Immunoreactive elastase I: clinical evaluation of a new noninvasive test of pancreatic function

Transcription

1 linicalhemistry 4: -6 (1996) mmunoreactive elastase : clinical evaluation of a new noninvasive test of pancreatic function J URGEN TEN,* MHAEL JUNG, ANDREA ZEGOLET, TEFAN ZEUZEM, WOLFGANG F APARY, and BERNHARD LEMBKE We have evaluated the diagnostic value of the fecal elastase test in comparison with the secretin-pancreozymin test in the diagnosis of exocrine pancreatic insufficiency Pancreatic elastase was measured immunologically nununoreactive elastase activity in spot stools from controls ranged from 136 to 444 g/g; 95% of all values were within 175 to 15 tg/g The elastase assay Vs ranged from 33% to 63% (intraassay) and from 41% to 1% (interassay) The output of elastase correlated well with those of amylase, lipase, and trypsin, yielding respective correlation coefficients of 83, 8, and 84 in controls and 86, 91, and 91 in patients with impaired pancreatic function n contrast to fecal chymotrypsin, the test results were unaffected by pancreatic enzyme replacement therapy These results indicate that fecal immunoreactive elastase may be recommended as a new, noninvasive tubeless test of pancreatic function NDEXNG TERM: secretin-pancreozymin test chymotrypsin fecal fat malabsorption steatorrhea Because invasive diagnostic approaches of impaired pancreatic function, such as the secretin-pancreozymin test with its several modifications or the Lundh test, are time-consuming, invasive, unpopular with patients, and expensive, numerous indirect or tubeless tests of pancreatic function (eg, fecal chymotrypsin, bentiromide, and pancreolauryl tests) have been introduced as alternatives [1]Among them, the 7-h fecal fat analysis still remains the standard test for diagnosing and quantifying fat malabsorption in chronic pancreatitis [], even though it is both insensitive and nonspecific in the diagnosis of chronic pancreatitis [3, 4] Furthermore, fecal fat analysis is also time consuming, requires special laboratory equipment, and is unpopular among the laboratory staff Among tubeless pancreatic function Division of Gastroenterology, nd Deparunent of ntemal Medicine, Johann- Wolfgang-Goethe-University, Theodor-tern-Kai 7, 695 Frankfurt, Germany, and nstitute of Medical Microbiology, University of Giessen, Giessen, Germany *uthor for correspondence Fax Received July, 1995; accepted November 3, 1995 tests, the determination of fecal chymotrypsin has been accepted as an indirect test for many years ts sensitivity reportedly ranges from 7% to 9%, and its specificity, which appears to depend on the selection of the control subjects studied, ranges from 49% to 9% [5-13] Previously ziegoleit detected human pancreatic elastase 1 [14], a member of the acidic elastase family, as a new endoprotease and sterol-binding protein present in both human pancreatic secretions and feces Quantitative studies by rocket immunoelectrophoresis indicated that this enzyme was unaffected during intestinal passage oncentrations in feces were five to six times greater than those determined in pancreatic juice [15,16] Therefore, we propose that concentrations of this enzyme in feces reflect pancreatic function The aim of this prospective study was to evaluate the specificity and sensitivity of fecal immunoreactive elastase as an accurate test of exocrine pancreatic function by comparing the results with direct measurement of pancreatic enzyme secretion PATENT Materials and Methods Fecal elastase was measured in stool samples from 164 consecutive patients presenting between August 199 and December 1993 (84 males, 8 females), ages 5-85 years (mean 41 years), who were admitted to the hospital (a) for the differential diagnosis of malabsorption syndromes (history of diarrhea, weight loss) or (b) with previously established causes of malabsorption or maldigestion (exocrine pancreatic insufficiency due to chronic pancreatitis or cystic fibrosis, inflammatory bowel disease, celiac sprue) and from a subgroup of extensively investigated patients with functional abdominal pain, but with entirely normal markers of gastrointestinal functions, who served as controls (Table 1) n each case, final diagnoses were established after thorough investigation according to generally accepted criteria The study was in accordance with ethical standards of the local committee PROEDURE Enzyme measurements Elastase was determined immunological! [16] with a new sandwich -type enzyme immunoassa (chebo-tech, Wettenberg, Germany) hymotrypsin activi

2 linicalhemistry 4, No, Table 1 Final diagnoses of patients studied (n = 164) ontrols Pancreatic insufficiency Due to chronic pancreatitis Due to cystic fibrosis Due to pancreatectomy Nonpancreatic disorders eliac disease Functional diarrhea rritable bowel disease Alcoholic liver disease Hemochromatosis Billroth l resection Pancreatic adenocarcinoma in stool samples was determined according to Kaspar et al [17], amylase according to Bernfeld [18],lipase according to Ziegenhorn et al [19],and trypsin according to Erlanger et al [], all with commercial test kits manufactured by Boehringer Mannheim (Mannheim, Germany) Fecalfat estimation Fecal fat was analyzed according to van de Kamer et a! [1]or by near-infrared analysis according to tein et al [] ; 1 U) 16! 14 9 ) GpO TAT! T because none of the data sets showed a gaussian distribution, omparison of paired differences was performed with the Wildayl day day ll Fig 1 hanges in immunoreactive elastase assayed in three different stools of six healthy controls () and of three patients with exocrine pancreatic insufficiency () during 3 days at room temperature coxon rank test, and comparison of unpaired data with the Mann-Whitney test orrelations were calculated with the method of least squares Limits of significance were P <5 in all tests ample preparationtool was collected for 7 h in plastic containers, and was either analyzed the same day or stored at - #{176} until analysis For fecal fat estimation, all stool samples were diluted, carefully homogenized (Ultraturrax-blender; Janke & Kunkel, taufen, Germany), and analyzed by the above-mentioned standard methods amples were analyzed immediately after homogenization or after freezing ecretin-pancreozymin testtudies were performed in the morning after overnight fasting [3-6]Briefly, after a 1-h fasting period, a double-lumen Lagerlof tube (R#{5}sch, Waiblingen, Germany) was placed into the duodenum under x-ray control After 15 mm of basal secretion, the subjects received intravenously secretin, 1 mg/kg body wt (Hoechst, Frankfurt, Germany), and 3 mm later cholecystokinin-pancreozymin (K-PZ; Ferring, Malm#{46}, weden), 1 lu/kg body wt To evaluate exocrine function, we measured the volume and concentration of bicarbonate produced 3 mm after stimulation with secretin Amylase, elastase, lipase, and trypsin output were measured 3 mm after K-PZ tabilityofimmunoreactiveelastase For stability studies we used stool samples stored at room temperature for various times (see egend to Fig 1) The degradation process was interrupted by reezing the sample at - #{176} The influence of enzyme replacement therapy on fecal mmunoreactive elastase excretion was studied in 4 patients with cystic fibrosis Results Precision The intraassay V of immunoreactive elastase, calculated from 1 consecutive assays of seven different fecal samples, was 33-63% (mean 48%) nterassay V, calculated over 7 days from seven different fecal samples, ranged from 41% to 1% (mean 77%) Refrrence intervals Fecal immunoreactive elastase concentration in controls ranged from 136 to 444 tg/g; 95% of all values (5th to 975th percentile) were within 175 to 5 Mg/g, and the mean ± D was 183 ± 193 zg/g n contrast to other pancreatic enzymes, immunoreactive elastase concentrations were not significantly lower in children than in adults (963 ± 143 jtg/g vs 113 ± 14 jtg/g) We and others [7] found an age-related increase of immunoreactive elastase only in the first 3 months after birth tabilityin stoolsamplesto determine the effect of bacterial degradation on elastase immunoreactivity in the time between passage of stool and analysis, we analyzed the changes in immunoreactive elastase assayed in three different stools of six healthy controls and of six patients with exocrine pancreatic insufficiency during 3 days at room temperature As shown in Fig 1, no remarkable rate of immunoreactivity loss was detectable in stool samples at 4 #{176} mmunological specificity No immunoreactive elastase could be detected in either porcine or bovine pancreatic enzyme preparations (< g/g granulate or tablet) ubstitution replacement therapy in 1 patients with exocrine pancreatic insufficiency due to cystic fibrosis resulted in a significantly greater output of chymotrypsin activity and a significantly decreased

3 4 tein et al: mmunoreactive elastase test of pancreatic function - - > U! 15 1 T J_ Lltat!ZZ chymotrypin eiastase without with enzyme replacement therapy Fig Fecal excretion (mean and D) of chymotrypsin, fat, and immunoreactive elastase during a 4-h collection period in 1 patients with cystic fibrosis and steatorrhea with and without enzyme replacement therapy fecal fat content, whereas fecal immunoreactive elastase was not affected by substitution therapy (Fig ) Elastase releasein duodenalaspirate There was a good correlation (Fig 3) between the output of elastase and that of amylase, 8 lipase, and trypsin The respective correlation coefficients were 83, 8, and 84 in normal controls (n = 5) and 86, 91, and 91 in subjects with impaired pancreatic function (n = ) Furthermore, the output of elastase in duodenal aspirate corre- 8 lated well with the amount of elastase in feces (r = 87, P <1) Figure 4 illustrates a receiver-operating characteristic (RO) curve based on calculations of sensitivity and specificity at several cutoff values for fecal elastase [81 The fecal elastase assay achieved optimal discrimination (defined as maximal likelihood ratio: the true-positive rate divided by the true-negative rate) at a cutoff value of 175 tg/g wet wt (specificity 94%, sensitivity 93%) When patients with pathological secretinpancreozymin test results were divided into groups with and without steatorrhea, correctly abnormal fecal elastase increased to 96% in severe exocrine pancreatic insufficiency n the less severely impaired patients, without steatorrhea, the diagnostic sensitivity decreased to 88% n the same groups of patients with chronic pancreatitis, the diagnostic sensitivity for chymotrypsin was 91%, whereas its diagnostic specificity was only 91% (with steatorrhea) and 56% (without steatorrhea) (Table ) Discussion The objective of the present study was the evaluation of fecal immunoreactive elastase for the diagnosis and differential diag- E 6 E? 4 a, 5 ( o 4 ocpo 6#{176} E ob od o 4 Q iooooo ( 5 c oo Elastase (g/3 mm) Elastase (llg/3 mm) Fig 3 orrelations between the output of elastase and that of amylase, lipase, and trypsin, and between elastase in stool and elastase pancreatic juice after K-PZ stimulation = normal function; = impaired function -

4 linical hemistry 4, No, c 8- U > > True negative fraction (1-pecificity) Fig 4 RO analysis for fecal elastase at various cutoff values A cutoff value <175 Mg (arrow showed the best test qualities nosis of pancreatic exocrine insufficiency This test is highly specific for human pancreatic elastase, because elastase does not interfere with pancreatic enzyme preparations Thus, in contrast to fecal chymotrypsin tests, the results are not affected by pancreatic enzyme replacement therapy A strong correlation was found between the duodenal secretion of elastase and that of amylase, lipase, and trypsin in the secretin-pancreozymin test onsidering the relationship between enzymatic outputs in feces and in pancreatic juice after K-PZ stimulation, we found a satisfactory correlation for elastase As with fecal chymotrypsin, the immunoreactive elastase test is stable in feces for several days, which offers the possibility that stool samples from outpatients can be mailed to diagnostic centers n contrast to chymotrypsin, however, which has a fecal recovery of only 5% of its activity in pancreatic juice, the degradation of immunoreactive elastase is negligible during intestinal passage, resulting in a five- to sixfold greater concentration than that in pancreatic juice [14] Determination of fecal chymotrypsin has been accepted as an indirect test for pancreatic function Measurement of fecal chymotrypsin can also be used to detect pancreatic insufficiency in children with cystic fibrosis The test has some advantages over other tubeless tests: The test is simple to perform when an Table Fecal elastase compared with secretin-pancreozymin test (PT) and fecal chymotrypsln n pancreatic nsufficiency with and without steatorrhea No (%) abnormal Fecal Fecal PT n elastase chymotrypsin Bicarbonate concentration#{176}, 7 6 (88) 4 (56) enzyme output abnormal, fecal fat normale Bicarbonate concentration8, enzyme 1 (96) (91) output, fecal fat abnormal Maximal concentration of bicarbonate: >7 mmol/l (normal) b Normal output 3 mm after stimulation: amylase >1 U/3 mm, trypsin >3 11/3 mm; lipase >65 11/3 mm Normal stool fat <7 g/day automatic titration is available, and stool samples from outpatients can be mailed to diagnostic centers that perform the assay, because chymotrypsin activity is very stable over several days at room temperature However, the sensitivity of fecal chymorrypsin determination as a test for chronic pancreatitis ranges from 45% to 1%, with most studies reporting values of 7-9% [5-1] False-positive results have been reported in a variety of nonpancreatic diseases: liver cirrhosis, Billroth gastrectomy, celiac sprue, rohn disease, and other gastrointestinal diseases associated with diarrhea and malabsorption [5,9, 3]tudies comparing results of the fecal chymotrypsin determination with those of the secreun-pancreozymin test (or analogous tests such as the secretin-cerulein test) showed that fecal chymotrypsin detects 85% of patients with advanced chronic pancreatitis but only 49% of those with mild or early manifestation of the disease [11-13] Previously Munch and Arnmann 1] used a fecal immunoreactive lipase assay as a new tubeless test in the diagnosis of pancreatic exocrine insufficiency by means of a new ELA technique n contrast to fecal chymotrypsin, the test results were unaffected by pancreatic enzyme replacement therapy Despite the excellent diagnostic specificity of 98%, however, the assay had the disadvantage of very low sensitivity (34%) n contrast, the measurement of fecal immunoreactive elastase has both high diagnostic sensitivity and specificity All subjectswith nonpancreauc diarrhea and celiac disease, except for two patients with prior gastric resection (Billroth ) and two with severe secretoric diarrhea due to enzyme dilution (water content 98%), had immunoreactive elastase values within normal limits (<175 ig/g stool) The overall sensitivity of the immunoreactive elastase test was also satisfactory, especially in patients with moderately or severely impaired pancreatic function Of the patients with severe pancreatic insufficiency (steatorrhea) or cystic fibrosis, none had a falsely normal test result Thus, compared with the fecal chymotrypsin or fecal lipase, the fecal elastase test showed a higher diagnostic sensitivity and specificity n conclusion, the simple performance, the noninvasiveness, and the diagnostic efficiency of the fecal immunoreactive elastase test make it one of the most satisfactory pancreatic function tests for screening of pancreatic insufficiency Furthermore, the test seems to be useful not only in adults, but also in children as a screening test for cystic fibrosis imilarly to all pancreatic function tests, the fecal test for immunoreactive elastase is unable to differentiate between pancreatic insufficiency due to chronic pancreatitis and that due to pancreatic cancer We thank U Bieniek and A chmitt for performing fecal fat and enzyme analyses and also to the staff of the Gastroenterological Laboratory for assistance with cholecystokinin stimulation tests We are also grateful to AF Hofinann for critical reading of the manuscript and his helpful comments Parts of this work were presented at the Annual Meeting of American

5 6 tein et al: mmunoreactive elastase test of pancreatic function Gastroenterological Association, New Orleans, May 15-18, 1994 References 1 DiMagno EP, Go VLW, ummerskill Hi Relations between pancreatic enzyme outputs and malabsorption in severe pancreatic insufficiency N Engl Med 1973;88:813-5 Bo-Linn G, Fordtran i Fecal fat concentration in patients with steatorrhea Gastroenterology 1984;87:319-3 Lembcke B, Grimm K, Lankisch PG Raised fecal fat concentration is not a valid indicator of pancreatic steatorrhea Am Gastroenterol 1987;8: Hofmann AF, Manier iw Fecal fat concentration Determinants and diagnostic value Gastroenterology 1985:89: Ammann RW, Akovbiantz A, Haecki W, Largiader F, chmid M Diagnostic value of the fecal chymotrypsmn test in pancreatic insufficiency, particularly chronic pancreatitis: correlation with the pancreozymmn-secretion test, fecal fat excretion and final clinical diagnosis Digestion 1981:1: Adler G, Weidenbach F Fecal chymotrypsin in chronic pancreatitis disease n: Malfertheimer P, Ditschuneit H, eds Diagnostic procedures in pancreatic diseases Berlin: pringer-verlag, 1986: Niederau, Grendell in Diagnosis of chronic pancreatitis [Reviewl Gastroenterology 1985;88: Dyck WP Titrimetric measurements of fecal trypsin and chymotrypsin in cystic fibrosis Am J Dig Dis 1967:1: Lami F, allegari, Miglioli M, Barbara L A single-specimen fecal chymotrypsin-test in the diagnosis of pancreatic insufficiency: correlation with the secretmn-cholecystokmnin and NBT-PABAtests Am Gastroenterol 1984;79: Duerr HK, Otte M, Forell MM, Bode Fecal chymotrypsmn: a study on its diagnostic value by comparison with the secretmn-cholecystokinin-test Digestion 1978;17: ale 1K, Goldberg DM, Thjodleifson B, Wormsley KG Trypsin and chymotrypsin in duodenal aspirate and feces in response to secretmn and cholecystokinin-pancreozymmn Gut 1974;15: Dyck W, Ammann RW Quantitative determination of fecal chymotrypsin as a screening test for pancreatic exocrine insufficiency Am J Dig Dis 1965:1: Muller L, Wisniewski Z, Hansky The measurement of fecal chymotrypsin: a screening test for pancreatic exocrine insufficiency Aust Ann Med 197:19: ziegoleit A A novel proteinase from human pancreas Biochem 1984;19: ziegoleit A, Krause E, Kl#{7}r HU, Kanacher L, Linder D Elastase and chymotrypsin B in pancreatic juice and feces lin Biochem 1989;: ziegoleit A, Linder D tudies on the sterol-bmnding capacity of human pancreatic elastase Gastroenterology 1991;1: Kaspar P, Moeller G, Wahlefeld A New photometric assay for chymotrypsmn in stool lin hem 1984:3: Bernfeld P Amylase a, J3 Methods Enzymol 1955;1: Ziegenhorn J, Neumann U, Knitsch KW, Zwez W Determination of serum lipase [Abstract] lin hem 1979:5:167 Erlanger BF, Kolowsky N, ohen W The action of chymotrypsin on two chromogenic substrates Arch Biochem 1961;95: van de Kamer JH, Huinik HB, Weyers HA Rapid method for the determination of fat in feces Biol hem 1949;177: tein, Purschian B, Bieniek U, aspary WF, Lembcke B Nearinfrared reflectance analysis (NRA): a new dimension in the investigation of malabsorption syndromes Eur Gastroenterol Hepatol 1994:6: reutzfeldt W Funktionsdiagnostik bel Erkrankungen des exokrinen Pankreas Verh Dtsch Ges nn Med 1964;7: Lankisch PG, chreiber A, Otto Pancreolauryl-test Evaluation of a tubeless pancreatic test in comparison with other indirect and direct tests for exocrine pancreatic function Dig Dis ci 1983; 8: LankischPG Exocrine pancreatic function tests [Reviewl Gut 198:3: Wormsley KG Pancreatic function tests [Reviewl lin Gastroenterol 1978;7: Terbrack HG, G#{5}rtler KH, Kl#{46}r H, Lindemann H Human pancreatic elastase concentration in faeces of healthy children and children with cystic fibrosis [Abstract] Gut 1995;37():53 8 Zweig MN, ampell G Receiver-operating characteristic (RO) plots: a functional evaluation tool in clinical medicine [Reviewl lin hem 1993:39: Dyck W, Ammann RW Quantitative determination of fecal chymotrypsin as screening test for pancreatic exocrine insufficiency Am Dig Dis 1965:1: Ammann RW, Tagwercher E, Dashiwagi H, Roasenmund H Diagnostic value of fecal chymotrypsin and trypsin assesment for detetection of pancreatic disease Am Dig Dis 1968:13: Munch R, Ammann RW Fecal immunoreactive lipase: a new tubeless pancreatic function test cand Gastroenterol 199; 7:89-94