Birthing experience and quality of life after vacuum delivery and second‐stage caesarean section: a prospective cohort study in Uganda

To assess perceptions of women undergoing vacuum extraction or second‐stage caesarean section (SSCS) in a tertiary referral hospital in sub‐Saharan Africa.


Introduction
Increased use of vacuum extraction or other modes of assisted vaginal birth could potentially prevent many maternal deaths caused by complications of prolonged labour or unsafe caesarean section and a large proportion of stillbirths and neonatal deaths in sub-Saharan Africa [1][2][3][4][5][6][7][8].
Indications for vacuum extraction are prolonged second stage of labour, foetal distress, maternal exhaustion or the need to avoid expulsive efforts in maternal conditions such as severe anaemia or heart failure [9][10][11][12]. Use of vacuum and forceps-assisted vaginal birth varies from more than 10% in northern Europe to <1% in many places in sub-Saharan Africa [2,13]. In many sub-Saharan African countries, surgery carries higher risks due to unsafe anaesthesia and unavailability of blood for transfusion [6,14,15]. A uterine scar may lead to uterine rupture and abnormal or invasive placentation in a subsequent pregnancy [16]. Women with a uterine scar may not be aware of these risks and try to deliver at home [6]. Recent publications suggest that caesarean section in the second stage of labour increases the risk of spontaneous preterm birth in the next pregnancy [17,18].

914
Therefore, particularly in these settings, preventing caesarean section is important: this view is supported by WHO and other international reproductive health and maternity care leaders [19][20][21]. Thus, vacuum extraction, the simplest method of assisted vaginal birth [9], was re-introduced in the main teaching hospital in Uganda in 2012. After re-introduction, clinical maternal and perinatal outcome improved significantly with considerably fewer intrapartum stillbirths and uterine ruptures [3].
In addition to clinical outcomes, birthing experience, pain, ability to work and pain-free sexual intercourse are also very important birth outcomes, particularly from a woman's perspective (and influencing her decision on where to give birth the next time). These outcomes, in women who had assisted vaginal births, have only been studied in high-income settings [22][23][24][25][26][27][28][29].
The aim of this study was to assess how vacuum extraction was experienced by women after its re-introduction in a tertiary referral hospital in sub-Saharan Africa, using women-centred outcomes such as birthing experience satisfaction; pain one day after birth; and quality of life, pain and dyspareunia six weeks and six months after birth. Outcome is compared with outcome after second-stage caesarean section (SSCS), which is another intervention in case of delay or failure to progress in the second stage of labour.
This study was part of a larger study that also investigated clinical maternal and perinatal outcome after vacuum extraction and SSCS [6]. Vacuum extraction (including failed vacuum extraction and subsequent caesarean section) had better maternal outcomes than SSCS and equivalent perinatal outcomes. The odds ratio for severe maternal complications after vacuum extraction vs. SSCS was 0.24 (95% CI 0.07-0.84). The odds ratio for perinatal death was 0.83 (95% CI 0.49À1.41).

Participants
This was an observational, prospective cohort study, comparing women-centred outcomes after vacuum extraction and SSCS. The study was conducted in the main labour ward of Mulago National Referral Hospital, Kampala, Uganda. The inclusion period was from 25 th of November 2014 to 8 th of July 2015, and women were interviewed one day, six weeks and six months after birth. Included were women who gave birth to a term singleton in cephalic presentation by vacuum extraction or SSCS and who consented to being included. Excluded were women who had a failed vacuum extraction followed by SSCS due to the inability to ascribe outcomes to either procedure, and women who had severe complications (defined as maternal death, uterine rupture, re-laparotomy, obstetric fistula and eclampsia) or perinatal death that had occurred before the moment of inclusion at one day after birth (Figure 1). Women whose babies had died before six-week or six-month follow-up were excluded from analysis of outcome at that timepoint. Clinical outcome of women and neonates excluded here is described elsewhere [6]. Vacuum extraction was compared to SSCS as the alternative treatment option. Forceps was hardly used in this hospital (fewer than five times during the study period).

Setting
Mulago Hospital is the national referral and main teaching hospital of Uganda. It is a government hospital with 2700 beds and more than 31 000 births annually, with the capital Kampala and surroundings as catchment area. Maternity services are free of charge. Vacuum extraction was by Kiwi vacuum extractor (Clinical Innovations, South Murray, Utah, US) or Bird and silicone cups with hand and foot pumps. Regional analgesia during labour or assisted vaginal birth was not used. Episiotomy for vacuum extraction was not routinely performed. If performed, local lidocaine infiltration was used, if available. SSCS was by lower abdominal transverse incision (Pfannenstiel, Joel-Cohen or modification) or subumbilical midline incision. SSCS was most often performed under spinal anaesthesia. Post-vacuum analgesia was by paracetamol and NSAIDs. Post-caesarean section analgesia was by pethidine, tramadol, paracetamol and NSAID's, when available. Women were normally discharged at the first day after vacuum extraction or the third day after SSCS.

Inclusion process
On the first day after birth, women who had given birth by vacuum extraction were identified from the hospital's birth register by a member of the research team. Women who had had a caesarean section were identified from the operating theatre register. To identify those women who had been in the second stage of labour during caesarean section, their medical records were examined. Women who fulfilled the inclusion criteria were asked to participate in the study on the first day after birth. After obtaining written informed consent, the woman was interviewed by a trained research assistant in either English or Luganda, which are the most commonly spoken languages in Kampala. Baseline characteristics were extracted from the medical records.

Birthing experience interview
The birthing experience assessment consisted of a short interview, based on four questions, each with four fixed response options. Questions addressed how women had experienced giving birth (Appendix S1). A numeric pain rating scale (NPRS, scale 0-10) was filled [30].

Follow-up
Postnatal consultations took place six weeks and six months after birth. During these visits, women were interviewed using structured questionnaires (Appendix S2). The questionnaires consisted of the SF-12v1 questions, pain scores (NPRS 0-10) and a question concerning dyspareunia [30][31][32]. Women who missed postnatal consultations were interviewed over the phone, using the same questionnaire, from which SF-12 questions were excluded. Therefore, groups with analysis of SF12-questions are smaller.

SF-12 questionnaire
The Short Form 12 (SF-12) questionnaire contains 12 items and is based on the original SF-36 Health Survey [30,31]. It is a norm-based generic measure to assess health-related quality of life, widely used and psychometrically robust [31]. SF-12 measures general physical and mental health status and is not specific for age, disease or health condition. It assesses eight physical and mental health dimensions: physical functioning, role limitations due to physical health problems, bodily pain, general health, vitality, social functioning, role limitations due to emotional health problems and mental health. From these eight dimensions, a physical component summary (PCS) and mental component summary (MCS) are calculated. The tool is designed such that mean PCS and MCS (in the US population) are 50 with a standard deviation (SD) of 10. High scores indicate better subjective health functioning (PCS) and emotional well-being (MCS). The SF-12 questionnaire is used worldwide and has been translated in more than 100 languages [31,33].

Outcome measures
Outcome measures were birthing experience satisfaction (based on the answers from the interview at the first day after birth), quality of life (mean PCS and MCS scores from the SF-12 questionnaire and answers per question at six weeks and six months after birth); pain during the 916 procedure and at one day, six weeks and six months after birth (using mean and stratified pain scores; 0: no pain; 1-4: mild pain; 5-7: moderate pain; 8-10: severe pain); percentage of women who reported they had resumed sexual intercourse at six weeks and six months, and percentage of women with dyspareunia at these timepoints.

Analysis
Baseline characteristics are reported in counts and percentages with P-values comparing vacuum extraction to SSCS. Outcome parameters are reported as means with standard deviations (SD) and P-values or counts with percentages and odds ratios (OR) with 95% confidence intervals (CI). P-values were calculated with two-sided Chi2 (or two-sided Fisher's exact test where total events were <10). Differences in means between groups were calculated using independent t-tests. Data were entered in Microsoft Excel, and SPSS version 24 was used for data analysis.

Sample size
A convenience sample was used, as this study was part of a larger study including clinical outcome after vacuum extraction and SSCS [6]. The sample size for that study was based on expected differences in perinatal deaths per mode of delivery.

Handling of missing data
Percentages were calculated by dividing the number of women with a certain characteristic or outcome parameter by the number of women with a valid response for that characteristic or outcome parameter at each timepoint. Therefore, the denominator in the fractions may differ slightly per characteristic or outcome parameter. In the tables, the total number of participating women and the number of women with a valid response per characteristic or outcome parameter are shown. The number of women with missing data can be calculated by subtracting the number with valid responses from the total number. Analysis was performed for included women at one day, six weeks and six months after birth. Groups were not exactly the same at six weeks and six months. Some women came for either the six-week or the six-month consultation, but not for both.

Ethical clearance
Ethical permission to conduct this study was obtained from the Mulago Hospital Research and Ethics Committee (refnr: MREC 489) and the Uganda National Council for Science and Technology (refnr: HS1752).

Results
During the study period, 289 women were enrolled after vacuum extraction and 357 women after SSCS (Figure 1). More women who had given birth by vacuum extraction were nulliparous or had their babies admitted to the neonatology unit, but these findings did not reach statistical significance (Table 1).
After six weeks, losses to follow-up were 29.8% (86/ 289) for vacuum extraction and 23.2% (83/357) for SSCS. Thirteen women who had given birth by vacuum extraction, and ten women who had had SSCS were excluded because their babies had died in the time elapsed between one day and six weeks postpartum. One woman who had had vacuum extraction was excluded because her baby was found to have a serious congenital syndrome and died after five months. After six months, losses to follow-up were 31.5% (91/289) for vacuum extraction and 28.3% (101/357) for SSCS. In both groups, one woman was excluded because the baby had died.

Birthing experience
One day after birth, women were feeling significantly better if they had given birth by vacuum extraction than by SSCS (Table 2). After either intervention, more than 90% of women were satisfied with their birthing experience, although more than 50% had been very concerned about their baby. Compared to SSCS, more women reportedly had been 'very scared' during vacuum extraction ( Table 2).

Pain
Women experienced more pain during vacuum extraction than during SSCS, but they reported less pain in the first 24 hours after birth (Table 3). One day after vacuum extraction 91.6% (263/287) of women had no pain or mild pain (NPRS 0-4), vs. 62.6% (223/356) after SSCS (OR 6.54; 95%CI 4.09-10.46). Women experienced more pain at one day after subumbilical midline incision than lower transverse incision: mean pain score 4.41 vs. 3.78 (P = 0.038). Perineal status (intact, episiotomy, tear) had no influence on pain scores during the procedure, or one day after birth in women who delivered vaginally.
Six months after birth, there was no significant difference in pain scores between both groups. Mean pain scores were 0.28 and 0.27 after vacuum extraction and SSCS, respectively (Table 3). After vacuum extraction, 89.2% (165/185) of women reported no pain at all vs. 85.2% (208/244) after SSCS. In 8.8% (10/113) of women after vacuum extraction, pain had interfered 'moderately' to 'extremely' with daily activities in the four weeks prior to the consultation, vs. 10.6% (11/104) in women after SSCS (difference not statistically significant).

Quality of life
During the six-week follow-up visit, the SF-12 questionnaire was completed by 112 women after vacuum extraction and 121 women after SSCS. At the six-month follow-up visit, the questionnaire was completed by 113 and 104 women, respectively.
Six weeks after birth the physical component summary (PCS), measured with the SF-12 questionnaire, was better after vacuum extraction than after SSCS ( Table 4). The mental component summary (MCS) was comparably good. After six months, PCS and MCS had improved and were comparable between both groups.
At six weeks, all 12 questions of the SF-12 questionnaire had better scores after vacuum extraction, seven of

Dyspareunia
At six weeks follow-up, more women reported to have resumed sexual intercourse after vacuum extraction than after SSCS (Table 5). At six months, almost all women had resumed sexual intercourse in both groups. Dyspareunia decreased over time and was comparably low for the groups at six weeks and six months after birth.

Main findings
This study shows that women-centred outcomes, such as pain scores, quality of life and absence of dyspareunia, were better in the first six weeks postpartum in women SSCS, second-stage caesarean section; OR, odds ratio; CI, confidence interval. *In n/N, n is the number of women with this characteristic and N is the number of valid responses for this variable. †'very good' or 'normal'. ‡'very much' or 'yes'. §'very scared' or 'afraid of dying'. ¶'very concerned' or 'afraid the baby would die' (Answers from questionnaire, Appendix S1).  who had given birth by vacuum extraction than in women who had given birth by SSCS. Six months after birth outcomes were similar. More than ninety per cent of women who had given birth by vacuum extraction were satisfied about their birthing experience.

Other findings
Similar to women after SSCS, more than half of the women after vacuum extraction reported to have been 'very concerned' about their baby during the intervention. Information about the procedure and its safety for mother and baby to women undergoing vacuum extraction is needed to prevent anxiety and concern. Debriefing after birth might reduce fear in future pregnancies [22]. Understandably, pain scores during vacuum extraction were higher than during SSCS, as spinal or general anaesthesia were used during SSCS. Pain scores during vacuum extraction in our study are comparable to pain scores during spontaneous vaginal delivery reported in the literature [34]. Pain was, however, significantly worse after SSCS at one day and six weeks after birth. The latter was also found in a study in the USA, where, in the first two months after birth 68% of women experienced significant pain after instrumental vaginal delivery and 79% after caesarean section. Six months after birth, these percentages had decreased to 2% after instrumental delivery and 18% after caesarean section [23].
Although mean PCS scores were better after vacuum extraction, we did not find differences in mean MCS scores of the SF-12 quality-of-life questionnaire per mode of birth. These findings were similar in a Norwegian study [24]. In a study from the UK, however, women who gave birth by forceps or unplanned caesarean section had a higher risk of reduced postnatal health and wellbeing (including increased risk of PTSS following forceps birth), while outcomes after vacuum extraction were comparable to those after spontaneous birth [25]. A study from Sweden reported better quality of life five years after instrumental vaginal birth compared to emergency caesarean section [26].
At six weeks after birth, pain, body weakness and emotional problems interfered significantly less with daily activities after vacuum extraction than SSCS (Appendix S3). This could be of particular importance in a setting where many women are self-employed and do not have a paid maternity leave, while their family is depending on their income.
Reports on dyspareunia after different modes of birth give a wide range of results. In a study from Germany, 13.9% of women had dyspareunia six months after instrumental birth vs. 3.4% after caesarean section [27]. In a study from Australia, 59.5% of women had dyspareunia at six months after vacuum extraction and 40.6% after emergency caesarean section. At 18 months, these percentages were 28.9% and 29.3% in that study [28]. Another study from Australia reported that at 12 months after birth, sexual function had returned to early pregnancy levels, irrespective of mode of birth [29].

Strengths and limitations
Strengths of this study are its prospective design and follow-up to six months after birth, use of validated questionnaires and large number of participants. Nearly all eligible women were included, preventing selection bias ( Figure 1). Follow-up at one day, six weeks and six months prevented recall bias. This study is the only study known to us about this subject in sub-Saharan Africa and the only study performed just after the reintroduction of vacuum extraction in a maternity care service.
Limitations of the study were its observational design, which may have introduced bias. However, not statistically different, nulliparity or having a baby in the neonatology unit were somewhat more frequent in the vacuum extraction group and this may have introduced bias. But the absence of this bias would probably have resulted in even larger differences in favour of vacuum extraction. Excluding women with a baby in the neonatology unit at the time of birthing experience interview did not have an effect on the results. Losses to follow-up may have caused bias but these were comparable between both groups and comparable to losses to follow-up in other settings [22,24,[26][27][28][29]. Recruitment from a single hospital compromises generalisability, but the situation in many resource-constrained high-volume hospitals in sub-Saharan Africa is likely to be similar.

Interpretation and implications
This study is the first study describing how vacuum extraction was experienced by women after its re-introduction in a hospital in sub-Saharan Africa. The major contribution is that it shows that aside from medical reasons to prevent caesarean section, there are several reasons in favour of vacuum extraction instead of caesarean section from the women's point of view. Main reasons are pain, quality of life and ability to perform daily activities in the first six weeks after birth.

Conclusion
Women-centred outcomes, such as pain scores, quality of life and absence of dyspareunia, were better in the first six weeks postpartum in women who had given birth by vacuum extraction, compared to women who had given birth by SSCS. Six months after birth outcomes were similar. Our findings suggest that women, even if they are not acquainted with the procedure, are satisfied with their birthing experience after vacuum extraction. These findings, in combination with medical reasons, support the use of vacuum extraction as first intervention to be considered in the second stage of labour to prevent complications and promote quick recovery.