Thesis (MSc)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: Background Caesarean section deliveries are the most common procedures performed by obstetricians in Zimbabwe. Surgical site infections (SSI) following caesarean section delivery result in increased hospital stay, treatment, cost, hospital readmission rates and related maternal morbidity and mortality. There is no national surveillance system for SSIs in Zimbabwe, however, information is available on number of cases of post-operative wound infection after caesarean section, but the denominator and definition used is not consistent. The objective of this study were develop and strengthen the surveillance system in Zimbabwe, to establish a clinical-based system in a setting with limited microbiological access, to measure post-operative SSI after caesarean section and to describe the associated risk factors and to determine whether feedback of SSI data has any effect on the surgical site infection incidence rate. Methodology This was a before and after study with two rolling cohort periods conducted at two Central hospitals in Harare, Zimbabwe. An Infection Prevention and Control (IPC) intervention was conducted in-between. During the pre-intervention period, baseline demographic and clinical data were collected using a structured questionnaire, and during the post-intervention period the impact of the interventions was measured. Convenience sampling was employed. Results A total of 290 women consented to participate in the study in the pre intervention period, 86.9% (n= 252) completed the 30-days post-operative follow-up and the incidence rate of SSI was 29.0% (n=73, 95% CI:23.4-35.0) Interventions developed included: training in Infection Prevention and Control for health workers; implementation of a protocol for cleaning surgical instruments; dissemination of information on post-operative wound management for the women. After implementation of the intervention, 314 women were recruited for the post-intervention, 92.3%(n= 290) completed the 30-day follow-up and there was a significant (p<0.001) reduction in the incidence rate of SSIs to 12.1 % (n=35, 95% CI: 8.3 -15.8) during this period. Development of SSI after caesarean section was found to be significantly associated with emergency surgery (p<0.001), surgical wound class IV (p=0.001) and shaving at home (p<0.001) at both pre- intervention and post-intervention periods. Stellenbosch University https://scholar.sun.ac.za iii Conclusion This study shows that caesarean section can be performed with low incidence of SSI if appropriate interventions such as training in IPC, adequate cleaning of equipment and education in wound-care for the mother are adhered to. It also demonstrated a simple surveillance data collection tool can be used on a wide scale in resource limited countries to assist policy makers with monitoring and evaluation of SSI rates as well as assessment of risk factors. / AFRIKAANSE OPSOMMING: Agtergrond Keisersnitte is die mees algemene prosedure wat uitgevoer word deur obstetriese dokters in Zimbabwe. Chirurgiese wond infeksies wat op keisersnitte volg lei tot verlengde hospitaal verblyf, behandeling, koste, heropname koerse en verwante moederlike morbiditeit en mortaliteit. Alhoewel daar geen nasionale waaktoesig sisteem vir chirurgiese wondinfeksies is nie, is informasie beskikbaar vir ‘n aantal gevalle wat post-operatiewe wondinfeksie na ‘n keisersnit onwikkel het, maar die noemer en definisie word inkonsekwent gebruik. Die doel van hierdie studie was om die waaktoesig sisteem in Zimbabwe te ontwikkel en te versterk, om ‘n klinies-gebasseerde sisteem te vestig in ‘n opset met beprekte mikrobiologiese toegang, om postoperatiewe chirurgiese wond infeksies na keisersnitte te meet en om die geassosieerde risikofaktore te beskryf en om vas te stel of terugvoering van chirurgiese wondinfeksie data enige effek op die infeksiekoerse na keisersnitverlossings gehad het. Metodologie Hierdie was ‘n voor-en-na studie met twee kohort periodes uitgevoer by twee sentrale hospitale in Harare, Zimbabwe. ‘n Infeksievoorkoming en –beheer intervensie was tussenin uitgevoer. Tydens die pre-intervensie periode was basislyn demografiese en kliniese data ingesamel deur middel van ‘n gestruktureerde vraeboog, en gedurende die post-intervensie fase was die impak van die intervensies gemeet. Gerieflikheidsteekproefneming was geimplementeer. Resultate ‘n Totaal van 290 vroue het toestemming verleen om aan die studie deel te neem in die pre-intervensie periode, waarvan 86.9% (n=252) die 30 day post-operatiewe opvolg voltooi het en die insidensiekoers van chirurgiese wondinfeksies was 29.0% (n=73, 95% CI:23.4-35.0) Intervensies wat onwikkel was het ingesluit: opleiding in Infeksie Voorkoming en -Beheer vir gesondheidswerkers; die implementering van ‘n protokol om chirurgiese instrumente skoon te maak; disseminering van informasie oor post-operatiewe wondhantering vir vroue. Na die implimentering van die intervensie was 314 vroue gewerf in die post-intervensie fase, waarvan 92.3% (n=290) die 30 dae opvolg voltooi het. Daar was ‘n beduidende (p<0.001) verlaging in die insidensiekoers van chirurgiese wondinfeksies na 12.1% (n=35, 95% CI: 8.3-15.8) gedurende hierdie periode. Stellenbosch University https://scholar.sun.ac.za v Daar was bevind dat chirurgiese wondinfeksies beduidend geassosieer was met noodchirurgie (p<0.001), chirurgiese wondklassifikasie IV (p=0.001) en skeer van hare by die huis (p<0.001) by beide die pre-intervensie en post-intervensie periodes. Gevolgtrekking Hierdie studie wys dat keisersnitte uitgevoer kan word met ‘n lae insidensie van chirurgiese wondinfeksies indien toepaslike intervensies, soos opleiding in infeksievoorkoming en beheer, voldoende skoonmaak van toerusting en opvoeding in wondsorg vir die moeders. Dit het ook aangedui dat ‘n eenvoudige data-insameling instrument op ‘n wye basis gebruik kan word in beperkte-hulpbron lande om beleidmakers te help met monitering en evaluering van chirurgiese wondinfeksie koerse, asook die assessering van risikofaktore.

Le ratio de mortalité maternelle en Afrique de l’Ouest est le plus élevé au monde et l’hémorragie du post-partum (HPP) y est la principale cause directe de mortalité maternelle. L’objectif général de ce travail était d’apporter des connaissances pour améliorer la prise en charge des HPP dans les structures hospitalières en Afrique de l’Ouest. Tout d’abord, nous avons montré que les facteurs de risques maternels de morbi-mortalité liée à l’HPP retrouvés dans les pays à ressources élevées étaient généralisables au contexte de l’Afrique de l’Ouest. Par ailleurs, nos analyses suggèrent que la qualification du médecin en charge du service de maternité (médecin généraliste versus gynécologue-obstétricien) a un impact déterminant sur la survie des femmes présentant une HPP dans ce contexte. Ensuite, nous avons montré que l'injection d'ocytocine dans les dix minutes qui suivent le diagnostic d’HPP était un facteur déterminant pour améliorer la santé maternelle au Bénin et au Mali. Le diagnostic rapide de l’HPP et une meilleure surveillance du post-partum immédiat étaient également importants. Enfin, nous avons mis en place une étude pilote pour évaluer la faisabilité d’un essai contrôlé randomisé (ECR) dont l’objectif est de tester l’efficacité du tamponnement utérin par condom catheter dans la prise en charge des hémorragies du post-partum dans cette région. Les résultats ont montré que la mise en place d’un ECR était faisable et ont apporté des connaissances supplémentaires pour optimiser le protocole de recherche. Ainsi, nous avons identifié un certain nombre de facteurs sur lesquels des actions doivent être menées pour améliorer la qualité de la prise en charge de l’HPP. / The maternal mortality ratio in West Africa is the highest in the world and postpartum hemorrhage (PPH) is the leading direct cause of maternal mortality. The general objective of this work was to provide knowledge to improve the management of PPH in hospitals in West Africa. Firstly, we have highlighted that maternal risk factors for morbidity and mortality related to the PPH found in high-resource countries were generalizable to the context of West Africa. Furthermore, our analysis suggests that the qualification of the doctor in charge of the maternity unit (general practitioner versus obstetrician) has a decisive impact on the maternal issue in this context. Then, we showed that injection of oxytocin within ten minutes after the diagnosis of PPH was a key factor for improving maternal health in Benin and Mali. Rapid diagnosis of PPH and better monitoring of the immediate postpartum period were also important. Finally, we set up a pilot study to assess the feasibility of a randomized controlled trial (RCT), which aims to test the effectiveness of uterine tamponade with condom catheter in the postpartum hemorrhage management in this region. The results showed that f a RCT was feasible and provided additional knowledge to optimize the research protocol. Thus, we have identified a number of factors on which action should be taken to improve the quality of PPH management.