More than a century after
its comprehensive description by Fitz [
It is generally
accepted that all patients with signs of moderate-to-severe acute pancreatitis should be
admitted to an intensive care unit and referred to specialised centres for
maximum supportive care [
To date, inhibition of
any known pathogenic step (by, e.g.,
octreotide, gabexate mesilate, and lexipafant) has not effectively reduced mortality
or increased long-term survival in severe acute pancreatitis [
UK guidelines [
Serum amylase and
lipase, the standard tests for acute pancreatitis diagnosis, are poor
predictors of severity [
E-cadherin is a 120-kDa transmembrane glycoprotein involved in the
calcium dependent adhesion of all epithelial cells.
Soluble E-cadherin is an 80-kDa peptide degradation product of the 120-kDa
E-cadherin molecule which is generated by a calcium ion dependent proteolytic
process [
(a) The zipper model for cadherin interaction showing homophilic interactions between E-cadherin molecules expressed on the cell membranes; (b) proteolytic cleavage of the extracellular portion of the E-cadherin fragment generating soluble E-cadherin.
Elevated serum
levels of sE-cadherin are found in patients with certain malignancies,
including pancreatic
[
In a previous
study by Pittard et al., significantly elevated levels of sE-cadherin were found in patients exhibiting a
systemic inflammatory response. Importantly, these elevated sE-cadherin levels
were found within an early time frame of the disease process and correlated to
the severity of the developed inflammatory response [
A systemic inflammatory response is seen in acute pancreatitis. Therefore, it is probable that sE-cadherin is cleaved from membrane-bound E-cadherin by inflammatory proteases. If this also occurs early in the inflammatory response in acute pancreatitis, and measurable serum levels of sE-cadherin relate to the severity of the developing disease, then this molecule could be used to predict disease severity. To test this hypothesis, we investigated, using a prospective pilot study, whether serum levels of sE-cadherin measured at an early time point following the onset of symptoms (within 12, 24, and 48 hours) in patients with acute pancreatitis were significantly different in patients who went on to develop mild, compared to severe disease. As serum levels of sE-cadherin are elevated in many different inflammatory processes, its use in the diagnosis of acute pancreatitis was not investigated as it is likely to be of limited value.
Patients admitted to the Newcastle NHS hospitals with acute pancreatitis were assessed for inclusion in the study. Acute pancreatitis was defined as acute abdominal pain with a typical clinical picture and a serum amylase level at least three times the upper limit of normal and/or typical findings on computed tomography. We excluded patients who had chronic pancreatitis, known malignancies, or were under 18 years. Clinical, pathological, and radiological patient data was collected prospectively on a database.
We enrolled by parallel recruitment a number of patients admitted with other abdominal inflammatory pathologies such as acute diverticulitis, perforated duodenal ulcer, cholangitis, acute appendicitis, and acute cholecystitis. A third group of healthy volunteers was also recruited. The same exclusion criteria applied to the patients with acute pancreatitis.
Ethical approval was obtained from the Newcastle and North Tyneside Health Authority, University of Newcastle upon Tyne and University of Northumbria at Newcastle, and Joint Ethics Committee.
Following informed
consent, three blood samples, for sE-cadherin measurement, were taken from each
individual. The first measurement was taken from the admission blood sample.
The subsequent three samples were taken at 12, 24, and 48 hours after the onset
of pain. Samples were subjected to centrifugation within 60 minutes and stored
at
Patients with local and/or organ failure were defined as having severe
acute pancreatitis according to the Atlanta classification [
Data was analysed
using GraphPad PRISM (version 3.0). The data for sE-cadherin levels were
compared using the Mann-Whitney
A total of 49 patients were recruited into the study; all met the eligibility criteria. 26 patients had acute pancreatitis, 12 had other acute abdominal pathologies, and 11 were healthy controls.
The overall median age of patients with acute pancreatitis was 56 (range 24–87); 12 were men. 19 patients had mild disease, and the remaining 7 had severe disease. The aetiology was gallstones in 13 patients (9 mild and 4 severe). Alcohol was the cause in 4 (3 mild and 1 severe). ERCP was the cause in 3 patients (2 mild and 1 severe). No obvious cause was found in 6 patients (5 mild and 1 severe). None of the mild cases and 1 of the severe cases died.
The median age of the 12 healthy controls (4 male) was 29 (range 21–54).
The median age of the 11 patients with other acute abdominal pathologies (5 male) was 59 (range 29–72). The final diagnoses for these patients were as follows: gastroenteritis 1; gastritis 1; acute cholecystitis 2; acute appendicitis 2; diverticulitis 1; cholangitis 1; perforated peptic ulcer 3.
At less than 12 hours
from onset of pain, the mean (standard deviation) concentration of sE-cadherin
in patients with severe acute pancreatitis was 17780 ng/mL (7853),
significantly higher than that of healthy volunteers 5180 ng/mL (1350),
sE-cadherin concentration in patients at 12 hours or less after onset of pain. Data are shown as a scatter plot with the mean represented by a solid line.
At 24 hours after
the onset of pain, the mean (SD) sE-cadherin concentration was 14320 ng/mL
(7532) in patients with severe acute pancreatitis, still significantly higher
than that of healthy volunteers 5518 ng/mL (1518),
sE-cadherin concentration in patients at 24 hours after onset of pain. Data are shown as a scatter plot with the mean represented by a solid line.
At 48 hours the
mean (SD) sE-cadherin concentration was 13360 ng/mL (6440) in patients with
severe acute pancreatitis, still significantly higher than that of healthy
volunteers 4928 ng/mL (1314)
sE-cadherin concentration in patients at 48 hours after onset of pain. Data are shown as a scatter plot with the mean represented by a solid line.
In the clinical setting of a patient with acute pancreatitis, initial therapy, accurate severity stratification, and an appropriate facility for patient management are of major interest for the admitting clinician. At present, there is no method by which to accurately predict severity. Many scoring systems have been proposed but all have their drawbacks. The possibility that an affordable, quick, single, and accurate test may exist has led clinicians to investigate numerous (mainly inflammatory mediators) biochemical molecules. Many have been assessed and detected either in serum or urine but for a number of reasons have failed to reach the clinical setting.
This study has
shown that serum levels of sE-cadherin can be used to predict severity of acute
pancreatitis at an early time point, with the mean differences in sE-cadherin
concentration being statistically different within 12 hours or less or at 24
hours after the onset of pain. Furthermore, although not presently widely available,
this test is quick, affordable (comparable to CRP), and could easily be
incorporated into hospital practice. It is thought
that sE-cadherin, elevated levels of which are found in inflammatory
conditions, originates
via cleavage from membrane-bound E-cadherin. The mechanism by which and at exactly which point during
an inflammatory process this occurs is not presently fully understood. Experimentally, molecules
such as metalloproteinases, kallikrein 7, plasmin, and trypsin have all been
shown to be capable of performing this cleavage process
[
In a recent study,
Steinhusen et al. [
The mean sE-cadherin concentrations were higher in our healthy
controls compared with those reported in one study (5180 ng/mL versus 2515 ng/mL,
resp.) [
Prior to our study, we suspected that raised sE-cadherin levels would be found in all patients experiencing an inflammatory event. From our results, it appeared that this was not the case, and that raised sE-cadherin levels were only found in patients with an inflammatory insult that resulted in significant multiorgan failure. The solitary high sE-cadherin level measured in the group of patients with other pathologies was from a patient with cholangitis who did go on to develop multiorgan failure.
Of the seven patients who had the severe form of acute pancreatitis, 6 had significantly elevated levels of sE-cadherin within 12 hours from the onset of pain. These patients required admission to critical care for organ support within 24 hours of their presentation to hospital and went on to develop MOF. The remaining patient, whose sE-cadherin level was not significantly raised within 12 hours after the onset of pain, is worthy of further discussion. This female patient presented with gallstone pancreatitis was managed initially on the general ward. A further bout of pain, however, developed on day 4 following admission for which she needed HDU care on day 5. She went on to develop multiorgan failure and pancreatic necrosis. It is possible that there may have been a second attack of pancreatitis induced on day 4 in an inpatient and that she subsequently developed severe acute pancreatitis and necrosis from this second attack. Unfortunately, we do not have serum sE-cadherin levels relating to this second bout of pain.
Despite the small sample size in this study, sE-cadherin seems to be an exciting potential very early marker (within 12 hours of onset of symptoms) of severity in acute pancreatitis. It is noted however that this is a small pilot study and that firm conclusions cannot be drawn until the hypothesis has been tested on a larger population. A larger sample size would permit more sophisticated statistical analysis. The predictive value of sE-cadherin could then be evaluated by receiver-operating characteristic (ROC) curves to determine the optimal cutoff value to predict severe AP. Sensitivity and specificity, as well as positive and negative predictive values, could then be calculated. This would also allow for a useful comparison of the prognostic value of E-cadherin with other predictive scores. We therefore feel that this molecule warrants further study in the form of a multicentre trial.