Gastrointestinal (GI) permeability is a measure of GI barrier function and is used frequently to assess the presence of GI mucosal injury in a number of disorders such as inflammatory bowel disease and celiac disease (1–3). It has also been used as a research tool to investigate the role of increased permeability in disorders such as inflammatory bowel disease, irritable bowel syndrome, and feeding intolerance in preterm infants (4–7). The test is safe, quantitative, and noninvasive.
The theoretical basis for the test has been reviewed extensively (1,8). In summary, the small intestinal epithelium contains a gradient of (probably 3) pore sizes, with the smallest being on the villus tip and the largest being in the crypt and an intermediary size along the villus base (9). These pores are passageways formed by tight junction proteins regulating movement of molecules based on size and molecular charge (8,9). Small molecules such as mannitol (a monosaccharide) are able to traverse the small pores on the villus tip, but larger molecules such as the disaccharide lactulose (formed of fructose and galactose) can only move through the larger pores in the villus base and crypts (1,9). Consequently, mannitol serves as a marker of epithelial surface area, whereas the ability of lactulose to permeate through the pores depends on their “leakiness” (1,8).
After absorption, the sugars enter the bloodstream; however, because the sugars are not metabolized significantly they are excreted in the urine in the same amount and ratio as they permeate the mucosa (1,10,11). If GI permeability is increased (ie, an increase in the size of the 2 larger pores), the ratio of lactulose to mannitol found in the urine will be greater than that was administered (1,10,11). By giving both sugars simultaneously and expressing the results as a ratio of lactulose to mannitol, other factors such as variations in gastric emptying and intestinal transit time, which may alter the amount of lactulose and mannitol appearing in the urine over a period of time, were they used alone, are obviated (1,10,11). Lactulose and mannitol are fermented rapidly in the colon to the same degree by the colonic microbiota and thus, measure small intestinal permeability (11).
More recently additional site-specific probes have been used. The disaccharide sucrose (formed by fructose and glucose) assesses gastric mucosal integrity because it is rapidly degraded by sucrase once it leaves the stomach (12,13). A small portion of intact sucrose is absorbed passively through the gastric mucosa (through the larger pores) and excreted unchanged in the urine (12,13).
Sucralose, a disaccharide that is used as a commercially available artificial sweetener, is synthesized by replacing 3 of the hydroxyl groups on sucrose with chlorine, which increases its sweetness 600 times but makes it impervious to sucrase hydrolysis (11). Thus, almost all of an oral dose is excreted unchanged in the feces with a small fraction being absorbed passively through the larger pores in the small intestine and colon, and excreted unchanged in the urine (14). Thus, sucralose recovery in the urine is a measure of small bowel and colonic permeability (11). If small bowel permeability is normal, then it can be used to detect an increase in colonic permeability alone (11).
Both sucrose and sucralose recoveries can be expressed as a ratio of the urinary recovery of lactulose to delineate further the state of GI mucosal permeability. The sucrose/lactulose ratio increases in the presence of proximal GI injury, and the sucralose/lactulose ratio increases with pure colonic injury (11–13). As in the case of the lactulose/mannitol ratio, the use of their ratios obviates issues related to gastric emptying and transit time (1,10,11); however, use of the sucrose/lactulose and sucralose/lactulose ratios presupposes that lactulose permeability is normal (ie, the increased permeability is limited to the stomach or colon, respectively) (11–13).
Despite the usefulness of GI permeability testing and its frequency of use, there is little consensus on the timing of urine collections. For example, in measuring gastric permeability, recommended collection times for sucrose have ranged from 3 to 10 hours (6,12,15,16). To assess small intestinal permeability, collection times for lactulose and mannitol (or similar sugars) have varied from 2 to 10 hours (16–18). In the case of colonic permeability measurements, collection times for sucralose have ranged from 5 to 26 hours (19–22). To our knowledge no data have been published either showing the 24-hour excretion profiles of the sugars or the simultaneous measurement of gastric, small intestinal, and colonic permeability in humans. Finally, the potential roles that age and sex may play in permeability testing results largely remain unexplored.
The goals of our study were to determine simultaneously the 24-hour temporal profiles of urinary recovery of sugars commonly used in GI permeability testing. Further, we sought to examine the potential role that age and sex may play in urinary recovery of the sugars.
SUBJECTS AND METHODS
Children and adults were healthy volunteers selected from family members of the faculty and staff at Baylor College of Medicine and Texas Children's Hospital in Houston, TX. They were excluded if they had GI disease, had taken nonsteroidal anti-inflammatory drugs within 2 weeks of the study, alcohol within 48 hours of the study, or antibiotics within 4 weeks of the study; were unable to drink the sugar solution; or if they had enuresis. The study was approved by the Human Investigations Institutional Review Board and consent was obtained from the adults and assent was obtained from the children.
Subjects were instructed to drink the sugar solution following a minimum of 4 hours of fasting after the evening meal. The sugar solution consisted of sucrose (10 g), lactulose (5 g), mannitol (1 g), and sucralose (1 g) in a volume of 127.5 mL. Following ingestion, the subjects drank an additional 240 mL of water.
Subjects voided before drinking the solution. They were instructed to try to void at 30, 60, and 90 minutes after ingestion of the solution (before bedtime) and then as needed. Each time the subject subsequently voided, the urine was placed in separate containers. Each container was labeled with the date and time of the void. Thymol (33 μL of a 10% solution) was added to each container provided to the subjects as a preservative. The urine was frozen until picked up by a courier and brought to the laboratory for analysis.
After the first morning void, subjects were allowed to eat and drink (with the exception of foods containing sucralose). Urine was collected for a total of 24 hours starting from the time of ingestion of the sugars.
High-performance liquid chromatography was carried out to quantify the individual sugars as we and others have described (7,11,23). The assay is sensitive to 1 μg/mL for sucrose, lactulose, and mannitol, and to 10 μg/mL for sucralose. The coefficient of variation is 5%.
Calculations and Data Analysis
Data are expressed as mean ± standard error of the mean (SEM). Permeability data are expressed per body surface area (18).
Percent urinary recovery was measured for the entire 24-hour period as well as for each time point in the study. The 24 hours were split into epochs of 30 minutes for the first 90 minutes and then into blocks of 2 to 3 hours so that each individual's data points were included.
The percent urinary recovery of sugar at each time point was determined by multiplying the concentration of sugar in the urine (mg/mL) by the volume of urine. The amount of excreted sugar was then divided by the weight (mg) of sugar ingested. Total percent urinary recovery for the entire 24-hour period was calculated by summing the percent recoveries at each time point. Because of the difficulty in excluding sucrose from the diet, its urinary recovery was calculated only during the fasting period that lasted overnight up to and including the first morning void.
We also calculated a cumulative percent urinary recovery for each sugar. This was accomplished by normalizing the total amount of each sugar obtained during the 24 hours to 100%. The fractional amount of each sugar at each time point was then determined (ie, fraction of 100). This allowed comparisons of urinary recovery across different subjects on the same scale (ie, regardless of total percent recovered).
In addition to the above calculations, we also determined the ratios of the sugars as commonly used in the analysis of GI permeability. The weight (mg) of each sugar was measured at each time point (concentration of sugar in the urine multiplied by the volume of urine) and ratios calculated by dividing by the appropriate sugar.
Differences in proportions were determined using chi-square analysis and differences in means using Student t test. General linear modeling techniques were used to assess the effect of sex, age, and the interaction of these on percent urinary recovery. Data not normally distributed were log transformed.
Seventeen adults and 15 children were studied. The median age for the adults was 47.5 years, with a range from 21 to 57 years. The median age for the children was 10 years, with a range from 5 to 17 years. The proportion of males and females in the adult and child groups did not differ (female: adults 44% and children 33%; P = 0.55). The number of individuals studied at each time point is shown in Table 1.
Urinary Recovery of Sugars
The mean percent sucrose urinary recovery during the overnight fasting period was similar for children and adults (0.14% ± 0.2% vs 0.17% ± 0.1%, respectively). Visibly the time point of peak percent urinary recovery appeared to occur later in adults versus children, although the peak was lower for the adults than the children (Fig. 1). For the majority of the adults, cumulative urinary recovery by 3 hours was greater than 50%, whereas it was greater than 70% in the children (Fig. 2). In contrast, by 6 hours cumulative urinary recovery in children and adults was comparable (approximately 95%, Fig. 2).
The mean total percent lactulose urinary recovery for 24 hours was similar for children and adults (0.54% ± 0.3% vs 0.68% ± 0.7%, respectively). Percent urinary recovery of lactulose began to rise and peaked earlier in adults than in children (Fig. 1); however, the time point of peak percent urinary recovery for the adults was lower and broader than in children. This fits with the observation that 90% of cumulative urinary recovery was achieved by 18 hours in adults and by 15 hours in children (Fig. 2).
The mean total percent mannitol urinary recovery for 24 hours was similar for children and adults (35.7% ± 10.0% vs 40.5% ± 19.0%, respectively). Percent urinary recovery of mannitol began to increase and visibly peaked earlier in adults than in children (Fig. 1); however, the time point of peak percent urinary recovery was lower and broader for the adults compared with the children. Again, this fits with the observation that, similar to lactulose, 90% cumulative urinary recovery occurred by 18 hours in adults and by 15 hours in children (Fig. 2).
The mean total percent sucralose recovered for 24 hours was similar in children and adults (2.2% ± 1.3% vs 2.9% ± 2.0%, respectively). Similar to the other sugars, percent urinary recovery of sucralose began to rise and visibly peaked earlier in adults than in children (Fig. 1). Also similar to the lactulose and mannitol, the percent urinary recovery at the peak was lower and broader for the adults compared with the children, coinciding with the observation that the cumulative percent urinary recovery was 90% or more by 21 hours in adults and by 18 hours in children (Fig. 2).
Urinary Sugar Ratios
The sucrose/lactulose ratio was highest at the first measurement (30 minutes) in both children and adults and declined thereafter (Fig. 3). In adults, the lactulose/mannitol ratio peaked 10 to 12 hours after ingestion. In contrast, in children the peak occurred at 7 to 9 hours after ingestion. The apparent rise at 1.5 to 3 hours after ingestion in children may or may not be spurious given that 3 children account for this time point (Fig. 3). The sucralose/lactulose ratio peaked later than the lactulose/mannitol ratio, 19 to 21 hours after ingestion for adults and 16 to 18 hours for children (Fig. 3).
Permeability Versus Age and Sex
There was no interaction between age and sex with regard to sucrose urinary recovery. Total percent urinary recovery in 12 hours declined with age but there was no influence of sex (Fig. 4).
In contrast, there was an interaction between age and sex with regard to lactulose urinary recovery. Total percent lactulose urinary recovery in 24 hours declined with age in females but not in males (Fig. 5). Similarly, the 24 hours' pooled lactulose/mannitol ratio declined with age in females but not in males (Fig. 5).
There was no interaction between age and sex with regard to total percent mannitol urinary recovery in 24 hours (data not shown). There was also no correlation between age or sex and total percent mannitol urinary recovery in 24 hours (data not shown).
Similar to the findings with lactulose, there was an interaction between age and sex with sucralose urinary recovery. There was a decrease in total percent sucralose urinary recovery in 24 hours with age in females but not in males (Fig. 6). The sucralose/lactulose ratio showed no interaction between age and sex and no correlation with age (data not shown).
To our knowledge, this is the first study in children and adults to delineate the temporal profiles over a 24-hour period of the urinary recovery of the sugars commonly used in permeability testing. Additionally, we evaluated the urinary recovery of the sugars simultaneously to compare temporal profiles among the different sugars and to evaluate concurrently permeability in the stomach, small intestine, and colon.
By expressing the data as urinary percent dose recovered per time, we could compare directly results between children and adults. Overall, the total percent urinary recovery of the sugars for 24 hours was similar between children and adults; however, peak urinary recovery could be seen to occur 1 hour or so earlier in adults, and the curves were lower in amplitude and broader than that in the children (Fig. 1). These observations cannot be because of differences in body surface area because the data were corrected for body size. We speculated that the decreased permeability with age in combination with a larger intestinal surface area that may include alterations in motility (ie, “functional area”) may account for the flatter, broader curves seen in adults as compared with children.
By calculating the cumulative percent urinary recovery and normalizing this to 100%, we not only could compare results between children and adults but also assess better the relative urinary recovery rates among the sugars (Fig. 2). As may be anticipated, sucrose urinary recovery occurred faster than the other sugars (Fig. 2). Urinary recovery was 90% or greater by 6 hours, whereas lactulose and mannitol required 15 hours and sucralose 21 hours to reach this point (Fig. 2). These findings are consistent with the physiological differences that are the basis of using these particular disaccharides. Sucrose primarily permeates the stomach, whereas lactulose and mannitol permeate the entire small bowel, and sucralose permeates the entire small and large bowel (11).
Our data explain some of the previous observations in the literature regarding the appropriate timing of urine collections. For example, there has been debate regarding how many hours are required for an “accurate” measurement of lactulose (17,18). Based on our data, a 2- to 5-hour collection, as has been proposed in the past, may provide some comparison data for these sugars but would represent the steep-slope regions for these sugars' recoveries and not include their point of maximal urinary recovery (Fig. 2). Furthermore, it can be seen from our data regarding the cumulative percent urinary recovery for these sugars that earlier collections represent a small percentage of what will ultimately be retrieved in a longer study (Fig. 2). Thus, collections completed before this time may not be as informative.
To obviate differences because of gastric emptying or transit time, the sugars often are expressed as a ratio. Thus, we also investigated the potential effects of time on the ratios. The sucrose/lactulose ratio, representing proximal bowel permeability, quickly declined after 30 minutes because of the rapid disappearance of sucrose (Fig. 3). Because a greater amount of sucrose than lactulose was given, the ratio is >1 despite the lower percent dose recovered at most time points (Fig. 1).
The lactulose/mannitol ratio, representing small intestinal permeability was the highest around 10 to 12 hours (Fig. 3). The sucralose/lactulose ratio that represents colonic permeability, if small bowel permeability is normal, peaked later (around 19–21 hours) (Fig. 3). Thus, the timing of the peak ratios appears to fit with the presumed site of absorption (11,12).
Some authors express the lactulose/mannitol ratio in terms of fractional excretion (eg, fractional excretion of lactulose divided by fractional excretion of mannitol) (24). To convert our data to this format requires that the lactulose/mannitol ratio be divided by 5 (lactulose result/5 g ingested and mannitol result/1 g ingested; ie, 5), the sucrose/lactulose ratio divided by 2 (sucrose result/10 g and lactulose result/5 g ingested; ie, 2), and multiplying the sucralose/lactulose ratio by 5 (sucralose result /1 g and lactulose result/5 g ingested; ie, 5). This conversion demonstrates similarity between our lactulose/mannitol ratio expressed as fractional excretion (0.015) and those of other investigators for normal adults (8–10 hours ratio <0.025) (J. Meddings, personal communication, 2008).
To our knowledge, our observation regarding the effect of age and sex on permeability has not been described previously (Figs 4–6). In infants, small intestinal permeability decreases with age but has been presumed to reach adult levels early in childhood (7). Sucrose, lactulose, and sucralose permeability decreased with age (Figs 4–6). In the case of sucrose, children appeared to have greater permeability than did adults (Fig. 4). Even if the 1 child with the highest sucrose permeability was excluded, the difference with age was still significant (Fig. 4). In the case of sucralose, the decline with age was clearest (Fig. 6).
We do not know the reason why the decline in lactulose and sucralose permeability with age was only seen in females (Figs 5 and 6). Presumably this is because of hormonal influences. It is known that the hypothalamic-pituitary-adrenocortical system including glucocorticoids and corticotropin-releasing factors can alter permeability in rats as well as in humans (25,26). Further studies are required to address this question directly; however, these results (Figs 4–6) stress the importance of considering age and sex in future studies of permeability.
A limitation of our study is that the individuals did not fast for the entire 24-hour period. Lactulose and mannitol can be found in small amounts in certain foods. Thus, we cannot exclude the possibility that some individuals ingested these sugars during the test. Inadvertent ingestion of lactulose or mannitol may account for some of the interindividual variation. However, we attempted to balance the need for compliance with practicality. Without a lengthy daytime fasting period we were able to include children as young as 5 years of age. Although optimal results may be achieved with a full 24-hour fast, subject acceptance of such a study undoubtedly would be low. Another potential limitation is the relatively small number of children who contributed to the 1.5- to 6-hour time frame; however, this potential limitation does not alter our basic findings.
The results from our study can serve as a basis for reexamining optimal collection times to compare healthy individuals to those with abnormal permeability (eg, inflammatory bowel disease, celiac disease). Although previous investigations have demonstrated differences in permeability between healthy individuals and those with GI and various other diseases, it is possible that by using different collection periods (eg, time to urinary recovery plateau, overnight fasted, 24 hours) greater differences may be seen between groups with less interindividual variation. On the basis of Figure 2, we would suggest first testing a urinary collection time of 4 to 6 hours for sucrose and 13 to 15 hours for lactulose, mannitol, and sucralose in both adults and children. If a sucralose/lactulose ratio is to be measured, then based on Figure 3 the collection time may be extended to 16 to 18 hours in adults and children. Sex and age differences also must be taken into account (Figs 4–6).
The authors wish to thank Beverly Vispo and Raheela Khan for technical assistance, Erica Baimbridge for organizational assistance, and Dr Jon Meddings for reviewing the manuscript.
1. Bjarnason I, Macpherson A, Hollander D. Intestinal permeability: an overview. Gastroenterology 1995; 108:1566–1581.
2. Welcker K, Martin A, Kolle P, et al
. Increased intestinal permeability in patients with inflammatory bowel disease. Eur J Med Res 2004; 9:456–460.
3. Smecuol E, Bai JC, Vazquez H, et al
. Gastrointestinal permeability in celiac disease. Gastroenterology 1997; 112:1129–1136.
4. Soderholm JD, Olaison G, Lindberg E, et al
. Different intestinal permeability patterns in relatives and spouses of patients with Crohn's disease: an inherited defect in mucosal defence? Gut 1999; 44:96–100.
5. Dunlop SP, Hebden J, Campbell E, et al
. Abnormal intestinal permeability in subgroups of diarrhea-predominant irritable bowel syndromes. Am J Gastroenterol 2006; 101:1288–1294.
6. Shulman RJ, Eakin MN, Czyzewski DI, et al
. Increased gastrointestinal permeability and gut inflammation in children with functional abdominal pain and irritable bowel syndrome. J Pediatr 2008; 153:646–650.
7. Shulman RJ, Schanler RJ, Lau C, et al
. Early feeding, antenatal glucocorticoids, and human milk decrease intestinal permeability in preterm infants. Pediatr Res 1998; 44:519–523.
8. Teshima CW, Meddings JB. The measurement and clinical significance of intestinal permeability. Curr Gastroenterol Rep 2008; 10:443–449.
9. Fihn BM, Sjoqvist A, Jodal M. Permeability of the rat small intestinal epithelium along the villus-crypt axis: effects of glucose transport. Gastroenterology 2000; 119:1029–1036.
10. Travis S, Menzies I. Intestinal permeability: functional assessment and significance. Clin Sci 1992; 82:471–488.
11. Meddings JB, Gibbons l. Descrimination of site-specific alterations in gastrointestinal permeability in the rat. Gastroenterology 1998; 114:83–92.
12. Meddings JB, Sutherland LR, Byles NI, et al
: a novel permeability marker for gastroduodenal disease. Gastroenterology 1993; 104:1619–1626.
13. Vera JF, Gotteland M, Chavez E, et al
permeability in children with gastric damage and Helicobacter pylori infection. J Pediatr Gastroenterol Nutr 1997; 24:506–511.
14. Roberts A, Renwick AG, Sims J, et al
. Sucralose metabolism and pharmacokinetics in man. Food Chem Toxicol 2000; 38(Suppl 2):S31–S41.
15. Ekenel M, Avsar E, Imeryuz N, et al
. Effects of selective COX-2 inhibitors on the gastric permeability of sucrose
: a controlled study with placebo and ibuprofen. Eur J Gastroenterol Hepatol 2003; 15:403–406.
16. Lowe CE, Depew WT, Vanner SJ, et al
. Upper gastrointestinal toxicity of alendronate. Am J Gastroenterol 2000; 95:634–640.
17. Akram S, Mourani S, Ou CN, et al
. Assessment of intestinal permeability with a two-hour urine collection. Dig Dis Sci 1998; 43:1946–1950.
18. Nathavitharana KA, Lloyd DR, Raafat F, et al
. Urinary mannitol:lactulose excretion ratios and jejunal mucosal structure. Arch Dis Child 1988; 63:1054–1059.
19. Anderson AD, Jain PK, Fleming S, et al
. Evaluation of a triple sugar test of colonic permeability in humans. Acta Physiol Scand 2004; 182:171–177.
20. Farhadi A, Gundlapalli S, Shaikh M, et al
. Susceptibility to gut leakiness: a possible mechanism for endotoxaemia in non-alcoholic steatohepatitis. Liver Int 2008; 28:1026–1033.
21. Sandek A, Bauditz J, Swidsinski A, et al
. Altered intestinal function in patients with chronic heart failure. J Am Coll Cardiol 2007; 50:1561–1569.
22. Suenaert P, Bulteel V, Den HE, et al
. In vivo influence of nicotine on human basal and NSAID-induced gut barrier function. Scand J Gastroenterol 2003; 38:399–408.
23. Catassi C, Pierani P, Natalini G, et al
. Clinical application of a simple HPLC method for the sugar intestinal permeability test. J Pediatr Gastroenterol Nutr 1991; 12:209–212.
24. Fries W, Renda MC, Lo Presti MA, et al
. Intestinal permeability and genetic determinants in patients, first-degree relatives, and controls in a high-incidence area of Crohn's disease in Southern Italy. Am J Gastroenterol 2005; 100:2730–2736.
25. Wallon C, Yang PC, Keita AV, et al
. Corticotropin-releasing hormone (CRH) regulates macromolecular permeability via mast cells in normal human colonic biopsies in vitro. Gut 2008; 57:50–58.
26. Filaretova L. The hypothalamic-pituitary-adrenocortical system: hormonal brain-gut interaction and gastroprotection. Auton Neurosci 2006; 125:86–93.