Pregnancy in sickle cell disease (SCD) is fraught with many complications including preeclampsia (PE) and intrauterine growth restriction (IUGR). Previously, the investigators found an abnormality in prostacyclin-thromboxane ratio in sickle cell pregnant women, a situation that is also found in non-sickle pregnancies with PE and unexplained IUGR. Low dose aspirin (LDA) has been found to reduce the incidence of PE and IUGR in high-risk women due to its reduction of vasoconstrictor thromboxane whilst sparing prostacyclin, in effect "correcting" the ratio. It has been found to be safe for use in pregnancy and is recommended in obstetric guidelines for this use but has not been tested in sickle cell pregnancy. The investigators hypothesize that LDA would reduce the incidence of IUGR and PE in pregnant haemoglobin (Hb)SS women. The investigators also plan to build a machine-learning model to predict severe maternal outcomes in them.

The investigators propose a multi-site, randomized, controlled, double blind trial comparing a daily dose of 100mg aspirin with placebo, from 12 - 28 weeks gestation until 36 weeks. The study sites are three teaching hospitals in Lagos and Ile-Ife, and twelve general hospitals and one federal medical centre within Lagos state, with the coordinating centre at the College of Medicine, University of Lagos (CMUL), Idi-Araba, Lagos. A total of 476 eligible pregnant HbSS and HbSC women will be recruited consecutively and randomly assigned to either group using a web-based app, sealed envelope. Each study group will comprise 238 pregnant women with SCD.

All participants will be followed from recruitment till six weeks postpartum. They will have their body weight, blood pressure and haematocrit checked at each antenatal visit. Their full blood count, vital signs and oxygen saturation will be checked and recorded at each visit. Primary outcome measure will be birth weight below 10th centile for gestational age on INTERGROWTH 21 birthweight charts, and incidence of miscarriage or perinatal death. Analysis will be by intention to treat, and the main treatment effects will be quantified by relative risk with 95% confidence intervals, at a 5% significance level. The investigators plan to develop a prediction model to predict the risk of complications in these women using machine learning. The prediction outcome will be severe maternal outcomes comprising maternal near miss or death.

STUDY RATIONALE Sickle cell disease (SCD) especially in pregnancy poses great burden and is thus designated a global public health problem by World Health Organization (WHO) and United Nations. In Nigeria, many SCD women die because of pregnancy complications. The physiological changes in pregnancy aggravate the severity of pregnancy complications in women with SCD. Maternal mortality ratio and perinatal mortality rate in Nigeria till date has remained high. Hypertensive disorders of pregnancy, which includes preeclampsia (PE) is a leading cause of maternal death and intrauterine growth restriction (IUGR) and low birth weight (LBW) are leading causes of perinatal deaths. Pregnancy outcomes in women with SCD follows a variable course and it is thus difficult to identify those who will develop complications during pregnancy. There is thus a need to develop cost-effective strategies to prevent complications in this category of women. However very few randomized controlled trials address their management, and guidelines for their treatment are supported by low quality evidence. Low dose aspirin (LDA) has been found to be beneficial in prevention of preeclampsia in non-SCD pregnant women. The NICE guideline lists several risk factors for preeclampsia for which LDA is recommended but no mention is made of women with SCD who are known to be at high risk of preeclampsia. This research is therefore targeted at determining if drugs such as LDA may be beneficial in reducing the contributions of this genetic condition to maternal and perinatal deaths in a low resource setting like ours, consequently saving economic as well as physical resources.

The aim of this clinical trial is to determine whether daily administration of 100mg LDA from 12 - 28 weeks of gestation till 36 weeks, reduces the risk of IUGR, PE, perinatal deaths or miscarriages and other complications (sickling and non-sickling related) in pregnant HbSS women compared with the use of placebo.

OBJECTIVES

1. To determine the effect of the use of LDA during pregnancy in HbSS and HbSC women on the risk of IUGR, perinatal death or miscarriage.
2. To determine the effect of the use of LDA during pregnancy in HbSS and HbSC women on the risk of other maternal complications including PE, preterm delivery, number of vaso-occlusive crises, need for blood transfusion, urinary tract infections, respiratory tract infections, acute chest syndrome, retained placenta, placental abruption, and vaginal bleeding.

METHODOLOGY Study design A multi-centre parallel, double blind, superiority randomized controlled trial, with women allocated in a 1:1 ratio into two groups, namely low dose aspirin (LDA) group and placebo group.

Settings and locations Study sites include 3 Teaching Hospitals in Lagos and Ife, Osun State, 12 General Hospitals in Lagos and 1 Federal Medical Centre within Lagos State. The coordinating centre is College of Medicine, University of Lagos (CMUL), Lagos, Nigeria.

Sample size Using the formula for proportions in superiority parallel trials and assuming an incidence of IUGR of 20% in SCD and expecting a 50% IUGR reduction with the use of LDA as detected by Bujold et al, 476 women in total (238 women per group) was calculated to give this study a 90% power of detecting a decrease in IUGR at the 5% significance level and allowing for 20% attrition.

Enrolment and data collection Each woman at 12 to 28 weeks gestation who self-reports or is found to have SCD on routine haemoglobin genotype testing, will have a confirmatory haemoglobin electrophoresis done. This is to have a uniform diagnosis and to be certain of the exact sickle cell phenotype. Every woman with haemoglobin SS or SC that fits the eligibility criteria and gives informed consent will be consecutively recruited. They will be randomized into either LDA group or placebo group using a web-based software, Sealed envelope, in a 1:1 ratio in blocks stratified according to centre. Only an unblinded pharmacist, will be aware of the actual codes and she will not be in contact with any of the study participants. Those in the LDA group will receive 100mg aspirin daily taken at once just before bedtime while those in the placebo group will receive a tablet that looks like the active drug in terms of size and thickness. The drugs will be continued till 36 weeks or delivery, whichever comes earlier. The women will have their haemoglobin fraction assayed by HPLC, mid-stream urine for microscopy, culture and sensitivity done and full blood count will be done at enrolment.

The women will be seen 2 - 4 weekly in the antenatal clinic up till 28 weeks and then weekly thereafter till delivery. Their body weight, blood pressures, haemoglobin concentration and urinalysis for protein and glucose will be measured at every visit. If proteinuria of ≥1+ (≥30mg/dl) is recorded in a participant with elevated blood pressure (hypertension), urine specimen will be collected for urinary protein to creatinine ratio estimation (UPCR) to confirm if proteinuria is significant, a value of 30mg/mmol will be considered significant. Their full blood count will be repeated at delivery. They will have an ultrasound scan at 20 weeks. Their oxygen saturation will be monitored in addition to other vital signs such as pulse rate, blood pressure, temperature, and respiratory rates whenever they are admitted to the hospital. At delivery, the babies will be weighed naked to detect low birth weight based on WHO INTERGROWTH-21st Chart. Placenta biopsies will be taken for histology to identify abnormalities that might be associated with some pregnancy complications such as PE and IUGR.

Adverse events will be recorded on the adverse event CRF as any symptom that starts or worsens after study drugs have been commenced. The severity (mild, moderate, severe) of adverse drug event, relationship to the study drug (suspected/not suspected), duration (start and end dates or if continuing at final exam) and whether it constitutes a serious adverse event (SAE) will be recorded at each visit. Pill count will be done at each visit. Reminder messages and call will be made to remind participants to take their drugs and to make their appointments. Participants who miss a visit will be tracked to their houses, if necessary, as a strategy to minimize loss to follow up. The participants were followed up till six weeks postpartum.

Data analysis plan Data analysis will be by intention-to-treat. Categorical variables will be expressed as frequencies and percentages. For continuous variables, a Shapiro-Wilk test of normality will be performed, and normally distributed data will be presented as means ± SD, while non-normally distributed data will be presented as median and interquartile range (IQR). Risk of occurrence of IUGR, perinatal death and other key outcome variables will be computed and compared in both groups. A multivariate regression analysis will be performed to determine the odds of each of the key outcomes among women who received the intervention with respect to those who did not, after controlling for common confounders. This will be presented as regression coefficients and their 95% confidence intervals. Level of significance will be set at 5%. Post-regression analysis will be performed to determine the goodness-of-fit of the final model. STATA version15.0 (Stata Corp LP, College Station, TX, USA) will be used for statistical analysis.

QUALITY CONTROL AND DATA MANAGEMENT Data will be double entered with assigned codes. Data will be captured in electronic case report forms (CRF) at various patient visit types and uploaded real time to the central server after being checked by site coordinators. They will follow guidelines specified in the SOP developed by the Data handling and communications committee for data collection, data entry, and transmission, data compilation and management and data quality and security.

ETHICAL AND ENVIRONMENTAL CONSIDERATIONS The clinical trial is registered in Pan African Clinical Trials Registry (PACTR202001787519553) and has ethical approvals from Health Research and Ethics committees of Lagos University Teaching Hospital, Idi-Araba (ADM/DST/HREC/APP/3301), Lagos State University Teaching Hospital, LASUTH (LREC/06/10/1318), Federal Medical Centre, Ebute Metta (FMCEB/RET/0052), Health Service Commission of Lagos State (LSHSC/2222/VOLIII), and National Health Research and Ethics Committee (NHREC/01/01/2007-04/12/2020).

Personal data of participants will be kept strictly confidential. All data will be stored securely in a central electronic database by the PI who will be the only one who will have access to the data of all participants collated centrally. The statistician will be granted access to the electronic database during statistical analysis or at any other time the PI might require her to review the data. None of the women will be made to pay for any aspect of the study as trial drugs and investigations will be offered at no cost. All participants will receive their routine medications (malaria prophylaxis, tetanus toxoid prophylaxis and folic acid supplementation) normally. Free malaria prophylaxis and folic acid will be given to them all through pregnancy as an incentive for participating in the research. Participants will have autonomy to participate and a right to withdraw from the study if they wish to do so. All participants will enjoy equal rights and quality care all through the duration of the research. Environmental issues are not applicable to this study.

MONITORING AND EVALUATION MECHANISM The M &E mechanism will include an Administrative Core consisting of the PI and a full-time project coordinator who will provide administrative oversight and management of the research study and coordinate all internal and external meetings of investigators and staff. A Steering Committee comprising PI, all co-investigators, site coordinators and a patient representative will be constituted, and will monitor project progress and ensure compliance with regulatory, fiscal, and reporting requirements. Clinical Trial Monitors who will be responsible for trial monitoring will also be appointed. A Data and Safety Monitoring Board (DSMB) will also be constituted.

Inclusion Criteria:

• Age 18 years and above
• Singleton fetus
• Women whose genotypes are Haemoglobin SS or SC.
• 28 weeks gestation or less at recruitment, estimated from the last menstrual period or by an early ultrasound scan.

Exclusion Criteria:

• Women with associated medical conditions in pregnancy e.g., HIV infection, diabetes mellitus, hypertension, renal disease, sickle nephropathy
• Multiple pregnancy
• Hypersensitivity to aspirin
• History of blood transfusion in the last 3 months
• Women who participated in the PIPSICKLE trial during their previous pregnancy.

Eligibility is based on self-representation of gender identity and confirmed pregnancy.

We will store the data and deposit it in 'Open Science Framework', after approval is obtained from the ethics committee. We will also provide metadata along with the data to describe it. No patient identifier will be included in data shared. Potential new users may access our data including the metadata on the 'Open Science Framework'. We will share the data at the time of publication of our first paper. The assigned DOI number, the OSF website details and our approach to data sharing will be included as an appendix to all publications emanating from this research to facilitate accessibility to our data and metadata. We will also share these at any conference presentation both international and local, and also on our study website to facilitate access to it by other researchers.

The IPD and supporting information will be made available at the time of first paper publication and will be made available for a period of 2 years

1. Governance of access The principal investigator will bear overall responsibility for this data and will be responsible for deciding whether to supply research data to a potential new user. The CMUL HREC will provide an independent oversight function. 2. The study team's exclusive use of the data Data will be made available at the time of publication, at the latest. Depending on the nature of the data itself, data may be made available earlier, either on an individual basis to interested researchers and/or potential new collaborators. 3. Restrictions or delays to sharing, with planned actions to limit such restrictions We will ensure that our informed consent forms clearly spell out and seek consent for future data sharing. 4. Regulation of responsibilities of users All external users will sign and be bound by our data sharing agreements and will not be allowed to use the data for reasons other than stated in their application.

Double blind design. An unblinded pharmacist does the drug packaging and has the code for drug identification. All other members of the research team and patients are blinded and do not know what each drug is.