Pathological esophageal acidification and pneumatic dilitation in achalasic patients. Too much or not enough?

Endoscopy, esophageal manometry and pH monitoring, gastric emptying test, and heartburn quantification on a visual analog scale were performed in 22 achalasic patients in order to clarify which events are associated with pathological esophageal acidification after successful LES dilatation. Five patients presented pathological acidification. Dilatation reduced LES tone from 38.3 +/- 4.2 to 14.6 +/- 1.1 mm Hg (mean +/- SEM); there was, however, no difference between nonrefluxers and refluxers (14.8 +/- 1.2 vs 13.8 +/- 2.5 mm Hg). The emptying time in achalasic patients was delayed compared to controls (315.9 +/- 20.9 min vs 209 +/- 10.4) due to prolonged lag-phase and reduced slope of the antral section-time curve, but, again, there was no difference between refluxers and nonrefluxers. The acid clearance was delayed in refluxers compared to nonrefluxers (15.9 +/- 4.5 vs 2.5 +/- 1.8 min, P<0.05). Two refluxers presented grade 1 esophagitis; one of them developed an esophageal ulcer. The heartburn score was the same in refluxers and nonrefluxers. Pathological acidification after pneumatic dilatation is associated with persistent problems in esophageal emptying rather than with excessive sphincter divulsion.

Medium-chain triglyceride absorption in patients with pancreatic insufficiency

BACKGROUND

The role of medium-chain triglycerides (MCTs) in the management of patients with pancreatic insufficiency is controversial. The aim of the study was to evaluate the absorption of MCTs in the presence of pancreatic insufficiency and the effect of pancreatic extracts on MCT absorption so as to clarify whether the replacement of usual dietary fats with MCTs is cost-effective.

METHODS

Six patients with severe pancreatic steatorrhea were for 5 days fed a low-fat diet to which butter (long-chain triglycerides (LCTs)) or MCT oil was added, with and without pancreatic extracts, in a crossover design.

RESULTS

Fecal weight and nitrogen losses were the same during MCT and LCT intake. Steatorrhea was substantial during both periods but was significantly lower during MCT than LCT intake. Fecal weight and nitrogen and fat losses were reduced by pancreatic extracts in both diets. Steatorrhea was the same when MCTs and LCTs were consumed together with pancreatic extracts.

CONCLUSIONS

MCTs are absorbed better than LCTs in the presence of pancreatic insufficiency but require pancreatic extracts for optimal absorption. No advantage is to be expected from replacing usual dietary fats with MCTs if pancreatic supplements are used.

Gastric emptying of a solid meal is accelerated by the removal of dietary fibre naturally present in food

Exogenous fibre added to liquid meals delays gastric emptying. Its effect on solid meals is uncertain, and nothing is known of the effect on gastric emptying of fibre naturally present in food. This study therefore looked at gastric emptying of two different solid meals in eight healthy subjects and their blood glucose responses. The meals were exactly equivalent except for the total dietary fibre content (high fibre 20 g, low fibre 4 g of dietary fibre per 1000 kcal) and supplied 870 kcal (700 kcal women), 47% of which was from carbohydrates, 36% from fats, and 17% from proteins. Ultrasonography was used to measure antral diameters before the meal (basal), immediately after it (time 0), and at 30, 60, 120, 180, 240, and 300 minutes. In addition, subjects filled in a questionnaire on their feelings of hunger, epigastric fullness, and satiety before the meal and at hourly intervals after it. Basal and maximal postprandial antral sections were similar for the two meals (basal section: 283.9 (29.5) v 340.9 (44.7) mm2 for the low and the high fibre meal, NS; maximal postprandial section: 1726 (101.9) v 1593 (120.4) mm2, NS). Total gastric emptying time was significantly reduced by fibre removal (186.0 (15.6) v 231.7 (17.3) minutes after the low and the high fibre meal, p < 0.05). Blood glucose was higher after the low fibre meal, and the area under the glycaemic curve significantly greater (226 (23.1) v 160 (20.0) mmol/min/dl-1, p < 0.05). No difference was found in satiety or fullness feelings, but hunger returned more rapidly after the low fibre meal. In conclusion, fibre naturally present in food delays gastric emptying of a solid meal, reduces the glycaemic response, and delays the return of hunger.

Effect of neutralized and native vinegar on blood glucose and acetate responses to a mixed meal in healthy subjects

OBJECTIVE

To investigate the influence of sodium acetate and acetic acid from vinegar on blood glucose and acetate response to a mixed meal in healthy subjects.

DESIGN

Five healthy subjects consumed in random order six test meals consisting of 100 g of sliced lettuce dressed with olive oil (Blank), olive oil plus 1 g acetic acid in the form of vinegar (AcOH), or olive oil plus sodium acetate in the form of vinegar neutralized to pH 6.0 with sodium bicarbonate (AcNa). On three occasions test meals were followed by a challenge consisting of 50 g carbohydrate portions of white bread (Bread). Glucose and acetate concentrations were measured in arterialized capillary blood before and until 95 min after the meals. Ultrasonography was performed in four other subjects to measure gastric emptying times after AcOH + Bread and AcNa + Bread.

RESULTS

Blood acetate response over 95 min was markedly reduced after AcOH and AcOH+Bread meals compared to AcNa and AcNa + Bread. Similarly, the glucose response was depressed by 31.4% (P = 0.0228) after AcOH+Bread with respect to AcNa + Bread and Blank + Bread. No difference was observed between gastric emptying times after AcOH + Bread and AcNa + Bread.

CONCLUSIONS

The results suggest that oral acetic acid and acetate might have a different effect on acetataemia and that a limited dose of vinegar, in the form of salad dressing, is sufficient to influence significantly the glycaemic response to a mixed meal in normal subjects by a mechanism related to acidity but not to gastric emptying.

Tussive effect of capsaicin in patients with gastroesophageal reflux without cough

The aim of this study was to clarify the influence of gastroesophageal reflux (GER) on cough threshold in patients with digestive symptoms but free from respiratory involvement. Of 57 consecutive subjects referred for 24-h esophageal pH monitoring because of digestive reflux symptoms, 29 patients free from respiratory disorders were studied. They underwent esophageal pH monitoring and manometry, upper gastrointestinal endoscopy, pulmonary function tests, and methacholine and capsaicin challenges. The methacholine test was performed by inhalation of increasing doses of methacholine up to 4,000 micrograms; the results were expressed as the dose causing a 20% decrease in FEV1 from baseline (PD20). The capsaicin threshold was evaluated by inhalation of increasing doses of capsaicin from 0.3 up to 9.84 nmol, expressing the results as the dose of capsaicin eliciting five coughs (PD5). Fifteen patients were considered refluxers on the basis of a total esophageal acid exposure time above 4.7%. Esophagitis grade 0 was found in 15 patients, grade 1 in seven patients, grade 2 in seven patients. PD5 was significantly lower in refluxers (median 0.51 micrograms, range 0.22 to 19.8) than in nonrefluxers (19.8 micrograms, range 0.31 to 19.8) (p < 0.001); there was no difference in baseline ventilatory parameters and in airway responsiveness to methacholine between the two groups. All patients with a pathologic acid exposure time but one had a low cough threshold, irrespective of the presence or absence of esophagitis.(ABSTRACT TRUNCATED AT 250 WORDS)

Gastric emptying of solids is markedly delayed when meals are fried

We studied the effect of heat-treated fats on gastric emptying. Eight healthy asymptomatic volunteers (five males; age 28-41 years) ate on different days and in random order two meals identical in contents (pasta, tomato, beef, olive oil, carrots, orange, water; 870 kcal males, 700 kcal females; 47% of calories from carbohydrate, 36% from fat, 17% from protein), but cooked differently (fats fried or not). Ultrasound measurement of antral diameters was used to calculate basal antral section, its maximal dilation after the meal, the time necessary for total emptying, and the percent retention at hourly intervals. No difference was found in basal and maximal antral diameters after the two meals. On the contrary, total gastric emptying was significantly delayed after the fried meal [317.1 (24.12) vs 226.7 (18.4) min, mean (1 SEM); P < 0.002]. A significantly greater percentage of maximal antral distension was still present between 120 and 240 min after the fried meal. The glycemic response and hunger feeling were the same after the two meals, whereas there was a longer persistence of satiety and epigastric fullness after the fried meal. In conclusion, gastric emptying can be influenced not only by the meal content, but also by the way it is cooked.

Acetylcholinesterase protection and the anti-diisopropylfluorophosphate efficacy of E2020

The reversible noncovalent inhibitor of acetylcholinesterase (R,S)-1-benzyl-4-[(5,6-dimethoxy-1-indanon)-2-yl]-methylpiperidine hydrochloride (E2020) was shown to inhibit electric eel acetylcholinesterase with high affinity in a mixed competitive-non-competitive way (Ki = 8.2 nM; Ki' = 13 nM). The pretreatment of electric eel acetylcholinesterase with E2020 dose-dependently prevented the inactivation of the enzyme by 40 microM diisopropylfluorophosphate. The EC50 for this protective effect (95% confidence limits) was 85 (76-96) nM, whereas under the same conditions E2020 IC50 was 12.3 (9.6-16) nM. E2020 injected together with atropine sulfate (17.4 mg/kg) into mice at doses in the range of 1.04-6.24 mg/kg 15 min before diisopropylfluorophosphate, caused a dose-dependent increase in diisopropylfluorophosphate LD50, resulting in protection ratios varying from 3.1 to 9.2. The effectiveness of E2020 antidotal effect was inversely correlated to the time between pretreatment and diisopropylfluorophosphate administration, being maximal when E2020 was injected 15 min, and possibly less than 15 min, before poisoning. From these experiments it is concluded that E2020 exerts a protective action against acute diisopropylfluorophosphate-poisoning in the mouse, presumably by protecting acetylcholinesterase from irreversible inactivation by this agent.

Gastric emptying of solid meals in achalasic patients after successful pneumatic dilatation of the cardia

Gastric emptying of a solid meal was measured by ultrasound scanning in 16 achalasic patients following successful pneumatic dilatation of the lower esophageal sphincter. The data were compared with those of a control group of 15 healthy subjects. Fasting and maximal postcibal antral sections were very similar in the two groups. On the contrary, the time interval before maximal antral dilatation, and the time necessary for the emptying of half or of the whole meal were significantly longer in the achalasic patients than in the controls. Half of the achalasic patients had longer emptying times than the upper normal limit. The percentage of the meal retained in the antrum at each hourly interval was significantly higher in the achalasic group. The finding of a high prevalence of gastric emptying disturbances suggests that the functional derangement in achalasia is not limited to the esophagus.

Pancreatic extracts are necessary for the absorption of elemental and polymeric enteral diets in severe pancreatic insufficiency

Tube feeding nutrition, either elemental or polymeric, is increasingly used in patients with digestive problems. Pancreatic insufficiency is a widely accepted indication for the use of an elemental formula, which requires less residual digestive capacity. To confirm this assumption, we have compared the absorption of elemental and polymeric diets and the effect of exogenous pancreatic enzymes in a patient on long-term total enteral feeding after total pancreatectomy. Malabsorption of both formulas was observed without enzyme supplementation. A marked improvement of fat and nitrogen absorption was obtained when pancreatic enzymes were added to both enteral diets. It is concluded that pancreatic enzymes should always be added to liquid diets in pancreatic insufficiency. No clear advantage is to be anticipated by the use of elemental as compared with polymeric diets.

[In vitro study of a new pancreatic enzyme with high lipase content in enteric coated microtablets]

In vitro behaviour of the enteric-coating of a new pancreatic enzyme containing preparation in enteric-coated microtablets was evaluated, by incubating at 37 degrees C, under gentle agitation, the preparation under study in buffered gastric (pH 2, 3, 4, 5, 6) and in buffered duodenal juice (pH 3, 4, 5, 6, 7, 8) for 30, 60, 90, 120 minutes. Lipase and chymotripsin activities were measured at each time and pH in the solution and in the undissolved microtablets. The drug under study showed a good enteric-coating, preserving about 100% of the enzyme content, when incubated in buffered gastric juice, up to pH 6, and releasing in the solution, during incubation in buffered duodenal juice, 75% of its enzyme content at pH 7 and 8. Therefore, the pancreatic enzymes contained in this new enteric-coated microtablet preparation are well protected against inactivation from acid and are bioavailable for digestion of alimentary substrates at optimal or near optimal pH in duodenal juice.

[A protocol for evaluating screen-film systems]

When a competition is made for the supply of sensitive photographic material, each USL is asked, in a time scan of one or more years, to evaluate the characteristics of the screen-film systems proposed by the various companies. In the last years the authors had several opportunities to check different screen-film systems for many Friuli-Venezia Giulia hospitals. For this reason they have set an evaluation form tested after following changes and recently employed in its final version. The check form is reported on in this paper with some personal considerations originated by the authors' experience. The technical parameters which must be evaluated are reported, together with the instrumentation needed to perform the technical examination, the examination modality, and the score-attributing criteria. The evaluation form is very flexible for it allows each Radiology Department to evaluate each parameter in a personal way and to fit it to its own needs.

Near infrared reflectance measurement of nitrogen faecal losses

Chemical methods of measuring nitrogen in stools are complex, unpleasant, and therefore rarely performed. Recently, near infrared reflectance (NIRA) has been suggested for stool analysis. The aim of this study was to evaluate the possible application of this method in routine faecal nitrogen measurement. Nitrogen concentration and daily output were measured in the stools of 83 patients using NIRA and, for comparison, the Kjeldahl method. Nitrogen concentration and output ranged between 0.4-2.72 g% and 0.45-8.96 g/day respectively. Correlation coefficients (r), of 0.89 and 0.97 were found between the two methods for concentration and output respectively, and similar values were found in patients on enteral nutrition. Repeated measurements from the same stool collection, requiring only a few minutes, allowed homogenisation to be avoided. NIRA seems to be an easy, fast, and reliable alternative to chemical assays of nitrogen measurement in the management of patients with digestive disorders.

Lipase latex test for acute abdominal pain: comparison with serum lipase, trypsin, elastase and amylase

The diagnostic capacity of a semiquantitative latex test for lipase measurement was compared with the measurement of other pancreatic enzymes in 100 consecutive patients admitted to a general hospital for recent onset of severe abdominal pain. Positive results of the test were found in two patients with acute pancreatitis, and in one out of three chronic pancreatitis relapses. The test yielded false-positive results only in two patients who had no apparent pancreatic involvement. A marginal increase in other pancreatic enzymes was found in a few patients with acute biliary or appendicular problems. In conclusion, the lipase latex test can be suggested in an emergency setting as a quick and reliable alternative to serum amylase to rule out a diagnosis of acute pancreatitis.

Fecal fat concentration in the screening of steatorrhea

We studied if the fecal fat concentration as measured by the near infrared reflectance analysis in a spot sample is an acceptable screening test for malabsorption. This measurement was compared with the more complex fat balance in 120 patients with a suspected malabsorption [53 with chronic pancreatic disorders (CP), 67 with other digestive disorders (nCP)]. The fecal fat concentration proved to be well correlated with steatorrhea in CP (r = 0.86) but not in nCP (r = 0.35). A fat concentration of 9 g% had a sensitivity and a specificity for steatorrhea of 88.8% and of 97.1% in CP, but only of 53.8% and of 94.4% respectively in nCP. The fecal fat concentration was significantly higher in CP than in nCP, even considering patients with steatorrhea only; however, the overlap between the two groups was too high to suggest a clinical usefulness of this test in the differential diagnosis of steatorrheas. It is concluded that the fat concentration in a small sample, easily obtained also in outpatients, is useful in the selection of patients with chronic pancreatitis to submit to a proper fat balance study.

Variations in time of serum pancreatic enzyme levels in chronic pancreatitis and clinical course of the disease

Thirty-four patients with chronic calcified pancreatitis were evaluated clinically and biochemically (at a time when painful relapses were not present) every 9 mo for 3 yr; 25 of them were also studied at 4 and 9 yr. Serum elastase-1, trypsin, lipase, and amylase in the same sera were measured at each visit; levels on entry and variations during the study were compared with the clinical and functional data of the patients. On entry, low levels of elastase-1 were found in 11.7% of the patients, high levels in 41.1%; in contrast, high levels of trypsin and lipase were found in only a small number of patients (5.8 and 11.7%, respectively), whereas low levels were present in a substantial number (47.8 and 32.3% for trypsin and lipase, respectively). Over time, we found a significant (p = 0.000002) reduction in elastase-1 levels. Such reduction was not found for trypsin, lipase, or amylase. The reduction of serum elastase-1 was significantly (p less than 0.003) more frequent in patients presenting a reduction in painful relapses than in patients with a stable or increased attach rate; this association was weaker (p less than 0.05) for lipase and trypsin, and absent for amylase. No correlation was found between circulating enzymes and either alcohol consumption or age of patients. In patients with severe exocrine impairment, low levels of elastase were found in only 20% of the cases, whereas trypsin and lipase were reduced in 73.3 and 53.3% of the cases, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Isoamylase determination by isoelectric focusing in pancreatic disorders. A potential clinical aid

Isoamylase analysis by isoelectric focusing was performed in the serum of 30 healthy volunteers, 65 patients with acute or chronic pancreatic diseases, nine with acute abdomen, four with macroamylasemia, and four with duodenal duplication. In controls, up to four fractions (2 salivary, 2 pancreatic) were found; the pancreatic fractions were as a mean 44.7% (SD 8.6) of total. In chronic pancreatitis, only patients with steatorrhea showed a significant reduction of pancreatic isoamylase (p less than 0.001). In all patients with acute pancreatitis or pseudocysts, an additional fraction (similar to the so-called P3 fraction) was resolved. Moreover, additional isoenzymes were found in all patients with severe acute pancreatitis or pseudocysts, and not in controls or patients with mild forms, acute abdomen or duodenal duplication. A similar pattern was shown in a stored control serum after 10 mo at -20 degrees C. These fractions disappeared after successful surgical drainage. No specific alteration was found in pancreatic cancer. Amylase fractionation by isoelectric focusing can be used to confirm an acute pancreatitis, and to monitor patients with pancreatic pseudocysts and collections after surgical drainage.

Pancreatic calcifications in patients with chronic pancreatitis. A sign of long-lasting or severe disease?

Pancreatic calcifications are particularly frequent in patients with severe pancreatic insufficiency and long-lasting chronic pancreatitis. To clarify whether calcifications point to a more severe form of the disease, irrespective of its duration, we have retrospectively analyzed patients with chronic pancreatitis submitted to the secretin-cerulein test in our center over a six-year period. Out of 120 patients, calcifications were found in 55. Higher alcohol intake and longer duration of the disease were found in patients with calcifications, compared with patients without calcifications (p less than 0.001). In both groups, lipase and chymotrypsin were more severely impaired than bicarbonate; a greater reduction of pancreatic exocrine function was found in patients with calcifications, compared to those without (p less than 0.001, Mann-Whitney U-test). When the patients were classified according to the duration of the disease or the severity of exocrine function impairment, higher percentages of patients with calcifications were found in the classes with more advanced disease. A log-linear analysis showed that the prevalence of calcifications was associated with pancreatic function impairment, even within the same class of duration of the disease. It is likely that calcifications mark more severe forms of chronic pancreatitis, even in the early phases of the disease.

Near infrared spectrometry for faecal fat measurement: comparison with conventional gravimetric and titrimetric methods

This investigation was aimed at comparing a new method for measuring faecal fat excretion, carried out with a semi-automated instrument by using near infrared analysis (NIRA), with the traditional titrimetric (Van de Kamer) and gravimetric (Sobel) methods. Near infrared analysis faecal fat was assayed on the three day stool collection from 118 patients (68 chronic pancreatitis, 19 organic diseases of the gastrointestinal tract, 19 alcoholic liver disease, 12 functional gastrointestinal disorders). A strict linear correlation was found between NIRA and both the titrimetric (r = 0.928, p less than 0.0001) and the gravimetric (r = 0.971, p less than 0.0001) methods. On homogenised faeces, a mean coefficient of variation of 2.1 (SD 1.71)% was found. Before homogenisation (where a mean coefficient of variation of 7% was found) accurate results were obtained when the mean of five measurements was considered. In conclusion, the assay of faecal fat excretion by the near infrared reflessometry appears a simple, rapid and reliable method for measuring steatorrhoea.

The fecal chymotrypsin photometric assay in the evaluation of exocrine pancreatic capacity. Comparison with other direct and indirect pancreatic function tests

Fecal chymotrypsin (FCT) has been measured by a new photometric method (Monotest Chymotrypsin; Boehringer, Mannheim) in 78 patients: 44 with chronic pancreatitis and 34 not affected by any pancreatic disease. The results were compared with those from other tests of pancreatic secretory (secretin-cerulein test) and digestive [serum and urinary p-aminobenzoic acid (PABA) and pancreolauryl] capacity. When FCT values were severely reduced (below 6.7 U/g), from 90 to 100% of the patients also presented abnormal pancreatic secretory and digestive capacity. On the other hand, 87% of the patients with normal FCT (above 20 U/g) presented normal secretory and digestive capacity. Patients with intermediate FCT values (between 6.7 and 20 U/g) showed normal or abnormal pancreatic secretory and digestive capacity with the same probability. Therefore, FCT, carried out as a first test, seems to identify subjects that need no further pancreatic function tests (normal and severely impaired FCT) and patients who need other more complex functional investigations (intermediate FCT values).

Urinary elastase 1 in chronic pancreatic disease

Serum and urine elastase 1, its renal output and clearance and urinary gamma-glutamyltransferase and ribonuclease excretions were measured in 16 patients with pancreatic cancer, 23 with chronic pancreatitis and in 22 healthy controls in order to evaluate elastase 1 plasma-urine transfer in chronic pancreatic disease and to investigate any factors that might influence the clearance of this enzyme. In an additional group of 17 patients with different pancreatic diseases the serum molecular size distribution of elastase 1 after chromatography was ascertained. An increased urinary elastase 1 output was found in 4/16 patients with pancreatic cancer and in 6/23 with chronic pancreatitis. No correlation was found between circulating elastase 1 and its urinary output; a negative correlation was detected between the serum levels of this enzyme and its clearance. The excretion of ribonuclease and gamma-glutamyltransferase was correlated with elastase 1 output and clearance. While the majority of elastase 1 in serum was accounted for by high molecular forms, probably the expression of complexes with serum inhibitors, free circulating enzyme was present in all patients with high serum elastase 1. Our findings suggest that elastase 1 urinary excretion increases in some patients with chronic pancreatic disease regardless of the neoplastic or inflammatory nature of the illness. Although the availability of different amounts of ultrafiltrable enzyme may play a role in influencing elastase 1 plasma-urine transfer, renal tubular damage appears to be the most important factor influencing the increase in the urinary output of elastase 1.

Fractional urinary clearance of immunoreactive lipase in chronic pancreatic disease

Using an immunoenzymatic method, we studied lipase in the serum and urine of 23 controls, 22 chronic pancreatitis patients in symptomatic remission, and in 9 patients with proven pancreatic cancer. Serum and urine lipase and its fractional urinary clearance were compared with those of amylase and immunoreactive trypsin. Lipase immunoreactivity was detectable in the urine of 81.5% of the studied subjects (controls: 82%, chronic pancreatitis: 86%, pancreatic cancer: 66%); its output was higher than the upper limit of controls in 31.8% of chronic pancreatitis and in only 1 of pancreatic cancer, and it was significantly correlated with the urinary output of trypsin (r = 0.487, P less than 0.001), but not with that of amylase. A significant correlation was found between urinary output and serum levels for lipase, but not for trypsin or amylase. Fractional clearance of lipase was of the same magnitude as that of trypsin but only 0.1% that of amylase. 19% of chronic pancreatitis and pancreatic cancer showed a fractional clearance of lipase above the upper limit of controls, compared with 45% for trypsin and 3.2% for amylase. No difference in urinary clearance of the three enzymes was found between chronic pancreatitis and pancreatic cancer. In conclusion, although of no diagnostic relevance in pain-free patients with chronic pancreatic disease, this measurement can provide information on the mechanisms of renal excretion of pancreatic enzymes.

A clinical evaluation of monoclonal (CA19-9, CA50, CA12-5) and polyclonal (CEA, TPA) antibody-defined antigens for the diagnosis of pancreatic cancer

We measured in 193 patients, admitted to our wards for symptoms and signs suggestive of pancreatic or digestive malignancy, the serum levels of five tumor-associated antigens (CA 19-9, CA 50, CA 125, TPA, CEA) and we evaluated their diagnostic accuracy both when used alone and in combination. For CA 19-9 and CA 50 a sensitivity for pancreatic cancer as high as 92 and 88%, respectively, and specificity of 91.8% were found. A lower sensitivity vs. pancreatic cancer was found for the other tumor markers, and vs. the other digestive and nondigestive malignancies for all tumor markers (apart for CA 19-9 and CA 50 vs. biliary carcinomas). As for the combined assays, the best figures were found vs. pancreatic cancer for CA 19-9 plus CA 50, CA 50 plus CEA, CA 50 plus CA 125; a sensitivity by far worse vs. the other gastrointestinal cancers was found for all the possible combinations. We conclude that in selected symptomatic patients some tumor-marker determinations can be useful in identifying those with a high probability of harboring a pancreatic cancer, to be further studied or operated upon. The clinical relevance of this in patients already symptomatic is at present unclear.