Search

Background: Laparoscopic cholecystectomy (LC) has become the treatment of choice for elective cholecystectomy, but controversy persists over use of this approach in the treatment of acute cholecystitis. We undertook a randomised comparison of the safety and outcome of LC and open cholecystectomy (OC) in patients with acute cholecystitis., Methods: 63 of 68 consecutive patients who met criteria for acute cholecystitis were randomly assigned OC (31 patients) or LC (32 patients). The primary endpoints were hospital mortality and morbidity, length of hospital stay, and length of sick leave from work. Analysis was by intention to treat. Suspected bile-duct stones were investigated by preoperative endoscopic retrograde cholangiography (LC group) or intraoperative cholangiography (OC group)., Findings: The two randomised groups were similar in demographic, physical, and clinical characteristics. 48% of the patients in the OC group and 59% in the LC group were older than 60 years. 13 patients in each group had gangrene or empyema, and one in each group had perforation of the gallbladder causing diffuse peritonitis. Five (16%) patients in the LC group required conversion to OC, in most because severe inflammation distorted the anatomy of Calot's triangle. There were no deaths or bile-duct lesions in either group, but the postoperative complication rate was significantly (p=0.0048) higher in the OC than in the LC group: seven (23%) patients had major and six (19%) minor complications after OC, whereas only one (3%) minor complication occurred after LC. The postoperative hospital stay was significantly shorter in the LC than the OC group (median 4 [IQR 2-5] vs 6 [5-8] days; p=0.0063). Mean length of sick leave was shorter in the LC group (13.9 vs 30.1 days; 95% CI for difference 10.9-21.7)., Interpretation: Even though LC for acute and gangrenous cholecystitis is technically demanding, in experienced hands it is safe and effective. It does not increase the mortality rate, and the morbidity rate seems to be even lower than that in OC. However, a moderately high conversion rate must be accepted.

We studied the adhesion mechanism of pancreatic carcinoma using in vitro adhesion and migration assays of stable cell lines and tumors grown from these cell lines in nude mice. We also compared the results with the expression profiles of laminins and their receptors in pancreatic carcinomas to evaluate the relevance of these mechanisms in vivo. All of the cell lines preferably adhered to laminin-5, irrespective of their capability to synthesize laminin-5. Cell migration was studied in the presence of hepatocyte growth factor, as it increased the speed of migration manyfold. Herbimycin A treatment and antibodies against the beta 1 and alpha 3 integrin subunits and laminin alpha 3 chain almost entirely blocked cell migration of the BxPC-3 cell line, whereas migration was nearly unaffected by RGD peptide and only moderately inhibited by antibody against the alpha 6 integrin subunit. Indirect immunofluorescence microscopy of wounded BxPC-3 cells suggested a rapid endocytosis of alpha 3 integrin subunit in the cells at the margin of the wound and a rapid, polarized rearrangement of the alpha 6 beta 4 integrin. Especially HGF-treated cultures showed a prominent cytoplasmic reaction for laminin-5 at the margin of the wound. Xenografted cells formed tumors that produced and deposited the same laminin chains as the in vitro cultures. Frozen sections of human pancreatic carcinomas showed reactivity for laminin chains suggestive for expression of laminin-1 and laminin-5. Both xenografted tumors and human pancreatic carcinomas also showed stromal reactivity for laminin-5. Electron microscopy of the human tumors suggested that this was due to an abundant reduplication the basement-membrane-like material around the nests of malignant cells. Our results suggest that pancreatic carcinomas synthesize and deposit laminin-5 in the basement membrane in an abnormal manner. Invading cells adhere to this newly produced basement membrane and migrate on it by using the alpha 3 beta 1 integrin receptor recognizing laminin-5.

We studied the expression and distribution of different laminin chains, the alpha 6 beta 4 integrin and type VII collagen, i.e., components of the epithelial adhesion complex, in gastric carcinomas and in suggested preneoplastic stages of this malignancy. Intestinal-type gastric carcinomas showed strong reactivity for laminin alpha 1, alpha 3, beta 1, and beta 3 chains, the components of laminin-1 and -5, at the interface between malignant cells and tumor stroma. The reactivities were continuous throughout the carcinomas, even in structures invading through the smooth muscle layers of the gastric wall. The expression of different laminin chains was accompanied by strong polarized reactivity for the alpha 6 beta 4 integrin, which is a receptor for both laminin-1 and laminin-5. Collagen type VII was only occasionally present at sites showing reactivity for laminin-5 and was totally absent from the cell islands invading through the gastric wall. Intestinalized gastric epithelium showed a similar expression pattern of laminins and the alpha 6 beta 4 integrin as the gastric carcinomas. Our results suggest that gastric carcinomas use the alpha 6 beta 4 integrin and newly deposited laminin-1 and -5, accompanied by the disappearance of type VII collagen, as their mechanism of adhesion during the invasion through surrounding tissues. Unlike in previous studies, the reactivity for the laminin-5 protein was not restricted to the invading cells but surrounded the malignant glandular structures throughout the tumor. Our results also show that both intestinal-type gastric carcinoma, and intestinal metaplasia mimic the gastric surface epithelium in the expression pattern of laminins and the beta 4 integrin subunit. This supports previous studies proposing a pathogenetic sequence from intestinal metaplasia to gastric carcinoma.

The proliferating cells of the gastric mucosa are found among the pit and mucous neck cells. These cells migrate upward to renew the surface epithelium and downward to restitute the glandular cells. As the epithelial basement membranes (BMs) function as substrate for cell adhesion and migration as well as signals for their differentiation, we studied, by indirect immunofluorescence microscopy, the distribution of different laminin chains and their integrin receptors in adult human stomach. The immunoreactivity for laminin alpha 2 chain localized to the BMs of glands and the lower parts of the gastric pits whereas the laminin alpha 3 chain (laminin-5/kalinin) immunoreactivity was strictly confined to BMs underneath the surface epithelium and the upper parts of the pits. Proliferating mucosal epithelial cells, identified by Ki-67 antibodies, were confined to the areas containing both alpha 2 and alpha 3 laminin chains. The alpha 1, beta 1, and gamma 1 laminin chains were found in all BMs of the mucosa whereas the beta 2 chain was prominent in mucosal blood vessels and also detectable in some glands. Among the laminin integrin receptors, the alpha 3 and beta 4 subunits were seen to be expressed in cells along the BMs with the alpha 3 laminin chain. The alpha 6 integrin, on the other hand, was seen in all gastric epithelia. The present results demonstrate that in the adult human stomach laminin alpha 2 and alpha 3 chains show zonal distribution in BM underlying gastric mucosal epithelium whereas other laminin chains show a more general distribution.

Pancreatic secretory trypsin inhibitor (PSTI) is a 6000-dalton peptide, that occurs in high concentrations in the pancreas and in pancreatic juice. It is thought to be synthesized by pancreatic acinar cells. We have recently reported the findings of an identical trypsin inhibitor at high concentrations in the urine of patients with gynecological malignancy. Therefore, we have named the inhibitor tumor-associated trypsin inhibitor (TATI). We have now studied patients who have undergone total pancreatoduodenectomy for pancreatic cancer or chronic pancreatitis. By radioimmunoassay (RIA), we found normal levels of this inhibitor in the serum and urine of pancreatectomized patients. The absence of pancreas was confirmed by measuring serum trypsin. By gel filtration and HPLC it was found that PSTI/TATI occurring in pancreatectomized patients was indistinguishable from that found in connection with pancreatitis and ovarian cancer.

Early postburn changes in central haemodynamics, organ blood flow distribution and morphology of the gastric mucosa were studied using a standarized thermal skin injury model. Organ blood flow and cardiac output were determined using radioactive microspheres. In the control animals no marked changes in cardiac output or organ blood flow were observed, and the gastric mucosa remained essentially undamaged. After burn injury and no fluid resuscitation, cardiac output decreased by 78 per cent, and blood flow to the stomach, pancreas, spleen, muscle, skin and kidneys also decreased markedly and to about the same degree as the cardiac output, however the adrenal flow remained roughly unchanged at the baseline level. Gross and microscopic lesions developed in the stomach, especially in the corpus. In animals given fluid resuscitation after burn injury cardiac output decreased by 38 per cent during the experiment, but blood flow in the stomach, brain, kidneys and spleen remained fairly constant, while pancreatic and muscle blood flow decreased and adrenal blood flow increased markedly. The gastric mucosa showed only minor microscopic, but no macroscopic lesions at the end of the experiment. The results indicate that acute thermal skin injury induces profound changes in central haemodynamics and organ blood flow which can, however, largely be overcome by adequate fluid resuscitation. The data also suggest that, as in other examples of 'stress ulceration', impaired mucosal blood flow may underlie the stress ulceration which complicates severe burns.