Exocrine pancreatic insufficiency has been reported in patients with celiac disease (CD) and may be so severe that it reduces the efficacy of the gluten-free-diet (GFD) and requires pancreatic enzyme supplementation (1). The most likely explanation for the impaired pancreatic function is a defective meal-stimulated pancreatic lipase output, because gut damage severely affects the intestinal release of secretin (2) and cholecystokinin (CCK) (3), which are potent stimulants of pancreatic secretion. Pancreatic dysfunction in CD is usually transient and reversible after repair of small bowel damage with a GFD (4). However, recent studies have shown that low tryptic and lipolytic intraduodenal activity also follows intravenous pancreatic stimulation with secretin and CCK (4), implying that mechanisms other than intestinal mucosal damage may account for the pancreatic insufficiency in CD.
The proportion of patients with CD and pancreatic insufficiency has been reported to vary from 10% to 42% (1,5). The large variation is explained by the invasiveness of the most sensitive methods to detect pancreatic insufficiency (i.e., the duodenal intubation techniques), which has limited the widespread evaluation of pancreatic involvement in CD, and by the low sensitivity of such noninvasive tests as fecal fat determination and urinary para-aminobenzoic acid excretion. Pitfalls of currently available pancreatic tests might be avoided by using the newly described, noninvasive, nonradioactive breath test, the 1,3-distearyl, 2[carboxyl-13C]-octanoylglyceride, or 13C mixed-triglyceride, breath test (MTBT) (6).
In the present study, the intraduodenal lipolytic activity in patients with CD has been evaluated by means of the MTBT. The 13C-mixed-triglyceride contains a 13C-labeled medium chain fatty acid in the 2-position and long chain fatty acids in the 1- and 3- positions: the hydrolysis of the two stearyl groups by pancreatic lipase is the rate-limiting step in the digestion of the triglyceride. When compared with the gold standard technique (intraduodenal lipase output after intravenous administration of 1 U/kg CCK; normal value, >90 kU/hr) the sensitivity and specificity of the MTBT are 89% and 81%, respectively (6). Our intent was to use the MTBT in evaluating the severity of pancreatic impairment in CD and in monitoring the pancreatic function after institution of a GFD.
MATERIALS AND METHODS
Seventeen consecutive newly diagnosed patients with CD (4 males and 13 females; mean age: 17.4 ± 10.5 years) were enrolled. The diagnosis was made by means of the revised ESPGHAN criteria (7). A sweat test was performed in all patients to exclude concomitant cystic fibrosis (8). Height, weight, height:weight ratio, and body surface area (BSA) (9) were recorded. Anthropometric measurements and the 13C-MTBT were performed in all patients at diagnosis, and at 6 and 12 months after institution of a GFD.
13C Mixed-Triglyceride Breath Test
All patients were admitted to the hospital and provided a standard fat-containing diet for 3 days (90 g fat per day). After an overnight fast, patients ate a standard test meal consisting of 100 g of white bread and 0.25 g of butter per kilogram of body weight, to which 16 mg di-stearyl-13C-octanoyl-glyceride (Eurisotop, Paris, France) had been added. Breath samples were taken twice before the meal and at 30-minute intervals for 6 hours after the meal. Patients were at rest and not allowed to eat or drink during the study period. 13C enrichment of carbon dioxide in the breath was determined by means of Isotope Ratio Mass Spectrometer [IRMS (ANCA-NT; Europa Scientific, Crewe, UK)]. All δ-values were expressed versus the Pee Dee Belemnite-international standard (PDB), after correcting for the oxygen isotope effect (10). The δ-value was calculated according to the following formula: δ = [(Rs/RPDB) - 1] × 1000 in which δ is the ratio of 13C:12C in the CO2 of the breath sample and RPDB is the ratio of 13C:12C in the CO2 of the standard. The results were expressed as the maximum percentage of 13C-dose per hour (maximum %DH) at any time (excretion peak), the time to reach 13C peak excretion (TTP), and the cumulative values over 6 hours (percentage of cumulative dose recovered in 6 hours [%CD 6h]). For this calculation, the formula of Schoeller et al. (11) was used, and the CO2 production was assumed to be 300 mmol/m2 BSA per hour. Body surface area was calculated by the weight-height formula of Haycock et al (9).
In normal people, the maximum %DH is 8% and the TTP is approximately 300 minutes (6). When expressed as %CD 6h, the normal value of the MTBT is more than 22% (6). When lipase output and 13CO2 cumulative excretion are compared with daily fat excretion, steatorrhea only occurs if lipase output declines below 40 kU/hr and the %CD 6h is below 7.5% (6). A pathologic value of %CD 6h between 7.5% and 22% denotes a lipase output between 40 and 90 kU/h, which is diagnostic for exocrine pancreatic disease even in the absence of steatorrhea (6).
Small Bowel Biopsy
Four endoscopic duodenal specimens were obtained by means of biopsy forceps (FB13V, open diameter 8 mm; Olympus, Lake Success, NY), routinely processed, and stained with hematoxylin and eosin. Histologic evaluation was performed by taking into account the criteria recommended by Marsh and Crowe (12). Before histologic processing, one biopsy specimen was spread on a ground-glass slide and examined under a conventional light microscope at low magnification. Intestinal specimens were obtained at the time of diagnosis and 6 months after gluten withdrawal.
Serum Antigliadin and Antiendomysial Antibodies Assay
Both serum antigliadin (AGA) and antiendomysial antibodies (EmA) were determined by means of commercial kits at diagnosis, and 6 and 12 months after gluten withdrawal. Antigliadin antibodies (IgG and IgA) were assayed by an enzymelinked immunosorbent assay (Alpha-Gliatest; Eurospital Pharma, Trieste, Italy). Results were expressed as a percentage of the reference serum: 15% was the upper normal limit for IgG and 7% for IgA antibodies (value equal to the mean ± 2 standard deviation of the value obtained from a large healthy population). The specificity and sensitivity of the AGA IgG assay were 91% and 96%, respectively, and the specificity and sensitivity of the AGA IgA assay were 99% and 87%, respectively. Antiendomysial antibody IgA was determined by an indirect immunofluorescence technique (Antiendomisio, Eurospital Pharma). Results were considered positive if an interstitial positivity in muscularis mucosa was seen in fluorescent microscopy (magnification, ×250-400). The specificity and sensitivity of the EmA assay were 96% and 94%, respectively. All results were always compared with the positive and negative serum controls supplied by the company.
Statistical analysis was carried out by means of a Wilcoxon paired test to determine significant differences in %CD 6h or maximum %DH before and after gluten withdrawal (Table 1).
Small Bowel Histology
All specimens but three showed subtotal villous atrophy with villous-to-crypt (V:C) ratio less than 1. The remaining three specimens showed severe partial villous atrophy with V:C ratio of 1.0 to 1.8. After institution of a GFD, intestinal mucosal histology returned to normal in all patients, assessed by a second biopsy performed 6 months later.
Serum Antigliadin and Antiendomysial Antibodies Assay: Sweat Test
All patients had high serum levels of IgA AGA, IgG AGA, and IgA EmA antibodies but one, who had normal values of IgA AGA. After 6 months of gluten withdrawal, all patients were negative for EmA and IgA AGA, except two, who had persistently increasing levels of IgG AGA. At the 12-month follow-up, all patients displayed negative serum antibodies. No patient had concomitant cystic fibrosis assessed by the sweat test.
Weight, weight:height ratio, and BSA significantly increased in all patients at 12-month follow-up (p<0.001; paired t-test; data not shown). The height significantly increased only in patients who were less than 14 years old at the time of diagnosis (p<0.001).
13C Mixed-Triglyceride Breath Test
The MTBT results were pathologic in three patients and at the lower limit of normal values in another (Table 1). In the remaining 13 patients, the %CD 6h was within the normal range. At the 6- and 12-month follow-ups, the hydrolysis of 13C-labeled triglyceride, as assessed by the MTBT, was normal in all patients. The maximum %DH and TTP were also evaluated (Table 1). Apart from the three patients with an abnormal %CD 6h, no significant difference in the maximum %DH or in the TTP was elicited among patients before and after beginning the GFD.
Celiac disease is a disorder characterized by small bowel villous atrophy, which impairs nutrient absorption and improves after gluten withdrawal. Most patients with CD have some degree of fat malabsorption as shown by results of fecal fat studies (13). The mechanism of steatorrhea in CD is still poorly understood, but several factors probably contribute. First, fat malabsorption is related to compromised absorptive capacity of the gut, and its severity correlates well with the severity and extent of intestinal mucosal damage (14). When the disease is limited to the proximal small bowel, steatorrhea is generally absent. In addition, fat malabsorption may occur in patients with CD with several (proposed) mechanisms (15): decreased secretin and CCK output resulting in suboptimal pancreatic stimulation (2), impaired CCK release with reduced bile salts delivered into the gut lumen (3), and pancreatic dysfunction caused by concomitant cystic fibrosis or secondary to long-standing protein-energy malnutrition (16).
Our results show that the intraduodenal pancreatic lipolytic activity is altered in approximately 24% of patients with CD and contributes to the development of diarrhea and other typical manifestations of CD. By using the 13C-MTBT, no reliable conclusion on the pathogenesis of this disorder can be made. Permanent pancreatic damage can be excluded, because the %CD 6h returned to normal in all patients while they were receiving a GFD. A temporary impairment of the exocrine pancreatic function, reported in severe malnutrition (16), has been advocated but is unlikely to have occurred in our study, because no patient was malnourished at the time of diagnosis. Moreover, when patients with CD with abnormal MTBT results are given cerulein intravenously, the results return to normal, which supports the claim that pancreatic function is intact (17). This finding sheds some light on the pathogenic mechanisms of fat maldigestion in CD. The defect might be related to the intestinal mucosal damage with impaired meal-stimulated release of secretin and CCK (13). These hormones are mainly produced by the proximal intestine, which is usually more markedly affected than the distal one in CD (3). Therefore, the exocrine pancreatic insufficiency observed in some patients with CD could be exclusively functional and will disappear as soon as the intestinal mucosa returns to normal.
False-positive MTBT results have been reported in patients with gastrectomy or metabolic diseases such as diabetes mellitus or liver disease (6). A false-positive result is unlikely to have occurred in our patients at diagnosis, because none of them had previously undergone abdominal surgery or had diabetes or liver disease.
For these reasons, we propose that the MTBT could be indicated as the appropriate test to assess the functional integrity of the gut-pancreatic axis in CD. By exploring the meal-stimulated intraduodenal lipolytic activity, the MTBT can be more sensitive than the cerulein test in detecting functional pancreatic impairment, at least in patients with extensive intestinal mucosal damage. Moreover, it is less invasive than the Lundh test. Finally, all techniques requiring duodenal intubation and collection of pancreatic secretions may underestimate the pancreatic lipase output, if juice aspiration is incomplete or inappropriate.
Because previous studies have shown that pancreatic enzyme supplement therapy could be useful in patients with CD with functional pancreatic insufficiency (18), the 13C-MTBT could be used not only for detecting fat maldigestion (even in absence of steatorrhea) but also for monitoring the duration of enzyme replacement therapy. Finally, the 13C-MTBT can be performed in each patient several times and repeated at short intervals without any biological hazard. Therefore, it can be proposed as the ideal test for evaluating the degree of pancreatic impairment in patients with CD or other diseases involving the integrity of the small bowel mucosa.
1. Regan PJ, DiMagno EP. Exocrine pancreatic insufficiency in celiac spruce: Cause of treatment failure. Gastroenterology
2. Kilander AF, Hanssen LE, Gilberg RE. Secretin release in celiac disease
. Scand J Gastroenterol
3. Polak JM, Bloom SR, McCrossan MV, et al. Cholecystokinin abnormalities in celiac disease
4. Carroccio A, Iacono G, Montato G, et al. Exocrine pancreatic function in children with celiac disease
before and after gluten-free diet. Gut
5. Ansaldi N, Oderda G. Exocrine pancreatic insufficiency in celiac sprue (letter). Gastroenterology
6. Vantrappen GR, Rutgeerts PJ, Ghoos YF, Hiele MI. Mixed triglyceride
breath test: A non invasive test of pancreatic lipase
activity in the duodenum. Gastroenterology
7. Walker-Smith JA, Guandalini S, Schimtz J, Shmerling DH, Visakorpi JK. Revised criteria for diagnosis of celiac disease
. Report of a Working Group of European Society of Paediatric Gastroenterology and Nutrition (ESPGAN). Arch Dis Child
8. Shwachman H, Mahmoodam A. Pilocarpine iontophoresis seat testing: results of seven years' experience. Med Problems Pediatr
9. Haycock G, Schwartz G, Wisotski D. Geometric method for measuring body surface area: A height-weight formula validated in infants, children and adults. J Paediatr
10. Mook WG, Grootes PM. The measuring procedures and correction for the high precision mass spectrometric analysis of isotopic abundance ratios, especially referring to carbon, oxygen and nitrogen. Int J Mass Spectrom Ion Phys
11. Schoeller DA, Klein PD, Maclean WC, Watkins JB, Van Santen E. Faecal 13
C analysis for detection and quantitation of intestinal malabsorption. J Lab Clin Med
12. Marsh MN, Crowe PT. Morphology of the mucosal lesion in gluten sensitivity. Baillieres Clin Gastroenterol
13. DiMagno EP, Go VLW, Summerskill WHJ. Impaired cholecystokinin-pancreozymin secretion, intraluminal dilution, and maldigestion of fat in sprue. Gastroenterology
14. Trier JS. Celiac sprue. In: Sleisenger MH, Fordtran JS, eds. Gastrointestinal disease.
Philadelphia: W.B. Saunders, 1993:1078-96.
15. Mann NS. Gallbladder and pancreatic dysfunction in celiac disease
(letter). Dig Dis Sci
16. Barbezat GO, Hansen JDL. The exocrine pancreas and protein-caloric malnutrition. Pediatrics
17. Ghoos YF, Vantrappen GR, Rutgeerts PJ, Schurmans PC. A mixed triglyceride
breath test for intraluminal fat digestive activity. Digestion
18. Carroccio A, Iacono G, Montalto G, et al. Pancreatic enzyme therapy in childhood celiac disease
. A double-blind prospective randomised study. Dig Dis Sci