The early prediction of mortality in acute pancreatitis: a large population-based study.

BACKGROUND: Identification of patients at risk for mortality early in the course of acute pancreatitis (AP) is an important step in improving outcome. METHODS: Using Classification and Regression Tree (CART) analysis, a clinical scoring system was developed for prediction of in-hospital mortality in AP. The scoring system was derived on data collected from 17,992 cases of AP from 212 hospitals in 2000-2001. The new scoring system was validated on data collected from 18,256 AP cases from 177 hospitals in 2004-2005. The accuracy of the scoring system for prediction of mortality was measured by the area under the receiver operating characteristic curve (AUC). The performance of the new scoring system was further validated by comparing its predictive accuracy with that of Acute Physiology and Chronic Health Examination (APACHE) II. RESULTS: CART analysis identified five variables for prediction of in-hospital mortality. One point is assigned for the presence of each of the following during the first 24 h: blood urea nitrogen (BUN) >25 mg/dl; impaired mental status; systemic inflammatory response syndrome (SIRS); age >60 years; or the presence of a pleural effusion (BISAP). Mortality ranged from >20% in the highest risk group to <1% in the lowest risk group. In the validation cohort, the BISAP AUC was 0.82 (95% CI 0.79 to 0.84) versus APACHE II AUC of 0.83 (95% CI 0.80 to 0.85). CONCLUSIONS: A new mortality-based prognostic scoring system for use in AP has been derived and validated. The BISAP is a simple and accurate method for the early identification of patients at increased risk for in-hospital mortality.

Cytomegalovirus infection of the colon associated with ulcerative colitis.

We report a 12th case in which infection of the colon with cytomegalovirus was demonstrated in association with ulcerative collitis. Rectal biopsy has proven to be reliable in identifying cells with cytomegaloviral inclusion bodies if careful histologic evaluation is performed. Thus far, prognosis has been serious when cytomegalovirus of the colon has been documented in ulcerative colitis. Nine of 12 patients have required a colectomy, and only seven of 12 have survived.

In vivo effects in mice produced by a 3'-branched homologue of alpha-2'-deoxythioguanosine.

Studies with a 3'-branched chain homolog (alpha-3'-BCTGdR) of 2'-deoxythioguanosine (alpha-TGdR) showed that it did not prolong the survival of mice bearing the Mecca lymphosarcoma. Host toxicity was quite profound and resembled that seen with 6-thioguanine (6-TG). Evidence was obtained that this nucleoside derivative was not appreciably converted to 6-TG in the mouse. Mice treated with toxic doses of 6-TG or alpha-3'-BCTGdR were found to have very similar pathological changes. The granulocytes were eliminated from the peripheral blood, bone marrow was acellular, and some more limited damage was seen in the intestinal crypts. Experiments with radiosulfur-labeled drugs demonstrated that alpha-3'-BCTGdR was incorporated into the DNA of mouse bone marrow, predominantly in the chain-terminating position, with the result that shorter chains of DNA accumulated. The new homolog, unlike alpha-TGdR, was phosphorylated in bone marrow as well as in tumor, and incorporated well into the DNA both of bone marrow and of the neoplastic cells. In devising other homologs attention must be given to the specificity of the kinases, i.e., to whether phosphorylation is superior in tumor cells or in the growing normal cells.

Management of pancreatic pain.

Treatment of pancreatic pain remains a very difficult problem. At present, when the main pancreatic duct is dilated, lateral pancreaticojejunostomy remains the treatment of choice for refractory pancreatic pain. When the disease is localized to the head or tail of the pancreas, resection of these segments has also proven to be effective in the majority of patients. It will be important to learn whether medical strategy (such as treatment with oral pancreatic enzymes or somatostatin) and endoscopic techniques (such as insertion of stents or shock-wave lithotripsy), compare favorably with surgical techniques. This evaluation will require randomized prospective trials.

Predictors of severity in acute pancreatitis.

Accurate predictors of severity in acute pancreatitis are sorely needed. At present, Ranson's scores provide useful information, some of which is recorded too late to be of maximal usefulness. APACHE-II scores on the day of admission and thereafter appear to provide important prognostic information that may enable the clinician to optimize patient care. CT scans, particularly those with bolus injection of i.v. contrast, help enormously in distinguishing interstitial from necrotizing pancreatitis. Most serum markers have not proven to be reliable in making this distinction.

Hemoconcentration is an early marker for organ failure and necrotizing pancreatitis.

In a previous retrospective case-control study, hemoconcentration was associated with the development of pancreatic necrosis. The aim of the present study was to determine in a cohort study whether hemoconcentration is a marker for both organ failure and necrotizing pancreatitis. A cohort study was performed on patients admitted with acute pancreatitis from February 1996 to April 1997. Pancreatic necrosis was defined by findings on dynamic contrast-enhanced computed tomography scan or magnetic resonance imaging. Of 128 total patients with acute pancreatitis, 53 underwent computed tomography or magnetic resonance imaging. Eighteen of 53 had necrotizing pancreatitis. Logistic regression identified an admission hematocrit > or = 44% and a failure of admission hematocrit to decrease at 24 hours as the best binary predictors of necrotizing pancreatitis and organ failure. By 24 hours, 17 of 18 patients with necrotizing pancreatitis versus 11 of 35 with interstitial pancreatitis met one or the other criterion for necrosis (p < 0.001). By 24 hours, 13 of 15 with organ failure versus 36 of 104 without organ failure met one or the other criterion (p < 0.001). The negative predictive value by 24 hours was 96% for necrotizing pancreatitis and 97% for organ failure. Hemoconcentration with an admission hematocrit > or = 44% and/or failure of admission hematocrit to decrease at approximately 24 hours was associated with the development of necrotizing pancreatitis and organ failure. Patients who did not experience hemoconcentration were very unlikely to develop pancreatic necrosis or organ failure.

Central cavitary necrosis: differentiation from pancreatic pseudocyst on CT scan.

Central cavitary necrosis of the pancreas has a computed tomography CT appearance of a well-defined sausage-shaped mass with a low-density center and convex margins, usually conforming to the pancreatic contour. Several other entities, including pancreatic pseudocyst, may have a similar appearance. Since the treatment of central cavitary necrosis differs considerably from that of these other entities, it is important to differentiate them. We present CT criteria that help distinguish central cavitary necrosis from pancreatic pseudocyst and from a variety of other intrapancreatic and peripancreatic masses.

Clinical features and natural history of central cavitary necrosis.

Central cavitary necrosis (CCN) is an unusual complication of acute pancreatitis in which the necrosis is confined almost entirely to the pancreatic parenchyma and there is little if any extrapancreatic necrosis. In our experience with 10 patients with CCN, clinical features suggested that the episodes of acute pancreatitis were initially severe, with high Ranson scores (mean, 4.2; range, 1-6), development of systemic complications, computed tomography (CT) grade of D or E by the Balthazar-Ranson scoring system, need for intensive care unit admission in 8 of 10 patients, and mean length of hospitalization of 56 days (range, 28-153 days). However, the incidence of infection was low (20%) and mortality was also low (10%). Factors that help explain a favorable prognosis were low APACHE-II scores at admission and at 48 h, absence of shock, paucity of extrapancreatic necrosis, and rapid resolution of clinical toxicity prior to the diagnosis of CCN by CT scan at a mean of 19.8 days (range, 9-63 days) after the onset of symptoms. Surgical debridement is indicated for complications such as secondary infection and ongoing pain. In the absence of complications, an attempt should be made to treat CCN medically.

Ultrastructure and elemental composition of human pancreatic calculi.

Scanning electron microscopy (SEM) and energy dispersive x-ray fluorescence (EDXRF) have been used in this study to investigate the morphology, nature, and arrangement of major, minor, and trace elements in human pancreatic calculi (PC) obtained from patients with alcoholic and tropical pancreatitis. The present study has confirmed previous observations that PC are rich in CaCO3 (as calcite) and many other elements. Further, we have shown for the first time that the amorphous nidus contained only iron, chromium, and nickel, whereas the outer shell contained calcium and 17 other elements. Based on our studies, we postulate the formation of PC to be taking place in multilayers and multistages. Formation of inner protein nidus in the form of a cobweb is the first stage, then calcite is deposited on this fibrous network as tiny crystals. Because of their high surface area and high surface activity, other metallic ions are incorporated onto it through co-precipitation, adsorption, and/or lattice substitution. This study has further shown that irrespective of the etiology for chronic pancreatitis, the structure and composition of PC are the same suggesting a common pathway for lithogenesis in the pancreatic duct.

Bacteriologic status of necrotic tissue in necrotizing pancreatitis.

To confirm the accuracy of guided percutaneous aspiration (GPA) in distinguishing sterile from infected pancreatic necrosis, we have performed Brown-Brenn tissue Gram stains on pancreatic and peripancreatic necrotic tissue removed operatively in 15 patients. In eight patients judged to have sterile necrosis on the basis of negative cultures of pancreatic exudate obtained first preoperatively (by GPA) and then intraoperatively, necrotic tissue debrided at surgery was also free of bacteria. In seven patients judged to have infected necrosis on the basis of positive cultures of pancreatic exudate obtained first preoperatively (by GPA) and then intraoperatively, necrotic tissue debrided at surgery harbored a considerable number of bacteria. We conclude that GPA targeted to areas of necrosis accurately distinguishes infected necrosis from sterile necrosis, and in infected necrosis, the solid necrotic tissue as well as the fluid component contains bacteria. We therefore believe that infected necrosis is not likely to be eradicated by catheter drainage and should be treated by surgical debridement.

Urinary trypsinogen activation peptides (TAP) are not increased in mild ERCP-induced pancreatitis.

Acute pancreatitis following endoscopic retrograde cholangiopancreatography (ERCP) occurs in 3-18% of patients undergoing either diagnostic or therapeutic ERCP. We prospectively measured urinary trypsinogen activation peptides (TAP) by an automated anti-TAP enzyme-linked immunoassay among 107 patients 4 h after ERCP to determine whether this measurement helps in the early diagnosis of ERCP-induced pancreatitis. Pancreatitis was documented in 10 of 107 patients (9.3%). All episodes were graded as mild. Urinary TAP was not significantly increased. We conclude that measurement of urinary TAP 4 h after ERCP is not helpful in documenting mild ERCP-induced acute pancreatitis.

Pleural effusion as a predictor of severity in acute pancreatitis.

Our objective was to determine whether pleural effusion is a predictor of severity in acute pancreatitis and, if so, whether it is an independent predictor. One hundred ninety-six consecutive cases of acute pancreatitis from October 1, 1994, to September 30, 1995, were reviewed. Medical records were analyzed for evidence of pleural effusion by chest radiograph and severe acute pancreatitis by identification of pancreatic necrosis or organ system dysfunction. Data were analyzed to determine if identification of pleural effusion provided an early sign of severity. Among 135 patients who underwent chest radiography, pleural effusion was seen in 16 of 19 (84.2%) with severe pancreatitis and 10 of 116 (8.6%) of patients with mild pancreatitis (p < 0.001). Pleural effusion was noted in severe pancreatitis prior to clinical or computed tomography evidence of severity in only 20% of cases. Pleural effusion is strongly associated with severity in acute pancreatitis but provides independent information on severity in only a minority of cases.

Modern concepts in pancreatitis.

A great deal remains to be learned about the pathophysiology, natural history, and treatment of acute and chronic pancreatitis. As we consider problems in treatment, we would be well advised to consider these words of Dr. Henry Janowitz: "Gastroenterologists are physicians who know how to think about and manage patients who have gastrointestinal problems." (44). To the extent that we heed these words and always give serious thought to what we are trying to achieve, we will provide greater help for our patients.

Infected necrosis: morbidity and therapeutic consequences.

Some 20% of cases of acute pancreatitis are associated with pancreatic and/or peripancreatic necrosis. Mortality of necrotizing pancreatitis is higher than that of acute interstitial pancreatitis, especially if there is secondary pancreatic infection. Despite the fact that patients with infected necrosis are in general more seriously ill than those with sterile necrosis, it is not possible at present by any individual laboratory test or constellation of tests to determine precisely which patients are infected or will develop pancreatic infection. CT-guided percutaneous aspiration with bacteriological sampling continues to be a very safe, reliable method of distinguishing severe sterile pancreatitis from pancreatic infection. Improved survival for infected necrosis depends on earlier recognition and prompt effective surgical debridement. Necrosectomy and post-operative local lavage appear to improve survival in infected necrosis.

Use and perceived effectiveness of non-analgesic medical therapies for chronic pancreatitis in the United States.

BACKGROUND: Effectiveness of medical therapies in chronic pancreatitis has been described in small studies of selected patients. AIM: To describe frequency and perceived effectiveness of non-analgesic medical therapies in chronic pancreatitis patients evaluated at US referral centres. METHODS: Using data on 516 chronic pancreatitis patients enrolled prospectively in the NAPS2 Study, we evaluated how often medical therapies [pancreatic enzyme replacement therapy (PERT), vitamins/antioxidants (AO), octreotide, coeliac plexus block (CPB)] were utilized and considered useful by physicians. RESULTS: Oral PERT was commonly used (70%), more frequently in the presence of exocrine insufficiency (EI) (88% vs. 61%, P < 0.001) and pain (74% vs. 59%, P < 0.002). On multivariable analyses, predictors of PERT usage were EI (OR 5.14, 95% CI 2.87-9.18), constant (OR 3.42, 95% CI 1.93-6.04) or intermittent pain (OR 1.98, 95% CI 1.14-3.45). Efficacy of PERT was predicted only by EI (OR 2.16, 95% CI 1.36-3.42). AO were tried less often (14%) and were more effective in idiopathic and obstructive vs. alcoholic chronic pancreatitis (25% vs. 4%, P = 0.03). Other therapies were infrequently used (CPB - 5%, octreotide - 7%) with efficacy generally <50%. CONCLUSIONS: Pancreatic enzyme replacement therapy is commonly utilized, but is considered useful in only subsets of chronic pancreatitis patients. Other medical therapies are used infrequently and have limited efficacy.