Does paraduodenal pancreatitis systematically need surgery?

Paraduodenal pancreatitis is a rare pathology, which is unfamiliar to most clinicians and thus usually underdiagnosed. The space between the duodenum, the head of the pancreas, and the main bile duct has been called “the groove” because of its anatomical appearance. This space is peculiar in that it allows the development of pathological situations of imprecise origin, of difficult diagnosis due to a large spectrum of clinical presentations and, above all, of difficult therapeutic approaches, which are still a matter for debate.

Different clinical names have been given to the disease in the past: “cystic dystrophy in heterotopic pancreas,” “cystic dystrophy of the duodenal wall,” or “groove pancreatitis.” Potet and Duclerc reported the first clinical description of this entity in the French literature as early as 1970 [1]. It was later identified from a large French clinical series of patients as “cystic dystrophy of heterotopic pancreas developed within the gastric or, more often the duodenal wall” [2]. The same disease was also identified as “groove pancreatitis” by German clinicians, due to the very localized site of the lesions between the duodenum and the head of the pancreas [3]. In 2004, Adsay and Zamboni proposed the term “paraduodenal pancreatitis,” which is now commonly accepted [4].

Paraduodenal pancreatitis is represented by the systematic association of some elementary morphological lesions: chronic or acute inflammation, cystic component, dystrophic alterations, malformative heterotopic development, and location in the duodenal wall around the minor papilla or the surrounding pancreas [5]. These lesions explain the histological association of dilated ducts lined by pancreatic epithelium in the duodenal wall, hypercellular granulation tissue with numerous giant cells, hyperplasia of Brunner’s glands, and fibromyoblastic dense proliferation with active fibrosis [6].

Physiopathology of paraduodenal pancreatitis is far from clear. In particular, it is not clear why the lesions are localized around the minor papilla. The role of protein plugs within the duct of Santorini, hyperplasia of the Brunner’s glands, and the presence of embryological vestiges of pancreatic glands within the duodenal wall have all been hypothesized, but a clear understanding of the origin of the disease has yet to be achieved [6]. Based on scarce available epidemiological data, patients are predominantly relatively young men (40 – 50 years old) with, in almost all cases, a history of chronic long-lasting alcohol abuse; no pediatric cases have been reported to date. Although the most frequent etiology of chronic pancreatitis (alcohol consumption) is frequently associated with paraduodenal pancreatitis, no other etiological factor of acute or chronic pancreatitis has been reported (IgG4 autoimmune disease, gallstone disease, hypercalcemia or familial chronic pancreatitis). However, the possibility of a drug-induced acute paraduodenal pancreatitis has recently been advocated with isoniazid [7].

The current issue of Endoscopy includes a retrospective analysis of a large series of paraduodenal pancreatitis cases by Arvanitakis et al. The paper describes the predominant symptoms, which are directly linked to pathological lesions: severe recurrent abdominal pain (as in the case of chronic pancreatitis), and vomiting related to the presence of cysts within the duodenal wall, which compress the upper part of the duodenum, in some cases even impeding the passage of the endoscope [8]. Weight loss is systematically present to a variable extent, which may falsely orientate the diagnosis toward malignancy when it is important and rapid. Compression of the lower part of the main bile duct may induce jaundice, which may also direct the diagnosis toward malignancy.

Due to the presence of variable degrees of pseudocystic ductal dilation and dense inflammatory fibrosis, the diagnosis may be considered as a differentiation between and/or a mix of two forms of paraduodenal pancreatitis, one characterized by multiple duodenal cysts and the other by a solid mass. The diagnosis is usually easy in cases of multiple cysts because of their size (usually small), their location (in between the pancreatic head and the duodenal wall or within the wall), and their number, which altogether render them quite unlikely to correspond to a malignant disease. In this situation, a diagnosis of paraduodenal pancreatitis can be made easily by various imaging techniques, which usually demonstrate the presence of multiple small cysts within the pancreatic groove, while some modifications of the remaining part of the pancreas are compatible with chronic pancreatitis changes (spots in pancreatic parenchyma, irregularities of the ducts or presence of stones). Only intrapancreatic mucinous neoplasia (IPMN) may create some confusion when it is located at the “branch ducts” in the upper region of the pancreatic head; however, in this case, cysts are located inside the pancreatic tissue and not in the duodenal wall, and communication between cysts and the main duct cannot be demonstrated in cases of paraduodenal pancreatitis. It should be noted that in both of these cases, surgery is not indicated.

Obviously the diagnostic distinction between a malignant tumor becomes very challenging when paraduodenal pancreatitis appears as a solid mass with scarce or no cystic formation. In this difficult situation, all imaging procedures may give some discriminative information between the two diagnoses. At ultrasonography, pseudo-tumoral paraduodenal pancreatitis appears like a tumor located in the pancreaticoduodenal groove: localized heterogeneous lesion with hypoechoic band-like pattern beside a thickened hyperechoic duodenal wall, while the pancreatic head appears slightly heterogeneous as in case of chronic pancreatitis [9]. On contrast-enhanced computed tomography scan, the suspicious mass of soft tissue corresponding to paraduodenal pancreatitis is hypoenhancing within the pancreaticoduodenal groove [10] [11]. The absence of vascular involvement and the pattern of the limits of the mass, which appears triangular between thickened duodenal wall in the case of paraduodenal pancreatitis rather than rounded, should also be considered. In most cases of paraduodenal pancreatitis, signs of chronic pancreatitis (calcifications, alterations of ducts, diffuse fibrosis) may also occur; however, chronic pancreatitis may also exist prior to cancer development, and therefore a clinical history of chronic pancreatitis may not rule out the diagnosis of carcinoma. In cases of paraduodenal pancreatitis, magnetic resonance imaging (MRI) also detects the fibrotic scar in the pancreatic groove (hypointense signal relative to pancreas in both T1 – or T2-weighted images) with a peripheral centripetal enhancement after gadolinium injection [12] [13].

Endoscopic ultrasound (EUS) provides most valuable information: location of cysts in the 4th hypoechoic layer (muscularis propria) of the duodenal wall, which is enlarged around the minor papilla area, lack of communication with pancreatic ducts, heterogeneous pattern of the lesion, which appears less hypoechoic than a carcinoma and the absence of vascular involvement around the tumor [14]. Lymph nodes do not achieve the same size and pattern as those in metastatic carcinoma. Linear EUS also allows the guiding of fine-needle aspiration (FNA) cytology (in both the tumoral lesion and lymph nodes), the negativity of which does not rule out malignancy but is an important argument favoring the diagnosis of a benign disease such as paraduodenal pancreatitis [15] [16]. If FNA cytology plays an important part in the decision to treat paraduodenal pancreatitis patients conservatively, one has to keep in mind that in some cases, FNA cytology may falsely orientate a diagnosis toward malignancy, such as an islet cell neoplasm, a spindle cell malignant proliferation, or an undifferentiated carcinoma with osteoclast-giant cells [17] [18]. This latter situation is particularly treacherous, as the presence of giant cells is one of the histological elements of paraduodenal pancreatitis diagnosis [6]. EUS may also participate in treatment by fenestrating or puncturing some cysts into the duodenal lumen. From a morphological viewpoint, a carcinoma that develops on a “main duct” IPMN can mimic this situation if the suspicious mass is located in the pancreatic groove; however, the association of a largely dilated main pancreatic duct is not frequently seen in cases of paraduodenal pancreatitis. As in other pathologies, endoscopic retrograde cholangiopancreatography no longer has a role in the diagnosis of paraduodenal pancreatitis but it is involved in the treatment.

As discussed in the paper by Arvanitakis et al. endoscopy has an important role to play in treating paraduodenal pancreatitis: cyst fenestration or drainage with double-pigtail plastic stents as used for chronic pancreatitis cysts (cystoduodenostomy or, less frequently, cystogastrostomy) or ductal drainage with plastic stents through the main papilla but also through the minor one [19]. All of these endoscopic treatments have to be attempted in order to avoid surgery, which in almost all cases is a pancreaticoduodenectomy; however, they have to be performed in a stepwise manner with particularly careful surveillance of the patient in order to recognize soon enough that surgery may become necessary. Furthermore, the Arvanitakis series demonstrates the great therapeutic benefit of combining two nonsurgical approaches to the treatment of paraduodenal pancreatitis: endoscopic treatments and medical therapy with long-acting somatostatin analogs. Our group first reported the efficacy of long-acting somatostatin analogs, which become most important when the papilla cannot be reached by the endoscope because of cyst number and/or compression [20]. The Arvanitakis results also show the high rate of recurrence after initial conservative treatment; however, even in these cases, conservative treatment could be repeated or prolonged successfully, with only 9 of 41 patients eventually needing surgery. In this series, patients underwent surgery not because of uncertainty over the diagnosis, as is often the case, but because of treatment failure despite a long stenting period (26 – 32 months), which was longer than that observed in usual cases of chronic pancreatitis.

In the Arvanitakis series, the diagnosis of paraduodenal pancreatitis was based mainly on MRI and EUS, when three signs were present (duodenal thickening, cyst in the duodenal wall, and mass in this area). The presence or absence of these three signs has recently been demonstrated to correctly identify paraduodenal pancreatitis or carcinoma, with a specificity of 88.2 % and 86.7 %, respectively [21]. These contributions strongly confirm that in cases of paraduodenal pancreatitis, malignancy can be reasonably ruled out thanks to modern imaging techniques, and that clinicians can become more confident in their treatment decisions, allowing nonsurgical therapies to be selected, including the combination of long-acting somatostatin analogs and endoscopic treatment [8] [15] [20]. Nevertheless, one has to remain cautious in the care of these patients: carcinoma can still become a possibility and surgery can still be indicated if diagnosis is not correctly based on MRI and EUS or if the evolution of the disease under conservative treatment is not satisfactory [22].

• References

• 1 Potet F, Duclerc N. [Cystic dystrophy on aberrant pancreas of the duodenal wall.] Article in French. Ach Fr Mal App Dig 1970; 59: 223-238
• 2 Fléjou JF, Potet F, Molas G et al. Cystic dystrophy of the gastric and duodenal wall developing in heterotopic pancreas: an unrecognized entity. Gut 1993; 334: 343-347
• 3 Stolte M, Weiss W, Volkholz H et al. A special form of segmental pancreatitis: “groove pancreatitis”. Hepatogastroenterology 1982; 29: 198-208
• 4 Adsay NW, Zamboni G. Paraduodenal pancreatitis: a clinico-pathologically distinct entity unifying “cystic dystrophy of heterotopic pancreas”, “paraduodenal wall cyst” and “groove pancreatitis”. Semin Diagn Pathol 2004; 21: 247-254
• 5 Fléjou JF. Paraduodenal pancreatitis: a new unifying term and its morphological characteristics. Diagn Histopathol 2011; 18: 31-36
• 6 Arora A, Dev A, Mukund A et al. Paraduodenal pancreatitis. Clin Radiol 2014; 69: 299-306
• 7 Yi PH, Veltre D, Kuttab J et al. Acute groove pancreatitis due to isoniazid. Neth J Med 2013; 71: 104
• 8 Arvanitakis M, Rigaux J, Toussaint E et al. Paraduodenal pancreatitis: is the noninvasive approach the way to go?. Endoscopy 2014; 46: 580-587
• 9 Wronski M, Karkocha D, Slodkowski M et al. Sonographic findings in groove pancreatitis. J Ultrasound Med 2011; 30: 111-115
• 10 Latham J, Sanjay P, Watt DG et al. Groove pancreatitis: a case series and review of the literature. Scott Med J 2013; 58: 28-31
• 11 Ishigami K, Tajima T, Nishie A et al. Differential diagnosis of groove pancreatic carcinomas vs. groove pancreatitis: usefulness of the portal phase. . Eur J Radiol 2010; 74: 95-100
• 12 Blasbalg R, Baroni RH, Costa DN et al. MRI features of groove pancreatitis. Am J Roentgenol 2007; 189: 73-80
• 13 Castell-Monsalve FJ, Sousa-Martin JM, Carranza-Carranza A. Groove pancreatitis: MRI and pathological findings. Abdom Imaging 2008; 33: 342-348
• 14 Tio TL, Luiken GJ, Tytgat G. Endosonography of groove pancreatitis. Endoscopy 1991; 23: 291-293
• 15 Brosens LA, Leguit R, Vleggaar F et al. EUS-guided FNA cytology diagnosis of paraduodenal pancreatitis (groove pancreatitis) with numerous giant cells: conservative management allowed by cytological and radiological correlation. In press 2014. [Epub ahead of print] Cytopathology DOI: 10.1111/cyt.12140.
• 16 Chute DJ, Stelow EB. Fine-needle aspiration features of paraduodenal pancreatitis (groove pancreatitis): a report of three cases. Diagn Cytopathol 2012; 40: 1116-1121
• 17 Tison C, Regenet N, Meurette G et al. Cystic dystrophy of the duodenal wall developing in heterotopic pancreas: report of 9 cases. Pancreas 2007; 34: 152-156
• 18 Levenick JM, Gordon J, Sutton J et al. A comprehensive, case-based review of groove pancreatitis. Pancreas 2009; 38: 169-175
• 19 Isayama H, Kawabe T, Komatsu Y et al. Successful treatment for groove pancreatitis by endoscopic drainage via the minor papilla. Gastrointest Endosc 2005; 61: 175-178
• 20 Basili E, Allemand I, Ville E et al. [Lanreotide acetate may cure cystic dystrophy in heterotopic pancreas of the duodenal wall.] Article in French. . Gastroenterol Clin Biol 2001; 25: 1108-1111
• 21 Kalb B, Martin D, Sarmiento JR et al. Paraduodenal pancreatitis: clinical performance of MR imaging in distinguishing from carcinoma. Radiology 2013; 269: 475-481
• 22 Patriti A, Castellani D, Partenzi A et al. Pancreatic adenocarcinoma in paraduodenal pancreatitis: a note of caution for conservative treatments. Updates Surg 2012; 64: 307-309