Abstract
Objectives
Skin carotenoid measurement by reflection spectroscopy offers a rapid, non-invasive assessment of carotenoid intake and status. However, feasibility, reliability, and validity have not been established in infants. We hypothesize that skin carotenoid measurement of infants’ heels and fingers will be feasible, reliable, and valid.
Methods
Skin carotenoid score (SCS) was measured by a modified, portable reflection spectroscopy device (Veggie Meter) at the index finger and heel of the foot in infants at 4 (n = 21), 6 (n = 12), and 8 (n = 9) months of age. Infant plasma, milk formula, and mother's breast milk carotenoid concentrations were measured by HPLC, while infant carotenoid intake was estimated from 7-day food diaries and 3 × 24-hour parent recalls. Feasibility was assessed by time to acquire triplicate reads and reliability was assessed by the coefficient of variation
(CV) of triplicates and the Pearson's correlation between mean heel and finger SCS. Validity was established in 4-month-olds if skin carotenoid scores were moderately correlated (r > 0.5) with breast milk carotenoid concentrations during exclusive milk-feeding (n = 16) and with infant plasma carotenoid concentrations (n = 11).
Results
Twenty-one infants (52% female) were enrolled (47% non-Hispanic white, 29% Hispanic white, 14% African American, and 10% Asian). Triplicate heel and finger SCSs were feasibly acquired in 56 ± 11 to 87 ± 32 seconds, respectively. Heel and finger SCS reliability was similar (mean CV range 18 ± 13 to 32 ± 22%), and mean within-subject heel and finger SCSs were highly correlated (r = 0.81, P = 1 × 10−5) in 4-month-olds. Finger and heel SCSs were valid correlates of total infant plasma carotenoid concentrations (r = 0.77, P = 0.006 and r = 0.66, P = 0.026, respectively). Finger SCS significantly correlated with breast milk carotenoids (r = 0.66, P = 0.006), while heel SCS non-significantly correlated (r = 0.40, P = 0.13).
Conclusions
This pilot suggests that infant skin carotenoids can be feasibly measured by reflection spectroscopy and that finger SCSs are correlated with both infant plasma carotenoid concentrations and maternal breast milk carotenoid concentrations. Skin carotenoid measurement should be further investigated to facilitate the assessment of infant dietary intake and health outcomes.
Funding Sources
Texas Children's Hospital, USDA-ARS.