Vitamin A deficiency (VAD) is still a serious public health problem in most developing countries. Several strategies are used to prevent and address the consequences of this deficiency and to reduce its prevalence, particularly in Africa. In Senegal, the prevalence of VAD, although low among women of reproductive age, is quite worrying among children under 5 years old. In 2009, the fortification of refined oil with vitamin A was made mandatory in addition to the strategies already in place. The study of the impact of these strategies on the vitamin A status of women and children, showed relatively stable prevalences between 2010 and 2018. However, this study used plasma retinol concentration as an indicator. It is known that evaluation of vitamin A status is relatively insensitive when based on changes in plasma retinol concentrations, which are homeostatically controlled and negatively affected by subclinical infections. Incremental studies in the Dakar region using the modified relative dose response (MRDR) test in children under 2 years of age have indicated adequate vitamin A stores and a low prevalence of vitamin A deficiency in these children. The various strategies to prevent and control vitamin A deficiency have reportedly improved and even increased vitamin A stores in women and children, particularly in the Dakar region. Indeed, the latter benefit from substantial intakes of preformed retinol through the fortification program, and the majority of children under 2 years of age are breastfed.

The aim of this study is to use a more sensitive method than plasma retinol, the retinol isotope dilution (RID) test, to assess the actual status of subjects following these different strategies and to better orient the policies implemented in Senegal.

Specifics objectives are:

To determine transfer of tracer to breast milk To determine the composite coefficient (FaS) using mathematical modeling To estimate TBS using the generated coefficients and the RID equation among mothers To estimate TBS from breast milk retinol using the mass balance equation among mothers To estimate TBS from serum retinol using the mass balance equation among mothers and their children To assess the agreement between TBS measured by different methods (the RID equation, the mass balance equation using serum or breast-milk derived retinol) among mothers To evaluate the relationship between mothers' and children's TBS according to their status To assess dietary intakes of vitamin A among mother-child pairs

The study is designed following a "Wonder women model". It will be implemented in Dakar city. A total of fifty-six (56) lactating mother-child pairs will be included in the study. The women will be divided into 8 groups of 7 women each. This sample size includes an attrition rate of 15%.

At baseline (day 0), blood samples will be obtained from all the women after an overnight fast. Women will also be asked to give casual milk samples. Then, each woman will be given an oral dose of 2.0 μmol 13C2 -retinyl acetate (Gannon et al., 2014; Kaliwile et al., 2021). Afterward, participants will consume a peanut butter-based snack to provide fat to improve dose absorption. The women will also have a follow-up sample taken at day 14. As for baseline, a follow-up milk sample will be collected. Short-term sampling of both blood and milk will occur between days 0 and 14 and between days 14 and 109. In total, each woman will have 4 blood and 4 milk samples taken over time.

For infants, the RID test will start at day 14. After the baseline blood, they will receive 1.0 μmol 13C2-retinyl acetate and will have a second blood sample at day 28. Thus, infants will have 2 blood samples.

The protocol will be explained to the mother and written consent will be obtained from her. Lactating mother-infant pairs will be dewormed before starting the research protocol.

The blood samples will be protected from light with aluminum foil and centrifuged at 3500 rpm for 15 minutes to separate serum from clot. All centrifugations will be done in the hour following the blood sampling and using a centrifuge. The serum will be aliquoted into sterile cryogenic vials and stored at -80 °C at the laboratory.

Regarding milk samples, 2 aliquots will be prepared in 2 cryovials and stored at +4°C in the field and at -80°C at the laboratory.

An interviewer-administered questionnaire will allow collection of background information and dietary data from mother-infant pairs. Dietary intakes of mother-infant pairs will be assessed using 2 quantitative 24-hour dietary recalls on non-consecutive days (FAO, 2018). The Minimum Dietary Diversity for Women (MDD-W) will also be determined from the 24-hour dietary recalls. For children, Infant and Young Child Feeding practices indicators will be assessed following the latest WHO/UNICEF recommendations (WHO/UNICEF, 2021).

Anthropometric measurements will be performed to evaluate the nutritional status of mother-infant pairs according to standard procedures.

Sample analysis:

Serum Retinol concentration will be done by HPLC. Natural abundance of 13C in serum and milk will be measured by Gas Chromatography-Combustion-Isotope Ratio Mass Spectrometry (GC-C-IRMS). CRP and AGP will be measured by Biosystems A15 automatic analyzer. Retinyl esters and serum carotenoids will be done by Ultra-Performance Liquid Chromatography (UPLC).

Anemia will be assessed by measuring hemoglobin (Hb) in whole blood using a HemoCue system (Hb-301) and malaria testing will be performed using SD BIOLINE Malaria Antigen P.f/Pan1 tests.

Further analysis will be done on saliva samples in a subsample of mother-child pair to measure milk production by the mother-dose deuterium isotope dilution method. Deuterium enrichment of saliva samples will be measured with a Fourier transform infrared spectrometer (FTIR Shimadzu IR Affinity, Kyoto, Japan).

Modelling retinol kinetics among mothers :

Serum and breast milk retinol concentrations and isotope enrichment data will be used to generate super women data sets. Model-based compartmental analysis will be applied to super women isotope tracer data for serum and breast milk enrichment using established models for vitamin A metabolism (Green et al., 2020; Gannon et al., 2018) and expanding these models to incorporate tracer transfer to breast milk.

Determination of Fa and S factors:

Outputs from the model-based compartmental analysis include estimates for vitamin A transfer rates among physiologically-based compartments and the resulting model outcomes; total traced mass, half-life, fractional catabolic rate, disposal rate, equilibration time, and serum/body tracer partitioning (factor S). Using absorption assumptions, the amount of tracer dose absorbed and stored at the time of sampling (related to factor Fa and other related equation coefficients) will be determined.

TBS calculation:

Mass balance equation: The isotopic ratio of tracer to tracee (TTR) (unlabeled retinol) in serum will be used to estimate total body VA stores (TBS) using the following equation with appropriate assumptions:

TBS (μmol)= a × 1/TTR× (factors for correction of absorption and storage) a is the dose amount RID equation: TBS (μmol)= Fa × S × (1/〖SAp〗) Fa is the fraction of the oral tracer dose absorbed and retained in stores S is the ratio of the specific activity of retinol (labeled retinol/unlabeled retinol) in plasma to that in liver after dose equilibration SAp is the specific activity of retinol (labeled retinol/unlabeled retinol) in plasma.

For mothers, Fa and S will be calculated using mathematical modelling while SAp will be analytically determined. Among children, generic values for Fa and S will be used.

Statistical analysis:

Data entry and quality control will be performed using Epi infoTM 7.2.3.1, Epi infoTM 3.5.1 (Centers for Disease Control and Prevention) and Microsoft Excel 2016 (Microsoft Corporation). Statistical analysis will be performed using STATA/SE 14.0 (STATA Corporation). Descriptive analysis will be used to tabulate the characteristics of the study population. Categorical variables will be expressed as percentages, and continuous variables as mean ± SD for normally distributed variables and median with interquartile range for skewed values. Two-tailed Student's t- test, Paired t-test or one-way analysis of variance (ANOVA) followed by Bonferroni's post-hoc comparison tests will be used to compare means while the Wilcoxon rank-sum test will be used to compare medians. Pearson's chi-squared test, Fisher's exact test or McNemar's chi-square test will be used to compare percentages. Appropriate agreement tests will be performed. All the statistical analysis stages will be performed by an experienced statistician using appropriate methodology to ensure proper and reliable outcomes.

Inclusion Criteria:

• For mothers: Willing to participate in the study and to give 4 blood and 4 milk samples during the whole duration of the study, Being aged 18-45 years, Not planning to travel during the study duration
• For children: Being aged 9-18 months, Being breastfed, Mother willing to authorize 2 bloods, Do not receive vitamin A supplements during the study

Exclusion Criteria:

• For mothers: Being pregnant, Consuming vitamin A and/or effector supplements, Smoke or drink alcohol during the study period
• For children: Have received a vitamin A supplement in the two months prior to the study
• All: Had fever or any obviously known or clinical diagnosed disease, malaria positive, severe anemia, severe acute malnutrition or severe obesity, being hemophiliac