To test the diagnostic accuracy of faecal elastase (FE), a new test of exocrine pancreatic function, in a large prospective population of patients with abdominal complaints. Between January 1994 and December 1995, 131 patients (age range 17-82 years) were submitted for exocrine pancreatic function testing. Sixty-three patients had a firm diagnosis of chronic pancreatitis (CP) at stage I-III according to endoscopic retrograde cholangiopancreatography (ERCP). Twenty patients suffered from other pancreatic diseases (PD), and 48 patients had various other gastrointestinal diseases (GD). Fifty-seven healthy controls (HC) were also investigated. Stool specimens were analysed for FE by enzyme-linked immunosorbent assay (ELISA) and for faecal chymotrypsin (FC). The pancreolauryl serum test (PLT) was also performed in 97 patients and 23 healthy controls. FE was 200; 45-500 micrograms/g (median; range) in CP-I (n = 19), 94; 0-400 micrograms/g in CP-II (n = 14) and 38; 0-135 micrograms/g in CP-III patients (n = 30). With a cutoff of 200 micrograms/g, abnormal test results were found in 47% of CP-I, 79% of CP-II and 100% of CP-III patients; in 30% of PD patients, in 38% of GD patients and in 7% of HC. Sensitivity of abnormal FE in diagnosing CP was 79% (FC: 48%; PLT: 71%). The specificity of only 62% (FC: 73%; PLT: 67%) in the GD group increased to 78% (FC: 81%; PLT: 77%) when patients with small bowel diseases and diarrhoea (n = 22) were excluded. Faecal elastase is more sensitive than chymotrypsin and comparable to the pancreolauryl test in the diagnosis of chronic pancreatitis. Indirect exocrine pancreatic function tests are not helpful to differentiate between pancreatic and small bowel diseases in a prospective population of patients with abdominal complaints.
Read full abstract