Abstract
Nutritional studies typically assess dietary intake of sugars using self-reporting techniques (i.e. diet records, 24-h recalls, and food frequency questionnaires). These techniques, however, are prone to misreporting, leading to attenuation and inconsistency in observed diet-disease relationships. Alternative assessment techniques are required to overcome these research limitations and to strengthen the evidence base for dietary intake recommendations. Recently, two biomarkers of sugar intake have been developed: urinary sugars excretions, used as a predictive biomarker of total sugars intake; and carbon stable isotope ratios (13C/12C; expressed as δ13C), used in various tissues as a predictive biomarker of corn- and cane-derived sugar intake. To date, much research has been carried out on populations where intake of sugars is predominately from corn. The research described in this thesis evaluates the use of urinary sugars excretion in 24-h urine collections and spot urine collections, and the use of carbon stable isotope ratios in red blood cells and hair, as biomarkers of sugars intake in populations where the majority of sugars consumed originates from sugar cane. Additionally, the comparative performances of urinary excretion of sugars and stable isotope ratios as biomarkers was evaluated.
Data was analysed from a controlled-feeding study (n = 12), and from two cross-sectional studies. The first cross-sectional study involved young Pacific people (n = 80), and the second study, in conjunction with Ngāti Porou Hauora, involved the Māori Tairāwhiti population (n = 168) around Gisborne and the East Coast of New Zealand. Dietary intakes were assessed using a 7-day diet record in the controlled-feeding study, while a culturally appropriate FFQ, designed to measure intake of sugars, was administered in the Pacific youth and the Māori Tairāwhiti population.
Urinary sucrose excretion was most strongly associated with total sugar intake, as measured using a 7-day weighed diet record (r = 0.87); however, urinary sugar excretion measured during the controlled-feeding experiment did not show a linear relationship between the sugars consumed and sugars excreted in the urine. A study conducted in the Māori Tairāwhiti population demonstrated that, urinary sugars in spot samples were associated with self-reported intakes of sucrose (r = 0.22; n = 168). Carbon stable isotopes measured in red blood cells and hair were not associated with self-reported intakes in the pilot study. In young Pacific people, a dual isotope model with δ13C and δ15N was correlated with self-reported intakes of sugar-sweetened beverage intake (R2 = 0.35) in normal-weight adolescents (body mass index (BMI): 19-25 kg/m2), but not in overweight and obese participants. Conversely, δ13Calanine was not found to be correlated with any of the self-reported sugar intakes overall, and in any of the BMI categories. In a subsample of the Māori Tairāwhiti population (n = 36) δ13Calanine was correlated with added sugar intake (r = 0.38). Bulk measures of δ13C and δ15N were not associated with self-reported sugar intakes.
The comparative performance of urinary sugars and stable isotope ratios as biomarkers of sugars intake does not give any firm indication of the superiority of using one over the other. Nevertheless, the research of this thesis has identified the potential use of both urinary sugar excretion and carbon stable isotope ratios as biomarkers of sugar intakes.