Pancreatic abscess following acute pancreatitis

Without surgical treatment, pancreatic abscess remains a highly lethal complication of acute pancreatitis. Many surgical series have reported mortality rates of 32 to 65 per cent in treated cases. Although pancreatic abscess is a rare condition, it is more common in patients with severe pancreatitis. A retrospective study of 130 patients admitted to our unit with severe acute pancreatitis during the period from 1965 to 1987 revealed 18 cases of pancreatic abscess. All pancreatic abscesses were primary in nature, and no infected pseudocysts were included in the series. Clinical surveillance, repeated laboratory tests, conventional radiology, and especially ultrasonography and CT scan all contributed to the preoperative diagnosis. The applied treatment was surgical debridement of all necrotic tissue and either local or extensive external drainage. In 12 cases this procedure was combined with other surgical interventions. The recorded mortality rate was 16.66 per cent. Factors adversely affecting survival include: 1) severity of precipitating pancreatitis; 2) difficulty in making early and accurate diagnosis of the pancreatic abscess; 3) marked tendency for recurrence of sepsis; and 4) life-threatening associated complications and/or diseases.     

Pancreatic remnant abscess after resection for acute necrotising pancreatitis

Pancreatic resection for acute necrotising pancreatitis was followed by abscess of the remnant in 14 out of 83 cases. Not even extensive pancreatic resection could prevent pancreatic remnant infection. The 14 cases of abscess are reviewed. Seven were fatal. Enterocutaneous fistula, commonly accompanied by sepsis and major bleeding, was identified in five patients, four of whom died.     

Pancreatic abscess.

Successful management of pancreatic abscess necessitates early diagnosis and prompt external surgical drainage. The infection is predominantly gram-negative and polymicrobic. Roentgenographic contrast studies are of particular diagnostic value. Prompt recognition and external drainage are associated most frequently with recovery. Multiple system organ failure is the typical pattern of death and should alert one to the possibility of occult sepsis, secondary to pancreatic abscess.     

Pancreatic necrosis and acute pancreatitis

The surgeon should take pains to section and study himself the operative specimen after excision for acute pancreatitis, in order to understand the true nature of the lesions, which the most attentive and competent pathological examination cannot describe as vividly as direct examination by the operator. Often he will be surprised to find that the lesions, predominant in the capsule, are less profound and less severe than he had thought at first sight. The necrosis, which is sometimes limited to the peripheral and interstitial tissue, sparing the gland itself (its prognosis is less serious and has led many surgeons to perform surgical excision). However the problems encountered postoperatively have given rise to doubts a posteriori as to whether this is legitimate. Reference is made to the decapsulation of the pancreas described by Romanian authors, and a method for future operative diagnosis of glandular necrosis is proposed.     

Pancreatic microcirculation in acute pancreatitis

We present a review of the microvascular morphology of the pancreas and microstructure of the pancreatic lobule, and report our experimental results of the investigation of pancreatic microcirculation following acute pancreatitis. Impairment of pancreatic microcirculation in the early phase of acute pancreatitis may play a key role in the progression of this disease. Possible contributory mechanisms include increased vascular permeability, reduced blood flow, leukocyte-endothelial cell interaction and intravascular thrombus formation. Using an in-vivo microscope system and off-line computer analysis, we achieved direct visualization and quantification of changes in microvascular permeability and leukocyte behavior in pancreas with acute pancreatitis. Bradykinin and oxygen radicals have been demonstrated to be involved in the increase of vascular permeability in the early stage of caerulein pancreatitis. Leukocyte adherence to the vessels in the pancreatic microcirculation is a secondary event following permeability changes in acute pancreatitis. Leukocyte infiltration during exacerbation of acute pancreatitis is mediated by leukocyte-endothelial cell interaction via leukocyte integrin CD11b/18. Received for publication on Jan. 29, 1997; accepted on April 24, 1997     

Pancreatic tissue perfusion in experimental acute pancreatitis.

OBJECTIVE: To investigate pancreatic tissue perfusion and oxygenation in severe and mild experimental acute pancreatitis in pigs. DESIGN: Randomised controlled experiment. SETTING: Animal laboratory, Finland. ANIMALS: 24 domestic pigs weighing 21-27 kg. INTERVENTIONS: 24 pigs were randomised into severe acute pancreatitis, mild acute pancreatitis and control groups (n = 8 in each). The pancreatic duct of eight anaesthetised and mechanically ventilated pigs was cannulated and taurocholic acid was infused into the pancreatic duct to induce severe acute pancreatitis. Eight animals received intraductally infused saline and developed mild acute pancreatitis. Eight pigs had their ducts cannulated alone, and served as controls. MAIN OUTCOME MEASURES: Pancreatic tissue oxygenation, laser Doppler red cell flux, central haemodynamics. RESULTS: Intraductally infused taurocholic acid rapidly induced macroscopically and histologically proven severe necrotising acute pancreatitis. Histological changes characterising mild acute pancreatitis were seen in animals after intraductal saline infusion. Pancreatic tissue oxygen tension decreased in the severe group and increased in the mild group during the six-hour study period. Laser Doppler red cell flux decreased in the severe group. Central haemodynamics, arterial blood gases, and acid base balances were stable throughout the study period in all groups. CONCLUSION: The present model of severe acute pancreatitis significantly impairs pancreatic oxygenation in the early phase. In mild acute pancreatitis, pancreatic oxygenation increases.     

Pancreatic resection for severe acute pancreatitis

Non-operative management of acute necrotizing pancreatitis carries a mortality of up to 80 per cent. Over the last 6 years we have pursued an aggressive policy of intensive supportive therapy followed by pancreatic resection in those patients with this severe form of the disease. We have managed 15 patients in this way, 14 by subtotal pancreatic resection (usually body and tail of the gland) and one by total pancreatectomy; 7 had early overwhelming multi-system failure with a median of 4 positive prognostic factors whilst 8 were operated on later between 3 and 8 weeks (plus one at 32 weeks) and had varying clinical pictures. Eight patients had ischaemia of the transverse colon which was noted at operation in four, and presented postoperatively in the remainder. Re-operation was necessary in 13 patients to remove further slough or resect ischaemic bowel. Five patients (33 per cent) died between 10 days and 4 weeks postoperatively, death being due to sepsis and multi-system failure in four and a massive retroperitoneal haemorrhage in one. Of the ten survivors, four require insulin. Timely excision of necrotic pancreatic tissue combined with intensive supportive therapy may help reduce the high mortality in this condition.     

Pancreatic microcirculation in acute pancreatitis of dogs

Pancreatic microcirculation in acute pancreatitis and the effect of dopamine and pancreatic protease inhibitor were investigated in 35 mongrel dogs. Acute pancreatitis was induced by the injection of autologous bile added trypsin into pancreatic duct. In acute pancreatitis dogs femoral artery pressure and pulse pressure gradually decreased and pancreatic microflow in basal state temporarily increased immediately after bile injection, however, thereafter continuously decreased during the experiments. Portal flow severely decreased just after onset of acute pancreatitis. By administration of dopamine femoral artery pressure was maintained during the first 90 minutes of experiments, however, thereafter decreased until the end of experiments. Pancreatic microflow, 56.1 +/- 15.3 ml/min/100g in basal level was shown 66.1 +/- 13.7 and 60.3 +/- 10.3 ml/min/100g at 1 and 2 hours, respectively, after bile injection, which were significantly high values as compared with those of non dopamine administration. However those values decreased at 5 hours of both experiments. Portal flow whose basal level was 237 +/- 67 ml/min was maintained during the first 1 hour however it decreased to 139 +/- 25 ml/min at 5 hours. By administration of pancreatic protease inhibitor femoral artery pressure and pulse pressure, temporarily decreased immediately after bile injection, however, they were maintained thereafter. Pancreatic microflow, 57.1 +/- 18.3 ml/min/100g in basal level, was maintained during the first 2 hours, however significantly decreased to 27.6 +/- 9.7 ml/min/100g at 5 hours. Portal flow significantly increased to 442 +/- 115 ml/min at 2 hours, however, thereafter decreased 219 +/- 93 ml/min at 5 hours.(ABSTRACT TRUNCATED AT 250 WORDS)     

Colonic complications of acute pancreatitis and pancreatic abscess

Colonic involvement should be suspected in patients with severe acute pancreatitis, especially in the following clinical settings: plain abdominal radiographs suggesting bowel ischemia, colonic obstruction, acute lower gastrointestinal hemorrhage, gram-negative septicemia, enteric bacteria on Gram stain or culture of peritoneal fluid, and feculent abdominal drainage from a previously drained pancreatic abscess. Intraoperatively, the pancreas should be widely drained and the fecal stream diverted. Colonic hemorrhage and nonviable bowel require immediate resection. Broad-spectrum antibiotic administration and vigorous nutritional support also are required in these critically ill patients. Although proximal diversion and pancreatic diversion alone may suffice, colonic resection may be required later for persistent obstruction or fistulization, but in a more elective setting. Colonic anastomoses should be performed only when pancreatic inflammation and associated sepsis have resolved completely.     

Computed tomography and the prediction of pancreatic abscess in acute pancreatitis.

Pancreatic abscess has become the most common cause of death from acute pancreatitis. Since computed tomography (CT) permits noninvasive imaging of the peripancreatic anatomy, the relationship of early CT findings to late pancreatic sepsis has been evaluated in 83 patients with acute pancreatitis. Pancreatic abscesses developed in 18 patients and were responsible for five of the six deaths in this study. Initial CT findings were graded: A = normal, in 12 patients; B = pancreatic enlargement alone, in 19; C = inflammation confined to pancreas and peripancreatic fat, in 17; D = one peripancreatic fluid collection, in 12; and E = two or more fluid collections, in 23. The incidence of pancreatic abscess in grades A and B was 0%; in grade C, 11.8%; in grade D, 16.7; and in grade E, 60.9%. The severity of pancreatitis was also graded by previously reported prognostic signs as "mild" (0-2 signs) in 56 patients, "moderate" (3-5 signs) in 22, and "severe" (greater than or equal to 6 signs) in five patients. The incidence of abscesses in mild disease was 12.5%; in moderate, 31.8%; and in severe, 80%. Fluid collections on CT resolved spontaneously in 19 of 35 (54.3%) patients. Abscess developed in two patients with no fluid collections on initial CT study. No abscess occurred in 31 patients with CT grades A or B, and in one of 22 patients (4.5%) with CT grade C or D and less than three positive prognostic signs. Among 30 patients with CT grade E or CT grade C or D and three or more positive prognostic signs, 17 (56.7%) developed abscesses. All deaths were in patients with five or more positive prognostic signs. Early imaging of the pancreas by CT identifies a group of patients with increased risk of pancreatic abscess. Identification of this group is improved further by use of early objective prognostic signs.     

Comprehensive management of acute necrotizing pancreatitis and pancreatic abscess

Achieving reduced mortality rates in patients with necrotizing pancreatitis and pancreatic abscess is possible by employing a comprehensive management plan. Components of the plan include (1) rapid evaluation and assessment of the degree of physiologic and anatomic derangement, the latter by the prompt use of vascular enhanced computed tomographic scan; (2) adequate fluid resuscitation determined by early institution of advanced hemodynamic monitoring; (3) attempts to identify and document septic foci via computed tomography-guided percutaneous aspiration; and (4) aggressive surgical debridement. Close adherence to these policies allowed us to keep mortality in this seriously ill group of patients to 14%. Most deaths occurred in patients who were referred to this service late in the course of their disease. The Acute Physiology and Chronic Health Enquiry (APACHE) II severity of illness index applied at the time of admission proved an accurate predictor of mortality. A score of 25 or greater was highly predictive of death, and a lesser score, of survival.     

Pancreatic pseudocyst as a complication of acute pancreatitis]

New hypothesis of the pancreatic pseudocyst (PPC) pathogenesis, intending detachment of nourishing "segment" of the organ as the source of the cystic cavity filling, is presented. An experiment on forming of internal anastomosis with cystic cavity and pancreatic duct was done. The model of suppurated PPC was elaborated, to treatment of which the ointments levosin and ophlotrimol, possessing noncorrect necrolytic activity, were applied. In experimental investigations the perspective of conduction of the PPC (already formated and "unripe" as well) curative decompression was shown and efficacy of the ointment compositions for the treatment of suppurated PPC too.     

Pancreatic duct pressure, duct permeability and acute pancreatitis

The relationship between pancreatic duct pressure, duct permeability to macromolecules and the development of acute pancreatitis was studied in a cat model. Perfusion of the pancreatic duct with 15 mM glycodeoxycholic acid, ethanol administration, or secretagogue-stimulated pancreatic secretion against greater than 50 per cent duct obstruction resulted in an increase in peak pancreatic duct pressure in all animals. Duct permeability to 20,000 molecular weight dextran molecules was increased in 22 of 29 experimental animals compared with two of 22 control animals (P less than 0.01). Perfusion of the pancreatic duct with activated pancreatic enzymes resulted in acute pancreatitis in 24 of 29 experimental animals compared with three of 22 control animals (P less than 0.01). These results suggest that pancreatic ductal hypertension, resulting in increased ductal permeability to large molecules, may be a common early event in gallstone and alcoholic pancreatitis.     

Pancreatic endocrine function in patients with acute pancreatitis]

The secretory function of the pancreas is also impaired in acute destructive pancreatitis. A four- to five-fold increase of serum insulin and glucagon concentrations during the development of the disease is evidence in favor of the development of pancreonecrosis. Diabetic, type disorders of glucose tolerance were encountered in 38% of patients with acute pancreatitis, the clinical form of diabetes mellitus was found in 8.3% of the examined patients.     

Splenic pseudoaneurysms following acute pancreatitis]

Splenic artery pseudoaneurysms are the most common of visceral artery pseudoaneurysms. Splenic pseudoaneurysms appear to have developed as a consequence of inflammatory processes adjacent to the splenic artery, particularly acute pancreatitis and chronic pancreatitis with associated pseudocysts. They are often asymptomatic and picked up on abdominal examination for ultrasound or CT scanning for other conditions. Complications include rupture with retroperitoneal hemorrhage or intraperitoneal hemorrhage. Two cases of splenic pseudoaneurysms, following acute pancreatitis, are reported between the years 1987 and 1996.