The improvement in imaging and pancreatic function testing has led to an increasing recognition of children with abnormal pancreatic function. Pancreatic exocrine dysfunction in children can present with chronic diarrhea, steatorrhea, abdominal pain, failure to thrive, and weight loss. Cystic fibrosis (CF) is the most frequent cause of severe exocrine pancreatic insufficiency in childhood. Approximately 85% of patients with CF are pancreatic insufficient. Other causes of pancreatic exocrine deficiency such as Shwachman-Diamond syndrome, Johanson-Blizzard syndrome, Pearson syndrome, chronic pancreatitis, and isolated enzyme deficiencies are less common.
Exocrine pancreatic function can be evaluated with direct or indirect testing. Direct testing using secretin, cholecystokinin, or both to stimulate pancreatic fluid secretion and analyze pancreatic enzyme activities in the collected fluid has the highest sensitivity and specificity. However, the direct pancreatic stimulation test is invasive, costly, time-consuming, and only performed at specialty centers. In contrast, the indirect tests are noninvasive, simple, and less time-consuming (1). Although the indirect tests are routinely used, they have limited sensitivity and specificity compared with direct testing. This is particularly evident in children and adults with mild-to-moderate exocrine dysfunction (2).
There are multiple indirect methods including serum, urine, breath, and stool tests (1). In the last decade the fecal elastase-1 (FE-1) test has become widely used and is the most sensitive indirect test for exocrine dysfunction. FE-1 is highly specific for the human pancreas and resistant to intestinal degradation (3). The FE-1 test uses specific monoclonal antibody against the human pancreatic elastase; therefore, pancreatic enzyme replacement therapy does not need to be discontinued.
The other advantage of FE-1 testing is stability; the stool specimen can be shipped to the laboratory without freezing. The stability of elastase-1 was assessed in the stool at different temperatures, from –25°C to room temperature for 1 week. The fecal elastase concentrations of stool samples varied only by a mean of 6.64% (4).
The aim of our study was to evaluate the performance of the FE-1 test in children in comparison with the direct pancreatic function testing with secretin stimulation.
The patients were selected upon review of FE-1 test results in the electronic medical records of the Alfred I. duPont Hospital for Children from January 2005 to April 2010. Patients were included only if they had direct pancreatic stimulation testing. A review of the electronic medical record for demographics and clinical data was completed on patients included in the present study.
Seventy children met the criteria of having both direct pancreatic function test with secretin stimulation and FE-1 testing. The average time between the stool collection and stimulation test was 22.8 ± 21 days (mean ± SD).
The age of the children was 4.60 ± 4.29 years (mean ± SD); 30 were girls and 40 were boys. Forty-six children had failure to thrive or weight loss, 21 had chronic diarrhea, 8 had chronic abdominal pain, 5 had allergic enteropathy, 4 had CF, and 1 had chronic pancreatitis.
Stool samples were collected either during hospital stay or home by parents and sent to the Mayo Clinic, which forwarded the samples for measurement to Genova Diagnostics (Asheville, NC). The FE-1 was measured from freeze-dried specimens, and the content was given as micrograms per dry stool weight. The FE-1 was measured with an enzyme-linked immunoabsorbent assay. The test uses monoclonal antibody highly specific to human pancreatic elastase-1. Normal values are above 500 μg/g, the intermediate range is 201 to 500 μg/g, and the pathologic values are below 200 μg/g.
Human or porcine synthetic secretin was used for stimulation (0.2 μg/kg body weight) and administered by intravenous push within 1 minute at the start of the endoscopy. The Secreflo (synthetic porcine secretin) was distributed by Repligen Co (Waltham, MA), and the Human Secretin by ChiRhoClin (Burtonsville, MD). During the insertion of the endoscope all of the gastric juice was removed before passing the endoscope into the duodenum. To further minimize gastric fluid contamination, approximately 1 to 2 mL of duodenal juice was suctioned out and discarded. Once the ampulla of Vater was visualized, a suction tube was attached to the endoscope and the sample collection was started. In smaller children (and occasionally in older ones), the fluid was removed through the endoscopy channel and collected by a fluid collection trap. Four fractions of duodenal fluid were collected between 5 and 10 minutes after the secretin administration. All of the samples were immediately placed on dry ice and sent to the gastroenterology laboratory for the assays. The peak enzyme activity of the 4 collected specimens was used for reporting.
The pH and protein content of the duodenal fluid were measured and samples with a pH <7 were discarded. Amylase, lipase, trypsin, chymotrypsin, and elastase activities were measured. A chromogen-linked peptide Suc-AAA-pNA was used as substrate for elastase measurement. Elastase hydrolyzes the peptide and releases the chromogen that gives a yellow color reaction. This chromogen (pNA, p-nitroaniline) is quantitatively measured by spectrophotometer at 405 nm, and it is proportional to elastase activity in the sample. Activity is expressed as the release of the chromogen (μmol) per minute, per milliliter sample (μmol · min−1 · ml−1). The cutoff activity for elastase is 10.5 μmol · min−1 · ml−1. It gives a sensitivity of 100% and specificity of 98.6% for pancreatic insufficiency based on the data of 33 patients with generalized pancreatic enzyme deficiency and 385 patients with normal enzyme activities. The institutional review board of the Alfred I. duPont Hospital for Children approved the present retrospective study.
Data were analyzed and compared between the FE-1 test and the PFT. Variables measured on the interval level were calculated by the mean and standard deviation and analyzed by Student t test. Variables measured on the nominal scale positive for a test were evaluated (yes = 1, no = 0), and the total number of the incidence was calculated and was analyzed by Fisher exact test. The 2×2 χ2 analysis was generated to obtain the specificity and sensitivity and correlation between the 2 tests. P < 0.05 was used to assess significance.
The characteristics of the patients with FE-1 <500 μg/g and >500 μg/g are summarized respectively in Table 1. The age between the 2 groups was statistically significant (P < 0.033), most likely because of the small sample size and variation with few older children in the <500 μg/g group. There was no significant difference in other clinical diagnoses between the 2 groups.
The average FE-1 concentration was 403 ± 142 μg/g, with a range from 33 to >500 μg/g. The average pancreatic elastase activity measured on direct stimulation was 49.1 ± 38.6 μmol · min−1 · ml−1. Fifty-nine children had normal direct pancreatic function test results and 11 children had values below the established cutoff (10.5 μmol · min−1 · ml−1). Of these 11 children, 6 had generalized enzyme deficiency, 3 had multiple enzyme deficiencies with normal trypsin activity, and 2 had multiple enzyme deficiencies with normal amylase and lipase activity. None of these children had a pathologic value (<200 μg/g) of FE-1 on testing.
After analysis, the correlation between the measured pancreatic elastase activity and FE-1 concentration was poor (r = 0.190). The sensitivity of the FE-1 test was found to be 41.7%, whereas the specificity was 49.2%. The positive predictive value of the FE-1 test was 14%. Table 2 shows the detailed comparison of the 2 tests.
The characteristics of the 11 patients with abnormal pancreatic enzymes on direct testing are summarized in Table 3. The average age in this group was 31 ± 22.3 months (mean ± SD). Nine of the 11 children had failure to thrive with a weight for height under the 5th percentile. These patients were treated with pancreatic enzyme supplementation and demonstrated weight gain and resolution of symptoms.
Indirect pancreatic function tests are easier to perform, more cost-efficient, and less invasive than direct pancreatic function testing. The 2 most widely used indirect stool tests are the fecal chymotrypsin and elastase-1 determinations. These pancreatic enzymes remain stable, avoiding degradation on transport through the gut. However, fecal chymotrypsin is variably affected during intestinal transport, and patients must discontinue pancreatic enzyme supplementation before the stool specimen collection.
Pancreatic elastase-1 is a pancreas-specific carboxy-endopeptidase that catalyzes the hydrolysis of native elastin. It is more stable than chymotrypsin during intestinal transport. A few publications compared the sensitivity of FE-1 with fecal chymotrypsin determination (5–11) and all of them concluded that the performance (sensitivity, specificity, or both) of FE-1 is better. The other advantage of the FE-1 is that the use of pancreatic enzyme supplementation does not need to be discontinued. Pancreatic elastase is concentrated 5 to 6 times in feces compared with pancreatic juice (3,12), and it remains stable in stool at room temperature for up to 1 week (4).
The FE-1 test has been evaluated for validity and reliability in multiple publications in adults in whom the main pancreatic disease is chronic pancreatitis. The sensitivity the FE-1 test varied between 22% to 63% in mild and 77% to 100% in moderate cases, and it was 100% in severe chronic pancreatitis cases in 3 adult studies (4–6). There are multiple publications on the value of FE-1 testing in children showing low sensitivity in children with CF with mild insufficiency (25%) and 100% sensitivity in moderate-to-severe cases (1,2,13–17).
Another disadvantage of FE-1 testing is the false positivity in children with malabsorption and infectious diarrhea. Nousia-Arvanitakis et al (18) reported decreased FE-1 concentrations in patients with active celiac disease and then normal values once on a gluten-free diet. Similar findings of decreased FE-1 in the presence of villous atrophy regardless of underlying etiology were reported by Walkowiak and Herzig (19). The FE-1 test can also give false-positive results with acute episodes of diarrhea as in a study in which 32 of the 112 children with acute enteritis had values below 200 μg/g, mainly in children with rotavirus and Salmonella enteritis(20).
In our study, of the 59 patients with normal elastase activity on direct testing, 11 had FE-1 activity <200 μg/g, leading to a high false-positive rate and a low specificity. On the contrary, of the 11 children with abnormal elastase activity on direct testing, 7 had normal FE-1 and none had values <200 μg/g. Therefore, the sensitivity and positive predictive value of the FE-1 test was poor. This is similar to data reported in mild cases of CF in children and chronic pancreatitis in adults. Only 5 of our patients had these diagnoses, whereas the majority had failure to thrive. Furthermore, we demonstrated that FE-1 testing does not identify children with isolated amylase, lipase, or colipase deficiency. Isolated pancreatic enzyme deficiencies in children have been reported in multiple cases in the pediatric literature (21–26).
Multiple factors may contribute to the poor sensitivity and specificity of the FE-1 test in children without CF. As discussed above, FE-1 can be transiently low in patients with villous atrophy and also in cases of bacterial and viral enteritis (18–20). A single specimen collection may lead to an unreliable result. In a study by Meyts et al (27), patients with CF had stool samples collected for 7 consecutive days and they concluded that the FE-1 test may be inconclusive as a result of significant intrapatient variability. This is less significant in patients with severe dysfunction; however, it could lead to poor sensitivity in mild-to-moderate dysfunction. In children without CF, such as the majority of patients in our study, there may be greater day-by-day variation in FE-1 concentration that can lead to false-positive test results. It may be necessary for children being evaluated for pancreatic insufficiency to have multiple stool samples for FE-1 testing to improve the overall sensitivity.
Our use of the secretin stimulation test with direct enzyme measurement has high sensitivity and specificity and helps to identify children with both general and isolated exocrine insufficiency. Two pediatric studies have compared the secretin-CCK stimulation test with the secretin test alone and reported that secretin alone resulted in no significant difference in enzyme activities with the exception of lipase that had lower absolute values, but the difference was not statistically significant (28,29).
In summary, FE-1 is the easiest available indirect test to assess exocrine pancreatic function in children; however, like other indirect tests it has limitations in the diagnosis of mild-to-moderate exocrine dysfunction and can give false-positive results in certain conditions as discussed above. Additionally, it can miss isolated amylase, lipase, and trypsin deficiencies. The ordering physician should be aware of these limitations of the FE-1 test and, if necessary, consider the use of direct pancreatic function testing.
1. Walkowiak J, Nousia-Arvanitakis S, Henker J, et al. Indirect pancreatic function tests in children. J Pediatr Gastroenterol Nutr 2005; 40:107–114.
2. Walkowiak J, Cichy WK, Herzig KH. Comparison of fecal elastase-1 determination with the secretin-cholecystokinin test in patients with cystic fibrosis. Scand J Gastroenterol 1999; 34:202–207.
3. Sziegoleit A, Krause E, Klor HU, et al. Elastase 1 and chymotrypsin B in pancreatic juice and feces. Clin Biochem 1989; 22:85–89.
4. Loser C, Mollgaard A, Folsch UR. Faecal elastase 1: a novel, highly sensitive, and specific tubeless pancreatic function test. Gut 1996; 39:580–586.
5. Glasbrenner B, Schon A, Klatt S, et al. Clinical evaluation of the faecal elastase test in the diagnosis and staging of chronic pancreatitis. Eur J Gastroenterol Hepatol 1996; 8:1117–1120.
6. Gullo L, Ventrucci M, Tomassetti P, et al. Fecal elastase 1 determination in chronic pancreatitis. Dig Dis Sci 1999; 44:210–213.
7. Carroccio A, Verghi F, Santini B, et al. Diagnostic accuracy of fecal elastase 1 assay in patients with pancreatic maldigestion or intestinal malabsorption: a collaborative study of the Italian Society of Pediatric Gastroenterology and Hepatology. Dig Dis Sci 2001; 46:1335–1342.
8. Dominguez-Munoz JE, Hieronymus C, Sauerbruch T, et al. Fecal elastase test: evaluation of a new noninvasive pancreatic function test. Am J Gastroenterol 1995; 90:1834–1837.
9. Katschinski M, Schirra J, Bross A, et al. Duodenal secretion and fecal excretion of pancreatic elastase-1 in healthy humans and patients with chronic pancreatitis. Pancreas 1997; 15:191–200.
10. Luth S, Teyssen S, Forssmann K, et al. Fecal elastase-1 determination: “gold standard” of indirect pancreatic function tests? Scand J Gastroenterol 2001; 36:1092–1099.
11. Walkowiak J, Herzig KH, Strzykala K, et al. Fecal elastase-1 is superior to fecal chymotrypsin in the assessment of pancreatic involvement in cystic fibrosis. Pediatrics 2002; 110 (1 Pt 1):e7.
12. Sziegoleit A, Linder D. Studies on the sterol-binding capacity of human pancreatic elastase 1. Gastroenterology 1991; 100:768–774.
13. Beharry S, Ellis L, Corey M, et al. How useful is fecal pancreatic elastase 1 as a marker of exocrine pancreatic disease? J Pediatr 2002; 141:84–90.
14. Nousia-Arvanitakis S. Fecal elastase-1 concentration: an indirect test of exocrine pancreatic function and a marker of an enteropathy regardless of cause. J Pediatr Gastroenterol Nutr 2003; 36:314–315.
15. Walkowiak J, Lisowska A. Re: fecal elastase: pancreatic status verification and influence on nutritional status in children with cystic fibrosis. J Pediatr Gastroenterol Nutr 2006; 42:117–118.
16. Walkowiak J, Nousia-Arvanitakis S, Agguridaki C, et al. Longitudinal follow-up of exocrine pancreatic function in pancreatic sufficient cystic fibrosis patients using the fecal elastase-1 test. J Pediatr Gastroenterol Nutr 2003; 36:474–478.
17. Walkowiak J, Lisowska A, Przyslawski J, et al. Faecal elastase-1 test is superior to faecal lipase test in the assessment of exocrine pancreatic function in cystic fibrosis. Acta Paediatr 2004; 93:1042–1045.
18. Nousia-Arvanitakis S, Karagiozoglou-Lamboudes T, Aggouridaki C, et al. Influence of jejunal morphology changes on exocrine pancreatic function in celiac disease. J Pediatr Gastroenterol Nutr 1999; 29:81–85.
19. Walkowiak J, Herzig KH. Fecal elastase-1 is decreased in villous atrophy regardless of the underlying disease. Eur J Clin Invest 2001; 31:425–430.
20. Salvatore S, Finazzi S, Barassi A, et al. Low fecal elastase: potentially related to transient small bowel damage resulting from enteric pathogens. J Pediatr Gastroenterol Nutr 2003; 36:392–396.
21. Ghishan FK, Lee PC, Lebenthal E, et al. Isolated congenital enterokinase deficiency. Recent findings and review of the literature. Gastroenterology 1983; 85:727–731.
22. Ghishan FK, Moran JR, Durie PR, et al. Isolated congenital lipase-colipase deficiency. Gastroenterology 1984; 86:1580–1582.
23. Hildebrand H, Borgstrom B, Bekassy A, et al. Isolated co-lipase deficiency in two brothers. Gut 1982; 23:243–246.
24. Ligumsky M, Granot E, Branski D, et al. Isolated lipase and colipase deficiency in two brothers. Gut 1990; 31:1416–1418.
25. Lilibridge CB, Townes PL. Physiologic deficiency of pancreatic amylase in infancy: a factor in iatrogenic diarrhea. J Pediatr 1973; 82:279–282.
26. Mehta DI, Wang HH, Akins RE, et al. Isolated pancreatic amylase deficiency: probable error in maturation. J Pediatr 2000; 136:844–846.
27. Meyts I, Wuyts W, Proesmans M, et al. Variability of fecal pancreatic elastase measurements in cystic fibrosis patients. J Cyst Fibros 2002; 1:265–268.
28. Del Rosario MA, Fitzgerald JF, Gupta SK, et al. Direct measurement of pancreatic enzymes after stimulation with secretin versus secretin plus cholecystokinin. J Pediatr Gastroenterol Nutr 2000; 31:28–32.
29. Madrazo-de la Garza JA, Gotthold M, Lu RB, et al. A new direct pancreatic function test in pediatrics. J Pediatr Gastroenterol Nutr 1991; 12:356–360.