[ Next ] [ Prev ] [ Abs ] [ Chi ] [ TOC ] [ Home ]
Chin Med J (Taipei) 1997;60:265-72.
1Department of Radiology; 2Department of Medicine, China Medical College Hospital, Taichung, Taiwan, R.O.C.
Background. Complicated pancreatic inflammatory diseases are often life-threatening, multifocal, and multibacterial disorders. Untreated, these lesions often prove fatal. The purpose of the study is to describe interventional radiologic techniques for diagnosis and treatment of complicated peripancreatic fluid collections.
Methods. Between January 1994 and December 1995, 17 patients with abdominal fluid collection from complicated acute pancreatitis underwent percutaneous drainage following initial diagnosis with computed tomography (CT). These patients required multiple CT examinations, multiple catheter insertions, multiple catheter manipulations, and long-term catheter drainage.
Results. Twenty-five infected fluid collections among these 17 patients were percutaneously drained. Sixteen (94%) patients were successfully treated with catheter drainage alone. Surgical treatment was necessary in one patient because of segmental stricture of transverse colon caused by inflammation and adhesion of transverse mesocolon. Catheter drainage duration averaged 31 days. Patients required an average of three catheter manipulations and four abdominal CT scans. Mean hospital stay was 42 days (range, 11-95 days), and mean ICU stay was 25 days (range, 1-70 days).
Conclusions. The study confirms that complicated peripancreatic fluid collections can be safely and effectively treated in most patients with percutaneous catheter technique.
[Chin Med J (Taipei) 1997;60:265-72.]
Keywords: interventional procedure, pan-creatitis, pancreatic abscess, percutaneous drainage
Received: June 27, 1996.
Accepted: August 22, 1997.
Address reprint requests to: Yung-Fang Chen, Department of Radiology, China Medical College Hospital, No. 2, Yute Road, Taichung, Taiwan, R.O.C.
Diagnosis and evaluation of the complications of acute and chronic pancreatitis are among the most important roles of the radiologist. These complications are responsible for a significant proportion of the morbidity and mortality of inflammatory disease of the pancreas. However, if detected early, they can often be treated appropriately without sequelae [1-5]Sepsis is a major complication of severe pancreatitis and can result from secondary infection of areas of pancreatic parenchymal necrosis, pancreatic phlegmon, or pancreatic fluid collections and pseudocysts [6,7].
Formerly, surgery was the only definitive method of managing these disorders, although it is fraught with problems because of the severity of the underlying disease or because the findings may be more complicated than preoperatively suspected [3,8-11]. Postoperative morbidity ranges from 32% to 92% [3,12], and reoperation is frequent (up to 57% of cases) [1,11,13,14]. However, interventional radiology has now been shown to be a valuable addition for both diagnosis (needle aspiration) and therapy of these disorders [4,5,15,16].
The paper presents the results of percutaneous catheter drainage of 25 infected abdominal fluid collections secondary to acute pancreatitis in 17 patients. This emphasizes the practical usefulness of computed tomography (CT) for patient selection, for accurate assessment of the morphology and location of the fluid collections, for careful planning of the percutaneous access route and for follow-up of the effectiveness of drainage. Treatment with percutaneous drainage often requires multiple large-bore catheters, numerous catheter manipulations and an extensive commitment of time and effort by the radiologist. The safety and efficacy of percutaneous catheter drainage of infected peripancreatic fluid collection are similar to those of drainage of nonpancreatic abscesses. The technical, clinical and catheter-related problems encountered in treatment of these types of patients are also outlined.
Seventeen patients with complicated pancreatic inflammatory disease were referred for percutaneous drainage during the past two years. There were 12 men and 5 women ranging in age from 30 to 73 years (average, 52 years); 365 other patients, treated with either surgery or medical therapy, were identified during the same period. Each of these 17 patients underwent a preliminary ultrasound and CT examination to determine the complication of pancreatitis. Causes of pancreatitis included alcohol in nine, idiopathic in five, and gallstone in three.
Sixty-eight CT scans were obtained from the seventeen patients (average, four per patient, range, two-eight). All the drainages were performed with fluoroscopic guidance using CT landmarks to indicate the percutaneous entry site, angle of needle entry, and depth. Catheter insertion and postprocedure care followed standard principles in all cases [17,18]. Initial entry into the fluid collections was accomplished by means of the Seldinger technique with use of 0.035-inch guide wire placed through an 18-gauge trocar needle. The catheter sizes were in the range of 8.3-28 F, depending on the content of fluid drained. Immediately after catheter placement and aspiration of as much purulent material as possible, the material was sent for Gram staining and culture. Broad-spectrum systemic antibiotics were started before the drainage procedures and were subsequently altered depending on the results of the cultures and antibiotics sensitivities. Follow-up care consisted of daily bedside inspection to assess clinical parameters (fever, white blood cell count), and monitoring the amount of catheter drainage. Subsequential follow-up CT was performed if fever did not defervensce, if a new fever occurred or if abdominal pain or discomfort persisted or developed.
Routine skin care and hygiene were maintained as long as catheters were in place. Care was taken to avoid tape burns and skin maceration from wet or moist bandages. Catheters were left in place until drainage stopped (< 10 mL/24 h). No attempt was made to withdraw the catheters incrementally. After drainage ceased, contrast study (fluoroscopy) with use of a 76% water-soluble iodinated contrast was performed to assess catheter patency and cavity size. If the catheter was patent and the cavity had collapsed, the tube was capped and left in place for 1-2 days. If no pain, fever or fluid accumulation occurred, the catheters would be removed. If a postprocedure CT scan did not show adequate decompression of a cavity or if a second cavity was seen, an additional catheter was placed. Catheter manipulation (i.e., exchange of catheters) would be performed if there were clinical evidence of catheter clogging or if injection of contrast material into the cavity showed that a large cavity was not adequately drained by a single catheter.
Twenty-five fluid collections were identified with CT in 17 patients. The locations of these collections were: pancreas, with extension into pararenal and/or pericolic tissues (n = 20); lesser sac and/or subphrenic areas (n = 3); and suprapubic area (n = 2). The fluid collections ranged from 3 to 13 cm in diameter. The amylase levels of the fluid drained from the collections ranged from 300 to 30000 U/L. All of the collections were grossly purulent and contained organisms at initial Gram staining. The cultures in five patients were negative. In the other twelve patients, cultures grew a variety of bacteria, the most common being Escherichia coli, Klebsiella pneumoniae, Enterococcus faecalis, Staphylococcus, Enterobacter (Table 1).
The use of large-bore catheters was essential to assure adequate drainage and evacuation of the thick fluid, pus and debris in the dependent portion of infected peripancreatic fluid collection (Figure 1). In the several cases in which recurrent catheter obstruction occurred, the catheters were removed and replaced with larger ones. More than one catheter was placed in 15 of 17 patients: 11 patients had 2 catheters, 2 had 3 catheters, and two patients 4. Catheter manipulation (defined as a catheter exchange with guide-wire techniques and fluoroscopic guidance) was performed an average of three times per patient (range, 1-6). The length of catheter indwelling drainage varied from 8 to 90 days, with a mean 31 days. The mean duration of hospital stay was 42 days (range, 11-95 days), and the mean time spent in the Intensive Care Unit was 25 days (range, 1-70 days). During the hospital stay, an average of four abdominal CT scans (range, 2-8) were obtained per patient. There was no complication of fistula formation during and after percutaneous catheter drainage.
Percutaneous catheter drainage resulted in complete resolution of the infected fluid collections in 16 (94%) of the 17 patients (Figure 2). Surgical debridement was performed in one patient because of segmental stricture of transverse colon caused by inflammation and adhesion of the transverse mesocolon (Figure 3). All of the patients underwent follow-up CT or ultrasound from one month to two years (average, 7 months) after catheter removal, with no evidence of recurrence.
Abdominal fluid collections complicated with pancreatitis is frequently multilocular or multiseptated, extensive and poorly defined; they may be associated with fistulae and necrotic tissue [4,15]. Percutaneous drainage of infected peripancreatic fluid collections is one of the formidable challenges facing the interventional radiologist.
Complicated pancreatitis is often life-threatening, associated with 80% to 100% mortality if untreated. Early diagnosis of pancreatic necrosis or infection is now feasible with contrast material-enhanced CT [19-21]and needle aspiration [5]. Treatment of such complicated pancreatitis, however, remains a difficult problem for both surgeons and radiologists. Surgeons appreciate the difficulties in diagnosis and management of complicated pancreatic inflammatory disease. Who should to be operated, when to operate, which operation to do, the likelihood of postoperative recurrent disease, and the inherent morbidity and mortality of both the disease and the operation are major controversies [1,8-11]. Similarly, interventional radiologists should ideally delay percutaneous intervention until collections are largely liquefied to increase the chances of a successful drainage. However, waiting for collections to liquefy is often impractical in critically ill patients with sepsis. Delayed diagnosis has been a factor which has increased mortality for this group of diseases [3,11]. In this institution, an aggressive approach, with percutaneous drainage, has been successful in managing infected abdominal fluid collection secondary to pancreatitis. We believe that early percutaneous drainage of pancreatic pus and extravasated pancreatic juice is necessary and beneficial so that the degree of the damage to multiple organs through digestion by the pancreatic enzyme can be reduced.
Sepsis is one of the most serious complications of pancreatitis. Pancreatic sepsis may be associated with infected parenchymal necrosis or secondary infection of a pancreatic phlegmon, a fluid collection, or pseudocyst. The mortality rate of patients with pancreatic sepsis varies from 100% if adequate or early treatment is not instituted to a range of 0%-66% following various forms of surgical therapy [6]. Beger et al. reported that patients undergoing necrosectomy for infected pancreatic necrosis had postoperative mortality rates ranging from 0% in those with less than 30% gland necrosis to as high as 67% in patients with total or near-total gland necrosis [22].
Both Beger et al. and Warshaw emphasized the need for concomitant debridement of necrotic tissue and for the use of large-bore catheters for drainage of infected fluid and pus [22-24]. There is little doubt that patients with extensive infected necrosis or phlegmon are unlikely to survive without surgical debridement [25,26]. However, according to experience, here, patients with loculated infected fluid collections can be treated successfully with percutaneous drainage with use of large-bore catheters. Thus, the safety and efficacy of treating pancreatic infected fluid collections with catheter techniques depend on the ability to differentiate between patients who need surgical debridement of solid necrotic tissue and those who need only drainage of infected fluid or pus.
Other important factors include precise anatomic definition and localization of the fluid collections, a safe percutaneous access route, use of multiple large-bore catheters and attentive postprocedural management of the catheters. It is believed that these can be accomplished with CT in most patients.
Accurate localization of the collections is critical for both diagnosis and for planning catheter placement. Fluid collection could be located in areas remote from the pancreas, including the pelvis, paracolic gutters and mesentery even the suprapubic region, as indicated by this series. Therefore, it is important to scan the entire abdomen and pelvis. Multiple collections are common, and they may be contiguous to or remote from one another. The plan of safest access route for percutaneous drainage to avoid traversing any normal solid or hollow organs with the catheter can be done according to CT images.
Repeated CT scanning is frequently required to follow-up patients with infected pancreatic fluid collections. These follow-up scans may indicate the need to alter therapy (i.e., use of more or larger catheters, catheter repositioning, or surgery).
The severely compromised condition of many of these patients creates reluctance among some surgeons to operate under these conditions. Morphologically, the thick, multiloculated, poorly defined inflammatory material has been a significant obstacle at surgery [1,3,10,13,14]. Postoperative morbidity is common, and mortality occurs in up to 56% [3,11]. Recurrence has been reported in 19% of patients with initial operation [10]; multiple operations are associated with increased mortality [27]. There is a trend to transfer these patients to the radiologist. Altemeier and Alexander [10] emphasized that early diagnosis and early percutaneous drainage reduced mortality. Sixteen of our patients were cured and significant temporary improvement was achieved, even in one patient who ultimately underwent surgery. Cure is defined as evacuation of infected fluid without surgery. A combined approach of percutaneous drainage and surgery is sometimes essential. Even if percutaneous drainage is not curative, a beneficial temporary effect should be achieved in almost all patients [4,15]. It is emphasize that early percutaneous drainage of pancreatic abscesses and pancreatic juice is necessary and rewarding. Thus, the degree of damage of multiple organs by the digestion of pancreatic enzyme can be reduced.
Common causes of failure of catheter drainage are incomplete evacuation of fluid collection because of unrecognized loculation, undetected and undrained remote collections and subsequent development of unrecognized new collections [28]. The use of pre-drainage CT and CT performed within 24-48 hours after catheter drainage can avoid these problems by initial detection of all fluid collections (peripancreatic and remote), confirmation of complete catheter evacuation of the content, and exclusion of undrained loculations. If evacuation is incomplete, or if undrained loculations are identified, additional catheters can be placed or existing catheters can be manipulated into appropriate position and changed to a larger bore catheter. Use of multiple large-bore catheters is the rule for peripancreatic fluid drainage, since the collections frequently are multiple and multilocular and contain thick necrotic material [4,15]. If the patient is responding to treatment, no additional scans are needed until catheter removal is planned or unless new symptoms or signs (pain, fever, rising white blood cell count) develop. These procedures were performed often in our series.
Catheter drainage time in the previously reported series averaged about 15 days (range, 2 days to 5 months) for infected fluid collections or pseudocysts and 20 days (range, 7-41 days) for pancreatic abscesses [5,15,16,28]. Catheters were capped prior to removal, and the patients were studied with CT within two to seven days to detect any evidence of fluid reaccumulation. This technique has been described previously by Torres et al. and van Sonnenberg et al. [5,15]. This technique is believed to be important, since fluid reaccumulating after catheter removal will likely become infected and thus lead to abscess formation. In previous reports, the incidence of fluid reaccumulation after catheter removal was 9% [4].
Percutaneous management of complicated pancreatitis involves substantial time, effort and cost. Fifteen of 17 patients had more than one catheter placed. Catheter exchange was more common than with other types of abscesses. In addition, follow-up CT examinations (not performed for most abscesses in other locations) were frequent. However, each patient benefited from the percutaneous procedure; previously there had been a mortality of 100% in untreated cases and 50% in those treated surgically [1,3,14]. The operative difficulties are evident from surgical series where hospitalization averaged 30 days for patients who survived (43% of 74 patients) [14] and 74 days in another series [29]. The results of percutaneous management of complicated pancreatitis, while not as good as with other abscesses, compare favorably with surgery: a cure rate of almost 70%, temporary beneficial effects in patients who were not cured, and a duration of catheter drainage of up to 32 days.
In summary, percutaneous catheter drainage was effective and safe. Complete resolution of the infected fluid collections was achieved in 16 (94%) of the 17 patients. Percutaneous drainage seems to be the method of choice in the management of complicated pancreatitis, providing either a temporized effect until surgery can be performed or, occasionally, a cure.
Interventional radiology procedures benefit surgeons by expediting accurate diagnosis, determining the need for operation, facilitating surgery by draining abscesses and improving patient condition and obviating the need for general anesthesia and major surgery in high-risk or postoperative patients.
Copyright: 1997, Chinese Medical Association (Taipei)