Enterococcal sepsis and lung microvascular injury in sheep.

In a common bile duct contamination model, we studied the effect of Streptococcus faecalis compared with Escherichia coli in sheep with chronic lymph fistulas to investigate the role of enterococcus in acute lung injury and acute sepsis. Early pulmonary hypertension in the E coli group was not expressed in the S faecalis group, probably due to a failure of S faecalis to illicit a thromboxane A2 response. In the late period, E coli was associated with significantly greater lung microvascular damage compared with S faecalis. The lack of difference between groups with respect to complement activation suggests the action of chemotactic factors, in addition to complement, mediating granulocyte aggregation, and neutropenia. In this model, S faecalis demonstrated limited pathogenicity as expressed in lung microvascular injury compared with E coli.

Effects of biliary obstruction on hepatic clearance of bacteria.

High surgical mortality in patients with obstructive jaundice and sepsis have been attributed to reticuloendothelial system (RES) depression. The purpose of this study was to clarify the effects of mechanical biliary obstruction on RES clearance of pathogenic bacteria by comparing the phagocytic index (K) with the directly measured hepatic uptake of indium 111-labeled bacteria injected into the portal vein of normal dogs and dogs with partial (PBO) or complete biliary obstruction (CBO). No significant difference was observed between the K in normal dogs (0.19 +/- 0.08; n = 6) and that in dogs with PBO (0.24 +/- 0.06; n = 5) or CBO (0.21 +/- 0.03; n = 4). There was no significant difference in uptake of radiolabel by the liver among the three groups of dogs. In our model, biliary obstruction had no effect on hepatic RES function and may not represent a significant determinant of mortality in patients with obstructive jaundice.

Pancreatic pseudocysts: clinical and endoscopic experience.

Records of 51 patients with proven pancreatic pseudocyst (PC) were retrospectively reviewed to assess clinical and endoscopic variables. Thirty-nine had surgery after preoperative endoscopic retrograde cholangiopancreatography (ERCP), six patients had ERCP only, and six patients had surgery without prior ERCP. Forty-two pancreatograms were obtained, and all were abnormal. PC communication with the pancreatic duct was present in 29 and ductal obstruction downstream from the PC was shown in 11 studies. Thirty-five cholangiograms were obtained and 19 were abnormal, including ductal narrowing in 16 and biliary calculi in three. PC contents were cultured intraoperatively in 29 patients, and 17 were positive. PC infection was highly correlated with the incidence of recurrent hospitalizations and with PC recurrence. PC infection did not occur significantly more often in patients with communicating PC or in those with preoperative ERCP. Factors that did not influence outcome included PC size, multiplicity, anatomic location, and PC communication with the pancreatic duct. Results from both ERCP and culture of PC contents add significantly to our knowledge of PC disease.

Nosocomial abscess. Results of an eight-year prospective study of 32,284 operations.

This article details the results of 114 abscesses found after 32,284 operations during a 30-day prospective infection surveillance. Abscesses occurred more often after gastric/esophageal (3.6%), colonic/appendiceal (2.8%), and pancreatic/biliary (1.5%) operations. Abscesses were intra-abdominal (58%), pelvic/perineal (31%), and retroperitoneal (3%). The most common pathogens were Escherichia coli, enterococci, and Bacteroides organisms. Abscess symptoms appeared 10 +/- 6 days (mean +/- SEM) after surgery and were drained 8 +/- 8 days after onset of symptoms. Hospital stay from onset of symptoms to discharge was four times greater than that for wound infection. Twenty-seven patients (28%) died. Mortality was increased in patients older than 50 years and in those with renal failure or multiple abscesses. Nosocomial abscesses had a greater morbidity than would infection. Excessive expense and high mortality warrant infection surveillance for both wound infection and ongoing nosocomial abscesses.

Partial biliary obstruction caused by chronic pancreatitis. An appraisal of indications for surgical biliary drainage.

This paper presents a retrospective review of 38 patients with intrapancreatic bile duct strictures secondary to chronic alcoholic pancreatitis. The strictures were identified by endoscopic retrograde cholangiopancreatography (ERCP). All patients with pancreatic cancer and gallstone pancreatitis were excluded. The mean alkaline phosphatase and total bilirubin values were 344 +/- 57 IU/dl and 4.4 +/- 0.7 mg/dl, respectively. The mean stricture length was 3.9 +/- 0.5 cm, and the mean common bile duct (CBD) diameter was 1.8 +/- 0.2 cm. The degree of bilirubin and alkaline phosphatase elevation did not correlate with stricture length or the severity of bile duct dilatation. Eighteen of the 38 patients received surgical biliary drainage (BD) as part of their initial therapy, and 20 patients did not. Liver function tests, intrapancreatic stricture length, and the degree of proximal CBD dilation were comparable in these two groups. Patients not undergoing BD did well clinically as only one patient required BD over an average follow-up period of 3.8 years. In conclusion, bypass of these strictures is usually unnecessary, and most patients may be safely treated without operation.

Complement in local biliary tract defense: dissociation between bile complement and acute phase reactants in cholecystitis.

Biliary complement concentrations and activity are lower in patients with infected bile than in those with sterile bile in cholecystitis. Plasma complement is increased during the acute phase response to inflammation. To determine whether low biliary complement in infected bile is a specific response to biliary tract infection or part of a general systemic reaction, we analyzed bile complement proteins (C3 and C4) and activity (C4H50) and acute phase reactants fibronectin, C-reactive protein, and alpha 1-antitrypsin concentrations in acute and chronic cholecystitis. Results were correlated with bile cultures and gallbladder histology using the Wilcoxon rank sum test. While biliary C3, C4, and C4H50 were significantly lower in infected bile than in sterile bile, none of the acute phase reactants were different. The biliary acute phase reactants were all significantly higher in acute cholecystitis than in chronic disease, but there was no difference in the biliary C3, C4, or C4H50 levels. There was no clear relationship between plasma levels of complement and the acute phase reactants. The dissociation between biliary complement and acute phase reactants indicates that bile complement is not a reflection of a systemic reaction to inflammation. We propose that biliary complement is a specific host defense mechanism against bacterial infection in the biliary tract.

Pseudomonas infection of the biliary system resulting from use of a contaminated endoscope.

Pseudomonas aeruginosa was present in bile cultures from 10 patients who had undergone previous endoscopic retrograde cholangiopancreatography in 1984. After environmental cultures and review of instrument disinfection, we traced the infections to a single endoscope contaminated with P. aeruginosa, serotype 10. Although the instrument had been cleaned repeatedly with an automatic endoscope cleaning machine, P. aeruginosa survived on residual moisture left in the channels of the endoscope. Contamination ended only after we began to manually suction alcohol through the endoscope before air drying. In 5 of 10 patients, P. aeruginosa caused clinical infections including gangrenous cholecystitis, abscesses, and death. We could identify no factor that distinguished symptomatic from asymptomatic patients. In asymptomatic patients, P. aeruginosa was recovered from gallbladder bile up to 2 mo after endoscopic retrograde cholangiopancreatography. As this P. aeruginosa epidemic was discovered retrospectively because we monitor bile cultures, we advocate this practice as part of endoscopic retrograde cholangiopancreatography procedures.

Giant gastric ulcers. A review of 49 cases.

Traditional therapy for giant gastric ulcers (greater than 3 cm) has been gastric resection because of a presumed high risk of hemorrhage and recurrence. To determine the validity of this approach and decide whether the need for resection has been altered by the introduction of H2-blockers, the records of 10,054 gastroduodenal endoscopies performed between 1971 and 1984 were reviewed. Forty-nine patients with giant gastric ulcers were identified. Five patients had malignant ulcers. Ten patients underwent gastric resection as initial therapy. Thirty-four patients were initially treated without surgery and were divided into Group I (no H2-blockers; 9 patients) and Group II (H2-blockers; 25 patients). Medical therapy was successful in three of nine patients (33%) in Group I and in 20 of 25 patients (80%) in Group II. Of 11 patients who failed medical therapy (7 intractability, 3 recurrence, and 1 fatal hemorrhage), 10 underwent subsequent gastric resection. Of the 20 patients treated surgically (10 initial and 10 medical failures), none were readmitted for recurrent ulcer disease. These data suggest that medical therapy of benign giant gastric ulcers is often effective and not associated with an excessive incidence of complications, as believed. Successful healing of these ulcers is greatly enhanced when H2-blockers are employed. Thus, the presence of an uncomplicated benign giant gastric ulcer is not an absolute indication for gastric resection.

In-111 Pseudomonas aeruginosa: a simple method of labeling live bacteria with a gamma-emitting radioisotope.

We describe a simple and reliable technique for labeling Pseudomonas aeruginosa with a readily available commercial preparation of indium-111 (111In) oxine. Labeling of a heavy bacterial suspension with 500 mu Ci of commercially prepared 111In-oxine resulted in a yield of 0.0004 mu Ci of cell-associated 111In per 10(6) colony-forming units (CFU). The label was 88% bacterially associated and did not effect viability of the organism. Radiolabeling a gram-negative organism with 111In-oxine provides a non-toxic, stable gamma-emitting bacterial tracer.