Association between platelet count and metabolic syndrome among healthy adolescents

Abstract

Background Platelet counts are suggested as one of the biomarkers for metabolic syndrome in adults. However, only limited studies evaluated the association in adolescents. The aim of this study was to investigate independent association between platelet counts and metabolic syndrome among healthy Korean adolescents.

Methods The JS High School (JSHS) study is a prospective cohort study which enrolled and screened 1,071 first-grader high school students between 2007 and 2012. This cross-sectional analysis was conducted for 1,044 healthy adolescents (544 males and 500 females) who aged 15-16 years, after excluding 7 for incomplete blood test and 20 for abnormal platelet counts (<150k/μL or >450k/μL). The metabolic syndrome was defined based on modified NCEP-ATP III, three or more of the following conditions: triglyceride ≥100mg/dL; HDL cholesterol<40mg/dL in male <50mg/dL in female; WC≥90th percentile (sex and age-specific); fasting glucose ≥100mg/dL; and systolic or diastolic blood pressure ≥90th percentile (sex, age and height-specific). Spearman correlation analysis, multiple linear regression models and multiple logistic regression were used to examine independent association between platelet count and metabolic syndrome and its components. Confounding variables were selected as follow: age, sex, body mass index, hs-CRP, total cholesterol, hemoglobin, smoking, drinking and regular exercise. All analyses were performed according to sex.

Results There was a significant positive correlation between platelet count and metabolic syndrome components. In unadjusted linear regression analysis, platelet count was significantly associated with waist circumference, SBP, DBP, insulin, HOMA-IR and triglyceride in male. The association of DBP and triglyceride still remained significant after adjusting for age, body mass index and even after additionally adjusting for C-reactive protein, total cholesterol, hemoglobin, smoking, drinking and regular exercise (DBP: standardized β=0.112, p=0.008; triglyceride: standardized β=0.097, p=0.014). In female, waist circumference, SBP, DBP, Glucose, insulin, HOMA-IR and triglyceride were related with platelet count in unadjusted model. Except waist circumference, most relationship with components still remained significant after adjusted for other confounders. The prevalence of metabolic syndrome was significantly higher in high platelet count than in low platelet count with unadjusted model both in male (OR 2.43; 95% CI 1.27-4.66) and female (OR 3.91; 65% CI 1.42-10.80), but the association in male was disappeared after adjustment for age and BMI. Platelet count was more predictable than hemoglobin for prevalence of metabolic syndrome in female (R2 0.187 in hemoglobin; R2 0.195 in platelet count).

Conclusions Our findings suggest that high platelet count is associated with metabolic syndrome independently in female adolescents.

연구 배경 및 목적: 혈소판 수는 성인에서 대사증후군과 관련 있다고 보고되고 있다. 그러나 청소년에서 이러한 연관성에 대한 연구는 부족한 실정이다. 이에 본 연구는 한국 청소년에서 혈소판 수와 대사증후군과의 독립적인 연관성을 살펴보고자 한다. 연구 방법: 본 연구는 2007년부터 2012년까지 1071명의 일개 고등학교 1학년들을 대상으로 진행된 JS High School (JSHS) 연구의 자료를 이용하여 진행하였다. 혈액검사를 받지 않은 7명과 혈소판 수가 정상 기준인 150k/μL 이상, 450 k/μL 이하에 해당하지 않는 20명을 제외한 15-16세 1044명 (남자 544명, 여자 500명)을 대상자로 선정하였다. 대사증후군은 NCEP-ATP III 기준을 변형하여 다음의 5가지 항목 중 3개 이상이 해당될 때로 정의하였다. 1) 중성지방≥100mg/dL; 2) HDL 콜레스테롤<40mg/dL (남자), <50mg/dL (여자); 3) 허리둘레≥연령별, 성별 표준치의 90분위, 4) 공복혈당≥100mg/dL; 5) 수축기 혈압 또는 이완기 혈압≥연령별, 성별, 신장별 표준치의 90분위. 대사증후군의 구성 요소들과의 연관성을 확인하기 위하여 상관분석과 다중 선형회귀분석, 다변량 로지스틱 회귀분석을 시행하였으며, 혼란변수로는 연령, 성별, 체질량지수, hs-CRP, 총콜레스테롤 수치, 헤모글로빈, 흡연, 음주, 운동량을 선정하여 보정하였다. 모든 분석에서 남자와 여자를 따로 구분하여 진행하였다.

연구결과: 혈소판 수는 대사증후군 구성요소들과 유의한 양의 상관성이 있었다. 보정하지 않은 선형 회귀분석에서, 혈소판 수는 남자 청소년에서 허리둘레, 수축기 혈압, 이완기 혈압, 인슐린, HOMA-IR, 중성지방과 각각 유의한 상관성을 가지고 있었으나, 연령과 체질량지수를 보정하면 남자 청소년에서는 이완기 혈압과 중성지방만 유의하였고, 이는 추가적으로 C-반응 단백질, 총콜레스테롤, 헤모글로빈, 흡연, 음주, 운동량을 보정하여도 같은 양상을 보였다 (이완기 혈압: 표준화β=0.112, p=0.008; 중성지방: 표준화β=0.097, p=0.014). 여자 청소년에서는 보정...