CardioVascular and Interventional Radiology

© Springer Science⫹Business Media, Inc., 2004 Published Online: 12 August 2004

Cardiovasc Intervent Radiol (2004) 27:567–580 DOI: 10.1007/s00270-004-0037-1

REVIEW ARTICLES

Imaging and Percutaneous Management of Acute Complicated Pancreatitis Sridhar Shankar,1,2,4 Eric vanSonnenberg,1,2 Stuart G. Silverman,1 Kemal Tuncali,1 Peter A. Banks3 1

Department Department 3 Department 4 Department 2

of of of of

Radiology, Brigham and Women’s Hospital, Boston, MA 02115, USA Radiology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115, USA Radiology, University of Massachusetts Memorial Medical Center, Worcester, MA 01655, USA

Abstract Acute pancreatitis varies from a mild, self-limited disease to one with significant morbidity and mortality in its most severe forms. While clinical criteria abound, imaging has become indispensable to diagnose the extent of the disease and its complications, as well as to guide and monitor therapy. Percutaneous interventional techniques offer options that can be life-saving, surgery-sparing or important adjuncts to operation. Close cooperation and communication between the surgeon, gastroenterologist and interventional radiologist enhance the likelihood of successful patient care. Key words: Acute pancreatitis—Pancreatic abscess—Pseudocyst—Endoscopic ultrasound—Endoscopic retrograde cholangiopancreatography—CT—Interventional Radiology Acute pancreatitis refers to inflammation of the pancreas with variable secondary involvement of adjacent tissues and/or remote organ systems [1]. Pancreatitis is characterized by upper abdominal pain, associated with elevated serum amylase and lipase levels [2, 3]. The incidence of acute pancreatitis is between 17 and 28 per 100,000 population. The pathologic severity of acute pancreatitis ranges from mild edema to extensive necrosis. The edematous form of the disease occurs in about 80 – 85% of patients and is self-limited; recovery typically occurs within a few days. In the 15–20% of patients with the more severe form of pancreatitis, hospitalization may be prolonged and associated with infection and other complications, including multiple organ failure. Although the causes of acute pancreatitis are varied, they all produce an inflammatory reaction that results Correspondence to: Sridhar Shankar, M.D., Department of Radiology, University of Massachusetts Medical Health Center, 55 Lake Avenue North, Worcester, MA 01655, USA

in glandular edema and a spectrum of extensive local and systemic effects that are mediated by cytokines [4, 5]. The underlying pathology of most causes of acute pancreatitis (i.e., alcohol, gallstones and drugs) is initial injury to peripheral acinar cells followed by fat necrosis and autodigestion. Peripheral cells, relatively more distant from the arterial supply of pancreatic lobules, undergo parenchymal damage first, possibly secondary to microcirculation abnormalities. Infectious agents, on the other hand, cause direct toxic acinar cell necrosis associated with an acute inflammatory infiltrate. The major features of necrotizing pancreatitis are macroscopic focal or diffuse necrosis of the pancreatic parenchyma, large areas of fat necrosis, and occasionally hemorrhage on the surface of the pancreas and in peripancreatic tissue. Pancreatic necrosis is present mostly in the periphery of the lobules, but may progress to involve most of the gland. Interstitial pancreatitis is characterized by interstitial edema associated with inflammatory cells within the parenchyma. Although parenchymal necrosis may occur, it is microscopic. Small foci of fat necrosis characteristically punctuate the surface of the gland [6, 7]. The management of acute pancreatitis continues to be challenging; most patients with mild pancreatitis have self-limited disease, and require only simple supportive measures. Conversely, severe pancreatitis remains a deadly disease, with an overall mortality of up to 10 –15% [3]. In this paper, we will review the spectrum of local complications of acute pancreatitis, their definitions and interventional radiology management techniques.

Etiology The commonest causes of pancreatitis in the United States are gallstones and alcohol ingestion. Pancreatitis is idiopathic in up to 20% of all cases, although many of these are

568

S. Shankar et al.: Management of Acute Complicated Pancreatitis

Table 1. Etiology of acute pancreatitis Gallstones Biliary sludge and microlithiasis Other causes of mechanical ampullary obstruction Alcohol Hypertriglyceridemia Hypercalcemia Drugs Infections and toxins Trauma including biopsy Pancreas divisum Vascular disease Pregnancy Post-ERCP Postoperative pancreatitis Hereditary pancreatitis Structural abnormalities of the duodenum/ampullary region, bile duct, sphincter of Oddi dysfunction, main pancreatic duct

now thought to be caused by biliary sludge or gallstones too small to be visualized on imaging [8, 9]. One hypothesis, implicating “obstruction-regurgitation” as the underlying principle of pancreatitis was mooted by Opie, in “Opie’s common channel theory.” Opie indicated that impaction of gallstones in the distal common bile duct would create a common biliopancreatic ductal segment proximal to the obstruction, allowing bile to reflux into the pancreatic duct [10, 11]. Based on Opie’s “obstruction theory” of 1901 and subsequent experimental data, it is now widely accepted that passage of gallstones into or through the terminal biliopancreatic ductal system triggers acute (necrotizing) pancreatitis by causing pancreatic ductal obstruction [12, 13]. This socalled big duct theory is, however, not widely accepted as a cause of chronic pancreatitis [14]. Less common etiologies of acute pancreatitis (Table 1) include hyperlipidemia [15], endoscopic retrograde cholangiopancreatography (which can be severe) [16 –18], trauma, viral infection (including HIV) [19, 20], hyperparathyroidism [21], surgery [22], autoimmune diseases such as systemic lupus erythematosus, genetic mutations such as cystic fibrosis transmembrane conductance regulator (CFTR) [23] and radiation. Drugs can cause pancreatitis, either by a hypersensitivity reaction or by the generation of a toxic metabolite such as occurs with certain antibiotics and chemotherapeutic agents [3]. Rare causes of pancreatitis include a hereditary type [24], ischemia [3], cardiac pump placement [25] and carcinoma of the pancreas [26]. The roles of pancreas divisum (congenital failure of the dorsal and ventral pancreatic ducts to fuse) and sphincter of Oddi dysfunction in acute pancreatitis are controversial [3, 27]. Percutaneous biopsy may result in acute pancreatitis in 4 –5% of cases. It appears to be more frequent when a “normal” pancreas that has a suspicion of carcinoma is biopsied [28]. Current improved CT and MRI scanning techniques likely obviate some biopsies, and this latter complication should occur less frequently in the future.

Complications Complications of acute pancreatitis may be systemic, local or both. Systemic complications run the gamut from mild to life-threatening. The latter include shock, renal failure and adult respiratory distress syndrome (ARDS). Local manifestations include acute fluid collections, pseudocysts, necrosis and abscess; the last three are potentially lethal. Acute pancreatitis is graded by clinical or CT severity scores that predict the development of major complications. The Apache II score [29], Ranson criteria [30 –32] and Balthazar CT [33, 34] grading are utilized most frequently. The Ranson criteria utilize laboratory measures to predict eventual clinical severity; the system is fairly old, having been developed in 1974, and consists of 11 signs, five of which are determined at admission and six that are determined after 48 hours of observation. The first five signs indicate to some extent the intensity of the local inflammatory process. The six signs that are measured after the initial 48 hours reflect third-space losses and systemic complications. A limitation of this method is the requirement for 48 hours of observation. In actuality, the grading system is criticized because it predicts reasonably well the low grades (⬍3) and very high grades (⬎6), but intermediate grades (3–5) are not as well differentiated. The Apache II (Acute Physiology and Chronic Health Evaluation) assesses 12 physiologic variables, age and chronic health status. These measures are documented usually at admission and then again after 48 hours. The Balthazar system uses imaging parameters that include inflammation of the pancreas, and number and distribution of fluid collections along with assessment of the amount of pancreatitic necrosis (0, ⬍33%, 33–⬍50%, ⬎50%) to generate a CT score to assess severity and prognosis [33, 34].

Clinical Management Current guidelines for surgical management of pancreatitis were codified in the meeting of the International Association of Pancreatology [35]. There were 11 guidelines, 10 of which were recommendations grade B and one (the second) grade A: 1. Mild acute pancreatitis is not an indication for pancreatic surgery. 2. Prophylactic broad-spectrum antibiotics reduce infection rates in CT-proven necrotizing pancreatitis but may not improve survival. The antibiotics that the guidelines suggest are imipenem/cilastin and fluconazole. 3. Fine-needle aspiration for bacteriology should be performed to differentiate between sterile and infected pancreatic necrosis in patients with sepsis syndrome. 4. Infected pancreatic necrosis in patients with clinical signs and symptoms of sepsis is an indication for intervention including surgery and/or radiologic drainage.

S. Shankar et al.: Management of Acute Complicated Pancreatitis

5. Patients with sterile pancreatic necrosis (with negative fine-needle aspiration for bacteriology) should be managed conservatively and undergo intervention only in selected cases. 6. Early surgery within 14 days after onset of the disease is not recommended in patients with necrotizing pancreatitis unless there are specific indications. 7. Surgical and other forms of interventional management should favor an organ-preserving approach, which involves debridement or necrosectomy combined with a postoperative management concept that maximizes postoperative evacuation of retroperitoneal debris and exudate. 8. Cholecystectomy should be performed to avoid recurrence of gallstone-associated acute pancreatitis. 9. In mild gallstone-associated acute pancreatitis, cholecystectomy should be performed as soon as the patient has recovered and ideally during the same hospital admission. 10. In severe gallstone-associated acute pancreatitis, cholecystectomy should be delayed until there is sufficient resolution of the inflammatory response and clinical recovery. 11. Endoscopic sphincterotomy is an alternative to cholecystectomy in those who are not fit to undergo surgery in order to lower the risk of recurrence of gallstone-associated acute pancreatitis. There is, however, a theoretical risk of introducing infection into sterile pancreatic necrosis. The authors concluded that these guidelines should now form the basis for audit studies in order to determine the quality of patient care delivery [35]. Current medical treatment of acute pancreatitis is mostly supportive and includes fluid resuscitation, pulmonary care and supervision in an intensive care unit. Major aims are to reduce morbidity and mortality by limiting systemic complications, preventing pancreatic necrosis and infection, treating pancreatic inflammation and correcting any underlying predisposing factors.

Imaging Several imaging techniques permit detection and evaluation of acute pancreatitis and its complications. These include ultrasound (US), CT, endoscopic retrograde cholangiopancreatography (ERCP), endoscopic ultrasound (EUS) and magnetic resonance imaging (MRI). While each technique has its advantages and disadvantages, CT is the predominant imaging modality to diagnose and evaluate pancreatitis.

Ultrasound (Including Endoscopic Ultrasound and Intra-Pancreatic Ultrasound) US may be used during the first 24 – 48 hr of hospitalization to determine whether the cause of pancreatitis is gallstones,

569

and to assess whether or not the bile ducts are dilated. While US reliably demonstrates gallstones, it is not as accurate for stones of the common bile duct; ERCP is more accurate in the latter situation. US otherwise has not had a major role in the diagnosis and evaluation of patients with pancreatitis because of the limitations presented by abdominal distension, overlying bowel gas, body habitus and catheters that obscure the pancreas. A large proportion of patients with acute pancreatitis have some degree of ileus, further compounding the problem. Other technical difficulties, such as presence of monitoring lines, bandages or anterior abdominal wounds in postoperative patients, relative immobility, and inability to hold the breath, even for short periods of time, make US examinations challenging and inadequate in some patients. Even when these difficulties have been surmounted, US cannot distinguish between viable and nonviable pancreatic parenchyma. A normal examination does not exclude pancreatitis, nor does abnormal echogenicity, which is highly variable, have any predictive value. EUS was developed to overcome the limitations of transabdominal US that are caused by intervening bone and air-filled bowel loops. Due to the proximity of the transducer to the structures being evaluated, higher-frequency transducers can be used, thereby improving the spatial resolution. EUS is useful to identify occult pancreatic tumors in patients with “idiopathic recurrent pancreatitis” [36]. In patients with acute pancreatitis of suspected biliary origin, EUS may identify biliary sludge and microlithiasis [37]. EUS is as accurate as ERCP and magnetic resonance cholangiopancreatography (MRCP) for the detection of common bile duct stones [38, 39]. A preliminary study performed to evaluate the prognostic value of EUS in acute pancreatitis concluded that EUS might reliably distinguish edematous and necrotizing pancreatitis. The EUS scores also correlated significantly with the Balthazar CT scoring system [40]. One study has shown that echo-enhanced power-Doppler EUS reliably differentiates pancreatic neoplasm and focal pancreatitis. The authors concluded that the diagnostic accuracy of this technique is comparable to CT, MRI, positron emission tomography and ERCP [41]. While not currently utilized clinically, intravenous SHU508A (Levovist, Schering, Romania) as a contrast agent has been found to significantly enhance color Doppler signals during EUS in a swine model [42]. Vascular contrast of this sort could potentially have a significant role in improving the accuracy of EUS in evaluating pancreatic necrosis and vascular complications. Intraductal pancreatic ultrasound (IPUS), a new technique for evaluation of the pancreatic duct, utilizes a rotating radial 20 MHz ultrasound transducer that can be passed through the biopsy channel of an ERCP endoscope and into the pancreatic duct. IPUS has been used to differentiate chronic pancreatitis from pancreatic carcinoma, and for more precise evaluation of pancreatic duct abnormalities depicted by ERCP [43]. IPUS is similar to intraductal US in the bile ducts for the detection of stones, tumors and strictures [44]. One report showed IPUS to be helpful to differentiate pan-

570

S. Shankar et al.: Management of Acute Complicated Pancreatitis

Table 2. CT grading system of Balthazar and Ranson

Table 3. Computed tomography severity index (0 –10)

Grade

Description

CT

A B C

Normal-appearing pancreas Focal or diffuse enlargement of the pancreas Pancreatic gland abnormalities accompanied by mild peripancreatic inflammatory changes Fluid collection in a single location, usually within the anterior pararenal space Two or more fluid collections near the pancreas or gas either within the pancreas or within peripancreatic inflammation

CT grade

Score

Amount of

Score

A B C D E

0 1 2 3 4

None ⬍33% 33–50% ⬎50%

0 2 4 6

D E

creatic duct strictures caused by pancreatic carcinoma from focal pancreatitis, and for delineation of intraductal papillary pancreatic tumors [45]. It should be emphasized that EUS and IPUS are not widely available; also, their role in acute pancreatitis is limited. In most instances, patients with an initial attack of acute pancreatitis would not be candidates for either of these tests. One indication is suspected underlying pancreatic tumor.

Computed Tomography CT is the preferred technique to diagnose and often to treat patients with acute pancreatitis [46]. There are three main indications for obtaining a CT scan in acute pancreatitis. First, if other serious intra-abdominal conditions such as mesenteric infarction or a perforated ulcer need to be excluded, CT scan is helpful to establish the correct diagnosis. Second, CT is of value to stage the severity of acute pancreatitis. Severity can be assessed by the extent of spread of pancreatic inflammation beyond the confines of the pancreas, coupled with the presence of pancreatic parenchymal necrosis. Third, CT scans delineate the presence of complications of pancreatitis, such as pancreatic necrosis, pseudocyst, abscess and ascites. The currently used CT grading system (Table 2) [33, 34] for assessing severity and predicting the outcome of acute pancreatitis grades pancreatitis from A to E, the latter being the most severe. Grade A is a radiologically normal-appearing pancreas, grade B indicates focal or diffuse enlargement of the pancreas while grades D and E indicate a high probability of evolution into abscess. The CT scoring system predicts the outcome by including the amount of pancreatic necrosis in the equation (Table 3) [47]. Infected and noninfected necrosis as well as pancreatic abscess carry a substantial mortality rate. The mortality with infected necrosis is 30 –35%, compared with 10 –15% in sterile necrosis [48]. The mortality from pancreatic abscess (from 10% to 25%) [49, 50] is less than that from infected necrosis. Early diagnosis of these conditions can dramatically alter therapy; for example, a diagnosis of infected necrosis may warrant surgery, or an aggressive attempt at transcatheter therapy to stabilize or definitively treat such a patient [50 –56]. In addition, intrapancreatic gas, albeit a rare

Necrosis

CT grade (0 – 4) ⫹ Necrosis (0 – 6) ⫽ Total score

finding, can be identified by CT, and often indicates infection. In the absence of gas detected on CT, percutaneous needle aspiration is used for confirmation of infection [54]. Spiral and multidetector (MDCT) CT are the preferred modalities to diagnose and evaluate acute pancreatitis [57– 59]. MDCT is a significant technologic breakthrough that modifies the way volumetric imaging data are collected and reconstructed. Apart from increased speed of acquisition, MDCT provides narrower collimation, scans larger volumes and increases image resolution. Its major strengths lie in angiographic applications and three-dimensional reconstructions. With current technology, scans from the lower chest to the iliac crest can be obtained with a 30 – 40 sec breath-hold (approximately 20 –25 sec with MDCT). Oral contrast (barium or gastrografin) is administered to ensure that unopacified bowel is not misinterpreted as a pancreatic fluid collection. Some authors prefer water to orally administered contrast [60]. Volumetric acquisition in spiral CT or MDCT allows postprocessing of images, including the rendering of three-dimensional models and multiplanar reconstruction. For MDCT, intravenous contrast (60% non-ionic iodinated contrast) is administered via a pressure injector in a total volume of 100 –120 ml at a constant rate of approximately 2 ml/sec, with a scan delay of 50 sec. Images are reconstructed at 5 mm intervals. For spiral CT (non-MDCT), a similar protocol is used, with the injection of 100 –120 mL of contrast, 3–5 mm slices and an approximately 40 sec scan delay. An arterial phase may be included to evaluate vascular complications such as arterial erosion or pseudoaneurysm [61]. The primary purpose of administering intravenous contrast during a CT scan to evaluate pancreatitis is to distinguish the interstitial from the necrotizing variant. When pancreatic necrosis occurs, often there is some degree of peripancreatic necrosis as well. Dynamic, contrast-enhanced CT depicts the distinction between necrotizing and interstitial pancreatitis by demonstrating non-enhancing, non-viable areas of the pancreas with necrosis. Subsequent to the administration of intravenous contrast, normal pancreatic parenchyma (40 –50 Hounsfield units [HU] on unenhanced CT) enhances to at least 80 –90 HU. If the Hounsfield units are less than 80, pancreatic necrosis should be strongly suspected, and is confirmed if the density does not exceed 50 HU [62].

S. Shankar et al.: Management of Acute Complicated Pancreatitis

Some concern exists regarding the use of intravenous contrast in acute pancreatitis, as shown in animal models and retrospective studies [63, 64]. However, multiple studies have not shown a significant difference in the clinical course of acute pancreatitis following administration of intravenous contrast medium [65, 66].

Magnetic Resonance Imaging The normal pancreas appears slightly brighter than the liver on T1-weighted images, and isointense or slightly darker on T2-weighted images. Following the administration of gadolinium chelates, normal pancreatic tissue enhances earlier than the liver and spleen. The arterial phase of imaging (approximately 10 sec after onset of mechanical power injection of 20 mL gadolinium at 2 mL/sec) is most useful to differentiate between viable and non-viable pancreatic tissue, and also to evaluate possible peripancreatic arterial complications. Due to increased contrast resolution on MR images, debris within fluid collections may be better identified than on CT [67, 68]. MRI to evaluate the pancreas is advantageous in patients who cannot undergo CT scanning due to contrast hypersensitivity or renal failure. However, MRI has not yet been established to be superior to CT scanning as a modality for baseline evaluation and diagnosis of acute pancreatitis. The examination time with MRI is longer, and the usual problems in patients with pacemakers, those with multiple support lines and the critically ill are disadvantages [43, 67, 68]. MRI without intravenous contrast may be used for evaluation of the pancreatic and biliary ductal systems (magnetic resonance cholangio-pancreatography; MRCP). Heavily T2weighted images are obtained in a relatively short period of time, and provide accurate information on the presence of gallstones within the biliary system, as well as the condition of the pancreatic duct [69]. MRCP is highly sensitive and specific for diagnosing choledocholithiasis and can eliminate the need for a diagnostic ERCP [43, 70 –74]. MRP (magnetic resonance pancreatography) is also useful in the depiction of the pancreatic ductal morphology in patients with suspected strictures, stones, tumors and pancreas divisum. Localized pancreatitis within the ventral segment has been demonstrated on MRI [75]. Annular pancreas also can be diagnosed by MRI [76].

Endoscopic Retrograde Cholangiopancreatography Endoscopic retrograde cholangiopancreatography (ERCP) is particularly useful in patients with acute pancreatitis in whom bile duct stones are suspected. There is evidence to suggest a statistically significant reduction in major complications of severe pancreatitis (12% versus 61%) by early ERCP coupled with sphincterotomy and stone removal (within 24 –72 hr) in patients with severe acute pancreatitis complicated by biliary sepsis [77, 78]. ERCP can also dem-

571

onstrate pancreatic duct abnormalities such as pancreas divisum and pseudocysts with ductal communication. Pancreatic duct stenting to traverse strictured or injured ducts may be beneficial in select cases [79 – 81]. In recent years, ERCP has lost much of its diagnostic importance to the noninvasive techniques such as MRCP and US. Together, these tests can often determine whether gallstones and/or biliary tract calculi are present. ERCP is used if endoscopic sphincterotomy or pancreatic stenting is contemplated.

Definitions and Interventions The terms to describe acute pancreatitis and its complications that are in common usage were defined at an International Symposium held in Atlanta in 1992 [1]. The terminology and definitions used in that Symposium are the guidelines for imaging and intervention today. The interventional radiologist must work closely with the surgeon and endoscopist to plan treatment strategies in patients with severe complicated pancreatitis, as many patients are gravely ill, and need a coordinated, labor-intensive series of procedures. While no widely accepted algorithm exists, a consensus of indications for interventional diagnosis and treatment has emerged [80, 82– 85]. The practice or algorithm at a particular center depends ultimately on the level of expertise brought to the clinical problem at hand by the three different services that can offer therapy for such complex patients: surgery, interventional radiology and gastroenterology. Also, the status of the patient influences which therapies may be best applied for that clinical situation. While many percutaneous drainage procedures are curative, some are performed to stabilize the critically ill patient prior to a definitive surgical procedure [33, 56, 86]. The converse is also true, i.e., that radiologic drainage often follows surgical therapy that has been only partially successful [56, 80, 82– 86]. The following briefly defines and describes the various complications of acute pancreatitis and their respective radiologic interventions.

Acute Fluid Collection Acute fluid collections occur early in the course of acute pancreatitis. They are located in or near the pancreas, and lack a wall of granulation or fibrous tissue. Fluid collections may be intrapancreatic and appear as one or more small low-attenuation areas. Acute fluid collections occur in 30 – 50% of cases of acute pancreatitis [87]; the majority resolve spontaneously. On CT, acute fluid collections usually appear as a poorly marginated collections of fluid of low attenuation that insinuate into pancreatic and peripancreatic areas; they have no recognizable capsule. Intervention is usually unnecessary, beyond occasional needle aspiration (Fig. 1) to document sterility or infection; the latter of course requires drainage. Generally, a 20 –22 G needle is used; sampling different

572

S. Shankar et al.: Management of Acute Complicated Pancreatitis

Fig. 1. Importance of diagnostic aspiration: a 55-year-old man with acute pancreatitis with multiple fluid collections. Diagnostic aspiration yielded cloudy material with white blood cells and bacteria. These findings necessitated percutaneous drainage, and catheter drainage was initiated. The patient recovered well.

Fig. 2A, B. Percutaneous drainage of pancreatic pseudocyst: an 88-year-old woman. A Large pancreatic pseudocyst following acute traumatic pancreatitis (seat belt injury 10 weeks previously). B The pseudocyst was drained successfully with a 7 Fr catheter. A transhepatic route was chosen, and this can be safely performed. Fig. 3. Importance of injecting the catheter (abscessogram/ sinogram): a 60-year-old man with a continually draining collection following acute pancreatitis. Injection of the catheter demonstrated communication to the colon (arrows).

parts of the collection may increase the yield, and the fluid is sent for gram staining, culture and sensitivity (aerobes, anaerobes and fungus). The complication rate is very low. If these collections persist for 4 – 6 weeks and become walled off, they then have evolved into pseudocysts, or abscesses if infected [1].

Pseudocyst A pseudocyst is defined as a collection of pancreatic juice enclosed by a wall of fibrous or granulation tissue. Pseudocysts may occur as a result of acute pancreatitis, chronic pancreatitis or pancreatic trauma. They usually contain a

high concentration of pancreatic enzymes and variable amounts of tissue debris. Most pseudocysts are sterile; those that are infected are included as abscesses, according to the nomenclature from the International Symposium [1]. Asymptomatic pseudocysts smaller than 5 cm are likely to resolve spontaneously and should be monitored by follow-up CT (or US) scans; larger pseudocysts, those that demonstrate interval growth or those associated with pain or obstruction can either be drained percutaneously or surgically decompressed [87–93]. Endoscopic drainage, with or without US assistance, is another option [94 –96]. Pseudocysts that are greater than 4 –5 cm in diameter and those that are present for more than 2 months rarely resolve

S. Shankar et al.: Management of Acute Complicated Pancreatitis

spontaneously; consequently, the threshold for percutaneous drainage in these patients is lower. Usual indications for percutaneous drainage of pseudocysts include infection, fever, pain, increasing size on sequential imaging, biliary or gastrointestinal obstruction, and chronicity of symptoms [87–93]. The preferred guidance modality for draining pseudocysts is CT, although US may be used if visualization is adequate. The usual technique is to access the collection via the simplest and most direct route, avoiding traversal of vital structures or organs, similar to conventional drainage of intraabdominal abscesses. However, sometimes a direct route may not be available, and a transgastric or other enteric (e.g., jejunum, ileum) or transhepatic route may be selected [97, 98]. Transgastric insertion of a double-J catheter with one end in the pseudocyst and the other in the stomach may be used by endoscopic or laparoscopic techniques [99 –102]; this method involves additional input by an endoscopist [94 –96]. A percutaneous approach for percutaneous cystogastrectomy under fluoroscopic guidance has been described, using double mushroom stents; removal of the stent requires endoscopy [103]. Needle aspiration alone for therapy, even if performed multiple times, often is not adequate for cure because of recurrence of the collection [83, 90]. Drainage catheters are more effective (Fig. 2). Catheterization continues until drainage ceases (defined as ⬍10 mL/day for 2 consecutive days) [84, 88 –93]. Repeat CT imaging to document absence of a residual collection may be performed prior to catheter removal. If no remaining collections are seen, and no communication to the pancreatic ductal system of gastrointestinal tract exists (Fig. 3), cure rates are usually in the 90% range [84, 88 –93]. An equal success rate with surgical and percutaneous drainage of pancreatic pseudocysts has been documented [91]. Clinical infection as a complication of sterile pseudocyst drainage occurs in about 5% of patients [88 –93], although subclinical colonization is more common. Percutaneous drainage including the direct approach to the collection, or a percutaneous trangsgastric approach, can easily be performed with CT or US guidance; one complication not seen with percutaneous transgastric drainage is the formation of a persistently draining pancreatico-cutaneous fistula because it mimics the surgical approach [104]. Octreotide, along with percutaneous drainage of pseudocysts, is indicated for prolonged drainage, high-output drainage, pseudocyst recurrence or pancreatic fistula [105]. Octreotide exerts pharmacologic actions similar to the natural hormone, somatostatin. It is an even more potent inhibitor of growth hormone, glucagon and insulin than is somatostatin. Like somatostatin, it also suppresses the LH response to GnRH, decreases splanchnic blood flow, and inhibits release of serotonin, gastrin, vasoactive intestinal peptide, secretin, motilin and pancreatic polypeptide. By virtue of these pharmacologic actions, octreotide decreases pancreatic secretions. Octreotide acetate (Sandostatin, Biochemie, Schaftenau, Austria) is administered subcutaneously in doses

573

of 50 –1,000 ␮g three times a day. Results suggest that this combined approach is effective in decreasing catheter output in percutaneously drained pseudocysts [105]. Pancreatic duct obstruction precludes the effectiveness of octreotide. A small subset of patients with pancreatic duct fistulas may not respond to prolonged catheter drainage. These patients ultimately undergo surgical resection of the distal pancreatic segment, or endoscopic stenting [79 – 81]. Percutaneous embolization of the distal pancreatic duct with cyanoacrylate tissue adhesive has been described in one patient with good long-term results [106].

Pancreatic Necrosis Pancreatic necrosis is defined as focal or diffuse areas of nonviable or dead pancreatic parenchyma and is often associated with peripancreatic fat necrosis. Lack of contrast enhancement at CT of a portion of the pancreas, or of the entire pancreas, is indicative of necrosis [43, 46, 57]. Another approach is to visually compare the pancreatic and splenic densities, which usually are similar. The liver should not be used for this comparison because of the frequent association of diffuse fatty infiltration in the liver in alcoholic patients that reduces overall hepatic density. The percentage of necrosis is a predictor of the degree of the patient’s pancreatitis [46, 57, 107], according to the CT severity index (Table 2) [57]. Drainage of sterile necrosis is controversial. The main rationale for not using percutaneous catheter drainage is potentially converting a sterile collection into an infected one. However, as with drainage of pancreatic pseudocysts, bacterial colonization is common; clinical infection less so [88 –93]. Percutaneous drainage may be used, similar to abscess drainage. While sterile necrosis usually is clinically less severe in its manifestations than infected necrosis or pancreatic abscess, nonetheless it can cause major morbidity and mortality. Selected patients with sterile necrosis can be improved using percutaneous drainage [108, 109]. Data on the treatment of sterile necrosis are sketchy; most studies report recovery with conservative treatment using intravenous fluids, hyperalimentation, antibiotics (controversial), pain control medication and nutritional support [50, 80, 91, 107, 110 –114]. Some patients who do not improve undergo surgical debridement [111, 112]. Debridement or necrosectomy may be complicated, partially successful or unsuccessful, and repeat operations are common [111, 112, 115, 116]. Liquefied necrosis that yields a brownish fluid that contains debris, is amenable to percutaneous image-guided percutaneous drainage [84, 85, 115, 117, 118]. There is a theoretical risk of colonization or infection when draining sterile necrosis, i.e., of converting a non-infected collection into an infected collection. However, with careful catheter care and judicious use of antimicrobial agents, secondary infection usually is avoided [84, 85, 115]. Percutaneous catheter drainage may be particularly useful to decrease morbidity by avoiding surgery, or to temporize prior to

574

S. Shankar et al.: Management of Acute Complicated Pancreatitis

Fig. 4A–F. Sterile necrosis successfully drained with multiple catheters: a 77-year-old man with acute pancreatitis, extensive necrosis and multiple collections. He had a history of recurrent acute pancreatitis and was admitted to the emergency room severely ill with abdominal pain, fever of 40°C and hypotension. A Coronal reformatted contrast-enhanced CT scan demonstrates multiple loculi of fluid. Almost the entire pancreas is non-enhancing and necrosed. B Needle

aspiration did not reveal any organisms. C–E Nonetheless, the patient’s fever and hypotension failed to improve; he was admitted in the intensive care unit and multiple catheters were inserted. F Follow-up CT scan 4 weeks later demonstrates complete resolution of the collections; the three catheters are seen in place (arrows). The patient’s clinical condition also improved remarkably, and he was discharged from hospital following removal of the catheters.

operation [56, 119]. Percutaneous drainage alone can cure some patients with liquefied sterile necrosis (Fig. 4). Infected necrosis of the pancreas is a spreading infection of devitalized tissue in the retroperitoneum that occurs

within the first 2–3 weeks of illness. Multisystem organ failure is a potential cause of death in these early stages of the disease [120]. Conversely, pancreatic abscess after acute pancreatitis does not usually occur until after the first month

S. Shankar et al.: Management of Acute Complicated Pancreatitis

of illness. Pancreatic infection can be documented by percutaneous guided aspiration with analysis by Gram stain and culture. Occasionally, pancreatic infection can be predicted by the presence of retroperitoneal or lesser sac gas on CT scan. Infected pancreatic necrosis is believed to arise from translocation of intestinal organisms to the pancreas. Frequently (75%), the infecting organism is E. coli, Klebsiella, or other gram-negative rods; 20% are Staphylococcus and Streptococcus [121]. More recent series emphasize fungal infections such as Candida [122]. When infection is documented, antibiotics are prescribed if they are not already in use. Mortality rates of infected necrosis vary from 15% to 60% [48 –50], and infected necrosis is considered the most severe complication of pancreatitis. Traditionally, infected necrosis is an indication for surgical debridement or necrosectomy [111, 112]. The rationale against the use of percutaneous drainage has been that it is ineffective due to large chunks of necrotic material that block the catheter. A trend towards minimally invasive therapy with aggressive percutaneous maneuvers heralds further involvement by the interventional radiologist. First, diagnostic aspiration (often multiple sites) documents or excludes infection. Second, complete success may be achieved in some cases of liquefied infected (or non-infected) necrosis by catheter drainage alone. Finally, catheter drainage may temporize preoperatively; it does not preclude operation, rather optimizes surgical timing [56]. Early studies have shown promise in selected groups of patients using percutaneous basketing (or other percutaneous devices such as snares or forceps) and mechanical removal of solid debris to treat necrosis. Results have generally varied between 40% and 60% in terms of response in patients treated in this manner [89, 114, 117, 123, 124]. One study [117] reported a 100% success rate in 20 patients, using large-bore catheters with enlarged side holes and the use of suction catheters, stone baskets, and copious amounts of lavage fluid, performed over multiple sessions. Several important issues are essential when draining infected necrosis percutaneously: 1. Multiple catheters are often needed for successful drainage. 2. Large-bore catheters (20 –30 Fr) are frequently necessary, as the collections may contain solid components that will not drain through narrow-lumen catheters. 3. Revision of catheter size and repositioning of catheters is often required (Figs. 5, 6). 4. Catheter care: Daily irrigation with saline until the returning fluid is clear is a crucial factor. The irrigation is performed three or more times a day. The technique includes first aspirating all the fluid that can be withdrawn, followed by gentle instillation of 20 mL aliquots of saline, usually multiple times. The fluid is withdrawn after each 20 mL infusion and discarded; this is repeated until the returning fluid is reasonably clear. Finally 5–10 mL of saline is instilled to ensure patency of the catheter.

575

The amount of saline used should be carefully documented and subtracted from the daily drainage charts. This catheter care must continue over weekends. Close cooperation with the admitting team is necessary, and education and participation of the nursing staff are important. 5. Frequent CT follow-up is utilized to check for residual and/or undrained pockets of fluid if the patient is not responding well. If there are undrained pockets of fluid on follow-up imaging, new additional catheters are placed. The location of catheter placement should be strategized such that the largest number of side holes is in the most dependent part of the collection. Non-draining catheters are removed if they do not drain more than 10 mL/day for two consecutive days, and follow-up imaging demonstrates resolution of the collection. 6. A dissociation between the clinical status of the patient and the CT scan is common [125], and must be communicated to all those caring for the patient. It is advisable that the interventional radiologists who performed the procedure interpret the CT scans (if in their purview and expertise), as extenuating factors such as irrigation and cavity contents may affect the appearance of the images. Increase of draining fluid volume must be interpreted with caution, because it may represent the residual saline irrigant in the cavity and not necessarily worsening of disease [125]. If there are two catheters, through-andthrough flow of irrigant may occur from one catheter to another. While the collections may look minimally improved or even worse on the CT scan, the clinical picture can be remarkably better than the imaging suggests [125].

Pancreatic Abscess Pancreatic abscess is a circumscribed collection of pus, typically in proximity to the pancreas, containing little or no pancreatic necrosis, that arises as a consequence of acute pancreatitis or pancreatic trauma [126 –129]. Pancreatic abscesses include infected pseudocysts, late infected acute fluid collections or postoperative collections. The CT appearance of pancreatic abscesses is variable. An abscess usually appears as a relatively discrete low-density area that may or may not contain gas, and is found in the pancreatic bed or in the peripancreatic tissues [130]. Debate and controversy still persist regarding the timing, the type of intervention (percutaneous versus surgical), and methods to treat pancreatic abscesses [115, 131]. With the steady advance in techniques, use of multiple large-bore catheters, sump systems and appropriate follow-up, the percutaneous approach has assumed a primary role [132, 133]. Percutaneous drainage of pancreatic abscesses frequently requires more than one catheter. Catheters ranging from 12 to 30 Fr are used routinely. Published results indicate a success rate of 65–90% in percutaneous drainage of pancreatic abscesses [86, 118, 128]. The preferred approach is to

Fig. 5A, B. Catheter upsizing to optimize drainage: a 62year-old man with multiple fluid collections following acute pancreatitis. The collections were drained with multiple percutaneous catheters. A Persistent collection in spite of good catheter positioning. B The catheter was upsized from 9 Fr to 14 Fr; subsequent recovery was rapid. Fig. 6A–C. Catheter repositioning: a 70-year-old man with persistent fluid collection (combination of acute fluid collection and sterile necrosis) near the head of the pancreas, despite 6 days of catheter drainage. A The collection is persistent and the catheter is too anterior. B A new, larger catheter (14 Fr) was inserted more deeply, and resulted in rapid drainage of the remainder of the collection. C Follow-up CT image 6 weeks after removal of catheters demonstrates marked improvement, and the patient was well clinically.

Fig. 7A,B. Percutaneous pancreatic abscess drainage: a 57year-old man with acute pancreatitis following partial resection for pancreatic cancer. He presented with fever, abdominal pain and increased white blood cell count to 27.2 ⫻ 103/␮L, 2 weeks following distal pancreatectomy and sple-

nectomy. A CT fluoroscopic image of the diagnostic aspiration needle. Microbiologic examination demonstrated E. coli infection. B CT scan demonstrates the drainage catheter in place. The patient recovered well.

S. Shankar et al.: Management of Acute Complicated Pancreatitis

treat each individual major loculation with its own catheter; post-catheter care is similar to that for necrosis (Fig. 7). Many factors can affect the outcome of percutaneous drainage in pancreatic abscess. The most important among them are: careful patient selection, strict and diligent adherence to guidelines for percutaneous abscess drainage, catheters of adequate size, number and location, careful clinical follow-up with appropriate catheter manipulations, and an integrated, cooperative approach with surgeons and gastroenterologists who are willing to allow catheter drainage to effect its benefits, rather than operating prematurely. Percutaneous abscess drainage has been recommended as the initial therapy for those patients whose collections are culture-positive or with sterile abscesses, while surgical intervention is reserved for those in whom treatment fails.

Ancillary Procedures and Nutrition in Pancreatitis Many different ancillary procedures, not directly related to the pancreas, may be required from time to time in patients with acute pancreatitis, and the interventional radiology service can assist in these. Image-guided pleural effusion tap and drainage, ascites drainage, angiographic embolization of pseudoaneurysms and placement of nasojejunal feeding tubes under fluoroscopic guidance are some of the procedures that are routine parts of the management. Several controlled studies support the use of enteral nutrition, and enteral feeding has fewer complications than parenteral nutrition [134 –136]. No significant differences were found between enteral and parenteral nutrition in terms of mortality, pain scores, days to normalization of the serum amylase or resumption of a normal diet. Disease severity also appears to decrease with enteral nutrition and to increase with parenteral nutrition during the first week of therapy, and acute-phase responses improve with enteral but not parenteral nutrition. Finally, the cost of parenteral nutrition is significantly greater than that of enteral nutrition. Hence, whenever possible, radiologic or endoscopic placement of a jejunal feeding tube should be attempted [134, 135]. Nasogastric and nasojejunal tubes can easily be placed under fluoroscopic guidance. Feeding tubes for enteral nutrition can be inserted through the skin into the stomach (percutaneous gastrostomy) or via the stomach into the small intestine (percutaneous gastrojejunostomy) using radiologic techniques. Imaging modalities utilized usually include fluoroscopy and US. Percutaneous gastrostomy is technically easier to perform than percutaneous gastrojejunostomy and percutaneous gastrojejunostomy is indicated if patients are prone to gastro-oesophageal reflux [137, 138]. A percutaneous jejunostomy can also be performed by puncturing the jejunum under fluoroscopic control. The technical success of these percutaneous procedures is close to 100% in most series [137, 138].

577

Conclusion In conclusion, acute pancreatitis has myriad forms and presentations, and varies from a mild and self-limited disease to more severe forms that have significant morbidity and mortality. Several clinical criteria are available and used to evaluate acute pancreatitis; however, imaging is essential to diagnose the extent and severity of the disease and its complications, as well as to guide and monitor therapy. Several percutaneous interventional techniques offer therapeutic options that can be the primary line of therapy in association with supportive measures. These interventions can be life-saving, surgery-sparing, or important adjuncts to operative therapy. In addition, several ancillary procedures, not directly related to the pancreas, aid in patient management. It cannot be overemphasized that drainage of pancreatic necrosis, abscess and pseudocysts, while technically similar to other intra-abdominal drainages, is not the same as draining, for example, a periappendiceal abscess or diverticular abscess. There is the added burden of daily interventional radiology rounds with flushing and irrigation of catheters, catheter changes, follow-up CT scans, etc., needed to effect a cure in this population of patients. Hence, close cooperation and communication between the surgeon, gastroenterologist and the interventional radiologist treating the patient increase the likelihood of successful patient care. References 1. Bradley EL 3rd (1993) A clinically based classification system for acute pancreatitis. Summary of the International Symposium on Acute Pancreatitis, Atlanta, Ga, September 11–13, 1992. Arch Surg 128: 586 –590 2. Karne S, Gorelick FS (1999) Etiopathogenesis of acute pancreatitis. Surg Clin North Am 79:699 –710 3. Banks PA (1998) Acute and chronic pancreatitis. In: Feldman M, et al. (eds) Sleisenger & Fordtran’s Gastrointestinal and liver disease. WB Saunders, Philadelphia, pp 810 – 862 4. Rinderknecht H (1994) Genetic determinants of mortality in acute necrotizing pancreatitis. Int J Pancreatol 16:11–15 5. Brivet FG, Emilie D, Galanaud P (1999) Pro- and anti-inflammatory cytokines during acute severe pancreatitis: An early and sustained response, although unpredictable of death. Parisian Study Group on Acute Pancreatitis. Crit Care Med 27:749 –755 ¨ ppel G, Maillet B (1993) Pathology of acute and chronic pancre6. KlO atitis. Pancreas 8:659 – 670 7. DiMagno EP, Chari S (2002) Acute Pancreatitis. In: Feldman M, et al. (eds) Sleisenger & Fordtran’s gastrointestinal and liver disease. WB Saunders, Philadelphia, pp 857–941 8. Lee SP, Nicholls JF, Park HZ (1992) Biliary sludge as a cause of acute pancreatitis. N Engl J Med 326:589 –593 9. Ros E, et al. (1991) Occult microlithiasis in ⬙idiopathic⬙ acute pancreatitis: Prevention of relapses by cholecystectomy or ursodeoxycholic acid therapy. Gastroenterology 101:1701–1709 10. Opie EL (1970) The theory of retrojection of bile into the pancreas. Rev Surg 27:1–7 11. Opie EL (1901) The eiology of acute hemorrhagic pancreatitis. Bull Johns Hopkins Hosp 12:182 12. Sakorafas GH, Tsiotou AG (2000) Etiology and pathogenesis of acute pancreatitis: Current concepts. J Clin Gastroenterol 30:342–356 13. Runzi M, Layer P (1997) Etiology, pathogenesis and pathophysiology of acute pancreatitis. Schweiz Med Wochenschr 127:849 – 853 14. Pitchumoni CS (2001) Pathogenesis of alcohol-induced chronic pancreatitis: Facts, perceptions, and misperceptions. Surg Clin North Am 81:379 –390

578

S. Shankar et al.: Management of Acute Complicated Pancreatitis

15. Fortson MR, Freedman SN, Webster PD 3rd (1995) Clinical assessment of hyperlipidemic pancreatitis. Am J Gastroenterol 90:2134 – 2139 16. Freeman ML, et al. (2001) Risk factors for post-ERCP pancreatitis: A prospective, multicenter study. Gastrointest Endosc 54:425– 434 17. Freeman ML (2002) Post-ERCP pancreatitis: Patient and techniquerelated risk factors. Jop 3:169 –176 18. Sherman S, Lehman GA (1991) ERCP- and endoscopic sphincterotomy-induced pancreatitis. Pancreas 6:350 –367 19. Cappell MS, Hassan T (1993) Pancreatic disease in AIDS: A review. J Clin Gastroenterol 17:254 –263 20. Dassopoulos T, Ehrenpreis ED (1999) Acute pancreatitis in human immunodeficiency virus-infected patients: A review. Am J Med 107: 78 – 84 21. Mithofer K, et al. (1995) Acute hypercalcemia causes acute pancreatitis and ectopic trypsinogen activation in the rat. Gastroenterology 109:239 –246 22. Camargo CA Jr, et al. (1995) Acute pancreatitis following liver transplantation. J Am Coll Surg 181:249 –256 23. Durno C, et al. (2002) Genotype and phenotype correlations in patients with cystic fibrosis and pancreatitis. Gastroenterology 123: 1857–1864 24. Shah S, Amarapurkar D, Pitchumoni CS (1994) Hereditary pancreatitis. Am J Gastroenterol 89:928 –930 25. Herline AJ, et al. (1999) Acute pancreatitis after cardiac transplantation and other cardiac procedures: Case-control analysis in 24,631 patients. Am Surg 65:819 – 825, discussion 826 26. Mujica VR, Barkin JS, Go VL (2000) Acute pancreatitis secondary to pancreatic carcinoma. Study Group Participants. Pancreas 21:329 – 332 27. Morgan DE, et al. (1999) Pancreas divisum: Implications for diagnostic and therapeutic pancreatography. AJR Am J Roentgenol 173:193– 198 28. Mueller PR, et al. (1988) Severe acute pancreatitis after percutaneous biopsy of the pancreas. AJR Am J Roentgenol 151:493– 494 29. Wilson C, Heath DI, Imrie CW (1990) Prediction of outcome in acute pancreatitis: A comparative study of APACHE II, clinical assessment and multiple factor scoring systems. Br J Surg 77:1260 –1264 30. Demmy TL, et al. (1988) Comparison of multiple-parameter prognostic systems in acute pancreatitis. Am J Surg 156:492– 496 31. Banks PA (1991) Predictors of severity in acute pancreatitis. Pancreas 6[Suppl 1]:S7–S12 32. McKay CJ, Imrie CW (1999) Staging of acute pancreatitis. Is it important. Surg Clin North Am 79:733–743 33. Balthazar EJ, Freeny PC, vanSonnenberg E (1994) Imaging and intervention in acute pancreatitis. Radiology 193:297–306 34. Balthazar EJ (1989) CT diagnosis and staging of acute pancreatitis. Radiol Clin North Am 27:19 –37 35. Uhl W, et al. (2002) IAP guidelines for the surgical management of acute pancreatitis. Pancreatology 2:565–573 36. Catalano MF, Alcocer E, Geenen JE (1998) Diagnostic evaluation of patients with idiopathic recurrent pancreatitis (IRP): Role of endoscopic ultrasonography (EUS). Gastrointest Endosc 37. Chak A, et al. (1999) Prospective assessment of the utility of EUS in the evaluation of gallstone pancreatitis. Gastrointest Endosc 49:599 – 604 38. de Ledinghen V, et al. (1999) Diagnosis of choledocholithiasis: EUS or magnetic resonance cholangiography? A prospective controlled study. Gastrointest Endosc 49:26 –31 39. Frakes JT (1999) Biliary pancreatitis: A review. Emphasizing appropriate endoscopic intervention. J Clin Gastroenterol 28:97–109 40. Schoefer M, Rathgeber A, Lang J (1998) Prognostic value of endoscopic ultrasound in acute pancreatitis. American Gastroenterological Association, Bethesda, MD 41. Becker D, et al. (2001) Echo-enhanced color- and power-Doppler EUS for the discrimination between focal pancreatitis and pancreatic carcinoma. Gastrointest Endosc 52:784 –789 42. Bhutani MS, et al. (1997) Contrast-enhanced endoscopic ultrasonography with galactose microparticles: SHU508 A (Levovist). Endoscopy 29:653– 659 43. Freeny PC (1999) Pancreatic imaging. New modalities. Gastroenterol Clin North Am 28:723–746 44. vanSonnenberg E, et al. (1992) Percutaneous intraluminal US in the gallbladder and bile ducts. Radiology 182:693– 696

45. Ariyama J, et al. (1998) Endoscopic ultrasound and intraductal ultrasound in the diagnosis of small pancreatic tumors. Abdom Imaging 23:380 –386 46. Freeny PC (1993) Incremental dynamic bolus computed tomography of acute pancreatitis. Int J Pancreatol 13:147–158 47. Balthazar EJ, et al. (1990) Acute pancreatitis: Value of CT in establishing prognosis. Radiology 174:331–336 48. Vesentini S, et al. (1993) Prospective comparison of C-reactive protein level, Ranson score and contrast-enhanced computed tomography in the prediction of septic complications of acute pancreatitis. Br J Surg 80:755–757 49. Fedorak IJ, et al. (1992) Secondary pancreatic infections: Are they distinct clinical entities. Surgery 112:824 – 830; discussion 830 – 831 50. Beger HG, Isenmann R (1999) Surgical management of necrotizing pancreatitis. Surg Clin North Am 79:783– 800, ix 51. Bassi C (1994) Infected pancreatic necrosis. Int J Pancreatol 16:1–10 52. Dubner H, et al. (1996) Infected pancreatic necrosis and peripancreatic fluid collections: Serendipitous response to antibiotics and medical therapy in three patients. Pancreas 12:298 –302 53. Dervenis C, et al. (1999) Diagnosis, objective assessment of severity, and management of acute pancreatitis. Santorini consensus conference. Int J Pancreatol 25:195–210 54. Gerzof SG, et al. (1987) Early diagnosis of pancreatic infection by computed tomography-guided aspiration. Gastroenterology 93:1315– 1320 55. Sainio V, et al. (1995) Early antibiotic treatment in acute necrotising pancreatitis. Lancet 346:663– 667 56. vanSonnenberg E, et al. (1984) Temporizing effect of percutaneous drainage of complicated abscesses in critically ill patients. AJR Am J Roentgenol 142:821– 826 57. Balthazar EJ (2002) Acute pancreatitis: Assessment of severity with clinical and CT evaluation. Radiology 223:603– 613 58. Elmas N (2001) The role of diagnostic radiology in pancreatitis. Eur J Radiol 38:120 –132 59. Fernandez-del Castillo C, Rattner DW, Warshaw AL (1993) Acute pancreatitis. Lancet 342:475– 479 60. Horton KM, Fishman EK (1998) Helical CT of the stomach: Evaluation with water as an oral contrast agent. AJR Am J Roentgenol 171:1373–1376 61. Flati G, et al. (2003) Potentially fatal bleeding in acute pancreatitis: Pathophysiology, prevention, and treatment. Pancreas 26:8 –14 62. Hollett M, Jorgensen M, Jeffrey R Jr. (1995) Quantitative evaluation of pancreatic enhancement during dual-phase helical CT. Radiology 195:359 –361 63. Schmidt J, et al. (1995) Intravenous contrast medium aggravates the impairment of pancreatic microcirculation in necrotizing pancreatitis in the rat. Ann Surg 221:257–264 64. McMenamin DA, Gates LKJ (1996) A retrospective analysis of the effect of contrast-enhanced CT on the outcome of acute pancreatitis. Am J Gastroenterol 91:1384 –1387 65. Uhl W, et al. (2002) Influence of contrast-enhanced computed tomography on course and outcome in patients with acute pancreatitis. Pancreas 24:191–197 66. Hwang TL, Chang KY, Ho YP (2000) Contrast-enhanced dynamic computed tomography does not aggravate the clinical severity of patients with severe acute pancreatitis: Reevaluation of the effect of intravenous contrast medium on the severity of acute pancreatitis. Arch Surg 135:287–290 67. Megibow AJ, Lavelle MT, Rofsky NM (2001) MR imaging of the pancreas. Surg Clin North Am 81:307–320, ix–x 68. Sica GT, et al. (2002) Magnetic resonance imaging in patients with pancreatitis: Evaluation of signal intensity and enhancement changes. J Magn Reson Imaging 15:275–284 69. Sica GT, et al. (1999) Comparison of endoscopic retrograde cholangiopancreatography with MR cholangiopancreatography in patients with pancreatitis. Radiology 210:605– 610 70. Fulcher AS, et al. (1998) Half-Fourier RARE MR cholangiopancreatography: Experience in 300 subjects. Radiology 207:21–32 71. Fulcher AS, Turner MA (1999) Magnetic resonance pancreatography (MRP). Crit Rev Diagn Imaging 40:285–322 72. Fulcher AS, Turner MA (1999) MR pancreatography: A useful tool for evaluating pancreatic disorders. Radiographics 19:5–24; discussion 41– 44; quiz 148 –149

S. Shankar et al.: Management of Acute Complicated Pancreatitis

73. Fulcher AS, Turner MA (1998) Pitfalls of MR cholangiopancreatography (MRCP). Comput Assist Tomogr 22:845– 850 74. Fulcher AS, Turner MA, Zfass AM (1998) Magnetic resonance cholangiopancreatography: A new technique for evaluating the biliary tract and pancreatic duct. Gastroenterologist 6:82– 87 75. Takeda T, et al. (1999) Ventral pancreatitis defined on MRI. J Gastroenterol 34:138 –140 76. Reinhart RD, et al. (1994) MR imaging of annular pancreas. Abdom Imaging 19:301–303 77. Mergener K, Baillie J (1999) Endoscopic treatment for acute biliary pancreatitis. When and in whom. Gastroenterol Clin North Am 28: 601– 613, ix 78. Neoptolemos JP, et al. (1988) Controlled trial of urgent endoscopic retrograde cholangiopancreatography and endoscopic sphincterotomy versus conservative treatment for acute pancreatitis due to gallstones. Lancet ii:979 –983 79. Kozarek RA (1998) Endoscopic therapy of complete and partial pancreatic duct disruptions. Gastrointest Endosc Clin North Am 8:39 –53 80. Banks PA (1997) Practice guidelines in acute pancreatitis. Am J Gastroenterol 92:377–386 81. Parks RW, et al. (2000) Management of pancreatic pseudocysts. Ann R Coll Surg Engl 82:383–387 82. Lee M, et al. (1992) Acute complicated pancreatitis: Redefining the role of interventional radiology. Radiology 183:171–174 83. Lee MJ, Wittich GR, Mueller PR (1998) Percutaneous intervention in acute pancreatitis. Radiographics 18:711–724; discussion 728 84. vanSonnenberg E, et al. (1985) Complicated pancreatic inflammatory disease: Diagnostic and therapeutic role of interventional radiology. Radiology 155:335–340 85. vanSonnenberg E, et al. (1989) Imaging and interventional radiology for pancreatitis and its complications. Radiol Clin North Am 27:65–72 86. vanSonnenberg E, et al. (2001) Percutaneous abscess drainage: update. World J Surg 25:362–369; discussion 370 –372 87. Siegelman SS, et al. (1980) CT of fluid collections associated with pancreatitis. AJR Am J Roentgenol 134:1121–1132 88. vanSonnenberg E, et al. (1982) Percutaneous drainage of abscesses and fluid collections: Technique, results, and applications. Radiology 142:1–10 89. vanSonnenberg E, Mueller PR, Ferrucci JT Jr. (1984) Percutaneous drainage of 250 abdominal abscesses and fluid collections. I. Results, failures, and complications. Radiology 151:337–341 90. vanSonnenberg E, et al. (1989) Percutaneous drainage of infected and noninfected pancreatic pseudocysts: Experience in 101 cases. Radiology 170:757–761 91. Adams DB, Anderson MC (1992) Changing concepts in the surgical management of pancreatic pseudocysts. Am Surg 58:173–180 92. Adams DB, Harvey TS, Anderson MC (1990) Percutaneous catheter drainage of infected pancreatic and peripancreatic fluid collections. Arch Surg 125:1554 –1557 93. Wittich GR, vanSonnenberg E (1996) When should radiologists intervene in management of pancreatic pseudocysts and other complications of acute pancreatitis. AJR Am J Roentgenol 166:211 94. Hershfield NB (1984) Drainage of a pancreatic pseudocyst at ERCP. Gastrointest Endosc 30:269 –270 95. Lo SK, Rowe A (1997) Endoscopic management of pancreatic pseudocysts. Gastroenterologist 5:10 –25 96. Vidyarthi G, Steinberg SE (2001) Endoscopic management of pancreatic pseudocysts. Surg Clin North Am 81:405– 410, xii 97. Wittich GR, et al. (1995) Anatomic access to pancreatic fluid collections. Semin Interv Radiol 12:191–198 98. Wittich GR, et al. (1987) Computed tomography of peripancreatic fat planes. Eur J Radiol 7:253–256 99. Bernardino ME, Amerson JR (1984) Percutaneous gastrocystostomy: A new approach to pancreatic pseudocyst drainage. AJR Am J Roentgenol 143:1096 –1097 100. Hancke S, Henriksen FW (1985) Percutaneous pancreatic cystogastrostomy guided by ultrasound scanning and gastroscopy. Br J Surg 72:916 –917 101. Matzinger FR, et al. (1988) Pancreatic pseudocysts drained through a percutaneous transgastric approach: Further experience. Radiology 167:431– 434 102. Grosso M, et al. (1989) Percutaneous treatment (including pseudocystogastrostomy) of 74 pancreatic pseudocysts. Radiology 173:493– 497 103. Davies RP, et al. (1996) Percutaneous cystogastrostomy with a new

104. 105. 106. 107. 108. 109. 110. 111. 112.

113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132.

579

catheter for drainage of pancreatic pseudocysts and fluid collections. Cardiovasc Intervent Radiol 19:128 –131 Sacks D, Robinson ML (1988) Transgastric percutaneous drainage of pancreatic pseudocysts. AJR Am J Roentgenol 151:303–306 D’Agostino HB, et al. (1993) Treatment of pancreatic pseudocysts with percutaneous drainage and octreotide. Work in progress. Radiology 187:685– 688 Findeiss LK, et al. (2003) Percutaneous embolization of the pancreatic duct with cyanoacrylate tissue adhesive in disconnected duct syndrome. J Vasc Interv Radiol 14:107–111 Karimgani I, et al. (1992) Prognostic factors in sterile pancreatic necrosis. Gastroenterology 103:1636 –1640 Uhl W, et al. (2002) IAP guidelines for the surgical management of acute pancreatitis. Pancreatology 2:565–573 Ashley SW, et al. (2001) Necrotizing pancreatitis: Contemporary analysis of 99 consecutive cases. Ann Surg 234:572–579; discussion 579 –580 Baril NB, et al. (2000) Does an infected peripancreatic fluid collection or abscess mandate operation. Ann Surg 231:361–367 Buchler MW, et al. (2000) Acute necrotizing pancreatitis: Treatment strategy according to the status of infection. Ann Surg 232:619 – 626 Buchler P, Reber HA (1999) Surgical approach in patients with acute pancreatitis. Is infected or sterile necrosis an indication—in whom should this be done, when, and why. Gastroenterol Clin North Am 28:661– 671 Bittner R, et al. (1987) Pancreatic abscess and infected pancreatic necrosis. Different local septic complications in acute pancreatitis. Dig Dis Sci 32:1082–1087 Carter CR, McKay CJ, Imrie CW (2000) Percutaneous necrosectomy and sinus tract endoscopy in the management of infected pancreatic necrosis: An initial experience. Ann Surg 232:175–180 Stanten R, Frey CF (1990) Comprehensive management of acute necrotizing pancreatitis and pancreatic abscess. Arch Surg 125:1269 – 1274; discussion 1274 –1275 Uomo G, et al. (1996) Nonsurgical treatment of acute necrotizing pancreatitis. Pancreas 12:142–148 Echenique AM, et al. (1998) Percutaneous catheter-directed debridement of infected pancreatic necrosis: Results in 20 patients. J Vasc Interv Radiol 9:565–571 Freeny PC, et al. (1988) Infected pancreatic fluid collections: Percutaneous catheter drainage. Radiology 167:435– 441 Meadors J, et al. (1997) Nonsurgical management of complicated pancreatic inflammatory disease. Practical Gastroenterol 3:43– 47 Perez A, et al. (2002) Is severity of necrotizing pancreatitis increased in extended necrosis and infected necrosis. Pancreas 25:229 –233 Bassi C (2001) Infections in pancreatic inflammatory disease: Clinical trials for antibiotic prophylaxis. Pancreatology 1:210 –212 De Waele JJ, et al. (2003) Fungal infections in patients with severe acute pancreatitis and the use of prophylactic. Clin Infect Dis 15:208 – 213 Freeny PC, et al. (1998) Percutaneous CT guided drainage of infected acute necrotizing pancreatitis: Techniques and results. AJR Am J Roentgenol 170:969 –975 Horvath KD, et al. (2001) A technique for laparoscopic-assisted perutaneous drainage of infected pancreatic necrosis and pancreatic abscess. Surg Endosc 15:1221–1225 Chon K, et al. (1998) How to read a CT scan post percutaneous drainage of pancreatic collections. Radiology Bradley EL 3rd, Murphy F, Ferguson C (1989) Prediction of pancreatic necrosis by dynamic pancreatography. Ann Surg 210:495–503; discussion 503–504 Katsohis CD, et al. (1989) Pancreatic abscess following acute pancreatitis. Am Surg 55:427– 434 vanSonnenberg E, et al. (1997) Percutaneous radiologic drainage of pancreatic abscesses. AJR Am J Roentgenol 168:979 –984 Malangoni MA, et al. (1986) Factors contributing to fatal outcome after treatment of pancreatic abscess. Ann Surg 203:605– 613 Ranson JH, et al. (1985) Computed tomography and the prediction of pancreatic abscess in acute pancreatitis. Ann Surg 201:656 – 665 Rattner DW, et al. (1992) Early surgical debridement of symptomatic pancreatic necrosis is beneficial irrespective of infection. Am J Surg 163:105–109; discussion 109 –110 vanSonnenberg E, et al. (1991) Percutaneous abscess drainage: Current concepts. Radiology 181:617– 626

580

S. Shankar et al.: Management of Acute Complicated Pancreatitis

133. vanSonnenberg E, et al. (1982) Percutaneous radiographically guided catheter drainage of abdominal abscesses. JAMA 247:190 –192 134. Windsor AC, et al. (1998) Compared with parenteral nutrition, enteral feeding attenuates the acute phase response and improves disease severity in acute pancreatitis. Gut 42:431– 435 135. McClave SA, et al. (1997) Comparison of the safety of early enteral vs. parenteral nutrition in mild acute pancreatitis. J Parenteral Enteral Nutr 21:14 –20

136. Kalfarentzos F, et al. (1997) Enteral nutrition is superior to parenteral nutrition in severe acute pancreatitis—results of a randomized prospective trial. Br J Surg 84:1665–1669 137. vanSonnenberg E, et al. (1986) Percutaneous gastrostomy and gastroenterostomy. 2. Clinical experience. AJR Am J Roentgenol 146:581– 586 138. Saini SJ, et al. (1990) Percutaneous gastrostomy with gastropexy: Experience in 125 patients. AJR Am J Roentgenol 154:1003–1006