RESUMO
BACKGROUND/OBJECTIVES: Zinc is a negative acute-phase reactant; hence, its concentration decreases in the presence of inflammation. There is no current consensus on how to control for the effect of inflammation on serum zinc, which has implications for accurate estimates of population-level zinc status. We aimed to measure the association between inflammation and serum zinc concentrations and to compare the means and the prevalence of zinc deficiency using unadjusted and inflammation-adjusted serum zinc concentrations among Congolese children. SUBJECTS/METHODS: Non-fasting blood was collected in the afternoon in trace element-free vacutainers from 744 apparently healthy children aged 6-59 months in the Democratic Republic of the Congo. Serum was analyzed for zinc, C-reactive protein (CRP) and α-1 acid glycoprotein (AGP) for 665 children with complete data for all three biomarkers. Linear regression was used to generate correction factors (CFs) based on three stages of inflammation: incubation (CRP >5 mg/l and normal AGP), early convalescence (CRP >5 mg/l and AGP >1 g/l) and late convalescence (AGP >1 g/l and normal CRP), relative to no inflammation. RESULTS: Overall unadjusted mean±s.d. serum zinc concentration was 9.4±2.1 µmol/l. Study-generated CFs (95% confidence interval) for incubation, early and late convalescence were 1.01 (0.88, 1.14), 1.15 (1.11, 1.21) and 1.07 (1.03, 1.11), respectively. After applying the CFs, overall adjusted mean±s.d. serum zinc concentration was 10.1±2.2 µmol/l, and prevalence of zinc deficiency (<8.7 µmol/l) decreased from 35% (n=234/665) to 24% (n=160/665). CONCLUSIONS: Adjustment of zinc concentrations for inflammation is warranted when assessing population-level zinc status.European Journal of Clinical Nutrition advance online publication, 23 August 2017; doi:10.1038/ejcn.2017.127.
RESUMO
INTRODUCTION: Hemoglobin (Hb) concentration is often measured in global health and nutrition surveys to determine anemia prevalence using a portable hemoglobinometer such as the Hemocue® Hb 201+. More recently, a newer model was released (Hemocue Hb 301) utilizing slightly different methods to measure Hb as compared to the older model. The objective was to measure bias and concordance between Hb concentrations using the Hemocue Hb 301 and Hb 201+ models in a rural field setting. METHODS: Hemoglobin (Hb) concentration was measured using one finger prick of blood (approximately 10 µL) from 175 Cambodian women (18-49 years) using three Hemocue Hb 201+ and three Hb 301 machines. Bias and concordance were measured and plotted. RESULTS: Overall, mean ± SD Hb concentration was 116 ± 13 g/L using the Hb 201+ and 118 ± 12 g/L using the Hb 301; and anemia prevalence (Hb < 120 g/L) was 58% (n = 102) and 58% (n = 101), respectively. Overall bias ± SD was 2.0 ± 10.5 g/L and concordance (95% CI) was 0.63 (0.54, 0.72). Despite the 2 g/L bias detected between models, anemia prevalence was very similar in both models. CONCLUSIONS: The two models measured anemia prevalence comparably in this population of women in rural Cambodia.