Measurement of oro-cecal transit time in young children using lactose [13C] ureide requires further validation
To the Editors:
Van Den Driessche and coworkers (1) recently described the use of the lactose [13C] ureide (LU) breath test to measure oro-cecal transit time (OCTT) in children. Breath tests in 20 children aged between 3 and 17 years showed a mean (range) OCTT of 255 (165–390) minutes. In 32 children aged 0 to 3 years they performed stool incubations and demonstrated that bacterial enzymatic activity hydrolized LU in the stools of children from 8 months and over, but not in the stools of infants aged less than 6 months.
We have performed 11 breath tests on 10 Italian children aged between 7.9 and 21.8 months (median 11.8 mos.). Eight children were exclusively breast fed for at least 4 months, one was bottle-fed and one received mixed feeds. One child was studied at 11.8 months and again at 15.6 months. Median (range) age of weaning was 6 (4.0–6.5) months. The day before the test, each child was pre-dosed with 100 mg of unlabeled LU. Following an overnight fast, each consumed 100 mg 13C-LU as part of a starchy breakfast. Duplicate breath samples were obtained 15 minutes and immediately before breakfast and every 30 minutes thereafter for 12 hours using a bag and mask or by slowly aspirating close to the nostrils if the children were sleeping. We have presented this methodology elsewhere to investigate starch digestion in young children (2).
Not all children produced a clear 13CO2 rise above baseline as can be seen from Figure 1. Four had a median (range) age of 12.8 (9.3–21.6) months with a median (range) peak breath 13CO2 enrichment of 111 (90–153) ppme. The remaining children, including both experiments for the child studied twice were non-responders with a similar age spread of 7.9 to 21.8 months (median 11.8 mos). They had peak breath 13CO2 enrichment of 27 (1–34) ppme that was similar to that of four children, median (range) age 12.7 (8.8–16.1) months in whom we performed baseline studies using the same study design. The range of enrichment observed without a 13C enriched tracer was –13.4 to 22.5 ppme.
Van Den Driessche et al.'s results suggest that many more children in our study should have responded. Because they did not carry out breath tests in their younger subjects there may be a number of explanations for these seemingly inconsistent observations. A difference in curve parameters is observed if pre-dosing with unlabeled LU is used (3) but no consensus exists about a pre-dosing regimen. Van Den Driessche et al. used three doses of 500 mg unlabeled LU the day before the breath test. For those children undergoing the stool analysis, they used a single dose of 500 mg LU one week before stool collection as an in vivo induction, which resulted in a peak mean 13CO2 production rate at 450 minutes compared to the peak mean 13CO2 production rate for non-induced stool samples of 900 minutes. They also incubated some of the stool slurries with 2.5 mg/ml unlabeled LU one hour before the 13CO2 collection commenced. These results are not reported.
The metabolism of LU to CO2 involves the action of a specific enzyme that has been proposed to (4) split the glucose from the ureide moieties, followed by a urease that hydrolyses the urea. In addition to enzymatic induction by upregulation of receptors, adaptation may occur over a period of days by selection of bacteria capable of metabolizing LU. The peak 13CO2 production rate seen in Van Den Driessche et al.'s blank stool experiments was seen later than the time to maximal breath 13CO2 enrichment (tmax) observed in the breath tests although the converse may have been expected, since there is no OCTT delay factor before the 13CO2 appearance from incubated stool samples. This may have been accounted for by a lag time while enzymatic induction took place from the tracer sample itself rather than simply that the stool samples were cold at the start of the experiment.
The composition of the colonic flora is affected by the diet (5). Differences in the breast-feeding rates and composition of the weaning diet between Belgian and Italian children may account for a differential ability to metabolize LU and there is a need to reproduce the findings of both studies in different population groups.
The 13C-LU breath test is an attractive means to measure OCTT that is more valid than the currently available hydrogen breath test, but it requires further validation in young children. Particular attention must be focused on the role of a pre-dose.
1. Van Den Driessche M, Van Malderen N, Geypens B, et al. Lactose-[13
C] ureide breath test: a new, non-invasive technique to determine orocecal transit time in children. J Pediatr Gastroenterol Nutr 2000; 31:433–8.
2. Christian MT, Amarri S, Franchini F, et al. Contribution of small intestine and large bowel to starch digestion in infancy. J Pediatr Gastroenterol Nutr 1998; 26:588.
3. Morrison DJ, Dodson B, Preston T, et al. The lactose [13
C]-ureide breath test for measuring oro-caecal transit time. Proc Nutr Soc 2000; 59:68A.
4. Mohr C, Heine WE, Wutzke KD. Clostridium innocuum;
a glucoseureide-splitting inhabitant of the human intestinal tract. Biochim Biophys Acta 1999; 1472:550–4.
5. Stark PL, Lee A. The microbial ecology of the large bowel of breast-fed and formula-fed infants during the first year of life. J Med Microbiol 1982; 15:189–99.