Serum trypsin concentrations in diabetes mellitus.

Fifty patients (age range 14-65 years) with diabetes mellitus were studied; none of them had clinical evidence of chronic pancreatitis, chronic diarrhoea, or renal dysfunction. Duration of diabetes ranged from two weeks to 30 years. Patients were classified according to type of treatment: 25 were insulindependent diabetics; 15 were treated with sulphonvlureas, including two newly discovered diabetics; in 10 patients, including one new patient, diabetes was stabilised with biguanides. Seventy-six healthy subjects and 20 hospitalised patients without pancreatitis, diabetes, or other gastrointestinal disease served as controls. Serum trypsin concentration was measured by radioimmunoassay, by a double-antibody technique.2 Highly purified trypsin was used for iodination and as an immunogen. Antibody was raised in rabbits. lodination was carried out with chloraminc. Within-assay variance was 7 5 °0 at low concentrations and 2 7 0(, at high concentrations. Repeat trypsin measurements in the same patients under similar conditions varied less than 20 at low trypsin concentrations. Blood samples were obtained at around 11 am, roughly two to three hours after breakfast and the injection of insulin or the intake of oral hypoglycaemic agents. Samples were allowed to clot, centrifuged, and the sera stored at -20'C until assay. Eighteen out of 25 insulin-dependent diabetics had subnormal trypsin concentrations. The mean serum trypsin concentration in the insulin-dependent patients was subnormal and was also significantly lower (P < 0 005) than that in patients treated with sulphonylureas (figure). Four out of 15 patients treated with sulphonylureas also had subnormal trypsin concentrations; the mean for this group was in the low-normal range. Serum trypsin concentration in the group treated with biguanides was similar to that in the normal subjects, and was significantly greater than that in patients who received sulphonylureas (P < 0 05).