Estimation of serum ferritin levels accounting for C reactive protein concentrations in Mexican population

Background Serum ferritin (SF) is a marker of iron stores and it is also an acute‐phase reactant. An elevated C‐reactive protein forces to exclude ferritin determinations because of inflammation; confounding iron deficiency (ID) prevalence. Aim To develop a reliable correction factor to avoid biased values of SF concentration. Methods Fasting SF and CRP data from the National Health Survey ‐2000 (ENSA) (n=2535) and National Nutrition Survey ‐2006 (ENSANUT) (n=5 810) from adults were collected. Iron deficiency was defined as SF<12 mcg/L. Using segmented regression model, the inflection point of CRP where SF concentration began to modify was identified. Models were adjusted by sex, age, BMI and chronic diseases. Results The correction factor for SF concentration was 0.75 IC95%[0.67, 0.83] and 0.81 IC95%[0.73, 0.90] for ENSA and ENSANUT respectly, when CRP was >6 mcg/L. This analysis did not modify the ID prevalence in both surveys. The correction factor estimated for ENSA was not statistically different from ENSANUT. Conclusions In a population level, CRP does not affect estimation of SF concentration or ID prevalence in Mexico.