Serum IGF-I and C-reactive protein in healthy black and white young men: The CARDIA male hormone study

Abstract

Objective

Animal and human studies suggest that C-reactive protein (CRP) may be inversely associated with serum insulin-like growth factor-I (IGF-I) concentrations. However, most human studies have not controlled adequately for confounding factors, particularly nutritional intake. This population-based study examined whether CRP is inversely associated with IGF-I and IGFBP-3 concentrations.

Methods

In cross-sectional analysis, multivariable linear regression with adjustment for age, BMI, smoking status, alcohol intake, and nutritional factors was used to relate log CRP, the independent variable, to IGF-I and IGFBP-3 in a sample of black (n = 364) and white men (n = 486) separately by race.

Results

Only black men had positive findings: log CRP was significantly associated with IGF-I (β = −13.1 ng/ml, p = 0.02) and the difference in mean IGF-I concentrations between the highest and lowest quartiles of CRP was 26 ng/ml. There was a statistically significant interaction between log CRP and smoking status (p = 0.02); the regression coefficient for IGF-I predicted from log CRP was significant in smokers (β = −39.8 ng/ml, p = 0.0001), but not in non-smokers. The difference in mean IGF-I concentrations between highest and lowest quartiles of CRP was 100 ng/ml for black smokers. There were no associations for IGFBP-3.

Conclusions

In our study, CRP levels are inversely associated with IGF-I concentrations in black male smokers; however, the causal nature of the association is unclear and should be studied further.

Introduction

Insulin-like growth factor-I (IGF-I) is a peptide hormone produced primarily by the liver and is regulated, in part, by growth hormone (GH). The predominate binding protein (BP), IGFBP-3, regulates IGF-I’s bioavailability and mediates its actions through the IGFBP-3 receptor [1], [2]. Early in life, the GH-IGF-axis has a critical role in normal postnatal growth. Disruption of this axis with reductions in IGF-I concentrations results in growth impairment. Stunted growth is a complication of childhood diseases characterized by inflammation such as inflammatory bowel disease (IBD) and systemic juvenile idiopathic arthritis [3]. It is hypothesized that one mechanism by which inflammatory diseases in children might lead to stunted growth is through the IGF-axis. That is, inflammatory cytokines, which stimulate C-reactive protein (CRP) production [4], reduce IGF-I concentrations, contributing to stunted growth. This hypothesis is also supported by several animal models [5], [6].

In addition to their significance to postnatal growth, the IGF-axis and inflammatory processes are both important in a number of disease contexts. In adults, serum IGF-I concentrations and markers of inflammation have been associated with risk of cardiovascular disease and the metabolic syndrome [7], [8], [9], [10], [11], type 2 diabetes [12], [13], and site-specific cancers [14], [15], [16], [17], [18], [19]. Most of these studies [7], [10], [12], [13], [14], [15], [16], [17], [18], [19] focused separately either on the IGF-axis or on inflammatory markers but did not consider their inter-relationships. On the other hand, in adult patients with cancer or IBD, inflammatory markers have been inversely correlated, albeit not causally related, with IGF-I levels [20], [21], [22]. However, these studies were limited in that nutritional deficiency concomitant with these diseases may have contributed to reduced IGF-I levels. It is well established that energy and protein restriction reduces IGF-I concentrations [23], [24]. Two population-based studies reported inverse correlations between CRP and IGF-I, but in neither study was it clear that potential confounding factors, such as body mass index (BMI) or smoking, were adjusted for in computing these correlations [9], [25].

Importantly, no large, population-based study in healthy adults has examined the association of IGF-I with inflammatory factors while controlling for diet [26], [27], [28] or other potential confounding factors. Given the independent importance of the IGF-axis and inflammatory processes in health and disease states, it is essential to determine whether the causal relation between inflammation and IGF-I suggested by animal studies [5], [6] can be observed at the population level. The Coronary Artery Risk Development in Young Adults (CARDIA) male hormone study (CMHS) provides a unique opportunity to test the hypothesis that CRP is inversely associated with IGF-I in healthy, well-nourished men. Because of racial differences in cytokine genotype distributions [29], mean serum IGF-I levels [30] and nicotine metabolism [31], and differential impact of cigarette smoking on hepatic drug metabolism [32], it is also necessary to examine the interactions of race and smoking with CRP on IGF-I. Indeed, a study reported inverse CRP–IGF-I correlations that were significant in only one of three ethnic groups [9].