The Body Size and Micronutrients Status Among the Bride-To Be in Probolinggo Districtof East Java

Micronutrients deficiency is prevalent among women in reproductive age, particularly in developing countries, including in Indonesia. Small body size reflects a chronic deficiency of nutrients intake. Both of those problems may correlate each other and making maternal nutrition more complicated. Brides-to be are the most important group of women in reproductive age who require better recognition, because they will become mothers. This study was to describe inter-correlation among anthropometric indices which reflected the maternal body size and several indicators of micronutrient status. A cross sectional study was conducted in selected sub-districts of Probolinggo District, East Java Province year 2012. The samples size of 115 brides-to be were randomly selected from list of brides-to be in Sub-district Office of Religious Affair. The observed variables were body weight, height, body mass index (BMI), mid upper arms circumference (MUAC), micronutrients status including concentration of hemoglobin, serum ferritin, serum retinol and serum zinc. Pearson correlation test was used to analyze data using SPSS for Window version 13.0. Result showed that the average of body weight was 47,3 kg, average stature was 150,4 cm, BMI was 20, MUAC was 25,3 cm. The average of hemoglobin level was 12,4 g/dL, serum ferritin was 50,6 µg/L, serum retinol level was 1,06 µmol/L, and serum Zn level was 12,9 µmol/L. Statistical analisis showed significant correlation between the body weight and hemoglobin concentration (r =0,25, p = 0,01) and serum retinol (r = 0,21, p = 0,03); between BMI and hemoglobin concentration (r = 0,31, p = 0,00) and serum ferritin (r = 0,23, p = 0,03), and serum retinol (r = 0,21, p = 0,02); between MUAC and hemoglobin (r = 0,32, p = 0,00) and serum retinol (r = 0,24, p = 0,02); between hemoglobin concentration and serum ferritin concentration (r = 0,30, p = 0,00). Stature did not correlate with any indicators of micronutrient. It implies that among the indices of body size, body weight and BMI as well as MUAC are important indicators relating to iron status, and vitamin A status, but not zinc status.


Introduction
Protein energy deficiency and micronutrients deficiency are a problem occurred in pregnant women and reproductiveage women in several countries, exclusively in region of South-Central Asia. In some Central Asia countries, more than 10% of reproductiveage women have a height less than 145 cm. Prevalence of maternal under nutrition that are women with body mass index (BMI) below 18.5 kg/m², varies from 10% to 19% in different countries. Meanwhile, in South-East Asia, the prevalence of nutritional deficiency in women group was more than 20% (Black, 2008).
Riskesdas data in 2013 showed the prevalence of chronic energy deficiency risk in pregnant women aged 15-49 years  Alamat korespondensi: Fakultas Kesehatan Masyarakat, Universitas Airlangga, Surabaya Email : msrisumarmi@gmail.com pISSN 1858-1196 eISSN  Prevalence data of vitamin A deficiency of Indonesian reproductive-age women was still limited. The data obtained from Helen Keller Indonesia, by clinical manifestations, showed that prevalence of night blindness was fairly high in reproductive-age women in rural area of East Java and of South Sulawesi, and in Surabaya shanty town was higher than 5% (Sari, 2004). Available prevalence data of vitamin A deficiency was in Indonesian toddlers. In 2006, the average of subclinical vitamin A deficiency prevalence in toddlers (Vitamin A serum < 20 ug/dl) of 7 provinces (West Sumatra, South Sumatra, Banten, Bali, West Nusa Tenggara, South Kalimantan, and South-East Sulawesi) was 11.4% (Herman, 2007) Several studies suggested an association between Iron deficiency or other micronutrients (such as Zinc (Zn), vitamin A, and vitamin C) and the body lipid deposition (Garcia, 2009;Aderibigbe, 2014). But the other studies showed the association between nutritional deficiency and anemia (Qin, 2013;Anticona, 2014). Sri , presented that underweight as the risk factor affecting the occurrence of depleting of Iron storage in the body. A Philippian study demonstrated that body mass index in girls correlated negatively with provitamin A level and vitamin E level (Ribaya-Mercado, 2008).
Body size of women included body weight, height, and nutritional status that represented as a BMI, as a predictor of the neonatal birth weight had been documented (Neggers, 2003). Iron status and several micronutrients status also affected a pregnancy and its outcomes. Both animal studies and human observational studies evidenced that micronutrients deficiency in pre-pregnancy or during pregnancy was associated with a poor pregnancy or neonate (Ronnenberg, 2004;Bodnar, 2006;Catove, 2007;Beard, 2007;Sri Sumarmi, 2015). The low quality of the newborn baby was probably caused by not a single micronutrient deficiency, but by multiple micronutrients deficiency (Sunawang, 2009;Bhutta, 2009;Cetin, 2010;Kawai, 2011;Black, 2013).
Considering the importance of maternal nutritional status and micronutrients status in pre-pregnancy or during pregnancy, this study aimed to analyze the correlation between body size, nutritional status, and micronutrients status in the brides-to-be. By studying the association between body size and micronutrient status, these parameters would explain the importance of micronutrients, especially Iron, Zinc, and vitamin A in regulating body composition. Besides, it was possible to estimate the deficiency of micronutrients based on feasibility of body size measurement.

Method
A cross-sectional study was conducted in 9 Sub-districts of Probolinggo District, East Java, in 2012. The study protocol had been approved by ethical commission of Faculty of Medicine-Universitas Gadjah Mada, Yogyakarta, by a release of Ethical Clearance No. KE/FK/202/ EC. We informed the respondents aboutstudy objectives. After listening and understanding, the respondents agreed to participate in the study by signing an informed consent.
Probolinggo District was selected as the location of study considering the high prevalence of nutritional problems in the married reproductive-age women, especially Chronic Energy Deficiency by 27.3% and anemia by 48.5% (Sri Sumarmi, 2010), while percentage of anemia in brides-to-be was 15% (Putri, 2013). Samples of 115 brides-tobe that selected randomly from the data lists documented at Religious Affairs Office(KUA), were met inclusion criteria: group aged 16 to 35 years old, physically health checked by medical examination in Primary health care (Puskesmas), and not pregnant. The sample size was calculated by Sample Size Software version 2.0.
The observed body size variables were body weight, height, BMI, upper arm circumference (UAC), whereas the observed micronutrient variables were hemoglobin levels, serum ferritin levels, serum retinol levels, and serum Zn levels. The body weight was measured using digital scale Seca® type 803 with a level of accuracy of 100 grams. Height was measured using microtoise with a level of accuracy of 0.1 cm. BMI was calculated as the ratio of weight in kilograms and height in meters per square (kg/ m 2 ). For the brides-to-be who were younger than 20, nutritional status was assessed by BMI for age (De Onis, 2007), while for the brides-to-be who were older than 20, was assessed by WHO criteria (Gibson, 2005). Hemoglobin levels were measured from fingertip capillary blood, analyzed using a portable device HemoCue® AB of Anglehome, Sweden. Serum ferritin levels were obtained from venous blood samples, and were analyzed by Micro Particle Enzyme Immunoassay (MEIA) using Axym MEIA. Serum retinol levels were analyzed by High Performance Liquid Chromatography (HPLC). Serum levels of Zn were analyzed using atomic absorption spectrophotometry (AAS) method. Analysis of serum ferritin levels was conducted at Laboratory of Prodia Surabaya, while the analysis of serum retinol levels and serum Zn levels were conducted at Laboratory of SEAMEO-RECFON Jakarta. Anemia was assessed from hemoglobin level less than 12 g/dL, while iron deficiency or depletion of iron storage condition was defined by serum ferritin level less than 15 mg/L (WHO, 2007). Vitamin A deficiency was assessed by serum retinol levels less than 0.7 mol/L (WHO, 2011), while Zn deficiency was assessed from serum Zn levels less than 10,1 μmol/L (Wessells, 2012).
Data was managed using the SPSS for Windows Version 13.0 Statistical Software Package (SPSS Inc., Chicago, IL, USA). Data normality test was performed to the entire data using a one-sample Kolmogorov-Smirnov test. The test result showed a normal distribution of data (p> 0.05). Bivariate correlation analysis using Pearson correlation was conducted to observe a strong correlation between body size variables (weight, height, BMI and UAC) and micronutrient status (the levels of hemoglobin, serum levels of ferritin, serum levels of retinol, and serum levels of zinc).

Results and Discussion
Characteristics of respondents observed in this study included age, education level, occupation, and income. Detailed data was presented in Table 1. Table 1 showed that most respondents classified as young or younger than 25 years, which was around 80%, while the rest were over 25 years old. Age considered as an important variable because of the diversity of age would affect the diversity of physiology (Guyton, 2006) and the rate of protein energy metabolism and other nutrients (Stipanuk, 2006). Therefore, age was one of the dependent variables. How to control the age diversity was by including age factor in the inclusion criteria for the respondents. Respondents in this study were women ranged from 16 to 35 years. This age restriction was regarding the reproductiveage of women.
The education level of most respondents was classified at the level of basic education: education of 9 years or until the Junior High School. Most respondents were unemployment, therefore they had noincome, and less than 40% of respondents were working. Most of the no income respondents were teachers and labors. The majority of employed respondents obtained low income ≤ 1 million. Table 2 presented the value of descriptive statistics of the body size variable which included weight, height, BMI, and UAC measurements, as well as the concentrations of several micronutrients in the serum, such as levels of hemoglobin in the blood, the serum levels of ferritin, serum levels of retinol, and serum levels of Zn. Data was presented in average value ± standard deviations (SD) and range.
Respondent's weight ranged from 29 kg to 72.2 kg, with an average value of 47.3 kg and SD of 8.6 kg. The respondents average height was 150.4 cm with SD of 5.6 cm. BMI ranged from 15.4 to 34.5 with an average value of 20.9 and SD of 3.7. The smallest UAC was 16.5 cm while the largest UAC was 35.3 cm, with an average value of 25.3 cm.
The weight was rarely used as a single parameter for assessing the nutritional status, but the women weight before pregnancy could be a predictor of neonatal birth weight. The combination of the weight to the height into a BMI was a good predictor of neonatal birth weight (Neggers, 2003;Nahar, 2007;IOM, 2009). Nahar (2007), used the weight as a single indicator in early pregnancy by under the threshold of 45 kg was sensitive to risk a low birth weight (LBW). Another study used women weight before pregnancy by threshold of 40 kg as a risk factor for low birth weight (Sri   (Black, 2014). Women with body height less than 145 cm deliver baby with lower birth weight than women with body height more than 145 cm (Sri, 2016). Nahar (2017), use 150 cm as cut off for body height. This study use hemoglobin level and serum feritin to determine iron body status. Respondent have hemoglobin level between 9,1 g/dL to 15,7 g/dL, with mean and standard deviation 12,4 g/dL and 1,14 g/dL, respectively. Serum feritin level means are categorized as good with 50,6 µg/L, and wide range from 3,2 µg/L to 188,3 µg/L. From hemoglobin and serum feritin means, we can conclude respondent have normal iron body status. Non anemia hemoglobin level, which is above 12 g/dL, whereas serum feritin means above 15 µg/L and below 60 µg/L mean respondent have enough iron body storage or normal repleted (WHO, 2007) Retinol and Zn serum levels represent vitamin A and Zn status. Respondents retinol serum level range are between 0,41 µmol/L to 1,81 µmol/L, with mean value 1,06 µmol/L. Whereas Zn serum level mean value is 12,9 µmol/L with range between from 4,6 µmol/L to 22,03 µmol/L. From Retinol and Zn means level, our respondent vitamin A and Zn status are categorized as good and have no deficiency. Vitamin A and Zn deficiency are a condition where retinol and Zn serum level are less than 0,7 µmol/L and 10,1 µmol/L respectively (Gibson,2005). Subclinical vitamin A deficiency happen when retinol serum level is less than 0,7 µmol/L without clinical sign evidence. (WHO, 2011)  Underweight percentage in this study (23,5%) was higher than national prevalence of underweight in adult women (9,4%) based on Riskesdas 2013 (Irawati, 2013). Compared with Utami (2012), study, our study underweight percentage was lower. In comparison overweight and obesity percentage on respondent in this study was lower (12,2%) than national prevalence in adult women (32,9%) based on Riskesdas 2013 (Irawati, 2013) Anemia percentage (hemoglobin less than 12 g/dL) in our respondent about 33%, mostly were mild anemia (hemoglobin between 10,1-11,9 g/dL) 30,4% and without severe anemia case found. Anemia prevalence in this study and female teenager in West Lombok, Nusa Tenggara Barat were similar (Masthalia, 2015), but higher compared with Riskesdas 2013, about 22,7% (Balitbangkes, 2010). 15,7% of respondent suffered low or depleted iron storage. Low iron storage in body (less than 15 µg/L) indicates iron deficiency states (WHO, 2017). In conclusion almost half of respondent who suffer anemia in this study (14,3%) were anemia non iron deficiency, which caused by other micro-nutrient deficiency.
About 7% of respondent suffer vitamin A deficiency (retinol serum level less than 0,7 µmol/L). This percentage was higher than vitamin A deficiency in reproductive female age in Cambodia which were 0,7% (Wieringa, 2016). But respondents had lower percentage of vitamin A deficiency than primary school age female in Marliyati study (2014), in the amount of 58,1%. Whereas Zn deficiency percentage in this study (26,1%) was higher than reproductive age female in Mexico, that were 17,4% (Garcia,2012). Table 4. shows the result of statistical analyze using Pearson correlation on body size that include body weight, body height, body mass index, and upper arm circumference with hemoglobin level, ferritin serum level, retinol serum level, and Zn serum level. The result showed correlation between body size parameter which statistically significant had correlation with hemoglobin level were body weight, body mass index, and upper arm circumference. Whereas that has statistically significant correlation correlation with ferritin serum levelwas only body mass index. While in retinol serum level case, body weight, body mass index, and upper arm circumference all have statistically significant correlation. While there were no body size parameter that had statistically significant correlation with Zn serum level.
Body mass index and upper arm circumference had significant positive correlation with hemoglobin level at level of confidence α = 0,01 with correlation coefficient r=0,31, and r=0,32 respectively and p value p<0,001. Which means the higher body mass index and upper arm circumference, hemoglobin level will also higher. This result was similar with other study in China (Qin, 2013), also in Bangladesh (Kamruzzaman, 2013). Theoretically, body mass index and upper arm circumference can reflect nutritional status, which is the manifestation of adequacy in energy and protein diet. The first step in heme synthesis is condensation of glycine and succynil Co-A to form δ-aminolevulinic acid (ALA) with ALA synthase enzyme help which activated by vitamin B6 in pyridoxal phosphate form (Ajioka, 2006). Succinyl Co-A is the result of carbohydrate and fat catabolism via tricarboxylate acid (TCA) cycle (Stipanuk, 2006). Body mass index had weak correlation with ferritin serum level (r=0,23; p=0,03). This result differed with some previous study. Aderibigbe (2014), explain that iron deficiency had correlation with fat deposit in body or called adiposity. Micronutrient deficiency also ascociated with obesity and fat deposit. Therefore, micronutrient deficiency may coexisted with obesity (Garcia, 2009). Adipose tissue produce pro-inflammatory cytokines such as interleukin-1, interleukin-6, and Tumor Necrosis Factor-α (TNFα) and also adipokines (leptin, adiponectin, resistin) that influence iron balance in the body (Zekanowska, 2011).
Some body size parameter such body weight, body mass index, and upper arm circumference have positive correlation with retinol serum level. This study differed from Ribaya-Mercado (2008), in Philippine that showed there was no correlation between body mass index with retinol serum level, because retinol serum that circulate in blood, physiologically controlled by vitamin A concentration in liver. Also body mass index has negative correlation with carotenoid and vitamin E level. This phenomenon can be explained that body mass index is a predictor of body fat composition while vitamin A and carotenoid is lipid-soluble compound that stored in liver and fat There were no body size parameter that had correlation with Zn serum level. This result correspond with Ibekwe (2013), study in Nigeria that showed that there was no relation between anthropometric parameter with Zn serum level in female teenager. As other study, also showed no relation between Zn level in plasm with body mass index in Mexico's women (Garcia, 2012).

Conclusion
High percentage of underweight, overweight (BMI > 25), and also micro-nutrient deficiency, especially anemia and Zn deficiency were found among future brides in Probolinggo district, East Java. Some of body size parameter, such as body weight, body mass index, and upper arm circumference have positive correlation with iron status (hemoglobin and ferritin level) and vitamin A status (retinol serum level), but have no correlation with Zn status. Body height has no correlation with micro-nutrient status. These can deduced body size parameter that correlate with body fat composition is sensitive to show change in micro-nutrient status, unlike body height that not correlated with body composition, moreover is a long term parameter that not sensitive to show short term change in micro-nutrient status.

Acknowledgement
Our gratitude belongs to Indonesian Danone Institute Foundation for funding this study. Head of Bappeda and head of Probolinggo district Public Health Office and his crew, also Prodia Surabaya Laboratory staff and Medicine Faculty UI SEAMEO-RECFON laboratory staff, and all other party that help this study proceed successfully.