Birthweight And Postnatal Growth As Predictors of Elevated Blood Pressure In Adolescents of Low Socioeconomic Condition: A Cohort Study In Northeast Brazil

25 Objective: To evaluate low birthweight and rapid postnatal weight gain as predictors of 26 elevated blood pressure in adolescence in a population of low socioeconomic status. 27 Methods: A cohort study was carried out with 208 adolescents, 78 born with low weight 28 and 130 born with appropriate weight. The infants were followed up during the first six 29 postnatal months and reassessed at 8 and 18 years of age. The main exposure variables 30 were birth weight and weight gain in the first six postnatal months. Rapid weight gain 31 was defined when above 0.67 z score. The investigated co-variables were: sex, maternal 32 height and family income at birth, breastfeeding duration from birth to six months, 33 nutritional status at eight years old, socioeconomic conditions, nutritional status, fat mass 34 index and physical activity level at 18 years. The outcome variable was the occurrence of 35 elevated blood pressure at 18 years old. 36 Results: The proportion of adolescents with elevated blood pressure was 37.5%. The 37 multivariable logistic regression analysis showed the variables independently associated 38 with a higher chance of elevated blood pressure in adolescence were rapid postnatal 39 weight gain (OR=2.74; 95% CI 1.22-6.14; p=0.014), male sex (OR=4.15; 95% CI 1.66- 40 10.38; p=0.002) and being physically active (OR=2.70; 95% CI 1.08-6.74; p=0.034). 41 Conclusions: The occurrence of rapid weight gain in the first six postnatal months was a 42 predictor for elevated blood pressure in adolescence. This result highlights the influence 43 of factors related to development in early childhood on health problems in the future. 44


INTRODUCTION
Systemic arterial hypertension is characterized by the sustained elevation of the 51 pressure exerted by the blood on the walls of arteries and affects approximately 1.13 52 billion people throughout the world [1]. It is the main chronic noncommunicable disease 53 [2] and the greatest risk factor for heart and cerebrovascular diseases [3]. 54 The symptoms of hypertension rarely appear in the early stages and emerge 55 throughout adulthood, resulting from diverse factors, such as the ageing process, a poor-56 quality diet, excessive alcohol intake, inadequate physical activity, excess weight, and [12]. This mechanism seems to have a beneficial effect on the newborn during these first 68 two years [13], but is related to metabolic disorders in adulthood, especially when 69 occurring in the first six months of life [14]. 70 As rapid postnatal weight gain seems to favor the development of chronic diseases 71 and obesity [15,16], it is important to follow up such individuals to ensure the control of 72 the development of chronic noncommunicable diseases and identify this association in 73 the early phases of the life cycle. Therefore, the aim of the present study was to evaluate 74 low birth weight and rapid postnatal weight gain as predictors of elevated blood pressure 75 among adolescents in a population with a low socioeconomic status. Birth cohort and follow-up data 86 The study began in 1993 and 1994 with the recruitment of 549 full-term children 87 with a gestational age greater than 37 weeks: 206 with low birth weight (LBW -1800 to 88 2499 g) and 343 with appropriate birth weight (ABW -3000 to 3500 g). The inclusion 89 criteria were belonging to families who earned up to four times the Brazilian monthly 90 minimum wage as well as the absence of infection, congenital anomalies, and genetic 91 syndromes [18]. 92 The mothers were interviewed using a form with pre-coded closed-ended 93 questions addressing socioeconomic status, mother's exposure to smoke during 94 pregnancy, and characteristics of the newborn. Anthropometric data and information on 95 the duration of breastfeeding were also collected. activity. Losses to follow-up accounted for 60.47% of the initial sample, as 217 108 adolescents were found at 18 years of age. Most losses occurred due to difficulty locating 109 individuals and changes of address to another state. Despite the losses, no statistically 110 significant differences in socioeconomic characteristics were found between those who 111 remained in the study and those who dropped out (Appendix 1).

112
The present paper involves data from 208 adolescents (71 with LBW and 137 with 113 ABW) whose BP was measured. The BP of nine adolescents was not analyzed because 114 the cuff of the equipment was inadequate for the arm circumference of these individuals 115 due to obesity. The adolescents in the sample were those with data from six months, eight 116 years, and 18 years of age.

118
Variables of interest 119 The independent variables were birth weight and postnatal weight gain. The 120 following were the covariables in each evaluation phase: at birth -child's sex, height, 121 and birthweight, mother's exposure to smoke during pregnancy, family income 122 (categorized as < or ≥ the Brazilian monthly minimum wage, which corresponded to 123 US$70 at the time), and total duration of breastfeeding using the median of 40 days as the 124 cutoff point; at eight years of age: body mass index (BMI), waist circumference and gain 125 in BMI between six months and eight years; at 18 years of age: family income per capita, 126 gain in BMI between eight and 18 years, height, BMI, fat mass index, and physical 127 activity level. The outcome variable was elevated blood pressure at 18 years of age.  The adolescents were instructed to remove any jewelry and accessories from the head, 155 stand erect with the head positioned parallel to the horizontal plane, knees extended, feet 156 together, arms alongside the body, palms turned toward the thighs, and ankles, buttocks, 157 and shoulders in contact with the stadiometer. Height was measured twice and the mean 158 of the two measurements was considered for the analysis. If a difference of more than 0.5 159 cm was found between the measurements, height was measured a third time and the two 160 closest measurements were used for the determination of the mean. Median height 161 according to sex was considered as the cutoff point to establish categories.

162
BMI at six months, eight years, and 18 years of age was transformed into z scores 163 based on the distribution curve of the population studied. Differences in BMI in z scores 164 between eight years and six months as well as between 18 years and eight years were 165 considered to determine the growth gain in these periods. The same cutoff point for 166 weight gain between birth and six months of age was adopted.

167
Waist circumference was measured on the adolescents in the same manner as 168 performed at eight years of age. The waist circumference of three female adolescents who 169 were pregnant for more than 12 weeks was not considered in the study.   and Pearson's chi-square test was used with the level of significance set at 5% (p ≤ 0.05).

214
The predictor effect of the independent variables on the outcomes was evaluated   The adolescents with the highest mean systolic pressure were males, those whose 244 mothers were taller, and those who had a family income equal to or higher than the 245 Brazilian monthly minimum wage at birth. A high mean systolic pressure was also found 246 in adolescents who exhibited rapid weight gain in the first six months of life as well as 247 those who, at eight years of age, had excess weight and a high waist circumference. At 248 18 years, having height above the median, excess weight, and being physical active were 249 also associated with a higher mean systolic arterial pressure (Table 1). 250 Higher mean diastolic arterial pressure was found among adolescents whose 251 mothers were taller, those who, at eight years of age, had a high waist circumference, and 252 those who, at 18 years of age, were taller, had excess weight, and had excess body fat 253 ( Table 1).

254
A greater frequency of elevated blood pressure in adolescence was found among 255 males, those who had rapid postnatal weight gain, excess body weight, a high weight 256 circumference, as well as those who were taller in adolescence (Table 2).

257
The results logistic regression analysis of the factors associated with elevated 258 blood pressure in adolescence are displayed in Table 3. The likelihood of having elevated