Serum C-reactive Protein and Thioredoxin Levels in Subjects with Mildly Reduced Glomerular Filtration Rate

Shoko Tsuchikura; Tetsuo Shoji; Naoko Shimomura; Ryusuke Kakiya; Masanori Emoto; Hidenori Koyama; Eiji Ishimura; Masaaki Inaba; Yoshiki Nishizawa


BMC Nephrology. 2010;11:7 

In This Article


The aim of this study was to compare the levels of CRP and TRX between subjects with normal and mildly reduced renal function. When the subjects were divided into two groups by eGFR, both CRP and TRX were higher in the subjects with mildly reduced eGFR. Also, eGFR showed significant inverse correlations with CRP and TRX in the total subjects. The inverse associations of eGFR with CRP and TRX remained significant after adjustment for age and sex. When further adjustment was done for 6 additional possible confounders, the inverse associations of eGFR with CRP and TRX became less significant. In such models, CRP was independently associated with systolic BP, non-HDL-C, and HDL-C levels. Also, TRX was associated with sex significantly, and with eGFR and systolic BP at border significance. These results suggest that the increased levels of CRP and TRX in subjects with mildly reduced eGFR were mediated, at least partly, by alterations in blood pressure and lipid levels in mildly decreased kidney function.

Previous studies reported that patients with advanced renal failure have increased levels of CRP and biomarkers for oxidative stress including TBARS,[14,15] phosphatidylcholine hydroperoxide,[14] F2-isoprostane,[31] and AOPP.[15] However, there are only a few studies that examined oxidative stress in those with mild reduction in renal function. According to Witko-Sarsat et al,[33] plasma AOPP levels were increased early in the course of CKD, and further increased in more advanced renal failure. Fortuno et al[34] showed that patients with stage 1–2 CKD had an increase in phagocytic NADPH oxidase-dependent superoxide production in as compared with healthy control subjects. Regarding antioxidant defense in early CKD, Yilmaz et al[35] reported that erythrocytes from patients with stage 1–2 CKD had lower activities of SOD and glutathione peroxidase than healthy controls. These previous studies suggested the increased oxidative products and impaired antioxidant defense even in early stages of CKD. However, no study examined possible changes in the biomarkers for inflammation and oxidative stress among subjects with mild reduction in eGFR as compared to those with normal eGFR. In addition, these previous studies did not made correction for possible confounding variables, due presumably to small number of subjects. The present study compared CRP and TRX levels between those with normal and mildly reduced eGFR, and showed that mild reduction in eGFR was associated with increased levels of CRP and TRX in dependent of age and sex using multivariate analyses in 182 subjects. These data provide further evidence supporting the notion that inflammation and oxidative stress are increased in a very early course of renal function loss.

In this study, eGFR, CRP, and TRX were correlated with each other. Importantly, the association of eGFR and CRP was not significant after adjustment for TRX in addition to age and sex. Similarly, the association of eGFR and TRX was not significant after adjustment for CRP in addition to age and sex. These results suggest the close association among inflammation, oxidative stress, and renal function. Subjects with early stages of CKD have increased NADPH oxidase activity[34] and compromised antioxidant defense mechanisms.[35] These data may indicate that impaired renal function is the cause of increased oxidative stress. Conversely, since increased oxidative stress causes organ damage, the increased oxidative stress due to reduced GFR could, in turn, further impair kidney function.[38] In addition, inflammation may increase oxidative stress,[15,31,39,40] and also promote loss of kidney function.[41] Furthermore, renal insufficiency results in sustained inflammation, since some inflammatory cytokines are excreted through kidneys.[42] Thus, these studies suggest the complex inter-relationship among decreased renal function, increased oxidative stress, and inflammation.

Furthermore, the present study indicates possible contributions of blood pressure and plasma lipids to the eGFR-CRP link and the eGFR-TRX link. In the fully-adjusted models, eGFR was not significantly associated with either CRP or TRX, whereas CRP was significantly associated with systolic BP, HDL-C, and non-HDL-C levels. TRX was associated with systolic BP at borderline significance. Since both blood pressure and plasma lipids are adversely affected by impaired kidney function, and these are well known risk factors for atherosclerosis, we speculate that the increased levels of CRP and TRX in subjects with mildly reduced eGFR were mediated, at least partly, by alterations in blood pressure, plasma lipids and presumably arterial wall in such subjects.

We interpret the increased TRX levels associated with mildly reduced eGFR to indicate that oxidative stress is increased in those with mildly reduced renal function. However, we cannot exclude other possibilities. The increased TRX may be due simply to retention of TRX in decreased renal function. So far, it is unknown to what extent glomerular filtration is involved in the elimination of TRX from the circulation. According to Kasuno et al,[18] TRX is detectable in urine of healthy individuals, and urinary TRX is increased in some kidney diseases. They also demonstrated the translocation of TRX from renal tubular cells into urinary lumen in response to ischemia/reperfusion in mice. Thus, urinary TRX may represent 'leak' of TRX from damaged kidney cells rather than glomerular filtration of the protein.

This study has several limitations. First, GFR was not directly determined but estimated by the formula that was developed for and validated in Japanese subjects. Therefore, direct GFR determination would be needed to obtain more solid conclusion. Second, because of the cross-sectional design of this study, the associations between parameters did not necessarily indicate causality. Prospective studies will be required for this purpose. Third, the subjects of this study do not represent the general population although we recruited them from the participants of a health check-up program. They included more women than men, and did not include those taking medications for the three common diseases.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.