C-reactive protein concentration in children: relationship to adiposity and other cardiovascular risk factors

Abstract

Whether or not C-reactive protein (CRP) predicts heart disease in adults because it is a marker of damage or atherosclerosis is difficult to assess. In children, there is no confounding with coronary disease or active smoking. We measured CRP in 699 children aged 10–11 years. CRP levels were 47% higher in girls than boys, and rose with age by 15%/year. CRP levels were 270% (95% CI, 155–439%) higher in the top fifth than the bottom fifth of Ponderal index (weight/height³). After adjustment, CRP levels remained 104% (95% CI, 23–236%) higher in the 56 children of South Asian origin. CRP was unrelated to: birth weight, height, social class, Helicobacter pylori infection or passive smoke exposure. CRP was correlated with several cardiovascular risk factors, but only fibrinogen (r=0.33, P=0.0001), HDL-cholesterol (r=−0.13, P=0.0006), heart rate (r=0.12, P=0.002) and systolic blood pressure (r=0.08, P=0.02) remained statistically significant after adjustment. We conclude that adiposity is the major determinant of CRP levels in children while physical fitness has a small independent effect. The strong relationships with fibrinogen and HDL-cholesterol suggest a role for inflammation throughout life in the development of atherosclerosis and cardiovascular disease. Longitudinal studies are needed to determine whether these associations reflect long term elevations of these risk factors in some individuals, or short term fluctuations in different individuals.

Introduction

C-reactive protein (CRP) is the major acute phase protein in humans. In subjects with normal liver function, plasma concentrations of CRP are a highly sensitive measure of overall inflammatory activity in the body, and serum levels of CRP are highly correlated with serum levels of pro-inflammatory cytokines IL-6 and TNFα [1]. Concentrations may rise several 100-fold in acute illness, but in healthy subjects, concentrations are usually low. Only recently have sensitive assays been developed which allow within and between subject variation in CRP within the normal range to be studied [2]. In adults, elevated levels of CRP within the normal range have been observed to be associated with smoking, increasing age, obesity, triglycerides, various markers of cardiovascular disease [3] and in subjects with chronic infections (Helicobacter pylori, C. Pneumoniae and Chronic Bronchitis) [4].

In prospective studies, C-reactive protein has been shown to be a predictor of cardiovascular events in subjects with angina [3], in high risk subjects without disease [5], and in apparently healthy US physicians [6]. In the latter study, CRP was independently predictive while in the ECAT study [3] it was not independent of fibrinogen. Whether or not CRP is predictive because it is a marker for the existing level of cardiovascular damage or atherosclerosis is difficult to assess in adulthood.

CRP levels have not previously been studied in healthy children. In this paper, we examine the relationship of CRP to potential determinants and to cardiovascular risk factors in children from five English and Welsh towns. The advantage of examining the interrelationships in children is that there is no confounding by coronary disease or active smoking. The potential importance lies in elucidating the mechanisms whereby differences in cardiovascular risk emerge early in life. While there has been considerable interest recently in the potential importance of foetal life in determining cardiovascular risk, we have recently argued that obesity has a more potent effect on a number of key cardiovascular risk factors during childhood including blood pressure [7], insulin and glucose metabolism [8] and fibrinogen and factor VII levels [9].