Unravelling the Mystery of Pancreatic Pain

Correspondence: Charles F. Frey, MD, Dept. of Surgery, UC Davis Medical Center, 4302 X St., Rm. 2310, Sacramento, CA 95817.

There are many unanswered questions regarding the pathogenesis of chronic pancreatitis, e.g., the role of alcohol, the frequently spotty distribution of the disease within the pancreas, the cause of pain, why some patients have pain and others do not, the mechanisms of perineural damage, the reason nerves are increased in size and number in chronic pancreatitis (CP), why ductal decompression provides pain relief in many patients, why section of the splanchnic nerves often provides little relief from pain in patients with CP, and why total pancreatectomy may not provide pain relief in as many as 20% of patients with CP. ^1,2

The report by Friess and colleagues in this issue of Annals of Surgery³ addresses two unresolved questions relating to pain, and is noteworthy for the various disciplines represented in the authorship and the meticulous and sophisticated experimental design, techniques, and controls employed by the authors to answer the questions proposed. The data support the authors’ conclusions and advance our knowledge of pancreatic nerves and pain in CP.

The authors’ questions were, first, why do most patients with CP have severe pain, and second, is there a relationship between pain and pancreatic nerves, which are enlarged in size and number in CP. Severe pain, one of the most defining features of CP, is often associated with adverse consequences to the patient and society, including narcotic addiction, physical and emotional disability, unemployment, and welfare dependency. Since the 1950s, clinicians have tried to relieve typical CP pain by interrupting the afferent sympathetic nerve supply to the pancreas. ⁴ These efforts met with little or no success, and even when attempted recently in a highly selective patient population, successful long-term pain relief eluded many patients. The presence of inflammatory infiltrates and fibrosis about the nerves ⁵ and of injury to the perineurium of pancreatic nerves observed by electron microscopy, ⁶ however, along with the presence of substance P, and calcitonin gene-related peptide and histamine release ⁷ have kept the focus on the pancreatic nerves and their relationship to the pain of CP.

How does one approach a problem that involves such disparate components as highly subjective and variable pain, the concrete objective pathologic changes of fibrosis and inflammation in pancreatic nerves, and the elusive biochemical bases for these pathologic changes? The authors have applied a remarkable array of sophisticated modern scientific techniques to search for possible causes of the nerve changes. They lead us carefully from the clinical to the molecular and make this study a model for the application of molecular biology to clinical problems.

In Advice to a Young Scientist, P. B. Medawar describes research as the “art of the soluble.” Very often a solution turns on devising some means of quantifying phenomena. Quantification as such has no merit except insofar as it helps solve problems; to quantify is not to be a scientist, but goodness, it does help! ⁸ These authors from three continents had the skills, knowledge ranging from clinicians to molecular biologists, and motivation to overcome the obstacles associated with coordinating a multidisciplinary, multiinstitutional, multinational group of experts whose focus was chronic pancreatitis and pain. The authors have followed the dictum to “quantitate, then correlate” in a laudable manner. This is a major strength of the report. Application of molecular techniques to the study of such an individualized, internalized, and variable condition as pancreatic pain requires first of all that pain be quantified just as much as nerve growth factor.

The authors chose a pain scale described by Di Sebastiano et al. ⁹ Utilizing Northern blot analysis, in situ, hybridization, and immunohistochemical staining, investigators measured levels of nerve growth factor (NGF) and its high-affinity receptor tyrosine kinase receptor (TrkA) from multiple sites in the pancreas of 24 patients with CP and 20 controls. The control pancreata came from multiple organ donor programs in which there were no recipients available. NGF and TrkA were measured in pancreatic nerves, blood vessels, acini, and ducts in pancreata from the patients with CP and the controls. In patients with CP, areas which appeared normal histologically were examined as well as those that were inflamed and fibrotic. The results of these studies were correlated with the clinical findings of pain severity and duration.

The authors found the high-affinity receptor for NGF (TrkA) most concentrated in the tissues in which it was synthesized, i.e., in the perineurium of the largest pancreatic nerves in CP patients, while NGF was most concentrated and synthesized in degenerating acinar cells, in metaplastic duct cells, and acinar cells dedifferentiating into tubular structures. NGF and TrkA were also present in the intrapancreatic ganglia. When these results were correlated with the history of pain and histologic findings in CP patients, a positive correlation was found between NGF mRNA expression and, respectively, the extent of pancreatic fibrosis, acinar cell changes, and ductal metaplasia, but not with the intensity or frequency of pain. TrkA mRNA levels correlated with the intensity, but not the frequency of pain, fibrosis, acinar cell degeneration, or ductal cell metaplasia. Missing from the report is a table showing the individual pain intensity and frequency correlations with NGF mRNA and TrkA mRNA levels, along with an explanation of the statistical manipulations of the NGF mRNA and TrkA mRNA data that when “taken together” showed a strong relationship to pain frequency and pain intensity, which would allow the reader to see and judge these correlations for themselves.

The authors postulate that NGF released into the interstitium of the pancreas in patients with CP from degenerating acinar cells and metaplastic ductal cells interacts with the TrkA receptor in the damaged perineurium, promoting nerve growth and repair (and elevated levels of the neuronal plasticity marker GAP-43). The authors also propose that the NGF/TrkA interaction and pathway may contribute to the pain of CP in 54% to 67% of patients, which implies that there are other mechanisms as yet unidentified that also may contribute to the pain experienced by patients with CP. Drawing on their own data and the literature, the authors have presented a compelling case for the role of the NGF/TrkA pathway in both the growth of nerves in CP and the intensity of pain in patients with CP. To their credit and credibility, despite the importance of the authors’ findings, if anything they understate their significance. Without the careful groundwork that characterizes the clinical and pathologic aspects of CP, the molecular measurements have no meaning. Fortunately for us, and eventually for patients with this vexing pain, the measurements and correlations have been made and significant relationships have been established.

Like any other advance in scientific knowledge, this study raises new issues which bear investigation. What other factors beside the NGF/TrkA pathway contribute to pain in CP patients? What is the size and number of nerves found in patients with CP who present not with pain, but with steatorrhea? Hopefully, this talented team will continue to contribute answers to the many still unresolved questions about the pathogenesis of chronic pancreatitis.