The EPIC-Norfolk Study
C-reactive protein and fracture risk: European prospective investigation into Cancer Norfolk Study.
Bone 2012 ; 56: 67-72.
Ahmadi-Abhari S, Luben RN, Wareham NJ, and Khaw KT
DOI : 10.1016/j.bone.2013.05.009
PubMed ID : 23702388
PMCID : 0
URL : https://pubmed.ncbi.nlm.nih.gov/23702388/
Abstract
Serum C-reactive protein (CRP) is a marker of general systemic inflammation. It has been suggested that inflammation may have a role in osteoporosis and subsequent fracture risk. We examined the prospective relationship between CRP and fracture risk in a large cohort study.
The study populations were 18,586 men and women aged 40-79 enrolled between 1993 and 1997 in the European Prospective Investigation into Cancer in Norfolk study (EPIC-Norfolk), who had high sensitivity serum CRP measured in frozen serum from baseline. After 261,563 person-years of follow up (Median (inter-quartile range) follow-up time was 14.8 (13.6-16.1) years), 792 incident fractures (291 hip fractures) were ascertained by hospital record linkage. All multivariate regression models were adjusted for age, sex, body mass index, smoking, alcohol intake, physical activity, past history of fractures, history of osteoporosis, use of NSAIDs and corticosteroid medication, medical history of arthritis, cancer, myocardial infarction, stroke and diabetes, and in women only menopausal status and postmenopausal Hormone Replacement Therapy.
We found a U-shaped association for CRP and fracture risk. The lowest risk of fracture was observed for men and women with baseline CRP levels in the range of 1.1-2 mg/L in fractional polynomial Cox-proportional hazards regression analysis. Hazard ratios of all-fractures for participants with CRP levels 0.1-0.5; 0.6-1; 1.1-2; 2.1-3; 3.1-10; and above 10 mg/L were 1.31 (1.03, 1.66); 1.21 (0.98, 1.51); 1.00 (referent category); 1.06 (0.82, 1.35); 1.22 (1.00, 1.51); and 1.29 (0.91, 1.81) respectively. Although a similar U-shaped association was observed for hip fractures, the corresponding hazard ratios were not statistically significant.
A U-shaped association was observed between CRP and fracture risk. The increased risk of fracture observed at lower end of CRP compared to intermediate levels require further exploration, confirmation in other populations, and investigation into potential biological mechanisms.