Poster Topical Area: Global Nutrition
Location: Hall D
Poster Board Number: 604
Objectives: Inflammation confounds assessment of iron status, and approaches to adjust iron biomarkers for inflammation are utilized in nutritional surveys. We adjusted ferritin and transferrin receptor (sTfR) concentrations for inflammation following the acute phase response within a norovirus infected cohort.
Methods: Healthy adults were exposed to norovirus and 26 volunteers became infected (norovirus detectable in stool by RT-PCR). Ferritin, sTfR, C-reactive protein (CRP) and alpha-1-acid glycoprotein (AGP) were measured at ten time points: baseline and post-exposure days (1, 2, 3, 4, 7, 14, 21, 28 and 35). To adjust ferritin and sTfR for inflammation among individuals with CRP or AGP above the 10th percentile of baseline CRP (0.14 mg/L) and AGP (0.37 µg/L), iron biomarkers were estimated using linear regression models that controlled for the effects of CRP and AGP (sTfR only included AGP). Beta coefficients were calculated daily (generating 10 regression equations) or pooled over 10 days. Median daily inflammation-adjusted concentrations over time were compared to baseline concentrations.
Results: Median inflammation-adjusted ferritin estimates corrected with regression equations using daily beta coefficients estimated lower ferritin during infection (d 2-4) and higher ferritin post-infection (d 14-35), compared to estimates using pooled beta coefficients. Median inflammation-adjusted sTfR estimates corrected using daily beta coefficients estimated higher sTfR during infection (d 2-4) and lower sTfR post-infection (d 14-35), compared to estimates using pooled beta coefficients. Inflammation-adjusted ferritin and sTfR concentrations using pooled beta coefficients were corrected to within 20% and 10% of baseline values during peak infection, respectively [median ferritin (µg/L), sTfR (mg/L) concentrations at baseline (56.6, 5.0), inflammation-corrected d2 (45.2, 4.9) and uncorrected d2 (72.3, 5.6)].
Conclusions: Among predominately iron-replete and non-inflamed (at baseline) adults, the best-suited linear regression method to adjust iron biomarkers for inflammation appears to depend on the stage of norovirus infection. This suggests that a single regression equation may not be appropriate for all stages of inflammation warranting additional investigation.