Chronic abdominal pain remains a difficult entity to treat in the pediatric population. A significant number of children with abdominal pain are shown to have underlying functional bowel disorders such as functional abdominal pain, irritable bowel syndrome, and functional dyspepsia. Fructose is a naturally occurring monosaccharide, which is increasingly being used in the form of a high-fructose corn syrup as a cheap, readily available, intensely sweet food additive. In 1978, Andersson and Nygren (1) reported 4 patients with chronic diarrhea and colic who were cured with a fructose-free diet. Subsequently, fructose malabsorption has been well described in adults, but the role of fructose intolerance as a cause for chronic abdominal pain in children is not clear.
We hypothesized that fructose intolerance is a significant and treatable etiology of abdominal pain in the pediatric population. The purpose of the present study was to ascertain whether pediatric patients with chronic abdominal pain had concurrent fructose intolerance as determined by breath hydrogen test (BHT) using a standardized dose, and whether symptoms will improve with low-fructose diet.
Institutional review board approval was granted by the MultiCare Health Systems after review of the study protocol. This is a retrospective chart review of consecutive patients. Data were collected on a series of patients seen in the pediatric gastroenterology clinic between May 2007 and August 2009 who presented with unexplained chronic abdominal pain alone, or associated with constipation, gas or bloating, and/or diarrhea. Patients underwent a standardized evaluation, including complete history and physical examination and further blood work, stool studies, and endoscopy as indicated. All of the patients were screened for celiac disease. Patients with persistent abdominal pain and no other pathology diagnosed were referred for BHT. These pediatric patients were tested for concurrent fructose intolerance as determined by BHT using a standardized dose and studied to determine whether symptoms would improve with low-fructose diet.
In preparation for the BHT, patients were instructed to avoid antibiotics and probiotics for 2 weeks, to avoid laxatives, antidiarrheals and fiber supplements for 1 week, and to consume only low fiber food and drinks 1 day before the test. They were instructed to stop eating and drinking 12 hours before the test. Patients were given a standard dose of 1 g/kg fructose to a maximum of 25 g. A Microlyzer Self-Correcting Model SC (QuinTron, Milwaukee, WI) machine was used to measure hydrogen (H2) and methane (CH4) at baseline, every 15 minutes for the first hour, and at time points 90, 120, and 150 minutes, yielding a total of 8 breath hydrogen values. The test was considered positive if breath hydrogen exceeded 20 ppm above baseline any time after 30 minutes from the start of the test.
Patients with BHT positive for fructose malabsorption had a 1-hour individual consultation with a dietitian. Assessment of the patients’ typical diet and presence of symptoms of abdominal pain/bloating, diarrhea, or constipation were ascertained. The principles and rationale of a low-fructose diet, including the scientific basis of malabsorption were discussed with all of the patients and their families. Dietary strategies reviewed include avoidance of foods with significant free fructose, foods with fructose in excess of glucose, and foods rich in sorbitol. In addition, patients were advised to avoid desserts, drinks, and sweet snacks made with high-fructose corn syrup and use dextrose (glucose) in place of sucrose to aid in fructose absorption. Sample menu plans were provided with attention to individual patient food preferences to assure nutritional adequacy of the diet. An example of the prescribed diet is provided in Table 1.
Medical records were reviewed for demographics, symptoms, comorbidities, prior workup, BHT results, and response to dietary modification. Responses to dietary changes were categorized as worse, same, or improved. Logistic regression and 2-tailed z tests for proportions were performed for statistical analysis of the response to the restricted diet; significance was assigned at P < 0.05.
A total of 238 eligible patients were identified. Sixteen patients (8 BHT positive and 8 BHT negative) were lost to follow-up. The final dataset evaluated in the present study consisted of 222 subjects. See Table 2 for demographic details.
BHT for fructose indicated fructose intolerance in 121 of 222 patients (54.5%). A total of 101 of 222 patients (45.5%) had a negative BHT for fructose intolerance. The entire 121 patients with a positive BHT for fructose had a nutrition consult with a registered dietitian and were placed on a low-fructose diet. Using a standard pain scale for children, 93 of 112 patients (76.9%) reported resolution of symptoms on a low-fructose diet. Furthermore, 55 of 101 patients (54.4%) with negative BHT for fructose reported resolution of symptoms without a low-fructose diet. After 2 months on this low-fructose diet, patients who tested positive for fructose intolerance (121) continued to report improvement of symptoms. Compliance with the diet was assessed through patient report. This group of patients reported near universal compliance with the dietary restrictions.
Of the patients with positive BHT, 76.9% (95% confidence interval [CI] 68.6–83.5, P < 0.0001) had improvement of symptoms when placed on a low-fructose diet. Although 54.5% (95% CI 44.8–63.8, P = 0.37) of the patients with a negative BHT had spontaneous resolution of symptoms, there is no evidence that this was the result of any dietary modification or treatment regimen. We used logistic regression to evaluate the association between symptom resolution in BHT-positive subjects who were prescribed a low-fructose diet and that in BHT-negative subjects for whom a low-fructose diet was not prescribed, adjusting for age and sex in the model. This yielded an odds ratio of 2.43 (95% CI 1.32–4.45, P = 0.004).
It is becoming increasingly clear that malabsorbed fructose may be a cause of abdominal pain in some patients labeled as functional. Symptoms include but are not limited to abdominal pain, bloating, nausea, flatulence, and vomiting. Whereas this has been well established in adults (2–8), it is not as well studied in children. The main problem in diagnosing fructose malabsorption is the uncertainty surrounding the normal absorption capacity of fructose in healthy patients.
The importance of fructose malabsorption in chronic abdominal pain is hypothesized to depend on the ratio of fructose to glucose. Two transport systems are understood to facilitate the absorption of fructose. The first, GLUT (glucose transport protein) 5, was discovered in the brush border membrane of human small intestine enterocytes (9). Fructose was transported passively, by facilitated diffusion, by GLUT 5 (10). This is a glucose-independent model of fructose absorption. The second mechanism is postulated as follows: a paracellular transport system with opening of tight junctions by glucose absorption. This would allow small solutes including fructose to move passively with water by osmotic drag through the channels between the enterocytes (11). The small bowel has a limited capacity to absorb fructose by facilitated diffusion per the first model, and glucose facilitates fructose absorption by the second model. Through these mechanisms with glucose it is suggested that fructose malabsorption only occurs when there is fructose present in excess of glucose, or absorption of fructose is facilitated by foods that have closer to a 1:1 ratio of glucose:fructose (by the second mechanism) than those wherein fructose is greater than glucose (12).
Carbohydrates such as fructose that are not efficiently absorbed in the small intestine may affect the patient in 2 ways. First, carbohydrates that remain in the small intestine may present a solvent drag drawing fluid into the lumen. This in turn would result in distension of the small bowel, producing abdominal pain and bloating (12) and increased small bowel and colonic motility (13). Second, the remaining fructose would be fermented, by colonic bacteria, into hydrogen, carbon dioxide, methane, and short-chain fatty acids (acetate, butyrate, propionate) (14,15). These gases then lead to abdominal pain, bloating, and flatulence (16). Hydrogen then diffuses across the intestinal mucosa, dissolves in the blood, and circulates through the lungs (12). The hydrogen is excreted in expired air in the breath or by flatulence (hydrogen is neither produced nor metabolized in humans). It can then be measured noninvasively in collected samples of expired breath after the ingestion of a fructose load. Therefore, the production of hydrogen serves as a proxy for bacterial fermentation of unabsorbed carbohydrate in the intestines. The hydrogen breath test was first used as a noninvasive test to detect lactose intolerance and has since been used in many studies to measure carbohydrate absorption, including fructose (17).
Consumption of fructose may include free fructose as a monosaccharide or sucrose, the disaccharide of fructose and glucose. Increasing fructose consumption from 25 to 50 g increased the prevalence of fructose malabsorption from approximately 20% to 60% (5–8) in adults. Additionally, increasing fructose concentration from 10% to 20% increased the malabsorption of 50 g fructose from 38% to 72% (6).
The values chosen in the present study correlate to those mentioned in the adult literature (7). Patients were considered to have positive BHT if they were able to generate a ≥20-ppm rise in hydrogen in breath testing after ingestion of 1 g/kg fructose not to exceed 25 g. Although there is no definite cutoff value for the rise in hydrogen, most studies have used an increase of more than 20 ppm above the baseline (18). It has been shown that all healthy adults are able to absorb 15 g of fructose (19), and some healthy adults can start having symptoms at 25 g of fructose (18) (up to 40%). Thus, 25 g was chosen as the cutoff dose to diagnose fructose malabsorption positive breath test at this dose. Induction of symptoms was not necessary to have a positive BHT.
Approximately half of the BHT-negative patients in the present study (47/99) did receive some instruction for diet modification; however, the prescribed diets for these patients were not strict (most were assigned a nonirritating or high-fiber diet), and there was no clear pattern in the prescription of diet modification or symptom resolution as related to dietary modification. Most notably, no BHT-negative patients were placed on the low-fructose diet. Patients in the present study were seen before the advent of the low FODMAP (fermentable oligosaccharides, disaccharides, monosaccharides, and polyols) diet (20–22), which is now prescribed in our clinics to patients diagnosed as having fructose malabsorption.
Because of the retrospective nature of the present study, certain limitations do exist. We were not able to randomize or blind BHT-positive patients into a low-fructose diet arm and an arm without dietary modification. Similarly, we were unable to control for any potential placebo effect. We relied entirely on the patient's report of symptom improvement and adherence to the restrictive diet without the use of a standardized tool. We have no data on the reintroduction of fructose to the patient's diet to determine recurrence of symptoms. Finally, we do not have data on gastrointestinal symptoms that were experienced during the administration of the BHT. We expect measures will be taken to prevent such limitations as this topic is researched prospectively in future studies at our institution and elsewhere.
One major limitation of the present study because of its retrospective nature is our inability to differentiate patients with small intestinal bacterial overgrowth (SIBO) from those with fructose intolerance or a combination of both. In a prospectively designed study, we would control for this. We acknowledge that treatment of SIBO may influence the patient's response to therapy, and this could add bias to the study; however, no patients presented in this study were found to be treated with metronidazole, the standard therapy for SIBO. Even within the trappings of a retrospective study, we feel that these data represent the results of dietary modification treating fructose intolerance.
The present study is not the first study to measure fructose malabsorption in the pediatric population (12), but it is the largest and the first to compare with a control population. Thus, to our knowledge, this is the first study to demonstrate causality of fructose malabsorption to a pediatric functional bowel syndrome. A potential flaw in the study design was not collecting data on reproduction of symptoms during the BHT. This is an area of high interest and can strengthen the argument of fructose malabsorption being a cause of chronic abdominal pain. It is, however, not necessary that the symptoms be present during the test to confirm fructose malabsorption (18). Furthermore, the symptoms the patients were experiencing chronically were well documented, and resolutions of the chronic symptoms were noted as well. Although chronic abdominal pain can certainly be multifactorial in the pediatric population, we believe the data demonstrate 1 clear and treatable cause of such pain in children.
I will at once present the first specimen. As you see it is a colon, but of surprising size considering it is from a child who died at 11 months of age. As the abdomen was opened, there appeared a pair of enormously distended coils of intestine, the S. Romanum and the ever still more dilated transverse colon. The rest of the large intestine was enlarged and the rectum was not only dilated but the seat of a small constriction. The described parts of the intestine are not only distended but the wall markedly thickened in all its layers and particularly in the muscular part. What symptoms did this remarkable disease of the colon cause?
Immediately after the birth of the child, which took place in a lying-in institution in Copenhagen, it showed the peculiarity that in spite of various laxatives, it had no stool. Only after repeated enemas did the bowels move. The same sluggish action of the bowel continued in the following months and the various remedies had to be used changing from one to the other, but when it was possible to cause a stool, it was of a normal consistence and appearance.*
Harald Hirschsprung (1830–1916), “Constipation in the newborn due to dilatation and hypertrophy of the colon,” Jahrbuch für Kinderheilkunde (1887)
*Colonic hypertrophy was described by Charles Michel Billard (1800–1832) in 1828, but the above description is regarded as the classic. Hirschsprung, chief physician at the Queen Louise Children's Hospital in Copenhagen, published his monograph on hypertrophy of the colon using data from autopsies of 2 babies, 7 and 11 months old. Additionally, he published 4 case histories of esophageal atresia with tracheoesophageal fistula (1861). His most important report was on controlled hydrostatic reduction of ileocecal intussusception (1876) using enema therapy. Of the 84 children Hirschsprung treated with only enema, 65% or 77% survived. Surgical intervention at the time carried an 80% mortality risk.
—Contributed by Angel R. Colón, MD
1. Andersson DE, Nygren A. Four cases of long-standing diarrhoea and colic pains cured by fructose-free diet: a pathogenetic discussion. Acta Med Scand
2. Choi YK, Johlin FC Jr, Summers RW, et al. Fructose intolerance
: an under-recognized problem. Am J Gastroenterol
3. Fernandez-Banares F, Esteve-Pardo M, Humbert P, et al. Role of fructose-sorbitol malabsorption in the irritable bowel syndrome. Gastroenterology
4. Rumessen JJ, Gudmand-Hoyer E. Functional bowel disease: malabsorption and abdominal distress after ingestion of fructose, sorbitol, and fructose-sorbitol mixtures. Gastroenterology
5. Skoog SM, Bharucha AE. Dietary fructose and gastrointestinal symptoms: a review. Am J Gastroenterol
6. Ravich WJ, Bayless TM, Thomas M. Fructose: incomplete intestinal absorption in humans. Gastroenterology
7. Rumessen JJ, Gudmand-Hoyer E. Absorption capacity of fructose in healthy adults. Comparison with sucrose and its constituent monosaccharides. Gut
8. Truswell AS, Seach JM, Thorburn AW. Incomplete absorption of pure fructose in healthy subjects and the facilitating effect of glucose. Am J Clin Nutr
9. Burant CF, Takeda J, Brot-Laroche E, et al. Fructose transporter in human spermatozoa and small intestine is GLUT5. J Biol Chem
10. Davidson NO, Hausman AM, Ifkovits CA, et al. Human intestinal glucose transporter expression and localization of GLUT5. Am J Physiol
1992; 262 (3 pt 1):C795–C800.
11. Shi X, Schedl HP, Summers RM, et al. Fructose transport mechanisms in humans. Gastroenterology
12. Gomara RE, Halata MS, Newman LJ, et al. Fructose intolerance
in children presenting with abdominal pain. J Pediatr Gastroenterol Nutr
13. Spiller R. Role of motility in chronic diarrhoea. Neurogastroenterol Motil
14. Born P, Zech J, Lehn H, et al. Colonic bacterial activity determines the symptoms in people with fructose-malabsorption. Hepatogastroenterology
15. Rumessen JJ. Hydrogen and methane breath tests for evaluation of resistant carbohydrates. Eur J Clin Nutr
1992; 46 (suppl 2):S77–S90.
16. Stone-Dorshow T, Levitt MD. Gaseous response to ingestion of a poorly absorbed fructo-oligosaccharide sweetener. Am J Clin Nutr
17. Calloway DH, Murphy EL, Bauer D. Determination of lactose intolerance by breath analysis. Am J Dig Dis
18. Kyaw MH, Mayberry JF. Fructose malabsorption
: true condition or a variance from normality. J Clin Gastroenterol
19. Rao SS, Attaluri A, Anderson L, et al. Ability of the normal human small intestine to absorb fructose: evaluation by breath testing. Clin Gastroenterol Hepatol
20. Gibson PR, Shepherd SJ. Evidence-based dietary management of functional gastrointestinal symptoms: the FODMAP approach. J Gastroenterol Hepatol
21. Gibson PR, Shepherd SJ. Food choice as a key management strategy for functional gastrointestinal symptoms. Am J Gastroenterol
2012; 107:657–666.quiz 667.
22. Staudacher HM, Whelan K, Irving PM, et al. Comparison of symptom response following advice for a diet low in fermentable carbohydrates (FODMAPs) versus standard dietary advice in patients with irritable bowel syndrome. J Hum Nutr Diet