Pancreatic pseudocyst as a cause of upper gastrointestinal bleeding

Pancreatic pseudocyst as a cause of upper gastrointestinal bleeding

The Netherlands JOURNAL OF MElDICINE3 ELSEVIER Netherlands Journal of Medicine 46 (1995) 90-94 Brief report Pancreatic pseudocyst as a cause of up...

3MB Sizes 2 Downloads 117 Views

The Netherlands

JOURNAL OF MElDICINE3 ELSEVIER

Netherlands Journal of Medicine 46 (1995) 90-94

Brief report

Pancreatic pseudocyst as a cause of upper gastrointestinal bleeding A.L. Zanen, A.C.M. van Vliet * , R. Beukers Department

of Internal

Medicine,

Drechtsteden

Ziekenhuis,

P.O. Box 444, 3300 AK Dordrecht,

Netherlands

Received 18 April 1994; revision received 30 May 1994; accepted 3 June 1994

Abstract A 67-year-old man with a history of an attack of pancreatitis was repeatedly investigated for recurrent gastrointestinal bleeding necessitating blood transfusions. Routine investigations did not reveal the source of bleeding. Repeated angiograms also were not diagnostic. A hot spot identified on a 99mTc-pertechnate-labelled erythrocyte scan prompted an endoscopic retrograde cholangio-pancreatography (ERCP), which showed bleeding through the papilla of Vater. The source of bleeding appeared to be a small pancreatic pseudocyst. The patient was treated with a duodenopancreatectomy in which the pylorus was preserved. No rebleeding occurred since the operation. Pancreatic pseudocysts must be considered as a source of upper gastrointestinal bleeding in patients with bleeding of “obscure” origin. 99mTc-pertechnate-labelled erythrocyte scanning and ERCP may be helpful, even when angiography is normal. Endoscopic retrograde cholangio-pancreatography (ERCP); Erythrocyte scanning; Duodenopancreatectomy; Upper gastrointestinal bleeding; Pancreatic pseudocyst

Keywords:

1. Introduction Gastrointestinal bleeding is a common reason for patients to be admitted to hospital. In about 95% of patients the cause is identified by routine investigations. The remaining patients with “obscure” gastrointestinal bleeding are frequently difficult to manage and often undergo extensive investigation and even laparotomy without diagnosis [ 1,2].

* Corresponding

author.

The medical history may give an important clue to the source of bleeding. The use of radionuclide scanning with 99mTc-pertechnatelabelled erythrocytes is rarely reported, but may also offer a lead [3].

2. Case report A 67-year-old man was repeatedly admitted to hospital because of recurrent gastrointestinal blood loss. He was known to have hypertension, which was treated with atenolol. He did not smoke

0300-2977/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDI 0300-2977(94)00056-F

A.L.

Zanen

et al. /Netherlands

Fig. 1. ‘%Tc-pertechnate erythrocyte scan with a hot spot in the left upper abdominal region.

and used alcohol moderately (2 units consumed daily). In July 1991 he had abdominal pain for several weeks. Laboratory results showed raised concentrations in serum of pancreas-amylase (maximum 102 U/l, normal < 3.5 U/l) and lipase (maximum 551 U/l, normal < 180 U/l). Bilirubin, alkaline phosphatase, y-GT, ASAT, ALAT and LDH were within normal limits. Ultrasound and CT scan of the abdomen showed two pseudocysts in the pancreas and a gallbladder with sludge and a small concrement. The bile ducts were not dilated. The diameters of the two pseudocysts were both 1.5 cm. We diagnosed pancreatitis caused by gallstone disease. The course of the illness was uncomplicated and the patient recovered. A laparoscopic cholecystectomy was performed 2 months later. In March 1992 he passed black stools mixed with dark red blood. The Hb fell from 8.4 to 5.6

Journal

of Medicine

46 (1995)

90-94

91

mmol/l. We made a presumptive diagnosis of acute upper gastrointestinal bleeding. Gastroduodenoscopy, performed within 24 h of admission, did not show blood in the stomach or duodenum and was otherwise unremarkable. Colonoscopy showed diverticulosis, but no source of bleeding was found. A barium follow-through of the small bowel was normal and without a Meckel’s diverticulum. A CT-scan of the pancreas showed persistent pseudocysts which had not changed in size. The patient received blood transfusions and ferrofumarate and was discharged. Two months later he was re-admitted after another episode of melaena with an Hb of 4.8 mmol/l. Gastroduodenoscopy was normal. A radio-isotope scan with 99”Tc-pertechnate-labelled erythrocytes identified a hot spot in the left upper abdominal quadrant (Fig. 1). A selective angiography of the coeliac axis and the superior and inferior mesenterial arteries, however, showed no abnormalities. The patient received blood transfusions and 300 mg ranitidine daily. He was discharged without a definite aetiological diagnosis. In September 1992 he again suffered several episodes of melaena. The Hb was 4.2 mmol/l. Gastroduodenoscopy was again normal, as was a selective angiography of the visceral arteries. A radio-isotope scan with 99mTc-pertechnatelabelled erythrocytes again showed a hot spot in the left upper abdominal quadrant. The patient then told us that meanwhile he could predict the time of bleeding because it was invariably preceded by a sharp pain in the left upper abdomen. The history of pancreatitis, the left upper abdominal pain preceding episodes of bleeding and the left upper quadrant hot spot on 99mTc-pertechnate erythrocyte scanning pointed to the pancreas as the source of bleeding. An ERCP was performed. Blood was seen streaming out of the papilla of Vater. The choledochal duct was normal. The pancreatic duct showed irregularities and changes in calibre. There was a defect in the contrast suggestive of a blood clot and a small pseudocyst connected to the pancreatic duct (Fig. 2).

The patient underwent a duodenopancreatectomy in which the pylorus was preserved. Microscopic examination showed fibrosis and chronic

92

A.L.

Zanen

et ul. /Netherlands

.lournal

of Medicine

46 il99Si

90-W

inflammation of the pancreas with blood in a pseudocyst and the pancreatic duct. There was no evidence of malignancy. After the operation the patient developed diabetes mellitus and steatorrhoea. He was treated successfully with insulin and pancreatic enzyme supplements and has not suffered any further bleeding since the operation now 1 year ago.

3. Discussion Haemorrhage associated with pancreatic pseudocysts is a rare cause of upper gastrointestinal bleeding. It comprises less than 1% of all patients admitted for upper gastrointestinal bleeding and occurs more often in men than in women [4-71. In a recent review of 180 patients referred for surgical management of pancreatic pseudocysts 13 patients (7.2%) presented with arterial bleeding, 4 of whom presented with upper gastrointestinal bleeding (2%). Most cysts which presented with bleeding were relatively small (< 6 cm) [6]. Bleeding can occur over any period from days to months after an attack of pancreatitis and can be massive. It is a major cause of death in patients who die from pancreatitis [8,9]. Related to pseudocysts are pseudo-aneurysms which develop in 12% of patients after an attack of acute pancreatitis. The splenic and gastroduodenal arteries are mainly involved [7,10]. Peptic ulceration, which is the most common origin of upper gastrointestinal bleeding, may coexist with pancreatitis and bleeding from a pseudocyst may be confused with a bleeding duodenal ulcer [ll]. The gold standard for diagnosis of arterial haemorrhage associated with pseudocysts is presumed to be visceral angiography, but it can fail when the patient is not actively bleeding during the procedure. Contrast-enhanced dynamic CTscanning is a good non-invasive method. As an elective investigation, ERCP seems to be more helpful [6,7,12]. In a review of 131 patients with “obscure” gastrointestinal bleeding 3 patients (2%) had chronic pancreatitis. Diagnosis of the source of bleeding was made by ERCP in all of them [2]. Bleeding from the papilla of Vater has

Fig. 2a. Schematic presentation of Fig. 2b. a = duodenal bulb; b = blood clot; c = pseudocyst with blood clot; d = pancreatic duct; e = choledochal duct; f = cystic duct.

been termed “hemosuccus pancreaticus” [13], “santorinirrhage” [141 or “wirsungorrhage” [15]. It is characteristically intermittent and repetitive, often accompanied by colicky pain in the epigastric region [6,16]. Our patient had a 6-month history of recurrent severe gastrointestinal bleeding. Nine months before the first period of bleeding he had a mild attack of pancreatitis. Pseudocysts were seen on a CT-scan, but there was no evidence of a pseudoaneurysm on either contrast-enhanced CT-scan or visceral angiography. The pseudocysts were considered too small to warrant surgery. 99mTc-pertechnate-labelled erythrocyte scanning pointed to a source in the left upper abdomen. Combined with the left upper abdominal pain preceding the periods of bleeding and the history of pancreatitis we suspected a relationship between the pseudocysts and the recurrent bleeding. ERCP confirmed the diagnosis: bleeding from a pancreatic pseudocyst.

A. L. Zanen

et al. /Netherlands

Journal

of Medicine

46 (1995)

90-94

93

Fig. 2b. ERCP showing a blood clot, a pseudocyst and the pancreatic duct with an irregular calibre.

Chronic pancreatitis, which is the major cause of pseudocysts, may be difficult to diagnose: in a series of 332 patients diagnosis was delayed for 62 f 4 months after the onset of symptoms [17]. In patients with bleeding of obscure origin pancreatic pseudocyst must be considered as a possible source. 99mTc-pertechnate-labelled erythrocyte scanning and ERCP may be diagnostic in such circumstances, even when visceral angiography is normal.

References [l] Steger AC, Spencer J. Obscure gastrointestinal bleeding. Br Med J 1988;296:3.

[2] Thompson JN, Salem RR, Hemmingway AP et al. Specialist investigation of obscure gastrointestinal bleeding. Gut 1987;28:47-51. [3] Brayko CM, Kozarek RA, Sanowski RA, Chinichian A. Pancreatic pseudocyst with hemorrhage into the gastrointestinal tract through the duct of Santorini. J Clin Gastroenterol 1985;7:318-322. [4] Sondenaa K, Soreide JA. Pancreatic pseudocyst causing spontaneous gastric haemorrhage. Eur J Surg 1992;158: 257-260. [5] Henne-Bruns D, Frijschle G, Grimm H, Kremer B. Acute gastrointestinal bleeding as a complication of pancreatic pseudocysts. Hepato-Gastroenterology 1991;38:75-77. [6] Adams DB, Zellner JL, Anderson MC. Arterial hemorrhage complicating pancreatic pseudocysts: role of angiography. J Surg Res 1993;54:150-156. [7] Grisendi A, Lonardo A, Della Casa G et al. Hemoductal pancreatitis secondary to gastroduodenal artery ruptured pseudoaneurysm: a rare cause of hematemesis. Am J Gastroenterol 1991;86:1654-1657.

94

AL.

Zanen

et al. /Netherlands

[8] Owen TD, Davies DGL. Massive haemorrhage in pancreatitis. Br J Clin Pratt 1991;45:33-34. [9] Bender JS, Levison MA. Massive hemorrhage associated with pancreatic pseudocyst: succesful treatment by pancreaticoduodenectomy. Am Surg 1991;57:653-655. [lo] Hofer BO, Ryan JA Jr, Freeny PC. Surgical significance of vascular changes in chronic pancreatitis. Surg Gynecol Obstet 1987;164:499-505. [ll] Muckart DJJ, Bade P. Pancreatic pseudocyst haemorrhage presenting as bleeding duodenal ulcer. Postgrad Med J 1989;65:748-749. [12] Hall RI, Lavelle MI, Venables CW. Chronic pancreatitis as a cause of gastrointestinal bleeding. Gut 1982;23:250255. [13] Fernandez-Cruz L, Pera M, Vilella A, Llovera JM, Navasa M, Teres J. Hemosuccus pancreaticus from a pseudoaneurysm of the hepatic artery proper in a patient

Journal

of Medicine

[14]

[15] [16]

[17]

46 (1995)

90-W

with a pancreatic pseudocyst. Hepato-Gastroenterology 1992;39:149-151. Vgzquez-Iglesias JL, Durana JA, Yahez J, Rodrigez FI, Garcia-Vallejo L, Arnal F. Santorinirrhage: hemosuccus pancreaticus in pancreas divisum. Am J Gastroenterol 1988;83:876-878. Pissas A, Aubert H, Bouchet Y, Treille C. Rachail M, Faure H. Wirsungorragie par pancreatite chronique. A propos d’un cas. J Chir 1981;118:587-589. van Rooyen W, van Blankenstein M, Eeftinck Schattenkerk M et al. Haemorrhage from the pancreatic duct: a rare form of upper gastrointestinal bleeding. Br J Surg 1984;71:137-140. Lankisch PG, Peiper M, tihr-Happe A, Otto J, Seidensticker F. Stiickman F. Delay in diagnosing chronic pancreatitis. Eur J Gastroenterol Hepatol 1993;5:713-714.