Fructose and / or sorbitol intolerance in a subgroup of lactose intolerant patients

The diagnosis and treatment of lactose intolerance often does not resolve all the symptoms of postcibal bloating and flatulence. Included in this study were 104 lactose intolerant patients (71 female, 33 male) who complained of residual postcibal discomfort in spite of adherence to and benefit from appropriate measures for their documented lactose intolerance (at least 20 ppm H2 after 25 g lactose as well as appropriate symptomatic discomfort). Clinical characteristics common to this group included: symptomatic diarrhea (12.5%), history of foreign travel (5.8%), endoscopic and pathological evidence of gastritis and helicobacter infection (19.2 and 8.7%, respectively), nonspecific abnormalities of small bowel follow-through (15.4%), Crohn’s disease (8.7%) and colonic cliverticulosis (14.4%). Intolerance co fructose (at least 10 ppm H2 after 25 g fructose plus appropriate symptoms) or sorbitol (at least 10 ppm H2 after 5 g sorbitol plus appropriate symptoms) was documented in 17.3 and 18.3%, respectively. Intolerance to both fructose and sorbicol (administered as separate challenges), more than twice as common as intolerance to either one alone, occurred in 41.4% and was independent of sex. In conclusion, additional carbohydrate intolerances contribute to postcibal discomfort in more than 75% of lactose intolerant patients who remain symptomatic in spite of adherence to appropriate measures for this condition. While 62% of all patients had benefited significantly (greater than 50%) from appropriate dietary measures and enzyme replacement for lactose intolerance, only 40% of those who were also fructose intolerant and 47% who were sorbitol intolerant benefited (greater than 50%) from appropriate dietary measures (no enzyme replacement yet available) for intolerance to these sugars.

T HE VALUE OF TESTING FOR SPEC- ific carbohydrate intolerances is the subject of much debate.It i generally accepted chat while patients with irritable bowel syndrome (!BS) will experience more discomfort after ingesting a substance to which they arc intolerant (l-3 ), the incidence of fructo e-sorbicol intolerance is not increased compared with normal controls ( 1) and hence cannot be implicated as an etiological factor in chis very common condition.In contrast to reactions observed in patients with food allergie and celiac disease, the specific offenders being considered in !BS patients do not produce any known deleterious effects on the gastrointestina l trace.Estimates of the incidence of fructose and sorbitol malabsorption in the general population range between 3 7 .5 and 50% and between 10 and 60%, respectively, depending on the dose challenge used (4)(5)(6).In view of the relatively high prevalence of these intolerance in normal populations, it appears that we are dealing with the saturation of low capacity absorptive systems rather than pathophysiological comlitions.Whether we are dealing with physiological or pathophysiological entities, malabsorbed fructose and sorbitol can lead co real symptoms.Bacterial fermentation of malabsorbed sugars leads to the production of distressing, albeit beneficial, gases and short chain fatty acids such as propionate, butyrate and acetic acid (7,8).
Les caracteristiques cliniques communes a ce groupe comprenaient : diarrhee symptomatique (12,5 %), antecedents de voyages a l'etranger (5, mine the incidence and ignificance of fructose and/or sorbitol intolerance in a group of lactose intolerant patients, all of whom had continued to experience clinically significant postcibal discomfort even after dietary guidance and the use of the appropriate enzyme replacement (beta-galactosidase) had provided variable degrees of relief from their lactose-related symptoms.Were there any clinical characteristics, including pathological entities, that could predict which patients would manifest additional carbohydrate intolerances, and would their postcibal discomfort benefit from a modification of the intake of the offending sugars?

PATIENTS
The lactose intolerant population studied continued to complain of postcibal discomfort even after appropriate dietary guidance and beta-galactosidase enzyme replacement, and was derived from the private practice of a single gastroenterologist.One hundred and four consecutive patients (at least 18 years old; 68.3% female) with documented lactose intolerance (at least a 20 ppm rise in breath Hz as well as symptomatic discomfort after ingesting 2 5 g lactose as a l 0% solution) (Table 1) were selecred.The majority of the patients could be classified as having !BS except for 8.7% with Crohn' diease of the distal small bowel ( terminal ileum).The predominance of fema les in the IBS population seeking medical evaluation has been discussed by Thompson (9).The diagno is of !BS was based on accepted clinical criteria only after organic gastrointestinal disease had been excluded.Each patient had undergone at least a complete hitory and physical examination, full blood cell count, biochemical profile including liver-related enzymes, stool analysis for parasites, ova, bacterial pathogens and blood, as well as a small bowel follow-through and flexible sigmoidoscopy.Proceeding from the premise that the incidence of intolerance to fructose and orbitol was not more prevalent in !BS patients ( l) the authors did not attempt to assemble a control group of nonlactose intolerant !BS patients.As explained in the discussion this would have been a most difficult task and it is doubtful that a valid 'control group' could have been chosen.
Following an overnight fast and avoidance of gas-producing foods as well as fibre and cigarettes for the prior  1).Te ting of patients and interpretation of results were carried out.Questionnaires were filled out in the presence of one of the investigators during a follow -up visit at least six months after completion of these investigations and the initiation of appropriate dietary recommendation .
Only patients with either fructose and/or sorbitol intolerance in addition to lactose intolerance were asked to complete the questionnaires.Sixty of 80 patients (75%) who fe ll into these categories completed the questionnaire.The remaining 24 patient were not approached because the etiology of their po tcibal discomfort had not been elucidated by the investigations.
Testing for equality of proportions was the statistical methodology employed and the P values of the significant tests are provided with the results.

RESULTS
Based upon a review of all patient charts, clinical characteristics documented in at least 5% of the patients studied were analyzed.A history of diarrhea (12.5%) was significantly more common in males (21.2%) than in females (8.5%) (P<0.065), while a history of foreign travel (5.8%) was elicited in 7% of females compared with 3% of males (not significant).Gastritis and the presence of helicobacter, as determined by biop y of the gastric ant rum at gastroscopy, were present in 19 .2% of males and 8. 7% of females, respectively.Nonspecific radiological abnormalities of the mall bowel (small bowel follow-through) (ie, 'disordered' small bowel pattern or mild nodularity) were noted in 15.4% of patient .Except for an additional 8.7% of patients who were diagnosed to have Crohn's disease of the distal small intestine, no cases of proven celiac di ease or other pathological entity of the small intestine were documented.Diverticulosis mostly of the distal colon was present in 14.4% of the patient population.No significant ex-related differences were obtained in the above-mentioned categories.Analysis of the female subgroups according to specific carbohydrate intolerances indicated less foreign travel in patients with all three intolerance compared with those with either fructose or sorbitol intolerance in addition to lactose intolerance.No other significant differences were noted among the female subgroups analyzed.The number of male patient (33) was too small to permit a similar analysis.
In this group of lactose intolerant patients studied (n=104), additional intolerance to either fructo e or sorbitol was documented in 15.2 and 24.2% of males and in 18.3 and 15.5% of females, respectively (Figure 1).The striking finding was that additional intolerance to both fructose and sorbitol was documented in 36.4% of males and 43.7% of females.The latter figure for females was approximately twice as great as that for intolerance to either fructose or sorbitol alone, and the hypothesis of equal proportions across the three subgroups wa significant at P<0.005.Although the same trend was apparent for males, the difference was not statistically significant.Irrespective of sex, therefore, more that 75% of the symptomatic lactose intolerant patients studied had at least one additional carbohydrate intolerance as documented by the testing.Patient feedback wa obtained with the help of a questionnaire completed approximately six to 12 months after the completion of investigation and dietetic counselling by 60 patients with at least two demonstrated sugar intolerances of which 78% were females (Table 2).Clearly, although adherence to appropriate dietary measures and the perception of symptomatic benefit were greatest with respect to symptoms of lactose intolerance, clinically significant postcibal symptoms were still experienced (a criterium for inclusion in this study).While 75 % of patients classified lactose intolerance as a major problem, symptoms resulting from intolerance to fructose and sorbitol were similarly rated in only 60 and 55% of cases, respectively.With respect to symptoms of lactose intolerance, 62% of patients felt that they had experienced a symptomatic improvement of at least 50% with adherence to appropriate measures, while 13% had not experienced any significant relief.The corresponding figures for symptoms related to fructose and/or sorbitol intolerance indicated similar value for no relief (13 to 16%) while significant relief (greater than 50%) was noted in only 40 to 47% of these patients.Ninety-two per cent of patient claimed adherence to appropriate dietary measures for lactose intolerance, while approximately 85% of patients with additional intolerances to fructose or sorbitol did the same.Analysis of the data for the different subgroups of patients did not reveal any significant differences in the responses.When a ked if the entire investigation and dietary guidance was worthwhile 93.0% of patients responded affirmatively.The consensus was that increased awareness of food offenders had either reduced their discomfort or had provided reassurance that no serious illness was responsible for their symptoms (except for patients with Crohn's disease).

DISCUSSION
The aim of this study was to determine whether intolerance to fructose and/or sorbitol was clinically relevant to those lactose intolerant patients whose symptoms had only been partially alleviated by adherence to appropriate measures for lactose intolerance.What percentage of patients in this category would be intolerant to fructose and/or sorbitol and what percentage would benefit from appropriate dietary measures in an attempt to alleviate the postcibal di comfort and other symptoms associateJ with these adJitional intolerances 1  In this study we have shown that at least 75% of lactose intolerant patients whose postcihal discomfort was not eliminated entirely by appropriate dietary measures have demonstrable intolerance to fructose and/or sorbitol.While must patients were free from demonstrable organic gastrointestinal disease and could be categorized as suffering from JBS, 8. 7% had Crohn's disease involving the d istal small intestine (terminal ileum).Our results agree with previous studies that failed to demonstrate any sex-relateJ differences in the incidence of fructose and/or sorbital intolerance ( l) even though the male:female sex ratio of subjects studied is 1:2-3 (9).The incidence of biopsy documented gastritis, minor nonspecific abnormalities on radiologic examination of the small bowel (small bowel follow-through) anJ colonic diverticulosis ranged from 12 to 21 % in both males and females.
A history of diarrhea was obtained more often in males (21.2% in males versus 8.5% in females) and foreign travel in females (7.0% in females versus 3.0% in males).Thus far the signifi-cance of the correlations observed remain uncertain, especially in the absence of a control group.While it would have been of interest to have a control group of nonlactose intolerant subjects who were tested for intolerance to fructose and/or sorbitol, any control group assembled would have to be matched for ethnic heritage in addition to age and sex.We have experienced great difficu lty in attempting to recruit such a control group.In a recent study designed to determine the incidence oflactose intolerance among 222 patients with inflammatory bowel disease, we noted that refusal to undergo lactose breath testing was highest among patients with the lowest risk of having lactose intolerance by virtue of their ethnic heritage ( 10).Based on these considerations it was decided to forego accumulating a less than valid control group.
Fructose anJ sorbitol absorption in humans appears to take place slowly by an energy independent low capacity facilitated transport system.Another pathway for fructose may he glucose dependent and of high capacity (8).Fructo e and sorbitol may compete with each other for absorption along these pathways while glucose will facilitate the absorption of fructose.It appears that in most cases 1rn1labsorption of fructose and/or sorbitol represents a physiological rather than a pathological condition in view of the very high incidences reported: up to 60% in healthy adults (4,6,11,12).Additional factors that can influence the absorption of snrbitnl and fructose tnclude osmolalities of solutions, velocity of gastric emptying, intestinal and whole gut transit, adaptation of the colonic flora (4,13) as well as bacterial overgrowth.The doses of fructose (25 g) and sorhitol (5 g) used in the stanJard breath testing challenges correspond to amounts normally consumed.It is estimated that the average daily intake of fructose in the United States is 7 to 10 g ( 13).With the 55% frucrnse content of high fructose syrup currently available, drinking about 500 ml of soda would result in the ingestion of 37.5 g fructose.The major natural source of fructose is fruit; dried figs, dates, prunes and grapes contain the largest amounts: 30.9, 23.9, 15.0 and 8.0 g/100 g of edible portion, respectively ( 4, 1 5 }.The sorbitol content of sugar-free gum and mints ranges from 1.3 to 2.2 g/piece while pears, prunes, peaches and apple juice contain 4.6, 2.4, 1.0, 0.9 g/100 g of dry matter or juice ( 15).Sorbitol is commonly used as a sweetener in dietetic foods (5), and the practice of incorporating 5 g sorbitol into the harium administered during small bowel follow-through examirn1tions may account for the laxative effect and severe discomfort experienced by some patients.Ironically this malahsorbed sorbitol is almost completely digested by the colonic flora yielding an energy value close to that of sucrose (13 ).
The incidence of incomplete absorption of fructose (based on a 2 5 g challenge and at least a l O ppm increa e in breath Hz} may approach 37 to 50% (1,6), while figures quoted for the incomplete absorption of sorbitnl ( 5 g challenge and at least a l O ppm increase in breath H2} range up to 60l\> of healthy adult subjects ( 14,16).The incidence of incomplete absorption or intolerance to both fructose (25 g} and sorbitol (5 g), administered as separate challenges, has not been studied directly.Thus the values recorded for the incidence of intolerance to fructose and sorbitol in our study -namely 17.3 and 18.3%, respectively -arc significantly lower than those mentioned above.The fact that our patients were labelled as intolerant only if they experienced symptoms in addition to appropriate elevation of breath Hz may partially explain these apparent discrepancies.
It should be noted that studies documenting the incidence of malabsorption following the administration of solutions containing both fructose (25 g) and sorbitol (5 g) (1) recorded incidences among 73 patients with IBS and 87 age-and sex-matched controls of 45.2 and 57.5%, respectively.The higher incidence of malabsorption noted after the combined administration of fructose and sorbitol ( instead of administering each sugar separately as was done in our study) was attributed to the fact that each sugar interferes with the absorption of the other (1,17 ).Based on the observation that glucose facilitates the ab orption of fructose (8) it would have been of interest to have tested our patients with sucrose challenges.Although this was not done, the use of glucose is recommended to patients with fructose intolerance.We also advise patients to spread out their intake of the offending sugars throughout the day, in view o( the fact that no enzyme supplement is yet available to facilitate the limited absorption of either fructose or sorbitol.
Clearly intolerance to fructose and/or sorbitol, although very common in the group of patients studied, did not explain the postcibal discomfort experienced in the 22.1 % of patients who did not have additional demonstrable carbohydrate intolerances in our study.In addition, the fact that at least 53% of patients who did not benefit significantly from modification of their intake of fructose and/or sorbitol (Table healthy  13.Beaugerie L, Flourie B, Marceau P, Pellier P, Franchisseur C, Rambaud JC.Fructose and/or sorbitol intolerance 2) suggests that other intolerances, as yet undefineJ, were operative in these fructose and/or sorbitol intolerant imlividuals.The lower pain threshold of IRS patients (2,18) and discomfort related to the malabsorption of complex carbohydrates (19) represent other avenues to be explored.Patient feedback obtained from a questionnaire completed six to 12 months following testing and dietary advice indicated that intolerance to lactose was associated with the most significant symptoms as well as the greatest benefit from appropriate dietary measures which included the use of beta-galactosidase enzyme supplements.Patients perceived that intolerance to fructose and/or sorbitol was less problematic and less responsive to dietary measures than was lacto •e intolerance.In spite of this, most patients (93%) felt that they had benefited from the investigation, dietary guidance and reassurance that, in most cases, no serious disease process was responsible for their discomfort.

TABLE l Criteria for carbohydrate intoler- ance: positive H2 breath test plus symptoms
of 3 h with a hand-held Jayco H2 analyzer (Portsmouth, United Kingdom) which was standardized monthly against a Quintron clinical microlyzer (model CMZ) (Wisconsin).In addition to Hz concentration, symptoms of nausea, flatulence, cramping and diarrhea were recorded during the ubsequent 3 h.Intolerance to fructose and/or sorbitol was based on a rise of at least 10 ppm in H2 concentration as well as the development of significant symptoms (Table

TABLE 2 Results of questionnaire completed by 60 patients (78% female) SEVERITY O F SYMPTOMS
"Includes the use of beta-galactosidase (no enzyme replacements are available yet for fructose and/or sorbitol intolerance)