Back to browse results
Application of retinol-binding protein enzyme immunoassay to dried blood spots to assess vitamin A deficiency in a population-based survey: The Uganda Demographic and Health Survey 2006.
Authors: Baingana RK, Kasozi Matovu DK, and Garrett D.
Source: Food and Nutrition Bulletin, 2008; 29 (4): 297-305
Topic(s): Vitamin A
Country: Africa
  Uganda
Published: DEC 2008
Abstract: Background: The Uganda Demographic and Health Survey (UDHS) 2006 assessed vitamin A deficiency (VAD) in women (aged 15 to 49 years) and children (aged 6 to 59 months) by testing for retinol binding protein (RBP) in dried blood spots (DBS). A limitation to using RBP to estimate VAD is that RBP is a negative acute phase protein and this may result in over-estimation of the prevalence of VAD in populations with a high burden of infection. Thus, in order to improve estimates of VAD in surveys that use RBP, the simultaneous assessment of indicators of inflammation is recommended. Aim: To estimate VAD in the UDHS 2006 more accurately by measuring C-reactive protein (CRP) and correcting RBP measurements for the effect of the acute phase response. Methods: We analyzed a systematically selected subset of the DBS collected for the UDHS 2006 for CRP using enzyme-linked immunoassay. RBP had previously been analyzed using RBP-EIA. Subjects were classified into categories of “normal CRP” and “raised CRP” on the basis of a CRP cutoff of 5 mg/L. Correction factors for women and children, respectively, were calculated to remove the influence of infection/inflammation from the RBP values. Findings: Mean CRP was 14.4 mg/L (95% CI: 12.9, 15.9 mg/L) in children and 6.9 mg/L (5.4, 8.0 mg/L) in women. Correction increased mean RBP in children from 25.4 µg/mL (24.7, 26.1 µg/mL) to 27.2 µg/mL (26.5, 28.0 µg/mL) (p < 0.000). The prevalence of VAD (RBP < 17.325 µg/mL) was reduced from 20.1% to 15.4%. The only background characteristic that was significantly different between the children in the “healthy” and “raised CRP” groups was rural/urban residence. Among women, mean RBP values for the whole group, “normal” and “raised CRP” were 40.2 µg/mL (38.9, 41.6 µg/mL), 39.8 µg/mL (38.1, 41.5 µg/mL), and 41.0 µg/mL (43.4, 38.6 µg/mL), respectively; the differences were not statistically significant. Conclusion: Measurement of acute phase proteins removes the influence of inflammation from nutritional biomarkers and could improve population-level assessment of vitamin A status.