Postpartum infections are among the leading causes of maternal mortality world-wide, particularly in under-resourced countries. Available data suggests that HIV infected women are at greater risk of postpartum complications than uninfected women. In South Africa, HIV/AIDS and related infections are now cumulatively the leading causes of maternal deaths (though indirectly), with puerperal sepsis among the 5 most common causes.

This was a prospective longitudinal cohort of HIV infected (n = 675) and uninfected (n = 648) women. These were women in whom vaginal delivery was anticipated, and were recruited at > 36 weeks of gestation during the antenatal period.

Hypothesis - HIV infected women are at increased risk of postpartum infectious morbidity and this morbidity can be reduced by use of prophylactic intrapartum antibiotics.

INTRODUCTION The World Health Organisation (WHO) estimates that 582 000 women die world-wide each year due to pregnancy related conditions that could be prevented. Most of these deaths occur in developing countries1. Some of the underlying causes of increased maternal morbidity and mortality are HIV-related. In South Africa the antenatal HIV seroprevalence surveys report an infection rate of 32%, the highest among the Sub-Saharan countries2. According to the Saving Mothers Report (NCCEMD) pregnancy-related sepsis was the fourth commonest cause of maternal mortality, while AIDS was the second leading cause3. Of the 67 women who died of HIV/AIDS in 1998, 41 died from puerperal sepsis of which 54% had a vaginal delivery and 46% had a caesarean section. It is not clear from the report to what extent HIV infection had contributed to the severity of sepsis.

The clinical impression is that women with HIV infection have a higher incidence of post delivery morbidity than HIV-negative women. Proposed contributory factors include, low CD4 count and associated impaired immunological function, other obstetric complications and sexually transmitted infections3.

The puerperal sepsis rate is about 2-6 % in most studies. It is however much higher in women with bacterial vaginosis (14%)2 and those who deliver out of health institutions (18%)3. Women with HIV infection are more likely to harbour BV and other sexually transmitted infections. For this and other reasons it is generally accepted that HIV-positive women are at increased risk of puerperal infectious morbidity and mortality independent of the mode of delivery and antiretroviral therapy.

Di Lieto et al4 reported a 33.8 % incidence of infectious morbidity following emergency caesarean delivery. The incidence of infectious morbidity has been reported to be higher in HIV positive women compared to HIV negative women (Semprini et al. 1995)5.

The role of prophylactic antibiotics in obstetric complications has been previously investigated especially in women with preterm labour, prelabour rupture of membranes, cardiac disease and women delivered by caesarean section. There is good evidence that prophylactic antibiotics significantly reduce postpartum morbidity. In these studies prophylactic antibiotics were largely administered intra-operatively. The question that remains is whether prophylactic antibiotics need to be given to only women identified as at risk or to all women with HIV infection.

The identification of biochemical (laboratory)/immune markers will be of value in the timing of clinical interventions. The markers that can be used to predict sepsis will be to determine the level of apoptosis in neutrophils, neutrophil function, C-Reactive protein levels, ESR and the levels of the following cytokines viz. TNF-, IL-1, IL-6, IL-8, IL-10. Less clear is the role that apoptosis and it's dysregulation may play in pathological immune states like sepsis, autoimmune disease, or immunosuppressed states. The potential role of apoptosis and the pathogenesis of sepsis have resulted in some unanswered questions- (1) whether impairment of apoptosis as a method of down-regulating pro-inflammatory cells causally results in inappropriate persistence of inflammation under septic conditions and (2) whether increased apoptosis of immune effector cells results in immunosuppression and increased susceptibility to overwhelming infection? Neutrophils are the first immune cells to migrate to sites of inflammation and major functions include phagocytosis of bacteria,apoptosis of neutrophils may be one of the mechanisms of limiting tissue injury at sites of inflammation and phagocytosis of apoptotic neutrophils inhibits the release of pro-inflammatory cytokines from monocytes. Lower absolute neutrophil counts are associated with increased risk of hospitalization for serious infection among HIV infected patients. Because neutrophils are functionally impaired, we hypothesize that sepsis observed in HIV infection might be a result of accelerated apoptosis. This study will therefore try to define the mechanisms, direct or indirect, by which HIV infection may dysregulate apoptosis and cytokine production in the pathogenesis of sepsis.

In summary puerperal sepsis remains a major cause of maternal mortality in South Africa. The clinical experience in Durban, is that postpartum infections are on the increase particularly in view of the high prevalence of HIV. Clear guidelines are needed for the role of antibiotics among women at risk for postdelivery morbidity associated with infections. Current recommendations by the NCCEMD of SA indicate that all women with AIDS and women having caesarean sections be treated prophylactically with antibiotics. However, guidelines regarding timing of antibitiotic prophylaxis remain unclear6. This study therefore sets out to establish reliable data on the incidence of postdelivery morbidity including puerperal sepsis, associated risk factors, laboratory markers and an effective antibiotic regimen to reduce the risk of postdelivery morbidity.

AIM To evaluate the role of intrapartum prophylactic antibiotics in HIV infected women in reducing the incidence of postdelivery morbidity as assessed by clinical and immunological markers.

HYPOTHESIS Antibiotics administered during labour can reduce postdelivery morbidity in HIV infected women STUDY OBJECTIVES Primary: To reduce the incidence of post-delivery infections by use of prophylactic antibiotics.

Secondary: To investigate immunological parameters as markers of post-delivery infections.

To identify risk factors for post-delivery infections. PRIMARY ENDPOINTS The primary efficacy endpoints for this study will be the incidence of post-delivery infectious morbidity diagnosed between delivery and 6 weeks post-delivery.

STUDY DESIGN This is a prospective double blind, randomised placebo-controlled trial of prophylactic antibiotics for delivery.

STUDY POPULATION The study will be conducted at King Edward VIII Hospital, Durban and Umtata Hospital, Transkei over a 2 year period.

Women who have been counselled and tested for HIV during the antenatal period, and who have agreed to know their results, will be enrolled during the antenatal period. Informed written consent for participation in the study will be obtained from all voluntary participants.

Immunological studies will be conducted in a subsample (25%) of patients in each study group.

Inclusion Criteria

• Women with known HIV status as documented by routine rapid HIV tests, following pre-test voluntary counselling and testing (VCT).
• Women who gave informed study consent. Exclusion Criteria
• Women on antibiotic therapy less than 2 weeks prior to study enrolment.
• Women planned for elective caesarean delivery.
• Obstetric complications such as preterm prelabour rupture of membranes, cardiac disease, antepartum haemorrhage.

Randomisation will occur on admission to the labour ward. The following tests will be performed prior to randomisation i.e at enrolment: full blood count(FBC), CD4 count and ano-genital swabs for microbiology. The women will be randomly assigned to one of the three groups, depending on HIV status:

Group 1: HIV-POSITIVE given antibiotics (Cefoxitin 2g IVI stat dose) Group 2: HIV-POSITIVE given placebo Group 3: HIV-NEGATIVE (no placebo or cefoxitin = control group) Rationale for placebo and choice of antibiotic: The current standard practice is not to give prophylactic antibiotics for delivery to women without risk factors irrespective of HIV status. This justifies a placebo-controlled design. In this unit the current standard practice is to administer intravenous cefoxitin (2g stat dose) to women for whom prophylactic antibiotic is indicated. We therefore intend to use cefoxitin for the study.

Sample Size: Considering a 15% puerperal sepsis rate among HIV positive women following vaginal and emergency caesarean section delivery a sample size of 343 will be required in each group (placebo and study group) to observe a 33% reduction in the sepsis rate. This would detect a significant difference at 5 % level (p < 0.05) with the power of 90%. Therefore the appropriate sample sizes are:

Group 1 ( HIV-positive, antibiotic): 343 Group 2 (HIV-positive, placebo): 343 Group 3 (HIV-negative, 686 )

INVESTIGATION PLAN Study procedure at each visit Visit 1 (baseline and enrolment) will take place at the antenatal clinic at the respective sites.

This will occur after the mother has received post-test counselling. At the baseline the mother will be informed of the study, both verbally and in writing, she will then sign the informed consent form.

Demographic data, maternal medical history and HIV related symptoms and signs will be documented and HIV disease staging will be carried out, pap smears and CD 4 counts will be done at enrolment. A physical and obstetrical examination will be done and the mothers will be randomised into one of 3 groups depending on their HIV status and the antibiotic regimen to be given.

Visit 2 (labour/delivery) A diagnosis of labour will be made clinically. Women with HIV and confirmed to be in labour will be randomised into either the antibiotic or placebo group. A single dose of mefoxin or a similarly looking placebo will be administered. All women with HIV infection will be given a single dose of nevirapine ( and their newborns will receive 0.6ml of nevirapine 24-72 hours post-birth). During labour artificial rupture of membranes will be delayed; and invasive monitoring, use of instruments for delivery and episiotomy will be avoided where possible. Women will also be advised on the benefits and risks of exclusive breastfeeding and formula feeding and they will be further informed on their choice of an appropriate feeding method.

Blood will be taken for the following: apoptosis of neutrophils, neutrophil function, FBC, C-reactive protein. Serum samples will also be stored for the determination of cytokines levels. A cervical vaginal swab will also be done, for microbilogical evaluation. Urine will be analysed by means of dipstix, and microscopy and culture will be reserved for symptomatic patients on significant findings on dipstix. CD4 counts will be done by a private pathologist. The reason for this is that the mothers will be enrolled during the day and night and will thus need a 24-hour service for the CD4 counts and the FBC.

Visit 3 (24-48 hours post delivery) An infection symptom review and physical examination will be done. In addition blood samples will be take for the evaluation of apoptosis, neutrophil function test, FBC and C-reactive protein. Serum will be stored for cytokines assays. A wound swab will be done where indicated.

Visit 4 This will take place 1-week post delivery. A physical and obstetrical examination will be carried . A wound swab will also be done. An infant evaluation including laboratory parameters will also be carried out during this visit.

Visit 5 This will take place 2 weeks post-delivery. A physical and examination and an infection symptoms review will be carried out at this visit. A wound swab will also be done, where indicated.

Visit 6 This will take place 6 weeks post-delivery. A physical and infection symptom review will be done. Blood samples will be taken for CD4 counts and FBC. When necessary, a pap smear and wound swab will also be done at this stage.

Adherence to protocol Mothers will be encouraged to make every attempt to complete the protocol as specified, and will be reimbursed for transport.

Dropouts and Withdrawals Patients have the right to withdraw from the study at any time or may be withdrawn if they fail to comply with the study requirements or investigators instructions.

OUTCOME MEASURES The primary outcome measure will be puerperal sepsis defined as a temperature of 38C on two separate occasions due to genital tract infection and endometritis. Secondary measures will include other clinical infections, neonatal sepsis, duration of hospital stay and surgery for puerperal sepsis. Clinical infections include urinary tract infections, pneumonia and wound sepsis.

DATA COLLECTION AND MONITORING A standard data form will be completed for each subject. Study participants will not be identified by any name on any of the study documents. Subjects will be identified by a Study Identification Number. The data will be entered into a computerised database EPI info version 6 software program. The same shall be used for statistical analysis.

The study co-ordinator will be responsible for the accuracy of the database and to determine that all regulatory requirements surrounding the trial are met. A data safety and monitoring team will meet quarterly to assess the safety and accuracy of the data collected.

Ethical Approval This protocol and the informed consent and any subsequent modifications will be reviewed by the Ethics Committee of the University of Natal. Written informed consent will be obtained from the participants. The informed consent form will describe the study, the procedures to be followed and the risks and benefits of the participation. A copy of the informed consent form will be given to the subjects.

Subject confidentiality All laboratory specimens, reports, study data collection, process and administrative forms will be identified by a study number to maintain confidentiality. All local databases must be secured with a password protected access systems. Forms, lists, appointment log books in a separate locked file in an area with limited access.

REFERENCES

1. National Committee on Confidential Enquiry into Maternal Deaths. S Afr Med J 2000; 90: 367-73.
2. Fans S, Lin Z, Chen C. Bacterial Vaginosis in pregnant women. (Abstract, Medline) Chinese Journal of Obstetrics and Gynaecology 1997; 32: 84 - 6
3. Duvekot E et al. A comparison between health centre deliveries and deliveries out side health institutions. Papua New Guinea Med J 1994;37:173.
4. Di Lieto A, Albano G, Cimmino E et al. Retrospective study of postoperative infectious morbidity following caesarean section. Min Gynecol 1996; 48: 85-92.
5. Semprini AE, Catagla C, Ralizza M et al. The incidence of complications after caesarean section in HIV-positive women. AIDS 1995; 9;1913-7.
6. Smaill F, Hofmeyer GJ. Antibiotic prophylaxis for caesarean section (Cochrane Review). In: The Cochrane Library. Issue 4, 2000. Oxford. Update Software.

Inclusion Criteria:

• Women with a pregnancy of > to 36 weeks of gestation
• Women with known HIV status as documented by routine rapid HIV tests, following pre-test voluntary counselling and testing (VCT).
• Women who gave informed study consent.
• Over the age of 18years
• Eligible for vaginal delivery

Exclusion Criteria:

• Women who received antibiotic therapy less than 2 weeks prior to study enrolment.
• Women planned for elective caesarean delivery.
• Obstetric complications such as preterm prelabour rupture of membranes, cardiac disease, diabetes and antepartum haemorrhage.