Haemorrhage into the pancreatic duct (Hemosuccus pancreaticus): recognition and management.

OBJECTIVE: To highlight the clinical presentation, investigation and treatment of haemorrhage into the pancreatic duct. DESIGN: Retrospective study and review of publications. SETTING: University hospital, Switzerland. SUBJECTS: All 4 cases from 1972 to 1993. INTERVENTIONS: 2 Whipple procedures, 1 resection of the pancreatic head, 1 exploratory laparotomy. Radiological embolisation in one case. MAIN OUTCOME MEASURES: Cessation of haemorrhage and survival. RESULTS: The diagnosis was made preoperatively in three cases by gastroduodenoscopy and arteriography. Operation was the primary treatment in all patients and was effective with low morbidity and no mortality in three of them. Embolisation stopped the haemorrhage in the fourth patient, who was alcoholic and died of progressive liver insufficiency and variceal haemorrhage. CONCLUSIONS: There is no specific indication for haemorrhage into the pancreatic duct. The diagnosis is suggested by endoscopy (absence of a more common cause, or blood in the second part of the duodenum). Arteriography is essential to confirm the site of the bleeding and to attempt embolization. Operation is usually the definitive treatment.

Biliary calculi caused by hemobilia.

Formation of biliary calculi caused by hemobilia is rare. Including the two cases reported here, there are only a total of four in the literature. The characteristics of these calculi in vitro, on computerized tomographic scan, and cholecystography are described. The condition for the occurrence seems to be that blood clots remain in the gallbladder sufficiently long (about 6 months) to become encrusted with bile constituents. Patients with hemobilia with clots in the gallbladder should be observed for this complication.

Molecular dynamics simulations of the holo and apo forms of retinol binding protein. Structural and dynamical changes induced by retinol removal.

The effects of removing retinol from the X-ray structure of holo-retinol binding protein are studied using the molecular dynamics technique. Structural and dynamical properties emerging from an 80 ps simulation of the apo form, for which no crystallographic structure is available, are compared with the results of a 70 ps trajectory of the holo-protein. Dynamical stationarity is attained after roughly 30 ps, and the resulting average structure is proposed as a reasonable model of the apo-protein. Conformational changes are observed for the loops at the beta-barrel entrance during the non-equilibrium part of the apo-trajectory. Tryptophan labelling experiments and retinoid reconstitution experiments point towards this part of the molecule as being involved in prealbumin binding. Structural changes in this region may therefore explain the differences in prealbumin affinity between the apo and holo forms. Furthermore, a change in the position of the alpha-helix, corresponding to a pivot around its C terminus, is observed for the apo-protein. The resulting conformation of the alpha-helix is found to be similar to that in apo-beta-lactoglobulin, which also can bind retinol and for which a crystal structure exists. The results from the holo simulation are compared to the crystallographic data and show good agreement. The dynamics of the secondary and tertiary structural elements are analysed and compared for the two forms. The beta-barrel is found to be extremely cooperative in its atomic motions in both simulations, and the top and bottom beta-sheets perform collective fluctuations with respect to each other in the low-frequency limit of the simulations. The dynamics of the alpha-helical region presents clear differences between the two forms; while the holo-protein has a well-defined spectrum for the longitudinal stretching mode, the apo form displays a fairly large bending of the alpha-helix at several points of the trajectory.

Structural changes in retinol binding protein induced by retinol removal. A molecular dynamics study.

Relationships between structure and function for retinol binding protein (RBP) are elucidated with help of a 2.0 A resolution X-ray structure of the holo-protein and an average molecular dynamics (MD) structure of the apo-form. Comparisons between MD simulations of both the apo- and holo-forms with the X-ray holo-structure show conformational changes in apo-RBP that may be functionally significant. The average three dimensional structure obtained for apo-RBP is compared to the related protein apo-beta-lactoglobulin. Available biochemical information is consistent with structure/function relationships derived here.

Iatrogenic hemobilia.

Biliary tract hemorrhage is increasingly reported as a complication of diagnostic and therapeutic interventions on the liver and the bile ducts. Major hemobilia with profuse, even life-threatening, hemorrhage which necessitates emergency surgical intervention is uncommon. Minor hemobilia with melena or clot production in the ducts, although more common, may cause diagnostic difficulties. Diagnosis is obtained either endoscopically to demonstrate bleeding through the papilla of Vater, or through hepatic arteriography, which has the advantage of indicating the location of the lesion. Operative treatment with resection of the lesion or ligature of the hepatic artery or a branch has been replaced by embolization of the affected artery. Diagnostic procedures that may cause hemobilia include percutaneous needle biopsy and transhepatic cholangiography. The risk of hemorrhage increases if a catheter is left in place for drainage. Difficult extractions of common duct stones may cause hemorrhage due to lesions of the vulnerable duct mucosa. Finally, spontaneous hemobilia may, like hematuria or epistaxis, be caused by an overdose of anticoagulants.

Stones or clots in the biliary tract. A diagnostic dilemma.

Clots, which not uncommonly form in the biliary tract, may cause diagnostic errors. They often result from minor hemobilia, small hemorrhages into the ducts, mostly occurring in connection with gallstone attacks or biliary surgery. Hemobilia may also, as hematuria and epistaxis, complicate anticoagulant therapy. The blood will generally flow inconspicuously into the intestine, and even if it coagulates, the clots tend to dissolve rapidly through the fibrinolytic activity of the bile. For various reasons they may occasionally escape dissolution and remain in the ducts. fibrin clots can also form in inflammatory biliary tract disease. Clots will easily be mistaken for gallstones and treated as such. Cases are presented which illustrate why they are so often overlooked or misinterpreted. Cholangiography may be of diagnostic help since the clots show certain characteristics, appearing as casts of the biliary tract, often adhering to the walls, sometimes with indistinct borderlines. During 1980 we performed 128 cholecystectomies for suspected gallstone disease. Common duct exploration was done in 18 because of contrast defects in the cholangiogram. In 15 cases these were due to gallstones, in 3--one out of six--they were caused by blood clots. The comparatively high frequency with which clot formation in the biliary tract is observed when attention is given to this possibility, compared to the number of misinterpretations that otherwise occur leads to the conclusion that common duct clots are more common and of greater significance than is generally assumed.

Minor hemobilia. Clinical significance and pathophysiological background.

Profuse hemorrhage into the biliary tract--major hemobilia,--is an alarming condition which attracts much attention. Minor hemobilia, often caused by gallstones or operative injury, is much more frequent, yet often neglected. Clinical observations indicate that minor hemobilia is not always an innocent condition with the blood remaining fluid and unobtrusively flowing into the intestine. Examples are given where clots from occult hemobilia have caused diagnostic errors or obstructed the bile flow, thus imitating gallstones for which they may be mistaken. Experiments have been performed to elucidate the pathophysiology of this clot formation: 1) A model of the biliary tract was constructed with bile flowing through it. When blood was injected forcefully to simulate a major hemorrhage, mixed clots of blood and bile were formed. When introduced gently, as in minor hemobilia, the blood flowed immiscibly to the lowest level where it formed a clot of pure blood. The clots dissolved under the influence of flowing bile, but remained intact when protected from the flow. 2) Mixed clots were mushy and dissolved spontaneously, while pure clots remained solid and stable. When clots containing increasing amounts of bile were incubated, increasing amounts of cleavage products of fibrinogen and fibrin were formed. 3) Blood clots were produced in gallbladders of 33 dogs. Pure clots remained solid and floating while mixed clots formed a mushy layer, strongly adherent to the mucosa. Both kinds of clots disappeared after two-four weeks, but remained if bile flow was diverted. These findings explain why under certain circumstances minor hemobilia acquires clinical significance by forming clots that may obstruct the flow or cause diagnostic errors. Successful dissolution of "retained stones" may occasionally have the simple explanation that blood clots, mistaken for calculi, have been fibrinolysed. Consequently, in biliary obstruction or when defects are found on cholangiography, the possibility of blood clots in the ducts should be considered even in the absence of overt gastrointestinal hemorrhage.

Formation and fate of fibrin clots in the biliary tract: a clinical and experimental study.

Fibrin clots may form in the biliary tract from hemobilia or in inflammatory disease. There is a wide variation in the clinical course of such clots which is exemplified by 9 patients. They may either dissolve through fibrinolysis, get ejected into the intestine, remain and obstruct the biliary tract, or may even transform into gallstones. In order to elucidate the mechanisms involved, the behavior of blood clots in bile was studied in vitro. A model was constructed of the biliary tract and, drained by a T-tube, where human bile circulated with a flow rate resembling that in vivo. When a small amount of human blood was injected, it flowed immiscibly to the lowest level, displaced the bile, and formed a clot of pure blood. Even a minor bleeding may thus form a coagulum. This is different from the mixed clot of blood and bile that forms in experiments simulating major hemorrhage. These findings are related to clinical experience and especially to the disappearance of "retained stones" with or without the use of dissolving agents.