Postpartum haemorrhage (PPH) is the most common cause of maternal mortality leading to an estimated 86, 000 deaths/year. The most common cause of PPH is failure of the uterus to contract properly (uterine atony). Several measures have been introduced to prevent and treat atonic PPH, but in spite of active management of the third stage of labour (AMTSL), maternal deaths from PPH still occur. PPH can kill rapidly within two hours or less. PPH has long been recognised as a dangerous complication for mothers. In order to optimise the prevention and treatment of PPH, different approaches have been introduced and modified over the last century. We reviewed the regimes used in the management of the third stage of labour between 1917 and 2011 as described in the successive editions of the ‘Ten Teachers’ books. Throughout the Ten Teachers series, uterotonic drugs have always been taught as being the best initial measure to manage PPH. However, the importance of bimanual uterine compression (BMC) has increased gradually, moving from third to first treatment option over the editions (Aflaifel and Weeks, 2012a). The components of the AMTSL package for PPH prophylaxis have recently been extensively examined in clinical trials. Its effectiveness in reducing blood loss is now known to be almost all due to the uterotonics (Aflaifel and Weeks, 2012b). However, clinical trials evaluating the efficacy of uterotonics in treating PPH are comparatively rare. Where present they usually compare two uterotonics with an absence of control group, as it is unethical to leave a bleeding woman untreated. A recent innovation is to model the likely outcomes in the absence of uterotonic therapy through histograms. This also allows an assessment of the efficiency of treatment by measuring the number of women who stop bleeding shortly after administering treatments. This model has never previously been applied to databases in which uterotonics were used for prophylaxis. In a secondary analysis of 4 large randomised trials, small secondary histogram peaks (primarily attributed to a treatment effect) were still present even if uterotonic therapy had not been used. Furthermore, the study revealed that women were commonly treated at low levels of blood loss (< 500 mls). It was also seen that, of those diagnosed with PPH (≥ 500 mls), most stopped bleeding at blood losses of around 700 mls even if they did not receive any uterotonic therapy. This should warn against ascribing all the effect to uterotonic therapy. As well as stopping spontaneously, other physical therapies may also have been used concurrently and may have had an effect. The evidence from the histogram study suggested that use of additional uterotonic is not a good surrogate for PPH in the research context. Chapter 4 reports on evaluations of the outcomes that are used by researchers in PPH trials. In the 121 studies evaluated, there was a huge diversity in choosing the outcomes (PPH prevention). The most common was ‘Incidence of PPH ≥ 500 mls’, which was mentioned in 21% (25/121) of trials. The study interestingly showed that use of additional uterotonic was used for sample size calculation in 6% (7/121) of studies as a surrogate for PPH. The above findings emphasise the importance of physical measures in the early treatment of PPH. BMC is thought to help in treating PPH, although there are no clinical trials on its effectiveness. A survey was therefore conducted amongst obstetric care providers in the UK to look at the frequency of BMC use in clinical practice and the attitudes towards its use. The survey found that, although clinicians find BMC effective, it is rarely used as the procedure is considered to be too tiring and too invasive. If, however, BMC could be performed in a less invasive manner, then it could act as an effective low-cost treatment for those PPHs arising from atony. The thesis concludes with an investigation into a new low cost intervention that might contribute to the early physical management of PPH. The ‘PPH Butterfly’ is a new device that is designed to make uterine compression simpler, less tiring and less invasive. It was compared to the standard BMC in a mannequin model. The main objective was to compare the efficacy of the PPH Butterfly to standard BMC in producing sustained uterine compression. The study revealed that the PPH Butterfly is simple to use on a mannequin model, even among obstetric care providers with little experience. It produces an equivalent amount of pressure to BMC, but neither method produced sustained compression over the 5 minutes of use. It also demonstrates the feasibility of using a mannequin model for teaching and performing BMC.

Background: Postnatal Depression (PND) is experienced by around 13% of women, who suffer a range of disabling symptoms that can have a negative effect on the mother and infant relationship, with significant consequences in terms of the child's later mental health. Research has shown that providing support to mothers experiencing PND can help reduce their depressive symptoms and improve their coping strategies. This study aimed to evaluate the impact of telephone peer-support for women experiencing PND. Methods/Design: A pilot RCT was conducted in which women who screened positive for postnatal depression using the Edinburgh Postnatal Depression Scale (EPDS > =10) were randomised to receive telephone-based support from peers who had recovered from PND, or standard care. Primary outcome measures included depressive symptomatology measured post-intervention and at six-months using the EPDS, and parent-infant interaction using the CARE-Index. Secondary outcome measures included anxiety and depression, dyadic adjustment, parenting stress, and self-efficacy. Maternal perceptions of the telephone peer-support were being assessed using semi-structured interviews. Quantitative and qualitative data was also collected from the peer-supporters to assess the impact on them of delivering the intervention. Results: Participants: twenty-eight participants were recruited to the study, and there was a fifty-percent dropout rate (intervention group n=6, control group n=8). While there were no differences in EPDS scores between the two groups at post-intervention, the intervention group continued to improve at six-month follow-up, whereas the control group showed signs of relapse. The intervention had no impact on mother-infant interaction. In-depth interview data show that women valued the support that was provided. Peer-Supporters: nineteen peer-supporters were recruited, of whom five left before supporting a participant, and eight left after supporting only one participant. The quantitative results showed a significant non-clinical increase in anxiety at post intervention. The qualitative results indicated that the peer-supporters found the majority of calls challenging, and that delivering the intervention had had a deleterious impact on some peer-supporters. Conclusion: While these findings suggest a positive impact of telephone-based peer-support, further research into ways of improving mother-infant interaction are urgently required. Research is also required into providing effective support for the peer supporters.

The Pathways project was undertaken to devise guidelines to facilitate rapid diagnosis of paediatric brain tumours. Methods: A systematic review and meta-analysis of published data on paediatric brain tumour presentation and analysis of the presentation of children newly diagnosed with a brain tumour at four oncology centres was undertaken. The results informed a professional consensus process. Results: 74 papers met the inclusion criteria for the meta-analysis. 56 symptoms and signs at diagnosis were identified. The most frequent symptoms and signs at diagnosis were: headache (33%), nausea and vomiting (32%), abnormalities of gait and coordination (27%), and papilloedema (13%). 139 patients were recruited to a multi-centre cohort study. Symptoms and signs at disease onset and at diagnosis and factors associated with a long and short symptom interval were determined. A shorter symptom interval was associated with nausea and vomiting and motor system abnormalities. A longer symptom interval was associated with head tilt, cranial nerve palsies, endocrine and growth abnormalities and reduced visual acuity. A multi-disciplinary workshop and Delphi consensus voting were used to translate the evidence into a clinical guideline comprising 76 statements advising on the identification and assessment of children who may have a brain tumour.

Obstetric anal sphincter injuries (OASIS) are serious complications of vaginal birth with a reported average worldwide incidence of 4%-6%. They are a recognised major risk factor for anal incontinence resulting in concern amongst women who sustain such injuries when considering the most suitable mode of birth in a subsequent pregnancy. This thesis contains three studies; a systematic review and meta-analysis of the published literature exploring the impact of a subsequent birth and it’s mode on bowel function and/or QoL for women with previous OASIS, a follow-up study on the long-term effects of OASIS on bowel function and QoL and finally a prospective cohort study of women with previous OASIS to assess the impact of subsequent birth and its mode on change in bowel function. The work in this thesis demonstrated an increase in incidence of bowel symptoms in women with previous OASIS over time and that short-term bowel symptoms were significantly associated with bowel symptoms and QoL. This thesis also showed that the mode of subsequent birth was not significantly associated with bowel symptoms or QoL and for women with previous OASIS who have normal bowel function and no anal sphincter disruption a subsequent vaginal birth is a suitable option.

Fetal position throughout labour exerts considerable influence on labour and delivery, with a mal-positioned fetus during active labour known to contribute towards fetal and maternal morbidity. In response there is a move towards promoting the Left Occipito-Anterior (LOA) position at labour onset as optimal. It is thought that the LOA position encourages anterior rotation thus reducing the likelihood of mal-rotation. A systematic review was undertaken which highlighted an absence of scientific evidence. A prospective cohort study was therefore conducted with 1250 nulliparous women who were scanned to accurately determine fetal position, specifically the LOA position at the onset of labour and the association with delivery mode and other birth outcomes was examined. The LOA position at the onset of labour was not associated with mode of delivery, nor were any of the other positions (p=0.39). Pain relief, labour duration, augmentation, and Apgar scores did not show any association with the LOA or other positions. The only association found was that women with a fetus in the posterior position were more likely to use pethidine (p=0.008). This study has shown that the LOA fetal position at labour onset was not associated with improved outcomes and therefore should not be promoted as optimum.

Background: Every year 287,000 women die from pregnancy related complications. Methods: Systematic reviews of interventions to reduce maternal mortality in developing countries with meta-analysis or meta-synthesis where appropriate. Results: Participatory learning and actions cycles with women’s groups significantly reduce maternal and neonatal mortality, training and supporting TBAs also reduces perinatal mortality. Clinical officers performing caesareans section do not seem to cause any more maternal or perinatal mortalities than doctors. Prophylactic antibiotics reduce infectious morbidity in surgical abortion, yet the effect on miscarriage surgery is unclear. Cell salvage in ectopic pregnancy and caesarean section appear to be a safe and effective alternative in the absence of homologous transfusion. Motivational interviews may have potential to improve contraceptive use short term. Symphysiotomy may be a safe alternative to caesarean section. The anti-shock garment may improve outcomes when used in addition to standard obstetric haemorrhage management. Potential solutions to emergency transport for pregnant women include motorcycle ambulance programmes, collaboration with local minibus taxis services, and community education and insurance schemes. Conclusion: Several interventions reviewed in this thesis can be utilised to aid reduction in maternal mortality, however the level of evidence available within each review varies, some allowing firm inferences with others more tentative.

This thesis investigated the effectiveness of exercise in the treatment of postnatal depression (PND). PND is highly prevalent, affecting mothers, families and child development. Exercise is currently recommended to mothers with PND, potentially offering fewer side effects than antidepressants and wider accessibility than psychological treatments. This thesis reported three studies. A randomised controlled trial (RCT) investigated the effectiveness of an exercise counselling intervention, in addition to usual care, in treating PND. This intervention provided a moderate, non-significant decrease in depression compared to usual care alone. A qualitative study found that exercise was viewed as acceptable and often preferable to antidepressants in the treatment of PND. A range of mechanisms via which exercise produced psychosocial benefits were proposed, including improving self-confidence and supporting personal identity after childbirth. A systematic review with meta-analysis of RCTs of exercise interventions for PND concluded that exercise can be effective in reducing depression in general and depressed postnatal populations. Preliminary findings suggested the importance of social support within such interventions. Exercise is likely to be effective in the treatment of PND and should therefore be recommended to mothers. However, further research investigating the relative effectiveness of different intervention designs would be valuable.

Immune mediated mechanisms are thought to contribute to recurrent pregnancy losses. A number of treatment options with limited evidence are being used in clinical practice to treat women with recurrent miscarriages. The objectives of this thesis was a. To summarise the available evidence for granulocyte colony stimulating factor (G-CSF) in reproductive medicine. b. To perform a randomised controlled study (RCT) to evaluate the efficacy and safety of recombinant human granulocyte colony stimulating factor (rhG-CSF) in women with unexplained recurrent miscarriages. The main conclusions from this thesis are: a. The systematic narrative review found that available evidence is of poor quality, but suggestive of benefit with granulocyte colony stimulating factor in women with recurrent miscarriages. b. The RCT concluded that administration of rhG-CSF does not improve pregnancy outcomes among women with a history of unexplained recurrent miscarriages. RhG-CSF appears to be safe for both mothers and their offspring/s.

The aim of this thesis was to investigate the role of outpatient hysteroscopy in modern gynaecological care by conducting a series of systematic reviews and meta-analyses to examine how the procedure can be optimised to reduce pain and by performing a cost effectiveness analysis. The systematic reviews concluded that women undergoing outpatient hysteroscopy should take simple analgesia beforehand and that the hysteroscopist should adopt a vaginoscopic approach using a small diameter, rigid hysteroscope and normal saline as the distension medium. If dilatation of the cervix is required this should be done under a paracervical block. These findings were incorporated into a clinical guideline and the quality of the evidence that the reviews provided was assessed using the SIGN and GRADE methods. A comparison of the assessments found that they gave varying estimates of the quality of evidence and that neither offered a perfect solution to the assessment of evidence quality when writing clinical guidance. The economic analysis found that initial testing with outpatient hysteroscopy was the most cost-effective testing strategy for investigation of heavy menstrual bleeding when compared to other diagnostic tests, regardless of a woman’s wish for future fertility or prior treatment with a levonorgestrel intrauterine system.

Differences in recovery from types of hysterectomy may be due to different advice, rather than different incisions alone. Aims: What are the beliefs and experiences of women who have a hysterectomy and the practices and beliefs of health care professionals? Explore the evidence for psychological preparation for surgery. Methods: .Retrospective structured questionnaires .Semi-structured interviews and validated quality of life questionnaires .Systematic review Results: No significant difference in return to work for type of employment and incapacity pay. Variation between the advice given for recovery by UK gynaecologists and nurses. Regardless of route of surgery, the expectations and fears of women are similar and rely on health care advice, with conflicting advice and varied recovery experiences. Women who had a vaginal hysterectomy had concerns around sitting, laparoscopic route had a lower length of stay and abdominal hysterectomy had higher anxiety scores (P 0.003). Mean quality of life scores by EQ5 were not different based on route of surgery (pre-surgery P 0.4446, 1 week P 0.447, 4 weeks P 0.876,12 weeks P 0.850). There was a reduced length of stay [P 0.03, 5.65 (-10 82 to -0.48)] and reduced trait anxiety intervention [P < 0.00001, mean difference 7.78 (7.19, 10.61)] for psychological interventions.