There are no pathognomonic symptoms in acute pancreatitis. In reaching a diagnosis the patient’s history, physical findings, and serum amylase or lipase concentrations have to be considered. Serum amylase is one of the oldest clinically useful laboratory tests, having been around for almost 70 years.1 However, too great a reliance on amylase or lipase will lead to underdiagnosis owing to the lack of sensitivity and specificity.2

Up to 20% of patients with acute pancreatitis run a severe clinical course and these patients must be identified as quickly as possible. The “gold standard” for staging patients with acute pancreatitis is dynamic contrast enhanced computed tomography.3 Disease severity, however, may only become apparent at laparotomy (or necropsy). Operative intervention is not needed in uncomplicated cases and should be avoided if at all posssible in severe acute pancreatitis. Researchers have also attempted to differentiate between mild and severe forms of acute pancreatitis using so-called indicators of necrosis in blood or urine. Examples of these parameters are C-reactive protein (CRP), PMN elastase, phospholipase A₂, antiproteases, and cytokines.4-8 CRP is the most useful in clinical practice, but its use is hampered by the fact that the differentiation between mild and severe disease is best three to four days after onset. Some of the others still have methodological problems, or are only used experimentally, or their measurement is too time consuming, or too expensive.

Recently a Finish group have reported on complex formation between trypsin 2 and α₁-antitrypsin in acute pancreatitis.9 They showed that the serum concentrations of the complex between trypsin 2 and α₁-antitrypsin was increased in all 28 cases of severe pancreatitis but in only three of the 82 with mild pancreatitis. This verifies the results of earlier studies on the pathophysiology of acute pancreatits, which indicated that it is the activated trypsin in the pancreas and its vicinity that causes the inflammatory reaction.10 As the reaction is local, it is impossible to measure, and therefore one must measure traces of trypsin activity in blood or urine. One way to do this is to measure the small amounts of active trypsin which bind to protease inhibitors making their way to the systemic circulation (trypsin 2–α₁-antitrypsin complex). This is technically possible today and there is good evidence for a correlation between the concentration of these protease inhibitors and disease severity.9 ,11 However, it is still not known whether these complexes are present in other acute intra-abdominal inflammatory conditions, and the method is not yet suitable for emergency analysis.

Another recently described method to facilitate the diagnosis of acute pancreatitis is the measurement of trypsinogen 2 (anodal trypsinogen) in urine.12 Its concentration in urine correlates strongly with disease severity, in contrast to cathodal trypsinogen and amylase. The explanation is probably that these enzymes are normally excreted via glomerular filtration in the primary urine and at least 90% are degradated in the kidney tubuli. In acute pancreatitis the tubuli are heavily loaded by an increased filtration of large amounts of pancreatic enzymes with the concomitant occurrence of renal insufficiency.13 Why this affects the anodal trypsin to a higher degree than the other enzymes is not known, but provides a good basis for fast immunological methods.

Another way of detecting trypsinogen activation is to measure the peptide that is cleaved from trypsinogen during activation to trypsin, a method first examined by Hermon-Taylor et al.14 The activation peptide of trypsinogen, TAP, can be measured in urinary samples using a radioimmunoassay. There is good evidence for a correlation between the TAP concentration and disease severity. However, it has not been proved that TAP is derived from inflamed pancreatic tissue, and antibody to TAP is difficult to produce as it is so small (just five amino acids). Even though the potential of TAP has been noticed for almost a decade now, there is still a lack of clinical data proving its superiority over CRP.

Appelros et al, in this issue (see page 97), present theories that are similar to those of Hermon-Taylor’s group, but they have based their assay on the activation peptide of carboxypeptidase B—that is, CAPAP, instead. This peptide (95 amino acids) is considerably larger than TAP, making its measurement more reliable. Moreover, the CAPAP is stable in both serum and urine, which is an advantage over TAP. The Malmö group has found CAPAP in large concentrations in the urine in cases of acute pancreatitis. This large peptide can be present only if trypsin is activated and procarboxypeptidase B is available as a substrate, and, therefore, is likely to reflect the activation of trypsin. Most interestingly, Appelros et al also showed that there is a strong correlation between the concentrations of CAPAP in serum and urine and the severity of acute pancreatitis. This offers for the first time a reliable staging method in acute pancreatitis because, in contrast to the other markers for the assessment of disease severity, the concentration of CAPAP directly reflects the degree of pancreatic damage. However, as only a few patients with severe acute pancreatitis have been investigated so far using CAPAP analysis, Appelros et al’s study should be seen as the first step in the evaluation of this new and interesting parameter. Additionally, the specificity of this test in serum and urine has yet to be defined.

Soon we may be able to diagnose and stage acute pancreatitis using a single test, and CAPAP analysis is a promising candidate.