Pancreas divisum

Histologic Findings and Their Association With Persistent Pain

* Objective.-Indications for major pancreatic resections have been expanded to include complicated chronic pancreatitis (CP). We assessed clinical findings and outcomes and evaluated histology in patients who had major pancreatic resections for CP. We also determined if histologic findings were associated with persistent postoperative pain.

Design.-We reviewed charts and slides from 44 patients who underwent major pancreatic resections for CP between 1989 and 1999.

Results.-The etiology for disease included alcohol (n = 15), hereditary (n = 5), idiopathic (n = 6), pancreas divisum (n = 3), stricture (n = 2), trauma (n = 2), systemic lupus erythematosus (n = 1), and unknown (n = 10). Patients included 20 men and 24 women; ages ranged from 22 to 76 years. Perioperative mortality and morbidity were 0% and 4.5%, respectively. Persistent pain was present in 25 (57%) of the 44 patients, and pain was encountered

more frequently in patients with alcoholic pancreatitis (67%) versus other etiologies (52%), and in those who underwent Whipple/Beger or total resections (68%) versus distal or subtotal pancreatectomy (42%). Metaplastic changes were present in 14 cases, and ductal atypia was seen in 9 cases. No malignancies were found. Acinar necrosis and acute inflammation were seen more often in patients with persistent pain than in those who were pain free (P = .081).

Conclusions.-Major pancreatic resection for CP can be performed with low morbidity and mortality. This procedure relieves pain in nearly half the patients. There is a wide spectrum of histopathologic changes seen in CP, including ductal atypia and metaplastic changes. Acute exacerbations of CP identified histologically at the time of surgery and alcohol as etiology for CP may be associated more frequently with intractable pain.

(Arch Pathol Lab Med. 2001;125:1051-1054)

Chronic pancreatitis (CP) is a disease of the pancreas characterized by progressive destruction of the gland C with accompanying irregular fibrosis and chronic inflammation (1988 Marseille-Rome classification).1,2 Clinically, CP manifests as episodic or continuous upper abdominal pain, and it can be complicated by cysts, pseudocysts, pancreatic calcifications, diabetes mellitus, or pancreatic exocrine insufficiency.2-4 An early description of CP pathology and pathogenesis was presented by Comfort et al in 1946.5 In 1966, at Marseille, France, the first histologic classification was adopted.6,7 Although alcohol is responsible for more than 80% of adult cases in the industrialized countries, other etiologies have been implicated, such as pancreas divisum and abnormalities of the duct system.8-10 Between 10% and 40% of all patients presenting with CP have no recognizable predisposing factor.

Intractable abdominal pain is the main feature in more than 95% of patients with CP and can manifest as repeated attacks of mild to severe pain or as persistent epigastric or back pain.11 The pathogenesis of abdominal pain remains unknown, although several possibilities have been suggested, such as ductal hypertension due to strictures or calculi, inflammation and scarring around the pancreatic nerves, alterations of the perineural sheath, pseudocysts, acute inflammation, cell-mediated cytotoxicity, and fat necrosis.4,11-14

In the past, major pancreatic resections (MPRs), such as cephalic pancreaticoduodenectomy (Whipple) and caudal and subtotal/total pancreatectomy, were performed almost exclusively for malignant tumors.15,16 Since the operative morbidity and mortality have significantly decreased in more recent years, resection of part or the entire pancreas is considered a treatment option in cases of intractable pain or other complications from severe CP.17-20 The optimal management of complicated CP remains controversial because many patients continue to experience persistent postoperative pain.21,22

We evaluated histopathologic and clinical findings in patients who had MPRs for CP, assessed outcome, and determined if disease etiology, operative procedure, and/ or histologic findings could help predict recurrent or persistent postoperative pain. Additionally, we assessed histopathologic changes in CP

METHODS

We reviewed charts and all slides from 44 consecutive patients who underwent MPRs for CP between 1989 and 1999; these resections included 18 Whipple procedures (cephalic pancreaticoduodenectomies), 4 Beger resections (head of the pancreas), 21 distal/subtotal pancreatectomies, and 1 total pancreatectomy. Age, sex, etiology for CP, morbidity, mortality, and length of hospitalization were noted. Hospital charts and follow-up clinical notes were reviewed retrospectively. The follow-up time after surgery ranged from 1 to 10 years. Patients were classified as postoperative pain free (no mention of pain at follow-up visits, no prescribed narcotic use, and no emergency room visits for pain) or persistent postoperative pain (pain complaints at the clinic, referral to pain management, continuous need for narcotics, or emergency room visits for pain).

Hematoxylin-eosin-stained slides were reviewed by 2 pathologists and were scored as follows: 0, normal; 1, less than 25% of the pancreas involved; 2, 25% to 50%; and 3, more than 50% for chronic inflammation, fibrosis, and perineural inflammation (Figure 1). Ductal metaplasia, hyperplasia, and atypia were noted. Other findings, such as presence of intraductal calculi, impacted secretions, parenchymal calcifications, associated acute inflammation, acinar and fat necrosis, as well as islet loss/hyperplasia, were documented as absent (0) or present (1). A Fisher exact test was used to test the independence of pain, surgical procedure, and etiology, as well as the association between pain and different histologic findings.

RESULTS

The etiology for disease included alcohol (n = 15), hereditary (n = 5), idiopathic (n = 6), pancreas divisum (n = 3), stricture (n = 2), trauma (n = 2), systemic lupus erythematosus (n = 1), and unknown (n = 10). Patient age ranged from 22 to 76 years, with a mean of 44 years. There were 20 men and 24 women. Mean duration of hospitalization was 20 days. Perioperative mortality and morbidity were 0% and 4.5%, respectively. Morbidity included 1 wound infection, 1 fistula from pancreatojejunostomy, 1 pneumonia, and 1 urinary tract infection.

Persistent pain was present in 25 (57%) of the 44 patients. It was more frequently encountered in patients with alcoholic pancreatitis (10/15, 67%) as compared with all other etiologies combined (15/29, 52%), and in patients who underwent Whipple/Beger resections and total pancreatectomy (17/25, 68%) than in those with distal or subtotal pancreatectomy (8/19, 42%). However, the differences were not statistically significant.

All cases showed evidence of CP on histologic review. Metaplastic duct epithelial changes were present in 14 cases (32%; 11 mucinous, 1 squamous, and 2 mixed) (Figure 2), papillary hyperplasia in 2 cases (5%) (Figure 3), and ductal atypia was seen in 9 cases (20%) (Figure 4). No carcinoma in situ or adenocarcinoma was found.

Intraductal impacted secretions were present in 18 cases (41%), calcifications in 6 (14%), calculi in 2 (5%), abscesses in 4 (10%), and pseudocysts in 5 (11%). Features of acute exacerbation of CP were identified in the form of associated acute inflammation in 17 cases (38%), acinar necrosis in 11 cases (25%), and fat necrosis in 16 cases (37%). The average scores for chronic inflammation, perineural inflammation, and fibrosis (mean standard deviation) were 1.1 +/- 0.5, 0.6 +/- 0.9, and 1.8 +/- 0.9, respectively, in patients with pain and 1.2 +/- 0.4, 0.8 +/- 1.0, and 2.2 +/- 0.8, respectively, in patients without pain (P > .5). Neuroendocrine hyperplasia was present in 9 cases (20%), with 2 cases showing significant islet hyperplasia that raised the differential diagnosis of endocrine tumor (Figure 5). Occasionally, single endocrine cells were noted penetrating the fibrous tissue and infiltrating around the nerves (Figures 6 and 7).

Patients with persistent pain were found to have an increased incidence of acinar necrosis and associated acute inflammation as opposed to patients free of pain (Fisher exact P value = .081). Chronic inflammation, fibrosis, perineural inflammation, fat necrosis, ductal impacted secretions, and calculi were not statistically associated with persistent postoperative pain (Table).

COMMENT

Major pancreatic resections for CP can be performed with very low perioperative mortality and morbidity and have an acceptable success rate (43% pain free). Using adequate selection criteria, MPR can be the treatment of choice for severe CP associated with intractable abdominal pain.17,20-22 Other studies have reported good to excellent results in two thirds of the patients, with perioperative mortality ranging between 0% and 1.7% and a 5-year survival rate of 85% to 93%17,20 Traverso and Kozarek17 reported that 76% of their patients who underwent pancreaticoduodenectomy were free of pain, while 100% noticed a significant improvement. Major abdominal pain was the primary reason for surgery, and pain relief was most commonly obtained in patients who had diabetes mellitus or pancreatic duct disruption.

In our patients, persistent pain was not significantly associated with any single disease etiology or operative procedure. Patients who underwent Whipple/Beger and total pancreatectomy procedures were more likely to have persistent pain, as compared to those who had distal/subtotal pancreatectomies; however, the difference was not statistically significant. This finding may be related to patient selection for the particular operative procedure, rather than the type of procedure performed. Persistent severe postoperative pain was seen more often in alcoholic CP than in CP from all other causes combined, but the difference was not statistically significant. Other studies have shown that excessive alcohol consumption is more frequent among patients with unsatisfactory surgical results and have considered patients with alcoholic pancreatitis poor candidates for surgery.2.21,22 The lack of abstinence with subsequent acute exacerbations of pancreatitis in the residual pancreas may explain persistence of postoperative intractable pain in alcoholic CP.2.3 It has been suggested that patients most likely to be pain free after MPR have no preoperative use of tranquilizers, little dilatation of the pancreatic duct, and no pseudocysts.22

Chronic inflammation, perineural inflammation, fibrosis, fat necrosis, inspissated secretions, and calculi did not significantly differ between patients with and without continued postoperative pain, and therefore these histologic findings could not help predict continued pain after MPR. The association between histopathologic findings and persistent abdominal pain in CP has been studied extensively, but the pathogenic mechanism for the pain has not been defined.4,10,23 Perineural scarring and inflammation have been described in both painful and painless CP.4,10,23 Previous studies have shown an increase in the number and diameter of the pancreatic nerve fibers in CP,2 as well as perineural infiltration by T lymphocytes, suggesting a cell-mediated cytotoxic mechanism for the neural alteration.12 Di Sebastiano et al13 proposed that the lymphocytic infiltration of pancreatic nerves is one of the pathogenic factors for the generation of pain. They found that the perineural inflammation scores were significantly higher in patients with persistent pain. In our patients, perineural chronic inflammation was present in 48% of cases, but there was no statistically significant association between the percentage of nerves with inflammation and persistent postoperative pain. The difference between our results and those of Di Sebastiano et al could be due to the different methods used to determine perineural inflammation scores. Additional studies are necessary to further evaluate the association between perineural inflammation and pain.

Acinar necrosis and acute inflammation showed a trend toward association with persistent pain (P = .081). The most prominent feature of exacerbated CP is considered to be acute inflammation, mainly affecting the pancreatic ducts.23 Therefore, our findings suggest that acute exacerbation of CP at the time of resection may predict an increased likelihood for persistent postoperative pain.

Irregular fibrosis, acinar atrophy, and chronic inflammation were common histologic features in our cases. Similar to other studies, we noted a wide spectrum of histologic findings in CP.4,10,23 Duct lesions, including dilatation, inspissated secretions, abnormal epithelium, and calculi, were common components of CP Epithelial atypia, metaplasia, and papillary hyperplasia without atypia were often present in the pancreatic ducts in CP and sometimes raised the differential diagnosis with intraductal papillary tumors and well-differentiated duct adenocarcinomas. The relative risk for a patient with CP to develop carcinoma is 50 times greater compared to other individuals.24 Duct lesions, particularly papillary hyperplasia with atypia, are more common in the vicinity of pancreatic cancer.24,25 Therefore, pathologists should show extreme caution when examining duct changes in CP. Epithelial duct metaplasia, proliferation, and atypia may be seen in CP in association with obstruction and inflammation, but severe dysplasia is not common and could represent a premalignant lesion that should alert the pathologist to submit more sections in order to exclude an adjacent invasive carcinoma. Duct atypia, metaplasia and hyperplasia, and pseudoneoplastic proliferation of islets are histologic features that can occur in CP and are important to be aware of.

Significant islet cell hyperplasia and single-cell fibrous tissue permeation with perineural infiltration were seen in 2 of the CP cases we reviewed. Usually, a neuroendocrine proliferation in the setting of CP is nonneoplastic. In some instances, however, the proliferation may be marked and may exhibit stromal or perineural invasion; these lesions may mimic neoplastic processes.26,27 Whereas identification of a single clonal cell population is helpful in confirming a neoplastic process, the presence of multiple cell populations is inconclusive, and therefore there are no absolute criteria that can distinguish with certainty between exuberant regeneration and incipient neoplasia.27 Stromal and perineural invasion are not necessarily helpful in diagnosing a malignant process in neuroendocrine proliferation in CP.27 Unequivocal criteria for malignancy in such difficult cases are local invasion and metastasis.26,27

In summary, MPRs have become a valid treatment option in patients with severe CP and intractable pain, and satisfactory results can be obtained in carefully selected surgical candidates. The exact causes for intractable pain in CP after MPR remain unclear, and histologic parameters evaluated in this study were not statistically associated with persistent postoperative pain. However, acute exacerbation of CP identified histologically at the time of resection and alcohol as the etiology for CP may be associated more frequently with intractable pain.

Accepted for publication March 22, 2001.

References

1. Sarles H, Adler G, Dani R, et al. The pancreatitis classification of Marseilles-- Rome. Scand] Gastroenterol. 1989;24:641-642.

2. Chad ST, Singer MV. The problem of classification and staging of chronic pancreatitis: proposals based on current knowledge of its natural history [see comments]. Scand] Gastroenterol. 1994;29:949-960.

3. Mallinson C. The pancreas: chronic pancreatitis. Br Hosp Med. 1977;18: 553-566.

4. Kloppel G. Pathology of chronic pancreatitis and pancreatic pain. Acta Chir Scand. 1990;156:261-265.

5. Comfort MW, Gambill EE, Bagenstos AH. Chronic relapsing pancreatitis, an analysis of 29 cases without associated disease of the biliary or gastrointestinal tract. Gastroenterology. 1946;6:239-285.

6. Sarles H. Revised classification of pancreatitis: Marseille 1984. Dig Dis Sci. 1985;30:573-574.

7. Gyr K, Singer MV, Sarles H. Pancreatitis: concepts and classifications. In: Proceedings of the Second International Symposium on the Classification of Pancreatitis, Marseille, France. New York, NY. Elsevier Science Publishing Co; 1984: 28-30.

8. Samer M, Cotton PB. Classification of pancreatitis. Gut. 1984;25:756-759. 9. Sidhu SS, Tandon RK. The pathogenesis of chronic pancreatitis. Postgrad Med]. 1995;71:67-70.

10. Singh SM, Reber HA. The pathology of chronic pancreatitis. World] Surg. 1990;14:2-10.

11. Jensen AR, Matzen P, Malcow-Moller A, et al. Pattern of pain, duct morphology, and pancreatic function in chronic pancreatitis: a comparative study. Scand Gastroenterol. 1984;19:334-338.

12. Ebert MP, Ademmer K, Muller-Ostermeyer F, et al. CD8+ CD103+ T cells analogous to intestinal intraepithelial lymphocytes infiltrate the pancreas in chronic pancreatitis. Am J Gastroeneterol. 1998;93:2141-2147.

13. Di Sebastiano P, Fink T, Weihe E, et al. Immune cell infiltration and growth-- associated protein 43 expression correlate with pain in chronic pancreatitis. Gastroenterology. 1997;112:1648-1655.

14. Bockman DE, Buchler M, Malfertheimer P, et al. Analysis of nerves in chronic pancreatitis. Gastroenterology. 1988;94:1459-1469.

15. Imaizumi T, Hanyu F, Hanada N, Hatori T, Fukuda A. Extended radical Whipple resection for cancer of the pancreatic head: operative procedure and results. Dig Surg. 1998;15:299-307.

16. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg. 1997;226:248-260.

17. Traverso LW, Kozarek RA. Pancreatoduodenectomy for chronic pancreatitis: anatomic selection criteria and subsequent long-term outcome analysis. Ann Surg. 1997;226:429-438.

18. Rao R, Prinz RA. Surgical therapy in chronic pancreatitis. Curr Opin Gen Surg. 1993;287-293.

19. Beger HG, Buchler M, Bittner R. The duodenum preserving resection of the head of the pancreas (DPRHP) in patients with chronic pancreatitis and an inflammatory mass in the head: an alternative surgical technique to the Whipple operation. Acta Chir Scand. 1990;156:309-315.

20. Howard JM, Zhang Z. Pancreaticoduodenectomy (Whipple resection) in the treatment of chronic pancreatitis. World JSurg. 1990;14:77-87.

21. Gebhardt C. Surgical treatment of pain in chronic pancreatitis; role of the Whipple procedure. Acta Chir Scand. 1990;156:303-307.

22. Ebbehoj N, Christensen E, Madsen P. Prediction of outcome of pancreaticogastrostomy for pain in chronic pancreatitis. Scand Gastroenterol. 1987;22: 337-342.

23. Shimizu M, Hirokawa M, Manabe T. Histological assessment of chronic pancreatitis at necropsy. J Clin Pathol. 1996;49:913-915.

24. Bradt Dj, Lillemoe KD, Jeo CJ, et al. Progression of pancreatic intraductal neoplasias to infiltrating adenocarcinoma of the pancreas. Am J Surg Pathol. 1998;22:163-169.

25. Castellano-Sanchez AA, Perez MT, Cabello-Inchausti B, et al. Intraductal carcinoma (carcinoma in situ) of the pancreas with microinvasion. Ann Diagn Pathol. 1999;3:39-47.

26. Solcia E, Capella C, Kloppel G. Tumors of the Pancreas. Washington, DC: Armed Forces Institute of Pathology; 1997:215-231, 237-244. Atlas of Tumor Pathology, 3rd series, fascicle 20.

27. Batow SA, Mukai K, Rosai J. Pseudoneoplastic proliferation of endocrine cells in pancreatic fibrosis. Cancer. 1981;47:2627-2633.

Daniela M. Proca, MD; E. Christopher Ellison, MD; Dan Hibbert, BS, Wendy L. Frankel, MD

From the Departments of Pathology (Drs Proca and Frankel and Mr Hibbert) and Surgery (Dr Ellison), The Ohio State University Medical Center, Columbus, Ohio.

Presented in part as a poster at the US and Canadian Academy of Pathology, March 2000.

Reprints: Wendy L. Frankel, MD, Department of Pathology, Ohio State University, 401E Doan Hall, 410 W 10th Ave, Columbus, OH 43210 (e-mail: frankel-1@medctr.osu.edu).

Copyright College of American Pathologists Aug 2001
Provided by ProQuest Information and Learning Company. All rights Reserved