Adenomas and carcinomas missed in routine colonoscopy: a prospective study in resected colon segments

Colonoscopy is the standard technique in the diagnosis and treatment of colorectal neoplasia, but small adenomas and even advanced lesions can be missed during the procedure. With large scale screening colonoscopy programs installed, information on quality of colonoscopy in primary care is essential, but scarcely available. Over a period of 45 months, we prospectively included all those patients in our study, who underwent major colonic surgery at our institution and who had undergone a colonoscopy within 42 days prior to the operation. 89 men and 100 women, median age 71 years, were included. The majority of these operations were performed for colorectal carcinoma (125), other malignant tumors (4), suspected malignancies (6) or large adenomas (14). The pathologist inspected the resected colonic segment, and we compared his findings with the colonoscopy report. Colonoscopies had been performed by 22 doctors in 13 institutions. Median length of the resected colonic segments was 20 cm (range 3 to 135 cm), total length was 41,21 metres. In 14 segments the pathologist identified 28 neoplastic lesions not described in the endoscopy report. Colonoscopy had missed 2 carcinomas, both in the right colon, and a 12 mm tubulo-villous adenoma with high-grade dysplasia. Another 25 tubular adenomas had been missed, 2 measuring 10 mm, 7 between 5 and 9 mm and 16 smaller than 5 mm. We conclude that primary care colonoscopy misses neoplastic lesions in a significant number of procedures. Most of the missed lesions in our high risk group of patients would have been of little clinical consequence. In a small, but clinically important number of cases, however, advanced adenomas and even colorectal carcinomas were missed by endoscopy.

Missed colonic adenomas in routine primary care endoscopy: a prospective tandem colonoscopy study

Colonoscopy is the gold standard in the diagnosis of colorectal neoplasia. Several lines of independent evidence, however, suggest that a significant number of small adenomas and also some advanced lesions are missed even by experienced endoscopists. With large-scale screening colonoscopy programmes installed, information on quality of colonoscopy in primary care is essential, but not available. Between July 2006 and December 2008, a total of 40 patients (23 men and 17 women, median age: 69 years) underwent a second colonoscopy within 42 days after the first endoscopy (median: 11 days), in all cases exclusively for clinical reasons. Index colonoscopy was performed by 14 endoscopists in 6 hospitals and 4 private practices. Data on all consecutive patients were collected prospectively. A total of 98 neoplastic lesions were identified in 34 patients at the index colonoscopy, an additional 53 adenomas were removed at the second colonoscopy, 33 of them smaller than 5 mm. 25 out of 53 missed adenomas were identified between the coecum and the right colonic flexure. 12 of the additional lesions were considered significant lesions (larger than 10 mm or tubulovillous adenoma), nine of these were located between the coecum and the right colonic flexure. In 24 patients repeat colonoscopy detected adenomas not described in the original report. In eight patients a total of 12 significant lesions were removed, nine of these were located between the coecum and the right colonic flexure. About one-third of adenomas were missed in 40 routine colonoscopies, most of them only small and therefore probably of little clinical significance. However, 12 significant lesions were missed in 8 patients. Adenomas in the right colon seem to be a particular problem.

Guiding manometry probes into the human small bowel without X-ray: validation of a new technique

BACKGROUND

In order to perform small bowel manometry studies in man, the probes have to be guided through the upper gastrointestinal tract. This is usually controlled by fluoroscopy, but a method that allows to place the tubes without X-rays is obviously desirable.

METHODS

A new method that controls tube placing by sonography and pH-metry instead of X-rays is described. In a total of 19 volunteers 40 tubes were placed, either under fluoroscopic control (n = 22) or without X-rays (n = 18).

RESULTS

Nose to antrum time was not significantly different between the two groups (8 +/- 2 vs. 10 +/- 2 min), but the documented transit of the tube from the antrum into the duodenum was significantly slower in the fluoroscopy group (80 +/- 20 vs. 60 +/- 12 min). Small bowel transit time was not significantly different between the groups (73 +/- 16 vs. 65 +/- 15 min).

CONCLUSION

Placing motility probes in the human small bowel can be controlled by sonography and pH-metry instead of fluoroscopy. It is also easily performed, faster and more convenient for the investigator. As far as basic gastrointestinal research is concerned, in Germany less bureaucratic paperwork is involved, as these studies need not be approved by the 'Bundesamt fur Strahlenschutz' any more.

Small Bowel Motor Activity in Response to Physiological Meals of Different Chemical Composition in Man

Meals disrupt the interdigestive pattern of small bowel motor activity and convert it into the more irregular postprandial pattern. Previous animal studies suggest that the duration of and contractile activity within the postprandial period depend on the chemical composition of a meal. It is not clear whether this is also true for man. In 8 healthy volunteers I investigated how physiological and isocaloric meals of different chemical composition affect small bowel motor activity. Volunteers underwent two separate, ambulatory 24-hour small bowel manometry studies. Volunteers had a total of four meals on the two study days. They ingested two identical fish meals rich in protein, a pasta meal rich in carbohydrates, and a meat meal rich in fat. Records were analyzed visually for the reappearance of phase III of the migrating motor complex, and a validated computer program calculated the incidence of contractions during the postprandial period, as well as the amplitude and propagation of contractions. The durations of the postprandial periods were similar after the two protein meals (238 +/- 35 and 227 +/- 25 min), the carbohydrate (220 +/- 23 min) and the fat meal (242 +/- 43 min). The incidence of contractions was not significantly different after the protein meals (1.6 +/- 0.3 and 1.0 +/- 0.2 contractions per min), the carbohydrate meal (1.0 +/- 0.2 contractions per min) and the fat meal (1.5 +/- 0.2 contractions per min). The amplitude of contractions was similar after the protein meals (14 +/- 0.1 and 13 +/- 0.1 mmHg), the carbohydrate meal (14 +/- 0.1 mmHg) and the fat meal (14 +/- 0.1 mmHg). Propagation of contractions was similar after the protein meals (13 +/- 3 and 18 +/- 3 %), after the carbohydrate meal (15 +/- 2 %) and after the fat meal (13 +/- 2 %). Contractile activity within consecutive 30-min periods of the postprandial period was also not different between the meals. I conclude that physiological, isocaloric meals of different chemical composition elicit a similar postprandial motor response in the human small bowel.

Endoscopic retrieval of a broken and migrated esophageal metal stent

In patients with inoperable malignant tumors of the esophagus or cardia, self-expanding metal stents are increasingly used to improve dysphagia. Usually, they are not difficult to place and, as compared to conventional plastic stents, complications such as stent migration or perforation, seem to occur less frequently. This is a report on a young patient with metastatic adenocarcinoma of the cardia, who was treated with a self expanding metal stent after endoscopic dilatation of a tumor stenosis in the distal esophagus. Immediately after the procedure, he was able to eat and gained weight. Within 6 weeks and while on continuous infusion of 5-fluorouracil, the patient complained about recurrent severe dysphagia. Plain x-ray demonstrated a broken and migrated stent, the 2 parts of which were seen in the stomach and the duodenum. The stent could be extracted endoscopically without any complication, but the procedure was difficult and lasted 4 h, as the stent broke 2 more times during retrieval.

Liver transplantation in a 29-year-old patient with gallbladder carcinoma complicating primary sclerosing cholangitis

Patients with primary sclerosing cholangitis (PSC) are at increased risk for cholangiocarcinoma. This tumor usually is a fatal complication, median survival after diagnosis is less than six months. In an asymptomatic 29-year-old patient with long-standing PSC and ulcerative colitis, routine abdominal ultrasound demonstrated an irregular mass, 11 x 13 mm, in the gallbladder. Cholecystectomy was performed, and histological examination demonstrated a moderately differentiated adenocarcinoma with infiltration of all layers of the gallbladder and invasion of local lymphatic vessels. Extensive diagnostic work-up failed to consistently demonstrate metastatic disease, and the patient was offered a liver transplantation. 24 months after the operation, the patient feels well and there is no indication of tumor recurrence. In carefully selected patients with gallbladder carcinoma complicating PSC, liver transplantation may be a therapeutic option.

Human small bowel motor activity in response to liquid meals of different caloric value and different chemical composition

Previous animal studies have shown that the nature and duration of postprandial motility in the small bowel depend both on the caloric load and the chemical composition of a meal. It is not clear whether this is also true for the human small bowel. Therefore we investigated the motor activity of the human small bowel in response to nutrient liquids of different caloric value and different chemical composition. Ten human volunteers underwent three separate, 24-hr ambulatory manometry studies. They drank water, a pure glucose solution, and Intralipid 10% in volumes of both 300 and 600 ml. The caloric value of the nutrient liquids was 330 and 660 kcal, respectively. Records were analyzed visually for the reappearance of phase III of the MMC after ingestion of a test liquid, and a validated computer program calculated the incidence and amplitude of contractions during the postprandial period. Neither duration of the postprandial interval nor the mean incidence or mean amplitude of contractions were different between the fat and the carbohydrate solutions, but phase III reappeared significantly later after ingestion of the nutrient liquids than after water (P = 0.0002). Duration of the postprandial interval also depended on the volume or the caloric load of a liquid meal (P = 0.0012). Mean incidence of contractions tended to be higher after ingestion of nutrient liquids than after water (P = 0.059). We conclude that in ambulant subjects, small bowel motor activity in response to chemically diverse liquid meals is remarkably uniform. This is true for the duration of the postprandial motor activity, as well as the incidence and amplitude of contractions during that period. The caloric value of a liquid meal, however, regulates the duration of the postprandial interval in the human small bowel.

Silibinin, a plant extract with antioxidant and membrane stabilizing properties, protects exocrine pancreas from cyclosporin A toxicity

Silymarin can be extracted from the milk thistle, and silibinin is the main component of the plant extract. Possibly due to their antioxidant and membrane-stabilizing properties, the compounds have been shown to protect different organs and cells against a number of insults. Thus liver, kidney, erythrocytes and platelets have been protected from the toxic effects of ethanol, carbon tetrachloride, cold ischemia and drugs, respectively. The effect of silibinin on endocrine and exocrine pancreas, however, has not been studied. We therefore investigated whether silibinin treatment attenuates cyclosporin A (CiA) toxicity on rat endocrine and exocrine pancreas. Groups of 15 male Wistar rats were treated for 8 days with CiA and/or silibinin. On day 9, endocrine and exocrine pancreatic functions were tested in vitro. At the end of the treatment period, blood glucose levels in vivo were significantly higher in rats treated with CiA while silibinin did not affect glucose levels. In vitro, insulin secretion was inhibited after treatment with silibinin, but amylase secretion was not affected. After treatment with CiA both insulin and amylase secretion were reduced. Silibinin and CiA had an additive inhibitory effect on insulin secretion, but silibinin attenuated CiA-induced inhibition of amylase secretion. Despite CiA treatment, amylase secretion was in fact restored to normal with the highest dose of silibinin. Thus silibinin inhibits glucose-stimulated insulin release in vitro, while not affecting blood glucose concentration in vivo. This combination of effects could be useful in the treatment of non-insulin-dependent diabetes mellitus. Furthermore, silibinin protects the exocrine pancreas from CiA toxicity. As this inhibitory effect is probably unspecific, silibinin may also protect the exocrine pancreas against other insult principles, such as alcohol.

Value of an extended monoethylglycinexylidide formation test and other dynamic liver function tests in liver transplant donors

Measuring monoethylglycinexylidide (MEGX) formation after intravenous administration of lidocaine in potential organ donors (MEGX test) has been advocated as a useful test to select donor livers for transplantation, but some groups have demonstrated a low test efficacy. We, therefore, investigated the value of an extended MEGX formation test and the value of other dynamic liver function tests, in selecting suitable human donor livers. In 51 human multi-organ donors, we measured elimination of galactose, indocyanine green, and lidocaine, as well as formation of MEGX, at 15, 30, and 60 min after administration of the test substances. In the early postoperative period, the function of the transplanted liver was then classified as good or poor, as defined by a prothrombin time above or below 65% by day 4 and fibrinogen concentration above or below 300 mg/dl by day 7. Donor characteristics and preservation modalities were very similar between the two groups. Galactose, indocyanine green, and lidocaine metabolism failed to predict good or poor graft function in the early postoperative period. MEGX serum concentrations, however, were significantly higher in the group of donors whose organs functioned well in the recipients, as compared with donors whose organs functioned poorly in the recipients. This was true for MEGX concentrations at 15 min (117+/-9 vs. 90+/-9 ng/ml; P=0.03), 30 min (108+/-8 vs. 86+/-8 ng/ml; P=0.04), and 60 min (100+/-6 vs. 73+/-5 ng/ml; P=0.006). Extending the MEGX formation test from 15 to 60 min improved test efficacy. Maximal MEGX concentration in 9 or up to 12 consecutive blood samples, drawn between 3 and 120 min after lidocaine infusion, was also significantly higher in donors whose organs functioned well, than in donors whose organs functioned poorly (129+/-10 vs. 101+/-10 ng/ml; P=0.03). Although the groups with good and poor organ function differed significantly with respect to their MEGX serum concentrations, and although efficacy of the MEGX test was improved by extending the test from 15 to 60 min, the overlap in individual MEGX serum concentrations was still so wide that it is virtually impossible to predict early graft function only on the basis of the MEGX test in the donor. Therefore, the MEGX test, although of potential scientific interest, does not predict early graft function with an accuracy necessary for clinical use.

Conventional and quantitative liver function tests after hepatic transplantation: a prospective long-term follow-up

In long-term survivors of liver transplantation, hepatic function is obviously of vital importance. Therefore, we prospectively performed conventional and quantitative liver function tests in patients who had survived a first transplantation for at least 4 years. Compared to 6 months after transplantation, serum bilirubin concentration and gamma GT activity were significantly lower after 3, 4, and 5 years (bilirubin 1.2 +/- 0.2 mg/dl at 6 months vs 1.0 +/- 0.1, 1.0 +/- 0.2, and 0.8 +/- 0.1 mg/dl respectively; gamma GT 106 +/- 0 33 U/l at 6 months vs 56 +/- 17, 67 +/- 35, 39 +/- 10 U/l respectively). At these points in time, blood levels of cyclosporin A were also significantly lower. Other parameters of liver cell function and liver cell integrity (AP, AST, ALT, GLDH, total protein, thromboplastin time, partial thromboplastin time) were unchanged over time. Serial quantitative liver function tests (indocyanine green half-life, galactose elimination capacity, lidocaine half-life, and MEGX formation) also remained stable. Thus, we conclude that hepatic function remains stable in long-term survivors of liver transplantation for at least several years.

Oesophageal acid and salivary secretion: is chewing gum a treatment option for gastro-oesophageal reflux?

The presence of acid in the oesophagus has been shown to stimulate salivary secretion, but the relevance of this oesophago-salivary reflex for acid clearance in physiological and pathological gastro-oesophageal reflux (GOR) is unknown. This study was designed to investigate the interrelation between oesophageal acid and both resting and stimulated salivary secretion. In 10 healthy volunteers, the acid clearance times after bolus infusion of 20 ml of 0.1 N hydrochloric acid were measured by means of ambulatory oesophageal pH monitoring. With a constant swallowing rate and resting salivary flow, the acid clearance time was significantly longer with dry as opposed to wet swallows (12.6 +/- 2.6 vs. 6.9 +/- 1.9 min; p = 0.01). When the salivary flow was doubled by chewing a gum base (26.0 +/- 3.4 vs. 13.2 +/- 2.0 ml/15 min; p = 0.005), the acid clearance time was markedly shortened (6.9 +/- 1.9 vs. 2.3 +/- 0.2 min; p = 0.02). As compared with water control, salivary flow, pH, and protein content were not affected by a bolus infusion of hydrochloric acid. This was true both for resting and gum-stimulated salivary secretion. Our study suggests that an oesophago-salivary reflex becomes effective only in prolonged episodes of GOR. This may explain why the water brash phenomenon is not regularly experienced by all reflux patients. Our study also suggests that chewing gum might be a non-pharmacological treatment option for some patients with symptomatic GOR.

[Hemolytic anemia as cause of a marked bilirubin increase in primary biliary cirrhosis]

HISTORY AND CLINICAL FINDINGS

16 years ago a now 53-year-old woman was found to have primary biliary cirrhosis. 5 years later, after bleeding from oesophageal varices, she had a portacaval shunt. For several years she had been taking ursodeoxycholic acid (750 mg daily). Because of steadily increasing jaundice over the past few years she presented for possible liver transplantation.

INVESTIGATIONS

There was a discrepancy between the markedly raised serum bilirubin concentration (7.8 mg/dl) and the only slightly raised or normal activities of alkaline phosphatase (247 U/l) and gamma-GT (21 U/l). Further tests confirmed that the patients had not only PBC but also Coombs-negative haemolytic anaemia (haemoglobin 10.7 g/dl, reticulocyte count 122/1000, indirect bilirubin 6.4 mg/dl, haptoglobin not demonstrated, lactate dehydrogenase 316 U/l). She had splenomegaly despite the portacaval shunt. Blood smear revealed spherocytes, but hereditary spherocytosis was not confirmed.

TREATMENT AND COURSE

A six-week interruption of taking ursodeoxycholic acid led, as expected, to a rise in the activities of serum alkaline phosphatase and gamma-GT, while haemolysis parameters were not affected.

CONCLUSION

Serum bilirubin concentration is a decisive prognostic factor in the course of primary biliary cirrhosis and is therefore of particular relevance for the indication of liver transplantation. The reported case demonstrates the importance of considering other causes of hyperbilirubinaemia.

Liver function in patients with pulmonary emphysema due to severe alpha-1-antitrypsin deficiency (Pi ZZ)

Alpha 1-Antitrypsin deficiency predisposes to pulmonary emphysema, liver cirrhosis and hepatocellular carcinoma. Anecdotal evidence and a large autopsy study suggest that severe lung and liver disease rarely coexist in the same subject, but this has not been studied in patients. Therefore we investigated 27 patients with severe alpha 1-deficiency (Pi ZZ) and pulmonary emphysema for signs of liver disease and impaired hepatic function. A subgroup of 7 patients underwent quantitative liver function tests. On physical examination or ultrasonography, cirrhosis or tumor was not suspected in any patient. Conventional liver function tests were completely normal in 17 patients. Elevated serum activities of gamma-glutamyltranspeptidase and/or aminotransferases were seen in 10 patients. In some, the elevation was only marginal and in none more than twice normal. The serum bilirubin concentration and activity of alkaline phosphatase were increased in 1 patient. Serum protein, albumin, fibrinogen, antithrombin III, alpha 1-fetoprotein concentrations, serum activities of cholinesterase and glutamate dehydrogenase, activated partial thromboplastin time and prothrombin time were normal in all patients. The indocyanine green half-life was abnormal only in 1 of 6 patients, suggesting that hepatic blood flow was not impaired in the study group. However, the lidocaine half-life and galactose elimination capacity, parameters of hepatic metabolization, were impaired in 4 and 6 of 7 patients, respectively. We conclude that liver disease or impaired liver function is not a clinically relevant problem in most patients with pulmonary emphysema due to alpha 1-antitrypsin deficiency. But results of quantitative liver function tests, although performed in only a small group of patients, suggest that hepatic metabolization might be impaired even in those patients who present with pulmonary disease.

Rat pancreas after long-term treatment with the somatostatin analogue octreotide

Somatostatin is a potent inhibitor of endocrine and exocrine pancreatic secretion. However, it is not clear whether it also inhibits pancreatic growth. Therefore we treated male Wistar rats with a somatostatin analogue, octreotide (12-192 micrograms/(kg body wt.day)), over a period of 14 days. In a dose-dependent manner, this potent and long-acting analogue caused a reduction in weight of the pancreas and a reduction in pancreatic content of protein, DNA, trypsin, chymotrypsin, amylase and lipase, as well as pancreatic content of insulin-, glucagon- and somatostatin-like immunoreactivities. When growth of rat pancreas was induced by oral administration of camostate (200 mg/(kg body wt. day) or by subcutaneous administration of cholecystokinin (2 x 10 micrograms/(kg body wt. day)) over a period of 14 days, octreotide (12-192 micrograms/(kg body wt.day)) had the same effects, but these were even more pronounced. We conclude that somatostatin is an important regulator of pancreatic growth.

The effect of barium sulphate on small bowel motility in man

OBJECTIVE

To investigate the effect of barium sulphate on small bowel motor activity.

METHODS

Nine healthy male volunteers underwent two separate, ambulatory 24 h manometry studies. Jejunal motor activity was recorded during fasting and after ingestion of 300 ml volumes of barium, water or a glucose solution (1380 kJ). Recordings were analysed visually for recurrence of phase III of the migrating motor complex, and a validated computer program was used to calculate the incidence and amplitude of contractions.

RESULTS

Phase III reappeared 71 +/- 10 min after ingestion of barium sulphate. This interval was not significantly different after the intake of water (60 +/- 8 min) or in the fasted state (88 +/- 21 min), but it was significantly shorter than after ingestion of a glucose solution (136 +/- 17 min). The mean incidence of contractions after ingestion of barium was 3.0 +/- 0.5/min. This was significantly higher than that observed after water (1.9 +/- 0.4/min) and also significantly higher than during fasting (1.5 +/- 0.3/min), but not significantly different from the incidence of contractions after ingestion of a glucose solution (2.2 +/- 0.4/min). The mean amplitude of contractions after ingestion of barium was 24.2 +/- 1.4 mmHg. This was significantly higher than the amplitude of contractions during fasting (19.5 +/- 1.0 mmHg), but not significantly different from that after water intake (23.7 +/- 1.3 mmHg) or after ingestion of a glucose solution (21.4 +/- 1.3 mmHg).

CONCLUSION

Small bowel motor activity after ingestion of barium sulphate differs distinctly both from the interdigestive and from the postprandial motor pattern. Barium suspensions do not interrupt the migrating motor complex with a typical 'fed' pattern, but contractions are more frequent after ingestion of barium than after the intake of water and are both more frequent and of greater force than the contractions observed during the fasted state.

Endocrine and exocrine pancreatic function after camostate-induced growth of the organ

It is well known that oral administration of camostate induces hyperplasia and hypertrophy of the rat pancreas. It is not clear, however, whether pancreatic hormone and enzyme secretion are affected by camostate treatment. In rats, daily administration of 200 mg camostate/kg b. wt for 14 days significantly increased pancreatic weight and pancreatic content of DNA, protein, amylase, lipase, trypsin and chymotrypsin, as well as the amount of insulin, glucagon and somatostatin. In the intact animal, blood glucose levels and serum concentrations of insulin and glucagon in response to an oral glucose load were not impaired after camostate treatment. In the isolated perfused pancreas, however, insulin and glucagon secretions were reduced, whereas somatostatin release was not affected. The volume of pancreatic juice produced by the unstimulated isolated perfused organ, as well as protein and enzyme secretion, were increased after camostate treatment. Likewise, the isolated perfused pancreas from camostate-treated rats secreted a larger volume of pancreatic juice and more protein in response to cholecystokinin (CCK), while enzyme secretion was affected in a non-parallel manner: amylase release was markedly reduced, lipase release was unchanged, and release of trypsin and chymotrypsin was increased.

[Amebic liver abscess with hepatobronchial fistula]

Four weeks after a holiday in Kenya a 57-year-old woman developed a fever up to 40 degrees C, right upper abdominal pain, icteric sclerae, nausea and vomiting. Laboratory tests revealed leukocytosis (24,400/microliters), markedly accelerated erythrocyte sedimentation rate (123 mm/h) and moderately increased activity of liver enzymes in serum. The liver was unremarkable on ultrasound. Four days after hospitalization the patient complained of dyspnoea and pleuritic pain. Ultrasound examination and computed tomography showed an abscess in the right lobe of the liver. Amoebic abscess of the liver being the most likely diagnosis, although the relevant serological tests were unremarkable and a titre increase occurred only later, treatment was started with metronidazole (four times 500 mg daily intravenously) and paromomycin (three times 10 mg/kg daily). Her condition significantly improved within a day. Two weeks later, however, she developed chest pain, dyspnoea and cough productive of large amounts of white-yellow sputum, even though antibiotic treatment was continuing. A transdiaphragmatic rupture of the abscess with formation of a hepatobronchial fistula proved to be the cause of these symptoms. The patient was treated surgically by drainage and suturing-over of the extensive diaphragmatic defect and after 2 weeks she was discharged symptom-free on a maintenance dose of diloxanide furoate (three times 500 mg/d orally).

The islet-acinar axis of the pancreas: is there a role for glucagon or a glucagon-like peptide?

Intravenous glucagon inhibits exocrine pancreatic secretion in vivo, but exogenous glucagon does not affect exocrine secretion in vitro. Recent work, however, suggested that endogenous glucagon may be involved in the regulation of exocrine secretion even in vitro. We therefore investigated the effects of exogenous and endogenous glucagon on exocrine secretion by the isolated perfused rat pancreas in the presence of 1.8 mM glucose. Exogenous glucagon did not affect CCK-stimulated amylase output. 20 mM arginine stimulated glucagon release, but did not affect basal enzyme secretion. CCK-stimulated amylase output, however, was significantly inhibited in the presence of arginine. This inhibitory effect of arginine on exocrine pancreatic secretion could be blocked by glucagon antibodies, but not by nonspecific gammaglobulins. Thus exogenous glucagon failed to affect exocrine pancreatic secretion in vitro, but endogenously released glucagon or a glucagon-like peptide inhibited amylase release in the isolated perfused pancreas. We conclude that glucagon or a glucagon-like peptide may be a mediator in the islet-acinar axis.

Role of somatostatin in regulation of insular-acinar axis

Cysteamine is known to deplete somatostatin from pancreatic D cells. In the isolated perfused rat pancreas we investigated its effects on somatostatin and glucagon release as well as exocrine pancreatic secretion in the presence of 1.8 mM glucose. Cysteamine, 10 mM, released somatostatin, but had no effect on CCK-stimulated amylase secretion. Arginine-stimulated glucagon release, however, was significantly inhibited by cysteamine. Concomitantly we still observed stimulation of somatostatin secretion, but also a potentiation of CCK-stimulated amylase secretion. Our results are consistent with a role of somatostatin in the regulation of exocrine pancreatic secretion via its effect on pancreatic A and B cells.

Pancreastatin--a mediator in the islet-acinar axis?

Pancreastatin was isolated from porcine pancreas in 1986 and has been shown to inhibit insulin release and exocrine pancreatic secretion in vivo. In the isolated perfused rat pancreas, we investigated its effect on the exocrine pancreas and evaluated its indirect effects mediated via the islet-acinar axis. In the presence of 16.7 mmol/L glucose, 20 pmol/L, 200 pmol/L, and 2 nmol/L pancreastatin reduced insulin release but did not affect exocrine pancreatic secretion stimulated by cholecystokinin (CCK), secretin, or bombesin. Pancreastatin also failed to affect unstimulated exocrine pancreatic secretion. In the presence of 1.7 mmol/L glucose, 200 pmol/L and 2 nmol/L pancreastatin inhibited glucagon release and potentiated CCK-stimulated exocrine pancreatic secretion. Inhibition of glucagon release and augmentation of exocrine pancreatic secretion may be independent phenomena, but they could be linked by the islet-acinar axis. Thus we speculate that a pancreastatin-induced inhibition of glucagon release may indirectly have caused augmentation of exocrine pancreatic secretion.

Effect of cysteamine on insulin release and exocrine pancreatic secretion in vitro

Cysteamine is known to deplete somatostatin from pancreatic D cells. In the isolated perfused rat pancreas we investigated its effects on somatostatin and insulin release as well as exocrine pancreatic secretion in the presence of 16.7 mM glucose and 180 pM CCK-8. At a concentration of 0.1 mM, cysteamine had no significant effect on pancreatic endocrine and exocrine functions. At 10 mM, however, cysteamine released somatostatin (380 +/- 70 vs 100 +/- 20 fmol/20 min), inhibited insulin output (890 +/- 120 vs 13210 +/- 3260 mu units/20 min) and reduced exocrine pancreatic secretion (volume: 12 +/- 2 vs 20 +/- 2 microliters/20 min; lipase: 31 +/- 3 vs 60 +/- 7 units/20 min). We conclude that the complex changes induced by cysteamine are consistent with a physiological role of endogenous somatostatin in the regulation of insulin release. The reduction of exocrine pancreatic secretion, however, was at least in part, if not completely, mediated via the insuloacinar axis rather than a direct effect of cysteamine-released somatostatin on pancreatic acinar cells.

Importance of endogenous prostaglandins for the toxicity of cyclosporin A to rat endocrine and exocrine pancreas?

Previous work has shown that cyclosporin A is toxic to the endocrine and exocrine pancreas. The aim of this study was to examine whether endogenous eicosanoids play a role in controlling cyclosporin A induced toxicity. Rats were treated for eight days with indomethacin (2 mg/kg, twice daily) in addition to cyclosporin A (5 or 10 mg/kg daily). Effects of drug treatments on exocrine (as assessed by amylase and protein secretion into the pancreatic juice) and endocrine (as assessed by the glucose dependent insulin release) pancreatic functions, and pancreatic formation of prostaglandins and thromboxane were evaluated. Treatment with cyclosporin A in the doses used did not inhibit eicosanoid formation by the pancreatic tissue ex vivo. Indomethacin caused significant inhibition of pancreatic formation of prostaglandin E2, 6k prostaglandin F1 alpha and thromboxane B2. Combined treatment with indomethacin and cyclosporin A (5 or 10 mg/kg) augmented cyclosporin A induced pancreatic toxicity with further impairment of insulin release, amylase secretion, and pancreatic juice protein content, but did not result in more pronounced inhibition of pancreatic eicosanoid formation. The increased toxicity of the combined treatment was, however, associated with raised cyclosporin A whole blood concentrations. The data suggest that the potentiation of pancreatic toxicity of cyclosporin A observed during coadministration of indomethacin is not the result of suppression of endogenous pancreatic eicosanoid biosynthesis, but more likely results from altered cyclosporin A pharmacokinetic which may be caused by an interference of indomethacin with the hepatic cytochrome P-450 dependent monooxygenase involved in cyclosporin A metabolism. The possibility that coadministration of non-steroidal antiinflammatory drugs aggravates toxic effects in cyclosporin A treated patients should be considered.

Stimulatory and inhibitory effects of galanin on exocrine and endocrine rat pancreas

The influence of the neuropeptide galanin, present in intrapancreatic nerve endings, on the endocrine pancreas is well known. The most potent effect of galanin is inhibition of insulin release. Little is known of its effect on the exocrine pancreas. Whether galanin plays a role in the regulation of exocrine pancreatic secretion and whether this effect is mediated directly on acinar cells or indirectly via the influence on insulin secretion is not clear. In the present study, we investigated these questions using the model of the isolated and arterially perfused rat pancreas with intact exocrine and endocrine secretion. In the presence of 15.8 mM glucose in a modified Krebs-Ringer buffer and during half-maximal stimulation of enzyme secretion with 100 pmol/ml cholecystokinin octapeptide (CCK-8), a dose-response study of 0.001-100 pmol/ml porcine galanin was performed. At concentrations of 0.001 and 0.01 pmol/ml, porcine galanin significantly stimulated insulin release (p < 0.05 and < 0.01, respectively) and also significantly enhanced CCK-8-stimulated amylase secretion (p < 0.05). Doses of 0.1 and 1 pmol/ml galanin resulted in a nonsignificant inhibition of insulin release, while 10 and 100 pmol/ml strongly inhibited the endocrine response (p < 0.001). However, concentration levels of 1-100 pmol/ml galanin did not affect CCK-8-stimulated amylase secretion. Rat galanin, tested at 0.01 and 10 pmol/ml, showed no significant difference from the effects of porcine galanin at the equipotent concentrations. It is concluded that the effect of galanin on exocrine pancreas, like the effect on endocrine functions, tends to be a direct one and that it could exert a modulatory influence on the level of neuronal transmission.

[Hemangiosarcoma of the liver. The diagnostic difficulties and therapeutic possibilities]

A 65-year-old man complained of decreasing physical capacity and weakness over the preceding six months, associated with marked painless jaundice and subsequently ascites. Despite extensive tests, some invasive, in three different hospitals no cause was found of the patient's symptoms and the marked though nonspecific abnormalities of various biochemical values (raised bilirubin concentration; increased alkaline phosphatase activity). Liver transplantation was performed because of progressive liver failure, without a firm diagnosis being established. At operation the liver was found to contain a haemangiosarcoma. The patient died 14 months after the transplantation of a suppurative cholangitis. At autopsy neither metastases nor recurrences were found.

Identification of Somatostatin-14 and -28 in rat pancreatic juice by a new HPLC method

Recently we could demonstrate that in rats with pancreatic hypertrophy, somatostatin is secreted in higher concentrations into the pancreatic juice than into the portal vein blood. For measurement of juice somatostatin and to characterize the molecular forms, we established a new reverse-phase HPLC method, which we describe herein. This HPLC method, using a linear gradient system consisting of 0.2% heptafluorbutyric acid in 10 mM sodium acetate and acetonitrile, showed a stable recovery rate of about 85%. Applying the pure juice to this gradient system, we detect somatostatin-14 to be the major form of immunoreactive somatostatin (IRSS) in the pancreatic juice of the rat (5% of total IRSS). The remaining 35% were found to be somatostatin-28. The role of somatostatin in pancreatic juice is not known. It raises the hypothesis that it possibly interacts with the influences intraluminal intestinal growth factors. This study supports the assumption for the existence of an insuloacinar portal system to regulate exocrine pancreatic functions by islet hormones.

Glucose-dependent effects of pancreastatin on insulin and glucagon release

Pancreastatin (PST), a peptide isolated from porcine pancreas in 1986, has been reported to inhibit insulin and to stimulate glucagon secretion. Since both of these effects have been questioned, we investigated the effect of PST (20, 200, or 2000 pM) on hormone release in the isolated perfused rat pancreas at different glucose levels (1.7, 5.5, 11.1, and 16.7 mM). At 1.7 mM glucose, 20 pM PST had no significant effect on glucagon secretion, whereas 200 pM and 2 nM PST significantly inhibited glucagon release. At a concentration of 5.5 mM glucose, insulin output was not affected by PST in any of the concentrations tested. At 11.1 mM glucose, however, 200 pM and 2 nM PST significantly inhibited insulin output. At 16.7 mM glucose, insulin secretion was significantly reduced by all concentrations of PST tested. Unstimulated exocrine pancreatic secretion was not affected by PST in any of the experimental settings. We conclude that PST inhibits glucagon and insulin secretion dose-dependently, and these effects apparently are glucose-dependent. PST does not influence basal exocrine pancreatic secretion in vitro.

[Effect of pancreastatin on insulin secretion and the exocrine pancreas in rats]

Pancreastatin, a 49-amino-acid C-terminal amidated peptide, was isolated from porcine pancreas in 1986. It has been reported to inhibit insulin release and exocrine pancreatic secretion, but both these effects have been disputed. In the isolated perfused rat pancreas we therefore studied the effect of pancreastatin on insulin and exocrine pancreatic secretion. Neither basal exocrine pancreatic secretion, nor exocrine secretion stimulated by CCK-8, bombesin or secretin were affected by pancreastatin. 20 or 200 pM pancreastatin, however, significantly inhibited stimulated insulin release. We conclude that pancreastatin seems to be yet another inhibitory peptide, which--for unknown reasons--inhibits insulin release both in vivo and in vitro, but exocrine pancreatic secretion only in vivo.

Independence of GIP-induced insulin secretion from sympathetic and parasympathetic innervation in the isolated perfused rat pancreas

The incretin candidate GIP (gastric inhibitory polypeptide) is released from the gut by nutrients and can stimulate insulin secretion. Metabolic and hormonal factors have been shown to modulate insulin response to GIP. It is unknown, however, whether the autonomic nervous system, which itself controls insulin secretion, can modulate the insulinotropic effect of GIP. In the isolated perfused rat pancreas, we therefore investigated the influence of sympathetic and parasympathetic agonists and antagonists on the insulin response to GIP. As compared to control (6990 +/- 890 microU/10 min), the effect of either acetylcholine (29030 +/- 4600 microU/10 min), atropine (5880 +/- 1740 microU/10 min), norepinephrine (2520 +/- 750 microU/10 min), phentolamine (11380 +/- 1910 microU/10 min), isoproterenol (12740 +/- 2090 microU/10 min), propranolol (5600 +/- 880 microU/10 min), or GIP (29660 +/- 4490 microU/10 min) on insulin secretion was consistent with previous reports. The effects of the combined administration of GIP and either acetylcholine (48140 +/- 7540 microU/10 min), phentolamine (43930 +/- 4490 microU/10 min), norepinephrine (9000 +/- 1740 microU/10 min), or isoproterenol (36280 +/- 5210 microU/10 min) on insulin release were additive. Insulin response to GIP was resistant to atropine (24210 +/- 9470 microU/10 min) and propranolol (26450 +/- 4930 mu/10 min). We conclude that both GIP and the autonomic nervous system influence insulin secretion, but that they do so independently from each other.

[Hereditary angioedema type II with predominantly abdominal symptoms]

For two years a 19-year-old patient had been complaining of recurrent abdominal symptoms consisting of severe colicky pain, nausea, vomiting and diarrhoea. Occasionally he noticed concomitant subcutaneous swellings in the limbs. There were no swellings in the face or the region of the upper airway. The symptoms occurred once or twice a month, spontaneously receding within 3-4 days. These signs and symptoms indicated hereditary angio-oedema. Serum concentration of C1-esterase inhibitor was normal (13 mg/dl), but its serum activity was reduced to 16% (normal range 80-125%). This is thus a case of type II C1-esterase inhibitor deficiency. Since starting low-dosage treatment with Danazol (initially, for two months, 200 mg daily, then 200 mg five times per week) the patient has been symptom-free.