Editor,—Some recent studies on the measurement of serum trypsin 2–α-1-antitrypsin complexes in the diagnosis of acute pancreatitis have piqued our interest,1 ,2 including that of Kemppainen et al (Gut1997;41:690–5). We published a couple of articles on trypsin 1–α-1-antitrypsin complexes in serum some years ago. Principally, we arrived at the same conclusion as the Finnish group—namely, that concentrations of these complexes in serum correlate with disease severity.3-5

We had some problems with our assay in that we very often found low but measurable concentrations (5–25 μg/l) of trypsin 1–α-1-antitrypsin complexes in serum samples obtained from patients with no signs of acute pancreatitis. Our method was not sensitive enough to allow detailed characterisation of this low immunoreactivity and we were therefore not sure what it was comprised of. We did not publish these results as we found them hard to believe. If they were true it would mean that a substantial amount of trypsinogen was normally activated in and around the pancreas during the normal turnover of pancreatic enzymes (see later). Our hypothesis has always been that …