deliveries or instrumental deliveries. The feelings of the mother were assessed on how they answered questions about the birth experience and whether they were disappointed and frightened about having another baby.

METHODS We collected, from the delivery record book, the names and addresses of 250 nulliparous women who had caesarean sections in Huddersfield Royal Infirmary. The data were sequential from January 1991. The details were checked and updated with the hospital computer records (Master Patient Index). There were two control groups: nulliparous women who had normal vaginal deliveries and nulliparous women who had vaginal instrumental deliveries. The details of these women were randomly selected over the same time period from the delivery record book. In the vaginal instrumental group it was necessary to include some women who were delivered in 1992 in order to collect 250 cases. We collected data from the delivery records on age, gestation at delivery and indication for delivery. Women who had had multiple pregnancies, stillbirths, neonatal deaths and home births were excluded. A separate hospital record of perinatal and infant deaths for 1991-1993 was also inspected to ensure that these mothers were excluded. A questionnaire was developed to assess each mother’s ‘feelings’ regarding the birth of their first child. The first part asked about the experience of the birth and the second part asked about subsequent children born in the next five years. At the end of the questionnaire mothers were invited to add any comments. The questionnaire was tested on a group of 20 mothers to assess suitability for the target population, and then

adjusted to ensure that all questions were unambiguous and simple to understand. Each questionnaire was posted with a stamped, addressed envelope and a covering letter explaining the study and advising the women that their answers would be treated confidentially. A translation in Urdu was available on request. The questionnaire was mailed again to those who had not replied after six weeks. Women who had still not returned the questionnaire after a further six weeks were then contacted by telephone, if a number was available from the Master Patient Index, and were told that their answers would be treated confidentially. The telephone interviews were conducted by one of the authors (J. J.) and followed the questions and fixed choice answers as in the questionnaire. The software package Epi info 6 was used for data management and statistical analysis. Groups were compared using the independent t test. With approximately 160 respondents in each group, the study had >95% power to detect a statistically significant difference at the 5% level between any two groups if the real difference in the answer to any question was, for example, 10%vs 30% (e.g. Question 1: normal delivery ws instrumental). The local hospital ethics committee approved the study.

RESULTS Of the 750 questionnaires posted, 523 replies were received after two mailings. Of these, 56 (7.5%)were returned unanswered because the woman was not known at that address. Telephone numbers for 68 women who had not replied after two mailings were obtained from the Master Patient Index; 18 of these women were contacted. Of these, 13 were willing to complete the questionnaire over the telephone. For the remainder, there was either no answer (n=26), the telephone line was disconnected ( n = 17) or they had moved (n = 7). The completed questionnaire response rate 64.0%. There were more women who responded following caesarean section 169 (67.6%)compared with instrumental delivery 163 (65.2%) and normal delivery 148 (59.2%), although this was not statistically significant. The mean age of women who did not respond tended to be less than women who did reply: normal (23.2 vs 26.0 years): instrumental (25-7 ws 27-9 years) and caesarean (25-2 vs 26.7 years). The mean gestation at delivery for women who did not respond was similar to those women who did reply: normal (39.1 vs 39.6 weeks); instrumental (39-7 ws 39.7 weeks) and caesarean (37.9 ws 38.5 weeks). Twenty-two women (9.1%) who returned the questionnaire had preterm deliveries (before 36 weeks of gestation) compared with 10 (12.0%) of nonresponders. The indications for 0 RCOG 1999 Br J Obstet Gynaecol 106,227-232

SUBSEQUENT CHILDBIRTH A N D P R I M A R Y M O D E OF DELIVERY

caesarean section were: fetal distress 65 (38.4%); breech presentation 46 (27.2%); failure to progress 44 (26.0%); hypertension 5 (2.9%); and others 9 (5.3%). Four women were further excluded from the analysis because they had failed instrumental deliveries followed by emergency caesarean section. Of the caesarean sections, 139 (82.2%) were recorded as emergency and 30 as elective (17.8%). Of the nonresponding women who had caesarean sections, 86.7% were recorded as emergency and 13.3% as elective. General anaesthetic was used in 136 cases (8004%). Regional anaesthesia, either spinal or epidural, was used in the other 33 cases. There were 107 (65.6%) deliveries with straight forceps (Neville Barnes, Wrigleys or Haigh Ferguson), 38 (23.3%) vacuum extractions, and 15 (9.2%) rotational forceps (Kjelland’s). Three babies were delivered by vacuum followed by Neville Barnes forceps. Table 1 shows details of age of the women and gestation. The mean age for the three groups was similar. The mean gestation at delivery was similar for normal and vaginal instrumental deliveries. The mean gestation at delivery for the caesarean section group was approximately one week earlier. Table 2 shows the women’s responses to questions regarding feelings about the birth of their first child. Women who had vaginal instrumental deliveries or caesarean sections were more likely to report a bad birth experience, were more disappointed, and were more frightened about having another baby than women who had had normal deliveries. All these differences were statistically significant (P < 0.02). There tended to be a higher proportion of women in the vaginal instrumental group who were initially disappointed, frightened and

229

reported a bad birth experience than the caesarean section group. However, over five years the numbers of mothers still frightened about having another baby decreased and seem to be similar between the instrumental and caesarean groups. Table 3 reports subsequent childbirth in the five years after the birth of the first child. Caesarean section delivery was associated with more women who had no further children after five years compared with normal delivery. This difference was statistically significant (P c 0.03). There were also fewer women who had two or more further children over five years following caesarean section. After a primary vaginal instrumental delivery, women who had a second child were just as likely to have a third as those women who delivered their first child normally. This did not seem to be the case following a primary caesarean section. In the caesarean section group 63.8% of women had subsequent children delivered by caesarean. This compares with 9.5% in the instrumental delivery group and 3.9% of mothers who had a normal delivery. Women who had not had any subsequent children were asked to tick a box for the most appropriate reason or specify another reason (Table 4).Mothers who had their only child by caesarean section were more likely to have tried but not been successful in having further children than mothers who had normal deliveries. Also they were less likely to want to go through childbirth again than normal deliveries. This difference was statistically significant (P < 0.05). This trend was similar for women who had vaginal instrumental deliveries but not to the same extent. Conversely, women who had normal deliveries were more likely to report relationship problems as their

Table 1. Population details. Values are given as mean (95%CI) [range]. Normal

Age (years) Gestation at delivery (weeks)

(n = 148)

Instrumental (n = 163)

Caesarean ( n = 165)

26.0 (25.3-26.7) [16-391 39.6 [3242]

27.9 (27.1-28.7) [I6431 39.7 [33-42]

26.7 (25.9-27.5) [16-391 38.5 [ 2 8 4 l ]

Table 2. Feelings about birth experience. Values are given as n (%) [95%CI]. Normal

(n = 148) 1. Bad birth experience

14 (9.5) [4.8-14.2]

2. Disappointed following birth

28 (18.9) [12&25.2]

3. Initially frightened about having another baby 4. Frightened nbout having another baby 5 years on

50 (33.8) [26.2-41.4] 15 (10.1) [5.27-15.0]

0 RCOG 1999 Br J Obstet Gynaecol 106,227-232

Instrumental (n = 163)

Caesarean (n = 165)

46 (28.2) [21.3-35.1] P < 0.0001 86 (52.8) [45.1-60.4] P < 0.000I 93 (57.1) [49.744.6] P c 0-000I 41 (25.2) [18.5-31.8] P < 0.001

37 (22.4) 116.1-28.81 P < 0.005 81 (49.1) [41.5-56.7] P < 0.0001 79 (47.9) 140.3-55.51 P < 0.02 43 (26.1) [194-32.8] P < 0.000 1

230

J . JOLLY ET A L

Table 3. Childbirth in the five years after the birth of the first child. Values are given as n (%) [95%CI].

1. None

2. One 3. Two or more 4. Total no. of children

Normal (n = 148)

Instrumental (n = 163)

Caesarean (n = 165)

43 (29.1) [21,7-36.4]

57 (35.0) 127,642.31

84 (56.8) [48.844.7] 18 (12.2) t6.9-17.41 I23

87 (53.4) [45.7-61.0] 18 (11.0)[6.2-15.91 I24

70 (42.4) [34.9-50.0] P < 0.03 85 (5 1.5) r43.9-59.11 9 (5.5) [2.0-8.9] 104

Table 4. Mothers' with one child, reasons for no further children. Values are given as n (a). Instrumental (n = 57)

Caesarean (n = 70)

7 (16.3) 2 (4-7)

16 (28.0) 6 (10.5)

3. Only planned one child

16 (37.2)

14 (24.6)

4. Relationship problems

12 (27.9)

5 (8.8) P < 0.03 16 (28.0)

21 (30.0) 13 (18.6) P < 0.05 I1 (15.7) P < 0.02 5 (7.1) P