DOI 10.1515/jpem-2013-0255 J Pediatr Endocr Met 2014; 27(5-6): 467–473
Ruth C. Bindler*, Kenn B. Daratha, Ross J. Bindler and Robert Short
Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year Abstract Objectives: To (1) elucidate the range of serum uric acid (UA) levels in adolescent sample; (2) examine the relationships of serum UA and 1-year change with gender, anthropometric and cardiometabolic factors. Materials and methods: Measurements (anthropometrics, fasting venipuncture and blood pressure) were performed at the beginning of the seventh and eighth grades. Descriptive data, differences according to weight, correlations and changes over time were examined. Findings: In 77 adolescents, BMI and serum UA had relationships with several cardiometabolic measures. Males had higher serum UA at follow-up compared to baseline; female change was minimal. Time-by-gender interaction was significant, as were the main effects of gender and BMI classification. Males had lower HDL-C at follow-up than at baseline; females had higher HDL-C at follow-up. Conclusions: Serum UA has importance in cardiometabolic examination of youth risk factors. It should be examined in youth with elevated BMI and/or hypertension. Early adolescence is the period when gender-related metabolic changes occur. Keywords: adolescent health; HDL-C; uric acid. *Corresponding author: Ruth C. Bindler, RNC, PhD, PO Box 1495, Spokane, WA 99210-1495, USA, Phone: +1 509 324-7403, Fax: +1 509 324-7341, E-mail: [email protected] Kenn B. Daratha: College of Nursing, Washington State University, Spokane, WA, USA; Providence Medical Research Center, Providence Sacred Heart Medical Center, Spokane, WA, USA; and Department of Medical Education and Biomedical Informatics, University of Washington, Seattle, WA, USA Ross J. Bindler: College of Pharmacy, Washington State University, Spokane, WA, USA Robert Short: College of Nursing, Washington State University, Spokane, WA, USA; and Providence Medical Research Center, Providence Sacred Heart Medical Center, Spokane, WA, USA
Introduction Over the past several decades, increasing prevalence of obesity, type 2 diabetes, metabolic syndrome (MetS) and
cardiovascular disease (CVD) have spurred interest in defining the metabolic links between various clinical conditions and biomarkers. Serum uric acid (UA) has emerged as an interesting and important biomarker defining such clinical conditions. Uric acid is an oxidation product of purine catabolism; purine sources are dietary proteins and endogenous nucleic acids. Many enzymatic processes are involved in purine catabolism so both dietary and genetic factors influence metabolism. Uric acid is distributed in the body as monosodium urate and cleared from plasma by glomerular filtration of the kidneys. There are various causes of hyperuricemia, including excessive intake of alcohol, cancer chemotherapy and a diet high in purines or proteins, all acute causes; and reduction in glomerular filtration rate, reduced UA excretion, or an increase in tubular absorption, which are considered chronic causes (1, 2). Consumption of large amounts of fructose can cause serum UA to increase through a hepatic pathway via influences on nucleotide catabolism (3). However, hyperuricemia is often observed in persons without these relatively rare disease states or conditions, but who demonstrate more common conditions such as hypertension and obesity (4, 5). Research in adults has examined the relationship of elevated serum UA levels with CVD (2, 6–8), renal disease and MetS (1, 9–13). Furthermore, elevated serum UA often precedes the onset of hypertension, obesity or kidney disease in adults (11). Less is known about serum UA during childhood. Although laboratories have measured serum UA levels in samples of children and published 2.5–97.5 percentiles for age and gender (14), recommended levels that are indicative of health in children have not been uniformly established. Serum UA levels from a representative group of articles, as well as the study described in this article, are summarized in Table 1. Serum UA appears to be higher in males than in females, in children with obesity and/ or hypertension, and levels increase with age (12–14, 18). Thresholds for healthy serum UA levels in children are elusive because gender and age/developmental stage influence UA levels; ethnic variations also occur (22, 23). In addition, metabolic health factors, such as adiposity, serum lipids and glucose, interact with gender and developmental levels to further influence serum UA.
Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 7:11 PM
468 Bindler et al.: Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year Table 1 Representative research study reports on serum uric acid levels in adolescents. Study Clausen et al. (13)
Feig and Johnson (15) Alper et al. (12) Franco et al. (16) Yoshida et al. (17) Oyama et al. (18) Ford et al. (10) Lee et al. (19) Pacifico et al. (20) Wasilewska et al. (21) Present study
Year reported
N
Age, years
1998
186 males 192 females 40 control 63 primary hypertension 40 secondary hypertension 22 white coat hypertension 244 males 333 females 36 normal birth weight 42 low birth weight 340 males 228 females 923 males 806 females 923 all subjects 1227 males 1057 females 120 obese 50 control 25 prehypertensive 57 hypertensive 25 control 171 all subjects 71 males 100 females
18–32
2003
2005 2006 2006 2006 2007 2007 2009 2010
2012
A link has been established between elevated serum UA levels and CVD risk factors in children and adolescents with obesity (20), and a strong relationship has been recognized between MetS and higher serum UA in children (10, 24). A cross-sectional study published by Pacifico et al. (20) found that adolescents with recommended body mass index (BMI) had lower serum UA compared to individuals whose BMI indicated obesity. Increased serum UA levels are linked to pediatric hypertension (1, 3, 6, 9, 11, 12, 15). There is debate about the exact pathological link between UA and hypertension (1), but the association of elevated serum UA and increased carotid intima media thickness may be explanatory (22). Elevated levels of serum UA in adolescence also potentially influence future health as they have been associated with hypertension in adulthood (12). While research has shown the relationship of obesity and serum UA levels in youth, there is a need for further research that will elucidate the normal ranges of serum UA in children and adolescents, levels associated with disease processes, and the associations of health conditions with hyperuricemia in youth. Until ranges and developmental variations are known for youth, meaningful evaluation of
6–18
5–17 8–13 7–10 9–15 12–17 6–12 9–11 10–19
13 13 13
Mean serum UA, mg/dL
5.6 ± 1.1 3.97 ± 0.9 3.6 ± 0.1 6.7 ± 1.3 4.3 ± 1.4 3.6 ± 0.7 5.1 ± 1.4 4.1 ± 1.1 3.2 ± 0.1 4.2 ± 0.1 4.8 4.7 5.3 4.3 5.1 6.1 5.8 4.7 3.2 5.5 6.3 4.6 4.9 ± 1.0 5.1 ± 1.1 4.8 ± 1.0
serum UA that can guide clinical intervention is not possible. There has also been very limited investigation of weight status and serum UA levels over the specific period of early adolescence. No longitudinal study has examined the age when gender differences in serum UA emerge or the relationship of BMI with serum UA over time in early adolescence. Therefore the purposes of this study among a group of early adolescents participating in the Teen Eating and Activity Mentoring in Schools (TEAMS) study were to (1) elucidate the range of serum UA levels at the start of the study and (2) examine the relationships of serum UA and longitudinal change in serum UA with gender, anthropometric and cardiometabolic factors over a 1-year period of time.
Materials and methods Participants Participants included middle school (seventh and eighth grade) students who volunteered to participate in a larger project called
Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 7:11 PM
Bindler et al.: Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year 469 TEAMS. This 2-year longitudinal study enlisted a convenience sample to measure the effects of the intervention on the health of students at four schools in the Inland Pacific Northwest of the United States. Students were recruited during open houses at the schools, schoolbased barbeques for families, school newsletters, and by school and study personnel in the first month of the student attendance in seventh grade. (In the USA, the average age of students at the beginning of seventh grade is 12.5 years.) Students with mental or physical conditions that would limit their participation in study activities, or those taking insulin or oral glycemic medications were excluded. Only one student of those screened was excluded based on the larger project’s exclusionary criteria.
Procedures University and school district institutional review board approvals, and signed parent permission and student assent, were obtained. Students participated in assessments at their respective schools. This project aligns with the World Medical Association guidelines for ethical conduct of research involving human subjects. Measurements were performed at the beginning of the seventh and eighth grades (average age 12.5 and 13.5 years, respectively). Fasting serum venipunctures were performed by licensed phlebotomists in the early morning, and students were then provided with a nutritious breakfast. Samples were transported within 1 h to the laboratory for analysis. Total cholesterol, LDL-C, HDL-C and triglycerides were measured by enzymatic method on a Siemens Advia 2400 chemistry analyzer (Siemens, Hoffman Estates, IL, USA). High-sensitivity CRP (hsCRP) was measured by nephelometry on a BeckmanCoulter Immage 800 immunochemistry system (Beckman-Coulter, Brea, CA, USA). Glucose was tested using the hexokinase method on the Siemens Advia 2400 chemistry analyzer, and insulin was measured by chemiluminescent immunoassay on an Immulite analyzer (Siemens, Hoffman Estates, IL, USA). Serum UA was measured by enzymatic method on the Siemens Advia 2400 chemistry analyzer. Insulin resistance (IR) is commonly measured by the homeostasis model assessment of insulin resistance (HOMA-IR), and so HOMA-IR was chosen as the method for measuring IR in this study. The HOMAIR method has been shown in past research studies to be a reliable test for measuring IR (25–28). IR is related to MetS, and previous work relates IR and MetS to serum UA levels (10, 20). Anthropometric measurement of students by trained personnel included height and weight. Height was measured on a standing stadiometer with shoes, hair ornaments, hats, jewelry and braids removed. Subjects were instructed to stand tall facing forward with feet together and heels, buttocks, shoulder blades and back of head in contact with the vertical backboard. The upper measure of the stadiometer was aligned and the subject stepped away. The height was measured and recorded in feet and inches with two decimal points as needed. Weight was measured on a Seca digital scale (Seca, Chino, CA, USA) with shoes, coats, and other heavy clothing removed. The electronic display was positioned so it was visible only to the examiner and the person being weighed. Weight was recorded in pounds with two decimal points as needed. Body mass index was then calculated by dividing weight in pounds by height in inches squared, and multiplying by a conversion factor of 703 (29). This is equivalent to a BMI calculated in kilograms/(meters squared). Exact BMI percentile for each participant was determined using the LMS method (http://www.cdc.gov/growthcharts/percentile_data_files.htm). The
LMS method provides normalized growth centile standards using a smoothed L curve, trends in mean M curve and coefficient of variation S curve. Students’ BMIs from the 85th to the 94th percentiles were considered as overweight, and those in the 95th percentile and above were considered obese. For descriptions in this paper, the overweight and obese groups were combined so that youth who were non-overweight/obese were compared with youth who were overweight/obese. Waist circumference was measured at the midpoint between the bottom of the rib cage and above the top of the iliac crest during minimal respiration to the nearest 0.1 cm. Systolic and diastolic blood pressures were measured using the protocols of the National Health and Nutrition Examination Survey (NHANES) (30). Research participants sat quietly for 5 min. The largest cuff that would fit on the upper arm and leave room below for the head of the stethoscope was chosen, and the participant sat with the right arm supported on a table at heart level. The radial pulse was palpated while inflating the cuff, and the level at which the pulse was not palpable was noted; the cuff was then fully deflated. With the stethoscope on the brachial artery, the blood pressure cuff was inflated to 20–30 mm Hg above the level of cessation of the radial pulse. The cuff was deflated 2 mm Hg/s, noting the systolic and diastolic blood pressure. For two additional times, the participant sat quietly for 2 min and the blood pressure was measured again. Per NHANES protocol, the first blood pressure measurement was not used in the calculation; the remaining two systolic and diastolic blood pressure readings were averaged for use in the study (31, 32).
Statistical analysis In order to elucidate the range of serum UA levels at the start of the study, all subjects who participated in the initial data collection of the larger study (n = 171) are included in Table 1. In support of the second purpose of the current study, to examine the longitudinal changes in serum UA and the relationships with anthropometric and cardiometabolic factors, only subjects who completed both baseline and follow-up testing (n = 77) were reported in subsequent tables and figures. Baseline data were collected when students entered seventh grade (with average age of 12.5 years), and follow-up data were recorded in eighth grade, an intervening time period of 10 ± 2 months. Descriptive data were computed for each baseline measure and reported for each weight classification group. Differences in measures between weight classification groups were tested using univariate analysis of covariance controlling for age (in months). Pearson’s product moment correlation coefficients were used to assess the direction and magnitude of association between serum UA, BMI and cardiometabolic markers. Changes in cardiometabolic markers over time were examined using repeated measures analysis of variance while controlling for age, weight classification (non-overweight/obese and overweight/obese) and gender.
Results There were 77 adolescents who participated in baseline and follow-up measurement, 40 females and 37 males. Baseline data indicated that 45 (58%) were of recommended weight, while 32 (42%) were overweight or obese;
Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 7:11 PM
470 Bindler et al.: Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year Table 2 Characteristics of participants at baseline for total sample and by weight status.
Age, months BMI, kg/m2 Waist, cm Serum UA, mg/dL Glucose, mg/dL Insulin, mU/mL HOMA-IR hsCRP, mg/dL Triglycerides, mg/dL Total cholesterol, mg/dL LDL-C, mg/dL HDL-C, mg/dL SBP, mm Hg DBP, mm Hg
Total sample (n = 77; 40 females, 37 males)
Not overweight/obese (
Ruth C. Bindler*, Kenn B. Daratha, Ross J. Bindler and Robert Short
Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year Abstract Objectives: To (1) elucidate the range of serum uric acid (UA) levels in adolescent sample; (2) examine the relationships of serum UA and 1-year change with gender, anthropometric and cardiometabolic factors. Materials and methods: Measurements (anthropometrics, fasting venipuncture and blood pressure) were performed at the beginning of the seventh and eighth grades. Descriptive data, differences according to weight, correlations and changes over time were examined. Findings: In 77 adolescents, BMI and serum UA had relationships with several cardiometabolic measures. Males had higher serum UA at follow-up compared to baseline; female change was minimal. Time-by-gender interaction was significant, as were the main effects of gender and BMI classification. Males had lower HDL-C at follow-up than at baseline; females had higher HDL-C at follow-up. Conclusions: Serum UA has importance in cardiometabolic examination of youth risk factors. It should be examined in youth with elevated BMI and/or hypertension. Early adolescence is the period when gender-related metabolic changes occur. Keywords: adolescent health; HDL-C; uric acid. *Corresponding author: Ruth C. Bindler, RNC, PhD, PO Box 1495, Spokane, WA 99210-1495, USA, Phone: +1 509 324-7403, Fax: +1 509 324-7341, E-mail: [email protected] Kenn B. Daratha: College of Nursing, Washington State University, Spokane, WA, USA; Providence Medical Research Center, Providence Sacred Heart Medical Center, Spokane, WA, USA; and Department of Medical Education and Biomedical Informatics, University of Washington, Seattle, WA, USA Ross J. Bindler: College of Pharmacy, Washington State University, Spokane, WA, USA Robert Short: College of Nursing, Washington State University, Spokane, WA, USA; and Providence Medical Research Center, Providence Sacred Heart Medical Center, Spokane, WA, USA
Introduction Over the past several decades, increasing prevalence of obesity, type 2 diabetes, metabolic syndrome (MetS) and
cardiovascular disease (CVD) have spurred interest in defining the metabolic links between various clinical conditions and biomarkers. Serum uric acid (UA) has emerged as an interesting and important biomarker defining such clinical conditions. Uric acid is an oxidation product of purine catabolism; purine sources are dietary proteins and endogenous nucleic acids. Many enzymatic processes are involved in purine catabolism so both dietary and genetic factors influence metabolism. Uric acid is distributed in the body as monosodium urate and cleared from plasma by glomerular filtration of the kidneys. There are various causes of hyperuricemia, including excessive intake of alcohol, cancer chemotherapy and a diet high in purines or proteins, all acute causes; and reduction in glomerular filtration rate, reduced UA excretion, or an increase in tubular absorption, which are considered chronic causes (1, 2). Consumption of large amounts of fructose can cause serum UA to increase through a hepatic pathway via influences on nucleotide catabolism (3). However, hyperuricemia is often observed in persons without these relatively rare disease states or conditions, but who demonstrate more common conditions such as hypertension and obesity (4, 5). Research in adults has examined the relationship of elevated serum UA levels with CVD (2, 6–8), renal disease and MetS (1, 9–13). Furthermore, elevated serum UA often precedes the onset of hypertension, obesity or kidney disease in adults (11). Less is known about serum UA during childhood. Although laboratories have measured serum UA levels in samples of children and published 2.5–97.5 percentiles for age and gender (14), recommended levels that are indicative of health in children have not been uniformly established. Serum UA levels from a representative group of articles, as well as the study described in this article, are summarized in Table 1. Serum UA appears to be higher in males than in females, in children with obesity and/ or hypertension, and levels increase with age (12–14, 18). Thresholds for healthy serum UA levels in children are elusive because gender and age/developmental stage influence UA levels; ethnic variations also occur (22, 23). In addition, metabolic health factors, such as adiposity, serum lipids and glucose, interact with gender and developmental levels to further influence serum UA.
Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 7:11 PM
468 Bindler et al.: Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year Table 1 Representative research study reports on serum uric acid levels in adolescents. Study Clausen et al. (13)
Feig and Johnson (15) Alper et al. (12) Franco et al. (16) Yoshida et al. (17) Oyama et al. (18) Ford et al. (10) Lee et al. (19) Pacifico et al. (20) Wasilewska et al. (21) Present study
Year reported
N
Age, years
1998
186 males 192 females 40 control 63 primary hypertension 40 secondary hypertension 22 white coat hypertension 244 males 333 females 36 normal birth weight 42 low birth weight 340 males 228 females 923 males 806 females 923 all subjects 1227 males 1057 females 120 obese 50 control 25 prehypertensive 57 hypertensive 25 control 171 all subjects 71 males 100 females
18–32
2003
2005 2006 2006 2006 2007 2007 2009 2010
2012
A link has been established between elevated serum UA levels and CVD risk factors in children and adolescents with obesity (20), and a strong relationship has been recognized between MetS and higher serum UA in children (10, 24). A cross-sectional study published by Pacifico et al. (20) found that adolescents with recommended body mass index (BMI) had lower serum UA compared to individuals whose BMI indicated obesity. Increased serum UA levels are linked to pediatric hypertension (1, 3, 6, 9, 11, 12, 15). There is debate about the exact pathological link between UA and hypertension (1), but the association of elevated serum UA and increased carotid intima media thickness may be explanatory (22). Elevated levels of serum UA in adolescence also potentially influence future health as they have been associated with hypertension in adulthood (12). While research has shown the relationship of obesity and serum UA levels in youth, there is a need for further research that will elucidate the normal ranges of serum UA in children and adolescents, levels associated with disease processes, and the associations of health conditions with hyperuricemia in youth. Until ranges and developmental variations are known for youth, meaningful evaluation of
6–18
5–17 8–13 7–10 9–15 12–17 6–12 9–11 10–19
13 13 13
Mean serum UA, mg/dL
5.6 ± 1.1 3.97 ± 0.9 3.6 ± 0.1 6.7 ± 1.3 4.3 ± 1.4 3.6 ± 0.7 5.1 ± 1.4 4.1 ± 1.1 3.2 ± 0.1 4.2 ± 0.1 4.8 4.7 5.3 4.3 5.1 6.1 5.8 4.7 3.2 5.5 6.3 4.6 4.9 ± 1.0 5.1 ± 1.1 4.8 ± 1.0
serum UA that can guide clinical intervention is not possible. There has also been very limited investigation of weight status and serum UA levels over the specific period of early adolescence. No longitudinal study has examined the age when gender differences in serum UA emerge or the relationship of BMI with serum UA over time in early adolescence. Therefore the purposes of this study among a group of early adolescents participating in the Teen Eating and Activity Mentoring in Schools (TEAMS) study were to (1) elucidate the range of serum UA levels at the start of the study and (2) examine the relationships of serum UA and longitudinal change in serum UA with gender, anthropometric and cardiometabolic factors over a 1-year period of time.
Materials and methods Participants Participants included middle school (seventh and eighth grade) students who volunteered to participate in a larger project called
Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 7:11 PM
Bindler et al.: Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year 469 TEAMS. This 2-year longitudinal study enlisted a convenience sample to measure the effects of the intervention on the health of students at four schools in the Inland Pacific Northwest of the United States. Students were recruited during open houses at the schools, schoolbased barbeques for families, school newsletters, and by school and study personnel in the first month of the student attendance in seventh grade. (In the USA, the average age of students at the beginning of seventh grade is 12.5 years.) Students with mental or physical conditions that would limit their participation in study activities, or those taking insulin or oral glycemic medications were excluded. Only one student of those screened was excluded based on the larger project’s exclusionary criteria.
Procedures University and school district institutional review board approvals, and signed parent permission and student assent, were obtained. Students participated in assessments at their respective schools. This project aligns with the World Medical Association guidelines for ethical conduct of research involving human subjects. Measurements were performed at the beginning of the seventh and eighth grades (average age 12.5 and 13.5 years, respectively). Fasting serum venipunctures were performed by licensed phlebotomists in the early morning, and students were then provided with a nutritious breakfast. Samples were transported within 1 h to the laboratory for analysis. Total cholesterol, LDL-C, HDL-C and triglycerides were measured by enzymatic method on a Siemens Advia 2400 chemistry analyzer (Siemens, Hoffman Estates, IL, USA). High-sensitivity CRP (hsCRP) was measured by nephelometry on a BeckmanCoulter Immage 800 immunochemistry system (Beckman-Coulter, Brea, CA, USA). Glucose was tested using the hexokinase method on the Siemens Advia 2400 chemistry analyzer, and insulin was measured by chemiluminescent immunoassay on an Immulite analyzer (Siemens, Hoffman Estates, IL, USA). Serum UA was measured by enzymatic method on the Siemens Advia 2400 chemistry analyzer. Insulin resistance (IR) is commonly measured by the homeostasis model assessment of insulin resistance (HOMA-IR), and so HOMA-IR was chosen as the method for measuring IR in this study. The HOMAIR method has been shown in past research studies to be a reliable test for measuring IR (25–28). IR is related to MetS, and previous work relates IR and MetS to serum UA levels (10, 20). Anthropometric measurement of students by trained personnel included height and weight. Height was measured on a standing stadiometer with shoes, hair ornaments, hats, jewelry and braids removed. Subjects were instructed to stand tall facing forward with feet together and heels, buttocks, shoulder blades and back of head in contact with the vertical backboard. The upper measure of the stadiometer was aligned and the subject stepped away. The height was measured and recorded in feet and inches with two decimal points as needed. Weight was measured on a Seca digital scale (Seca, Chino, CA, USA) with shoes, coats, and other heavy clothing removed. The electronic display was positioned so it was visible only to the examiner and the person being weighed. Weight was recorded in pounds with two decimal points as needed. Body mass index was then calculated by dividing weight in pounds by height in inches squared, and multiplying by a conversion factor of 703 (29). This is equivalent to a BMI calculated in kilograms/(meters squared). Exact BMI percentile for each participant was determined using the LMS method (http://www.cdc.gov/growthcharts/percentile_data_files.htm). The
LMS method provides normalized growth centile standards using a smoothed L curve, trends in mean M curve and coefficient of variation S curve. Students’ BMIs from the 85th to the 94th percentiles were considered as overweight, and those in the 95th percentile and above were considered obese. For descriptions in this paper, the overweight and obese groups were combined so that youth who were non-overweight/obese were compared with youth who were overweight/obese. Waist circumference was measured at the midpoint between the bottom of the rib cage and above the top of the iliac crest during minimal respiration to the nearest 0.1 cm. Systolic and diastolic blood pressures were measured using the protocols of the National Health and Nutrition Examination Survey (NHANES) (30). Research participants sat quietly for 5 min. The largest cuff that would fit on the upper arm and leave room below for the head of the stethoscope was chosen, and the participant sat with the right arm supported on a table at heart level. The radial pulse was palpated while inflating the cuff, and the level at which the pulse was not palpable was noted; the cuff was then fully deflated. With the stethoscope on the brachial artery, the blood pressure cuff was inflated to 20–30 mm Hg above the level of cessation of the radial pulse. The cuff was deflated 2 mm Hg/s, noting the systolic and diastolic blood pressure. For two additional times, the participant sat quietly for 2 min and the blood pressure was measured again. Per NHANES protocol, the first blood pressure measurement was not used in the calculation; the remaining two systolic and diastolic blood pressure readings were averaged for use in the study (31, 32).
Statistical analysis In order to elucidate the range of serum UA levels at the start of the study, all subjects who participated in the initial data collection of the larger study (n = 171) are included in Table 1. In support of the second purpose of the current study, to examine the longitudinal changes in serum UA and the relationships with anthropometric and cardiometabolic factors, only subjects who completed both baseline and follow-up testing (n = 77) were reported in subsequent tables and figures. Baseline data were collected when students entered seventh grade (with average age of 12.5 years), and follow-up data were recorded in eighth grade, an intervening time period of 10 ± 2 months. Descriptive data were computed for each baseline measure and reported for each weight classification group. Differences in measures between weight classification groups were tested using univariate analysis of covariance controlling for age (in months). Pearson’s product moment correlation coefficients were used to assess the direction and magnitude of association between serum UA, BMI and cardiometabolic markers. Changes in cardiometabolic markers over time were examined using repeated measures analysis of variance while controlling for age, weight classification (non-overweight/obese and overweight/obese) and gender.
Results There were 77 adolescents who participated in baseline and follow-up measurement, 40 females and 37 males. Baseline data indicated that 45 (58%) were of recommended weight, while 32 (42%) were overweight or obese;
Brought to you by | University of California - San Francisco Authenticated Download Date | 2/17/15 7:11 PM
470 Bindler et al.: Serum uric acid: relationships with biomarkers in adolescents and changes over 1 year Table 2 Characteristics of participants at baseline for total sample and by weight status.
Age, months BMI, kg/m2 Waist, cm Serum UA, mg/dL Glucose, mg/dL Insulin, mU/mL HOMA-IR hsCRP, mg/dL Triglycerides, mg/dL Total cholesterol, mg/dL LDL-C, mg/dL HDL-C, mg/dL SBP, mm Hg DBP, mm Hg
Total sample (n = 77; 40 females, 37 males)
Not overweight/obese (
