Secondary Logo

Journal Logo

BRIEF REPORTS

Weight Gain and Blood Pressure in Toddlers Born Very Preterm

Rodriguez, Jeannie; Adams-Chapman, Ira; Affuso, Olivia; Azuero, Andres; Downs, Charles A.; Turner-Henson, Anne; Rice, Marti

Author Information
doi: 10.1097/NNR.0000000000000415
  • Free

Abstract

Heart disease is the leading cause of death in the United States, and hypertension is a key contributor to its development (Centers for Disease Control and Prevention, 2017). Hypertension affects about 3.5% of children, with the incidence much higher in children with certain chronic conditions, including obesity and prematurity (Flynn et al., 2017). Furthermore, hypertension and elevated blood pressure readings in childhood often continue into adulthood (Shen et al., 2017).

Preterm infants exhibit a weight gain pattern consisting of initial slow or poor weight gain followed by a period of rapid weight gain. This pattern is thought to contribute to the development of cardiovascular disease and is a primary postulate of what is now referred to as the Developmental Origins of Health and Disease (Hoffman, Reynolds, & Hardy, 2017; Wadhwa, Buss, Entringer, & Swanson, 2009). Metabolic and epigenetic changes are required to promote the survival of the premature infant both in utero and shortly after birth during a period of malnutrition and psychobiological stress (Hoffman et al., 2017). The preterm infant adapts physically to survive in a stressful and nutritionally inadequate environment. These adaptations become problematic when the infant begins demonstrating rapid weight gain. The sequelae of malnutrition and psychobiological stress followed by rapid weight gain in individuals born preterm has been associated with higher rates of obesity, hypertension, and cardiovascular disease in childhood and adulthood (Hoffman et al., 2017).

Reynolds (2013) suggested that the hypothalamic pituitary adrenal axis is affected during the adaptation process, resulting in a “reset” of cortisol to an elevated basal rate. Preterm infants have been noted to have elevated basal cortisol rates up to 18 months of age (Grunau et al., 2007). Winchester, Sullivan, Roberts, and Granger (2016) noted atypical diurnal cortisol patterns in young adults born preterm. Elevated cortisol levels result in an increase in cardiac contractility, cardiac output, and systemic blood pressure and, when maintained over time, can result in hypertension through wear and tear on the cardiovascular system (Reynolds, 2013).

The objective of this pilot study was to examine the effect sizes of the relationships between weight gain and blood pressure in toddlers born very preterm. A secondary aim was to note any mediating effect of cortisol on the relationships between weight gain and blood pressure.

METHODS

Research Design, Subjects, and Setting

A cross-sectional, correlational design was used to examine the effect sizes of the relationships between weight gain, cortisol, and blood pressure in a group of toddlers who were born very preterm (less than 32 weeks gestation). A convenience sample was recruited from two developmental clinics located in the southeastern United States. Participants had to meet the following inclusion criteria: (a) 18- to 24-month-old toddlers, corrected age; (b) born with a gestational age of less than 32 weeks; (c) born with a birth weight of less than 1,500 g; and (d) had a parent who could understand, speak, and read English and who was legally able to consent for the minor child. Exclusion criteria included toddlers with a current or chronic history of (a) congenital heart disease, (b) renal disease, (c) continuous oxygen therapy at home within the last 6 months, (d) endocrine diseases/disorders such as Cushings syndrome or Addison’s disease that are directly related to cortisol levels, and (e) a current diagnosis of failure to thrive.

Data Collection Procedures

The study protocol was reviewed and approved by the institutional review boards and research oversight committees at University of Alabama at Birmingham, Emory University, Children’s Healthcare of Atlanta, and Grady Health System. Informed consent was obtained from the parent of the participant. The study consisted of a one-time study visit and a chart review. The parent completed a demographic questionnaire, and a blood pressure reading and a saliva sample were obtained from the participant. All data on anthropometrics, inclusive of birth through current anthropometrics, were obtained via chart review. All measures (questionnaires, blood pressure, and saliva samples) were collected during morning clinic (8:30 a.m. to 1:00 p.m.) over a 9-month-period (summer–winter).

Weight Gain Calculation

Weight gain was calculated using the method described by Keijzer-Veen et al. (2005). First, a regression equation was computed with birth weight as the independent variable and current weight as the dependent variable (Keijzer-Veen et al., 2005). The outcome equation represented expected weight at 18–24 months corrected age (Keijzer-Veen et al., 2005). Expected weight (ew) was then subtracted from actual weight (aw) for each participant, which was termed the residuals (Xres = Xaw – Xew; Keijzer-Veen et al., 2005). The residuals represent weight gain that is more or less than expected at 18–24 months (Keijzer-Veen et al., 2005). This method was chosen because it best estimates the effect of growing more (or less) in weight than expected (Keijzer-Veen et al., 2005) and is most consistent with the conceptual definition of weight gain used in this study.

Saliva Collection and Cortisol Measurement

Salivary cortisol was collected and analyzed according to the protocol established by Salimetrics (2019), which has been validated in similar populations. Saliva was collected by placing a swab under the tongue for 60–90 seconds (Salimetrics, 2019). Samples were stored at −20°C (Salimetrics, 2019) until shipped to Salimetrics for determination of cortisol levels. Each sample was assayed in duplicate. The time of saliva collection, time of last meal, and time of morning awakening was noted. Because cortisol has a diurnal pattern with peaks prior to waking and a gradual decline throughout the day (Kliegman, Stanton, St. Geme, Schor, & Behrman, 2015), sampling was done during morning clinic, which occurred between 8:30 a.m. and 1:00 p.m.

Blood Pressure Measurement

Blood pressure was measured using a protocol established in a similar population by the American Heart Association (Flynn et al., 2017). Blood pressure was obtained via the right arm, with the participant seated, after approximately 5 minutes of rest and quiet activity (Flynn et al., 2017) with an automated device. The blood pressure measurement was repeated with a 5-minute rest period between each measure (Flynn et al., 2017). Measures of systolic and diastolic blood pressure were averaged, and the average was used in the analysis.

In order to address the validity of the blood pressure measurements, an ordinal scale was used to assess the participant’s behavior state at the time of blood pressure measurement. Categories included sleeping, awake and calm, awake and fussy/restless, and awake and vigorously crying/screaming (Duncan, Rosenfeld, Morgan, Ahmad, & Heyne, 2008). Participants were rated on a scale of 1–4 (1 = sleeping, 4 = awake and vigorously crying/screaming; Duncan et al., 2008).

Data Analysis

Data analysis was conducted using the computer software IBM SPSS Version 21. Data analysis included the use of descriptive statistics—and because the sample size was small and underpowered for inferential purposes, parameter estimates, effect sizes, and model fit estimates were the primary quantities of interest. Models were built using general linear models and regression techniques to allow for inclusion of control variables. The product of coefficients method (Tabachnick & Fidell, 2007) was used for estimation of indirect (e.g., mediated) and direct effects between weight gain and systolic blood pressure assuming cortisol as the mediator. Prior to running an analysis, the data were visually inspected for outliers, normality, and linearity (Tabachnick & Fidell, 2007). Multicollinearity was also assessed prior to running analyses on the full models.

RESULTS

The final sample after data collection was composed of 36 toddlers with corrected age of 18–24-months, which represented a participation rate of 71.2%. However, systolic and diastolic blood pressures were available for 18 and 23 participants, respectively (see Figure 1).

FIGURE 1
FIGURE 1:
Participants with blood pressure measurements. Note. aFive participants were too upset during blood pressure measurement to register a reading. bNine participants scored a 3 or higher on their state during blood pressure measurement and were excluded due to likely inaccurate readings. cFive participants had manual blood pressures taken instead of automated as per study protocol. To note any difference between means between those with manual blood pressures and those with automated blood pressures, the standardized mean difference between the five manual systolic and diastolic blood pressures and the 18 automated systolic and diastolic blood pressures were calculated. The effect size for the standardized mean difference for systolic blood pressure was large, whereas it was small for diastolic blood pressure. The five manual systolic blood pressures were removed.

Descriptive statistics and weight gain for the total sample (n = 36) have been previously reported (Rodriguez et al., 2018). Descriptive statistics for the sample in which blood pressures were obtained and used in the analysis (n = 23) are presented in Table 1. The mean corrected age at the time of data collection was 21 months. Most of the toddlers were female and Black/African American. Most had mothers who obtained a high school diploma or General Education Development. The mean gestational age at birth was 27 weeks, with a mean birth weight of 936 g.

TABLE 1
TABLE 1:
Descriptive Statistics for the Study Sample (n = 23)

The mean cortisol value was 0.279 μg/dl. The mean systolic blood pressure was 94.0 mm Hg. The mean diastolic blood pressure was 56.6 mm Hg. For systolic blood pressure, most of the participants were charted at less than the 90th percentile. For diastolic blood pressure, a slight majority of participants were charted at or above the 90th percentile, which is considered elevated (Flynn et al., 2017). Data on cortisol and blood pressure are reported in Table 2.

TABLE 2
TABLE 2:
Descriptive Statistics for Cortisol and Blood Pressure

The unstandardized regression coefficient for weight gain (in grams) and systolic blood pressure given the control variables (corrected age, race, gender, and socioeconomic status) in the model was .001, indicating a positive relationship. For every 1-kg increase in weight gain above what was anticipated, there was an associated 1-mm Hg increase in systolic blood pressure. Cohen’s r was estimated at .158, which is small and not clinically relevant. R2 and adjusted R2 for the entire model including control variables were .321 and .112, respectively.

The unstandardized regression coefficient for weight gain and diastolic blood pressure was .001, indicating a positive relationship. For every 1-kg increase in weight gain above what was expected, there is an associated 1-mm Hg increase in diastolic blood pressure. Cohen’s r was estimated at .366, which is a medium effect size. However, the R2 and adjusted R2 for the full model were .27 and .055, respectively, indicating that the model might explain little variance in diastolic blood pressure if applied to a different sample.

Given the medium effect size between weight gain and diastolic blood pressure, there was sufficient evidence to continue the analysis considering a mediating relationship between conditional weight gain, cortisol, and diastolic blood pressure. The unstandardized regression coefficient for weight gain on cortisol was .001, and Cohen’s r was estimated at .15, which is a small effect size. The R2 and adjusted R2 for the cortisol model were .25 and .03, respectively, indicating that the model might explain very little of the variance in cortisol if applied to a different sample. Using a third regression model for diastolic blood pressure, including both weight gain and cortisol as predictors, the estimated standardized mediated effect of weight gain through cortisol was .046, and the standardized direct effect was .32. The small magnitude of the mediated effect as well as the small relationship between weight gain and cortisol indicated that mediation was not supported in the sample.

DISCUSSION

In this study, 16.7% and 52.1% of the participants had elevated systolic and diastolic blood pressure readings, respectively. Although caution must be taken in interpreting these findings, researchers have noted that even single incident blood pressure elevations can continue into adulthood (Gauer & Qiu, 2012). The prevalence of elevated blood pressure readings among children has been reported between 5.4% and 21.0% across studies (Urrutia-Rojas et al., 2006). However, children born very preterm with a very low birth weight (VLBW) are typically excluded from these large, national studies. Research on blood pressure outcomes in early childhood in individuals born very preterm with a VLBW is scant. Bonamy, Källén, and Norman (2012) reported significantly higher z scores for systolic and diastolic blood pressures in toddlers born extremely preterm compared to those born at term. In particular, median diastolic blood pressure was significantly higher in those born extremely preterm compared to controls, with those born extremely preterm having higher odds of exhibiting elevated systolic and diastolic readings (greater than the 90th percentile; Bonamy et al., 2012). Furthermore, de Jong, Monuteaux, van Elburg, Gillman, and Belfort (2012) conducted a systematic review and meta-analysis of five studies comparing blood pressure between a preterm and term born cohort. They found that those born preterm had a 3.8-mm Hg higher systolic blood pressure in late adolescence/early adulthood compared to those born at term.

These studies do not directly describe the number of individuals with elevated readings; however, they do confirm that those born preterm have higher systolic and diastolic blood pressures compared to a term-born cohort during early and late childhood, adolescence, and early adulthood. It is well documented that individuals born very preterm with a VLBW carry a higher risk of developing hypertension in adulthood compared to their term counterparts (Barker, 2007), and some have suggested that blood pressure readings may amplify over time (Bonamy et al., 2012). If this is the case, the fact that elevations are already present during the toddler period could indicate more significant cardiovascular disease in adulthood for this population.

The effect size between weight gain and diastolic blood pressure was medium, although weight gain explained very little of the variance in diastolic blood pressure. Most of the literature on the topic of weight gain and blood pressure focus on systolic blood pressure only. However, Jones et al. (2012) and Hemachandra, Howards, Furth, and Klebanoff (2007) did note positive relationships between weight gain and diastolic blood pressure in school-aged children. If blood pressure continues to amplify as the individual ages, then the risk could be more evident in the adult years.

Race was considered only as a control variable in this study, but it is notable that most of the participants in this study were Black/African American. The prevalence of hypertension is higher among Black/African Americans (Centers for Disease Control and Prevention, 2017), and researchers have noted that Black/African American infants had higher cortisol levels at 12 months compared to White infants (Dismukes et al., 2018). Future research should consider racial differences as differences noted this early in life may magnify and contribute further to future health disparities.

Limitations

This study had several limitations. First, the small number of participants limits our ability to note relationships beyond effect sizes between variables of interest and make any inferential statements. Second, the convenience sample may mean that the sample characteristics are dissimilar to the actual very preterm, VLBW population in the United States. Third, both cortisol and blood pressure were only measured once. Ideally, we would work to measure the entire diurnal pattern of cortisol and repeated measures of blood pressure on at least three separate occasions. Finally, as a cross-sectional observational study, no causality can be inferred.

Conclusion

In this study, we noted higher than expected elevations in blood pressure, especially diastolic blood pressure and some evidence of a relationship between conditional weight gain and diastolic blood pressure. Both of these findings are troubling given the already existing vulnerability of the preterm population. Future research will focus on recruitment of a larger sample of individuals born very preterm with a VLBW noting diurnal cortisol patterns. In addition, we will seek to incorporate -omics measures inclusive of the gut microbiome and metabolomics to note additional influences and perturbations of metabolic pathways that may further explain relationships between weight gain and blood pressure.

REFERENCES

Barker D. J. P. (2007). The origins of the developmental origins theory. Journal of Internal Medicine, 261, 412–417. doi:
Bonamy A. K., Källén K., & Norman M. (2012). High blood pressure in 2.5-year-old children born extremely preterm. Pediatrics, 129, e1199–e1204. doi:
Centers for Disease Control and Prevention. (2017). Heart disease facts. Retrieved January 2019 from https://www.cdc.gov/heartdisease/facts.htm
de Jong F., Monuteaux M. C., van Elburg R. M., Gillman M. W., & Belfort M. B. (2012). Systematic review and meta-analysis of preterm birth and later systolic blood pressure. Hypertension, 59, 226–234. doi:
Dismukes A., Shirtcliff E., Jones C. W., Zeanah C., Theall K., & Drury S. (2018). The development of the cortisol response to dyadic stressors in Black and White infants. Development and Psychopathology, 30, 1995–2008. doi:
Duncan A. F., Rosenfeld C. R., Morgan J. S., Ahmad N., & Heyne R. J. (2008). Interrater reliability and effect of state on blood pressure measurements in infants 1 to 3 years of age. Pediatrics, 122, e590–e594. doi:
Flynn J. T., Kaelber D. C., Baker-Smith C. M., Blowey D., Carroll A. E., Daniels S. R., … Gidding S. S. (2017). Clinical practice guideline for screening and management of high blood pressure in children and adolescents. Pediatrics, 140, e20171904. doi:
Gauer R., & Qiu K. (2012). Clinical inquiries. Does blood pressure screening benefit children? Journal of Family Practice, 61, 425–426.
Grunau R. E., Haley D. W., Whitfield M. F., Weinberg J., Yu W., & Thiessen P. (2007). Altered basal cortisol levels at 3, 6, 8 and 18 months in infants born at extremely low gestational age. Journal of Pediatrics, 150, 151–156. doi:
Hemachandra A. H., Howards P. P., Furth S. L., & Klebanoff M. A. (2007). Birth weight, postnatal growth, and risk for high blood pressure at 7 years of age: Results from the Collaborative Perinatal Project. Pediatrics, 119, e1264–e1270. doi:
Hoffman D. J., Reynolds R. M., & Hardy D. B. (2017). Developmental origins of health and disease: Current knowledge and potential mechanisms. Nutrition Reviews, 75, 951–970. doi:
Jones A., Charakida M., Falaschetti E., Hingorani A. D., Finer N., Masi S., … Deanfield J. E. (2012). Adipose and height growth through childhood and blood pressure status in a large prospective cohort study. Hypertension, 59, 919–925. doi:
Keijzer-Veen M. G., Euser A. M., van Montfoort N., Dekker F. W., Vandenbroucke J. P., & Van Houwelingen H. C. (2005). A regression model with unexplained residuals was preferred in the analysis of the fetal origins of adult diseases hypothesis. Journal of Clinical Epidemiology, 58, 1320–1324. doi:
Kliegman R. M., Stanton B. F., St. Geme J. W. III, Schor N. F., & Behrman R. E. (2015). Nelson textbook of pediatrics (20th ed.). Philadelphia, PA: Elsevier.
Reynolds R. M. (2013). Glucocorticoid excess and the developmental origins of disease: Two decades of testing the hypothesis—2012 Curt Richter award winner. Psychoneuroendocrinology, 38, 1–11. doi:
Rodriguez J., Affuso O., Azuero A., Downs C. A., Turner-Henson A., & Rice M. (2018). Infant feeding practices and weight gain in toddlers born very preterm: A pilot study. Journal of Pediatric Nursing, 43, 29–35. doi:
Salimetrics. (2019). Cortisol saliva collection. Retrieved from https://salimetrics.com/analyte/salivary-cortisol/saliva-collection/
Shen W., Zhang T., Li S., Zhang H., Xi B., Shen H., … Chen W. (2017). Race and sex differences of long-term blood pressure profiles from childhood and adult hypertension: The Bogalusa Heart Study. Hypertension, 70, 66–74. doi:
Tabachnick B. G., & Fidell L. S. (2007). Using multivariate statistics (5th ed.). New York, NY: Pearson.
Urrutia-Rojas X., Egbuchunam C. U., Bae S., Menchaca J., Bayona M., Rivers P. A., & Singh K. P. (2006). High blood pressure in school children: Prevalence and risk factors. BMC Pediatrics, 6, 32. doi:
Wadhwa P. D., Buss C., Entringer S., & Swanson J. M. (2009). Developmental origins of health and disease: Brief history of the approach and current focus on epigenetic mechanisms. Seminars in Reproductive Medicine, 27, 358–368. doi:
Winchester S. B., Sullivan M. C., Roberts M. B., & Granger D. A. (2016). Prematurity, birth weight, and socioeconomic status are linked to atypical diurnal hypothalamic–pituitary–adrenal axis activity in young adults. Research in Nursing & Health, 39, 15–29. doi:
Keywords:

blood pressure; very low birth weight; very preterm; weight gain

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved