dietary habits

A 2017 UK study assessed the efficacy of proton nuclear magnetic resonance spectroscopy for monitoring dietary habits using urine samples containing digestion by-products. The study found this technique useful for identifying diet composition based on patterns of by-product excretion in urine.


Despite efforts to determine associations between everyday dietary habits and the risk of various diseases, unreliable self-reporting of food consumption has made it difficult to draw definitive conclusions. As such, researchers have sought objective methods of monitoring food consumption to overcome this limitation.

In a 2017 UK study published in The Lancet Diabetes and Endocrinology, researchers assessed the efficacy of proton nuclear magnetic resonance (1H-NMR) spectroscopy – a technique which can identify hundreds of molecules within a sample, in this case, the by-products of digested food passed through urine – for the determination of dietary habits. In total, 20 volunteers aged 21-65 with a body mass index (BMI) of 20-35 kg/m2 were recruited from a clinical research facility in the UK between August 2013 and May 2014. Volunteers were non-smokers, free of serious illness, were not on any prescription medications, and were not pregnant or breastfeeding.

Each subject had four 72-hour visits, spaced at least 5 days apart. During visits, each subject was randomly assigned 1 of 4 diets. Patients were instructed to consume all the food provided. Any leftover food was weighed. Water was provided freely, and light walking was permitted as exercise. Diet 1 (2260 kcal, 1.2 kcal/g) consisted of 24% protein, 51% carbohydrates, 23% fat, 2367 mg of sodium, 14 g of sugar, 45.9 g of fibre, 600 g of fruits and vegetables. Diet 2 (2259 kcal, 1.5 kcal/g) consisted of 22% protein, 51% carbohydrates, 24% fat, 2261 mg of sodium, 18 g of sugar, 32.1 g of fibre, and 300 g of fruits and vegetables. Diet 3 (2427 kcal, 1.6 kcal/g) consisted of 16% protein, 46% carbohydrate, 35% fat, 3812 mg of sodium, 22 g of sugar, 31.5 g of fibre, and 180 g of fruits and vegetables. Diet 4 (2490 kcal, 1.9 kcal/g) consisted of 13% protein, 44% carbohydrates, 42% fat, 3066 mg of sodium, 25 g of sugar, 13.6 g of fibre, and 100 g of fruits and vegetables. Urine samples were collected during the morning, evening, and afternoon.

To determine the applicability of the study’s findings to everyday conditions, findings were measured against the Dietary Approaches to Stop Hypertension (DASH) index scores – assessments of diet quality, with higher scores representing higher diet quality – of 225 patients from another UK study and 66 patients from a Danish study.

Of the 20 volunteers, 19 completed the study. Significant changes in the levels of 28 digestion by-products, including hippurate (indicative of fruits and vegetables), carnitine (indicative of red meat), and tartrate (indicative of grapes) were noted. The most pronounced differences were observed between those assigned to Diet 1 and those assigned to Diet 4. Though there was great variation in the levels of hippurate and carnitine produced by different people, hippurate levels were consistently higher in Diet 1 than in Diet 4, and carnitine levels were consistently higher in Diet 4 than in Diet 1. As Diet 4 did not contain any grape products, though tartrate levels varied between people for Diet 1, no tartrate was found in urine samples from those assigned to Diet 4. The by-product levels of Diet 2 were most similar to those of Diet 1, and the by-product levels of Diet 3 were most similar to those of Diet 4. The by-product levels for Diet 1 most closely resembled those of the UK patients with DASH scores in the 90th-100th percentile, while levels for Diet 4 correlated highly with the 0th-10th percentile. Those with higher DASH scores also had higher levels of hippurate, 4-hydroxyhippurate (indicative of fruit), and S-methyl-L-cysteine-sulfoxide (SMLCS; indicative of cruciferous vegetables such as broccoli or cabbage) than those with low DASH scores. SMLCS and hippurate levels, however, were not significantly different between those with higher and intermediate scores, nor were 4-hydroxyhippurate or hippurate levels significantly different between those with lower and intermediate scores. DASH scores in the Danish group were high, relating to Diets 1 and 2.

dietary habits

Overall, the findings suggest the analysis of digestion by-products in urine using1H-NMR spectroscopy may be useful as an objective indicator of everyday dietary habits. However, with this technique, monitoring dietary habits using urine samples may be more effective when examining by-product patterns than individual by-products. As the DASH index assesses a diet’s quality with respect to its ability to reduce the risk of high blood pressure (which predisposes to and contributes to a number of health complications), in lieu of a truly comprehensive assessment of diet quality, future research may benefit from incorporating a variety of health indexes, such as the Mediterranean Diet index or Alternate Healthy Eating index, into the study design. Future research into the by-product patterns of individuals who consume high amounts of coffee, on various medications, of different backgrounds, with conditions such as pregnancy, and with various diseases would provide a stronger indication of the breadth and usefulness of this technique.

Written By: Raishard Haynes, MBS

Facebook Comments