Hemorrhage into a hepatic adenoma and type Ia glycogen storage disease: a case report and review of the literature.

Older patients with type I glycogen storage disease (GSD) develop hepatic adenomas that may undergo malignant transformation. Despite their similarity to oral contraceptive-related hepatic tumors, only one previous report has even mentioned hemorrhage in GSD-related hepatic tumors. We recently followed a 20-year-old patient with type Ia GSD and a 10 cm focal defect in the left lobe of the liver; angiography suggested that this was a benign adenoma. At 22 years of age, after an acute symptomatic episode, repeat studies (ultrasonography and angiography) revealed a 2 cm increase in diameter of the hepatic mass. Imminent tumor rupture was of grave concern; thus the patient was admitted to the hospital and given 2 weeks of constant glucose administration by central venous line in the hope of improving her metabolic abnormalities. After resolution of the coagulopathy and metabolic disorders, the patient safely underwent surgical enucleation of the tumor. Pathologic examination of the tumor revealed that the patient had indeed hemorrhaged into a typical hepatic adenoma that had focuses of hepatocellular dysplasia. She has done well without evidence of tumor recurrence for 3 years since the operation. We conclude that hemorrhage and malignant transformation are potential complications of GSD-related hepatic adenomas. This conclusion underscores the importance of following these patients closely as they age. Nocturnal nasogastric feeding should be considered in the hope of preventing a tumor or inducing regression. Acute symptomatic attacks should be evaluated promptly for possible tumor hemorrhage.

Percutaneous aspiration of peripancreatic fluid collections: a safe method to detect infection.

During the past 5 years, we have used percutaneous aspiration of peripancreatic fluid collections guided by computed tomography (CT) or ultrasonography (US) to facilitate diagnosis of infection in selected cases. Fifteen of 18 patients undergoing guided needle aspiration had persistent fevers (greater than 38.3 degrees C). The three afebrile patients all had abdominal pain and leukocytosis, and two of the three also had elevated serum amylase levels. Percutaneous aspiration was guided by CT in 14 patients and by US in four. On the basis of aspirate Gram stains and cultures, as well as surgical (15) and percutaneous drainage (1) findings, the final diagnosis was pancreatic abscess in nine patients, infected pseudocyst in four, uninfected pseudocyst in four, and cystadenoma in one. Diagnosis based on percutaneous aspiration was correct in 17 of 18 patients (94%), and no complications could be directly attributed to the procedure. We conclude that CT- or US-guided percutaneous aspiration is a safe and accurate diagnostic procedure for patients with peripancreatic fluid collections in whom secondary infection is suspected.

Pancreatic denervation does not influence glucose-induced insulin response.

BACKGROUND: Pancreatic transplantation results in denervation and loss of splanchnic venous drainage and inflicts numerous metabolic abnormalities. However, it is unclear whether denervation or loss of splanchnic venous drainage is responsible for the observed metabolic abnormalities. METHODS: To discern denervation's role in these abnormalities, four mongrel dogs underwent extrinsic pancreatic denervation with preservation of splanchnic venous drainage. These animals, as well as four innervated control subjects, underwent standardized enteral and intravenous glucose tolerance testing. In addition, hyperglycemic clamps that maintained stable serum glucose elevations at either 2.8 or 8.3 mmol/L above basal were also performed. RESULTS: Prestimulated glucose (90.4 +/- 2.7 vs 92.6 +/- 4.9 mg/dl) and insulin levels (6.8 +/- 1.7 vs 8.5 +/- 1.4 muU/ml) did not differ between innervated and denervated groups. Integrated incremental enteral glucose (5320 +/- 1900 vs 7790 +/- 2000 mg/dl) and insulin (2565 +/- 350 vs 2836 +/- 598 muU/ml) levels did not differ between groups. Integrated incremental intravenous glucose (3680 +/- 400 vs 3950 +/- 1000 mg/dl) and insulin (741 +/- 70 vs 1053 +/- 326 muU/ml) levels also did not differ. During glucose clamp studies, time-weighted 60 to 120-minute insulin levels (2.8 mmol/L, 30 +/- 5.0 vs 24 +/- 4.8 muU/ml; 8.3 mmol/L, 57 +/- 5.9 vs 50 +/- 9.8 muU/ml) did not differ between groups. In addition, glucose disposal, cyclic insulin release, and insulin sensitivity indexes were unchanged by denervation. CONCLUSIONS: Extrinsic pancreatic neural elements are not necessary for cyclic insulin release in response to enteral or parenteral glucose challenge or physiologic and pharmacologic hyperglycemia. These findings suggest that the previously described posttransplantation glucose and insulin abnormalities are not attributable to denervation.

Release of human pancreatic polypeptide and gastrin in response to intraduodenal stimuli: a case report.

A recent clinical case afforded an opportunity to study the effects of duodenal stimulation on plasma human pancreatic polypeptide and gastrin concentrations, independent of gastric stimulation. A distension stimulus was provided by rapid injection of 100 ml of water and saline via a T-tube into an isolated duodenal afferent limb. In a third experiment, the saline contained 200 pg/ml of heptadecapeptide human gastrin. Within 2 min after each injection, a rapid rise in circulating human pancreatic polypeptide levels appeared that fell promptly towards basal thereafter. Injections of 100 ml of Flexical, a supplemental tube feeding, resulted in a biphasic human pancreatic polypeptide response, the initial peak comparable to that seen following distension with water, saline, or saline containing gastrin, and a second peak of much greater magnitude and duration followed the initial peak. Plasma gastrin concentrations were not influenced following any of the stimuli. Duodenal distension alone may induce an early transient increase in plasma human pancreatic polypeptide concentrations, while intraduodenal nutrients per se may induce a later increment of greater magnitude and duration.

Pancreatic polypeptide release by intraluminal fatty acids.

The question of whether the response of pancreatic polypeptide to intestinal fatty acids is influenced by the site of intestinal perfusion or the chain length of the fatty acid was investigated. Six dogs with chronic gastric, pancreatic, and intestinal fistulas were studied. Proximal perfusates were administered at the pylorus and diverted via a Foley catheter in the orad stoma of an intestinal fistula placed 45 cm beyond the pancreatic cannula. Distal perfusates were administered into the caudal stoma of the intestinal cannula. Three experimental protocols were used: proximal fatty acid perfusion (20, 40, or 80 mmol/L) combined with distal saline perfusion; distal fatty acid perfusion (20, 40, or 80 mmol/l) combined with proximal saline perfusion; or distal fatty acid perfusion combined with proximal fatty acid perfusion of 80 mmol/L. Each dose of fatty acid was given in random order and the two fatty acids (dodecanoate and oleate) were tested on different days. Blood samples were drawn for pancreatic polypeptide radioimmunoassay, and pancreatic secretion was collected for determination of bicarbonate and protein outputs. Pancreatic polypeptide responses to perfusion of both proximal and distal segments with oleate exceeded (P less than 0.05) those evoked by dodecanoate. The responses of pancreatic polypeptide to dodecanoate administration into either the proximal segment or the distal intestine were not significantly different. In contrast, perfusion of the proximal intestinal segment with oleate release significantly (P less than 0.05) less pancreatic polypeptide than did distal intestinal perfusion with oleate.(ABSTRACT TRUNCATED AT 250 WORDS)

Indolent presentation of pancreatic abscess. Experience with 100 cases.

One hundred cases of pancreatic abscess were identified at five hospitals affiliated with UCLA between 1973 and 1985. Patients were included if a pancreatic mass or phlegmon followed an episode of pancreatitis, if the clinical impression was pancreatic abscess, and if drainage cultures were positive. Less than three Ranson's signs were present on admission in 72% of patients. The admission temperature was less than 38.3 degrees C in 71% of patients, and 27% of patients never had a fever. Abdominal tenderness was absent in 40% of patients. The admission amylase concentrations and white blood cell counts were normal in 36% and 23% of patients, respectively. Extensive débridement, external drainage, and a low threshold for reoperation were the mainstays of surgical therapy. Twenty patients (20%) died, but Ranson's signs did not predict outcome. pancreatic abscess may have an insidious presentation. A high index of suspicion, early computed tomographic scanning, and diagnostic needle aspiration may be necessary to establish this diagnosis.

Endoscopic intervention in pancreatitis.

This overview clearly documents the critical diagnostic and therapeutic role for endoscopic intervention in pancreatitis. Further, although at the forefront of endoscopic development, the new therapeutic techniques discussed offer promise, suggesting cautious optimism. Their ultimate role in treatment of pancreatic disorders must await clarification of appropriate indications, safety, efficacy, and long-term benefit.

Exogenous insulin does not influence CCK- and meal-stimulated pancreatic secretion.

Although previous reports suggest interactions between the endocrine and the exocrine pancreas, insulin's effect on pancreatic exocrine function remains unclear. Chronic pancreatic fistulae were created in five dogs; these animals were studied using the euglycemic, hyperinsulinemic clamp technique. After a 30-min unstimulated period, both groups received a 60-min, 1.5 mU/kg/min insulin (clamp) or vehicle (control) infusion. Cholecystokinin (CCK) or meal stimulation was then begun. Intravenous CCK was initiated at 12.5 ng/kg/h; the CCK dose was doubled every 30 min until 100 ng/kg/h was achieved. The intraduodenal liquid test meal (1.5 kcal/ml; 15% protein, 32% fat, 53% carbohydrate) was administered at 100 ml/h. Unstimulated (0- to 30-min) serum glucose and insulin levels and pancreatic bicarbonate and protein outputs did not differ between groups. Clamp (30- to 90-min) and stimulated (90- to 210-min) insulins were significantly elevated in clamp groups (p < 0.001); glucose and bicarbonate were unchanged. Exocrine outputs during clamp periods were unaffected by insulin. Neither CCK- nor meal-stimulated pancreatic secretion (90-210 min) was influenced by insulin administration. These data suggest that hyperinsulinemia does not alter pancreatic acinar cell secretion in the intact animal.

The effect of pancreatic polypeptide and peptide YY on pancreatic blood flow and pancreatic exocrine secretion in the anesthetized dog.

Pancreatic polypeptide (PP) and peptide YY (PYY) are inhibitors of pancreatic exocrine secretion in vivo but not in vitro, which suggests intermediate mechanisms of action. To examine the role of pancreatic blood flow in these inhibitory effects, xenon-133 gas clearance was used to measure pancreatic blood flow while simultaneously measuring pancreatic exocrine secretion. PP or PYY (400 pmol/kg/h) was administered during the intermediate hour of a 3-h secretin (125 ng/kg/h)/cholecystokinin octapeptide (CCK-8) (50 ng/kg/h) infusion. Exocrine secretion and pancreatic blood flow during the PP or PYY hours were compared with that observed in the first and third hours of the secretin/CCK-8 infusion. PP and PYY significantly inhibited secretin/CCK-8-induced pancreatic exocrine secretion. In addition, PYY (but not PP) significantly reduced pancreatic blood flow during secretin/CCK-8 stimulation. Nevertheless, there was no correlation between pancreatic blood flow and bicarbonate or protein outputs. It is concluded that changes in pancreatic blood flow do not mediate the inhibitory effects of PP or PYY on the exocrine pancreas.

Comparison of gas clearance and radioactive microspheres for pancreatic blood flow measurement.

Measurement of pancreatic blood flow (PBF) is technically demanding. Although radiolabeled microspheres are considered the "gold standard" for PBF assessment, they have practical limitations. In the current study, H2 and xenon-133 gas clearance techniques were adapted to PBF measurement and compared to radiolabeled microsphere techniques. Simultaneous measurements of PBF were made using either hydrogen or xenon gas washout and radiolabeled microspheres. Measurements were made under basal, vasoconstricted (vasopressin 2U i.v. or nicotine 4 micrograms/kg/h) and stimulated (secretin 125 ng/kg/h or 2 U/kg i.v.) conditions (random order). Mean PBF was 26.9 +/- 5.3, 50.5 +/- 2.3 and 27.6 +/- 5.2 ml/min/100 g basally, 36.9 +/- 8.0, 90.1 +/- 18.9, and 81.7 +/- 14.5 ml/min/100 g in the stimulated state, and 24.2 +/- 7.8, 25.0 +/- 3.5, and 14.9 +/- 7.5 ml/min/100 g in the vasoconstricted state for hydrogen gas clearance, xenon gas clearance, and radiolabeled microspheres, respectively. The H2 clearance technique resulted in tissue trauma, was complicated by frequent electrode displacement, and correlated poorly (r2 = 0.36, p greater than 0.05) with microsphere values. In contrast, xenon clearance measurement had no apparent effect on the pancreas and correlated well (r2 = 0.83, p less than 0.01) with microsphere data. We conclude that xenon clearance offers an attractive, validated alternative to radiolabeled microspheres for measuring pancreatic blood flow.

Pancreatic polypeptide and peptide YY inhibit the denervated canine pancreas.

Pancreatic polypeptide and peptide YY are inhibitors of pancreatic exocrine secretion in vivo but not in vitro, which suggests secondarily mediated mechanism(s) of action. To determine the role of extrinsic neural and intrinsic cholinergic elements on this inhibitory effect, a total of nine dogs underwent two-stage extrapancreatic denervation and creation of a chronic pancreatic fistula. After recovery, pancreatic polypeptide or peptide YY (400 pmol/kg/h) was administered during the intermediate hour of a 3-hour secretion (125 ng/kg/h)/cholecystokinin (50 ng/kg/h) infusion. Exocrine secretion pancreatic polypeptide or peptide YY hours was compared with that of the first and third hours. The studies were then repeated during infusion of atropine (10 micrograms/kg/h). Despite extrinsic denervation, pancreatic polypeptide and peptide YY significantly inhibited secretin/cholecystokinin-induced pancreatic output. Although less profound, significant inhibition persisted in the presence of an atropine background. Pancreatic polypeptide or peptide YY infusion also decreased the exocrine response to meal stimulation. We conclude that the inhibitory effects of pancreatic polypeptide and peptide YY are not mediated by extrapancreatic, and possibly not by intrapancreatic cholinergic, neural pathways.

Potential methodologic problems with in vivo immunoneutralization of pancreatic polypeptide.

Dogs with chronic pancreatic fistulae were given 0.5 ml of nonimmune rabbit serum or antibody S5, an antibody raised against the C-terminal pancreatic polypeptide (PP) hexapeptide. A 3-h infusion of secretin (125 ng/kg/h) and CCK8 (50 ng/kg/h) was started 30 min after injecting serum. Exogenous BPP (400 pmol/kg/h) was administered during the middle secretin/CCK hour. In a second protocol, 30 min after injecting nonimmune serum or PP-anti-serum, the animals were fed 15 g/kg cooked ground beef. Pretreatment with S5 enhanced secretin/CCK-induced bicarbonate outputs; protein outputs did not differ. Exogenous BPP inhibited pancreatic secretion, even in S5-treated animals. Meal-induced pancreatic secretion was not altered by S5 pretreatment. Significant increments in PP were measured by radioimmunoassay during administration of secretin/CCK and during BPP infusion. Anti-PP pretreatment abolished the former and significantly decreased, but did not abolish, the latter. The meal evoked significant postprandial increments in PP which were essentially abolished following S5 pretreatment. A physiological role for PP cannot be proved or refuted because antiserum pretreatment failed to block the effects of exogenous hormone. The latter must be established before excluding a peptide's physiological role based on negative in vivo immunoneutralization data.

Effect of pancreatic denervation and atropine on the pancreatic response to secretin.

To study the influence of extrapancreatic nerves and intrapancreatic cholinergic activity on the pancreatic response to secretin, six dogs underwent extrapancreatic denervation and creation of pancreatic fistulae. A second group of six dogs had pancreatic fistulae created without pancreatic denervation. The pancreatic exocrine response to graded doses of secretin (16-500 ng/kg/h) was determined, both alone and during a background infusion of cholecystokinin-octapeptide (CCK8, 50 ng/kg/h). All studies were replicated during administration of atropine (10 micrograms/kg/h). Secretin-induced bicarbonate output was significantly inhibited by atropine in both the innervated and denervated groups. Combined secretin and CCK8 elicited a dose-dependent increase in bicarbonate output and a sustained increase in protein output in both groups, regardless of atropine. In addition, potentiation of secretin-induced bicarbonate output by CCK8 was observed despite both extrinsic pancreatic denervation and administration of atropine. We conclude that endogenous intrapancreatic cholinergic activity influences the pancreatic response to secretin. Potentiation of secretin-induced bicarbonate output by CCK, however, is not dependent on neural mediation.

Inhibition of pancreatic exocrine secretion by galanin.

The influence of extrapancreatic nerves on the inhibition of meal- and secretogogue-induced pancreatic secretion by galanin was studied in conscious dogs. Chronic pancreatic fistulae were created in five mongrel dogs and a second group of five dogs also underwent complete pancreatic denervation. After recovery, galanin dose response (150-1,200 pmol/kg/h) revealed that 600 pmol/kg/h was the lowest dose of galanin to significantly inhibit pancreatic exocrine secretion. Pancreatic responses to a mixed meal, cholecystokinin (CCK) dose response (12.5-200 ng/kg/h), and secretin dose response (16-500 ng/kg/h) were determined. The experiments were then replicated with a continuous background infusion of galanin (600 pmol/kg/h). Galanin inhibited meal-, CCK-, and secretin-induced bicarbonate outputs in both the innervated and denervated pancreas. Galanin also inhibited meal- and CCK-induced protein responses in both groups. We conclude that extrapancreatic nerves do not mediate the inhibitory effects of galanin.

Hyperglycemia alone does not inhibit secretin-induced pancreatic bicarbonate secretion.

Administration of exogenous insulin (INS) inhibits secretin-stimulated pancreatic bicarbonate (HCO3) output via a dose-dependent, neurally mediated mechanism. To determine whether this effect was due to systemic hyperinsulinemia or to reduced endogenous insulin production, we examined the effect of hyperglycemia on secretin-stimulated pancreatic secretion. Chronic pancreatic fistulae were created in six dogs. After 30 minutes of equilibration, a computer-assisted hyperglycemic clamp protocol was used to maintain glucose (GLU) levels 100 or 150 mg/dL above basal in clamp animals; control animals received volume- and rate-matched infusions of 0.9% saline. One hour after beginning the clamp period, intravenous secretin dose-response (16-125 ng/kg/h) was begun, doubling the dose every half hour. Unstimulated (0-30 minutes) HCO3, GLU, and INS levels did not differ between groups. INS and GLU levels in clamp animals were significantly elevated during clamp (30-90 minutes) and stimulated (90-210 minutes) periods. For the same periods, HCO3 secretion was not significantly changed despite profound hyperinsulinemia. We conclude that systemic hyperinsulinemia alone does not inhibit secretin-stimulated HCO3 output. Since exogenous INS exerts feedback regulation on the pancreas, we propose that suppression of endogenous INS secretion mediates the previously reported inhibitory response.

Meal-induced pancreatic polypeptide release in a validated pancreatic denervation model: a role for the distal pancreas?

In previous studies, postprandial pancreatic polypeptide (PP) release was diminished when pancreatic denervation was combined with distal pancreatectomy in animals without chronic fistulae. In contrast, postprandial PP release was reported to be unchanged when the pancreas was denervated without performing a distal pancreatectomy. These findings were unexpected given that the distal pancreas is a PP-poor region. To clarify this issue, we performed extrinsic pancreatic denervation, distal pancreatectomy, and insertion of chronic pancreatic fistulae in nine mongrel dogs. Insulin (0.5 U/kg) and meal-stimulated PP release were measured pre- and postoperatively. In addition, insulin- and meal-induced exocrine secretion was measured postoperatively. Preoperatively, insulin-induced hypoglycemia stimulated significant PP release (80,553 +/- 18,540 pg/min/ml). This response was completely abolished postoperatively (-1,669 +/- 5,054 pg/min/ml). Exocrine secretion did not increase above basal levels after administration of insulin in postoperative animals. These findings suggest adequate pancreatic denervation. Ingestion of a meal evoked significant PP response preoperatively (97,909 +/- 18,394 pg/min/ml). Postoperatively, the response was significantly blunted (17,231 +/- 6,407 pg/min/ml). This finding is in contrast to a previous report using a similar experimental preparation without distal pancreatic resection. We speculate that although the distal pancreas is PP-poor, it may play a role in the regulation of PP release from the PP-rich pancreatic head.

The effect of pancreatic denervation and atropine on the cholecystokinin-induced pancreatic exocrine response.

To study the influence of extrapancreatic neural and cholinergic activity on the pancreatic response to cholecystokinin (CCK), six dogs underwent creation of Herrera pancreatic fistulas, placement of Thomas gastric cannulas, and distal pancreatectomies (innervated; INN). Six additional dogs were prepared similarly, with the addition of total extrinsic pancreatic denervation (denervated; DEN). The pancreatic protein and bicarbonate response to graded 12.5 to 200 ng/kg/h CCK doses was determined for INN and DEN animals both alone and with 10 micrograms/kg/h atropine infusion. The influence of extra-pancreatic neural and cholinergic activity on secretin's potentiation of the CCK-induced pancreatic response was then determined by repeating the studies with a 125 ng/kg/h secretin infusion. The latter results were compared to those predicted by summating the responses seen during separate 12.5-200 ng/kg/h CCK dose-response and 125 ng/kg/h secretin studies. Unstimulated protein output was diminished by atropine in INN animals (78 +/- 21 vs. 39 +/- 9 mg/15 min; p < 0.05) but not in DEN animals. Unstimulated bicarbonate outputs, integrated bicarbonate and protein outputs, and bicarbonate and protein dose-response curves were unaffected by denervation or atropine. Potentiation of CCK-induced bicarbonate output by secretin was also unaffected by atropine and denervation. We conclude that cholinergic elements are involved in unstimulated, but not CCK-induced, enzyme secretion. Further, potentiation of CCK-induced bicarbonate output by secretin does not depend on extrinsic neural or cholinergic elements.

Role of the distal pancreas in pancreatic polypeptide release.

In two previous studies, postprandial pancreatic polypeptide (PP) release was inhibited when pancreatic denervation was combined with distal pancreatectomy. In contrast, postprandial PP release was unaffected by pancreatic denervation without a distal pancreatectomy. These findings suggested a role for the distal pancreas in regulation of postprandial PP release. To examine this possibility, we performed distal pancreatectomy on four mongrel dogs. Pancreatic polypeptide response to i.v. insulin (0.5 U/kg), a meal, and an infusion of cholecystokinin octapeptide (CCK-8; 50 ng/kg/h) were measured in conscious dogs before and after distal pancreatectomy. Insulin-induced hypoglycemia stimulated PP release both preoperatively (47.0 +/- 8.8 ng.[0-120]min/ml) and postoperatively (61.6 +/- 9.2 ng.[0-120]min/ml). Ingestion of a meal also evoked significant PP release preoperatively (85.8 +/- 22.1 ng.[0-180]min/ml) and postoperatively (105.8 +/- 42.2 ng.[0-180]min/ml). CCK-8 elicited only a small increase in circulating PP, which was not influenced by distal pancreatectomy. These findings demonstrate that distal pancreatectomy does not alter PP response to insulin, a meal, or CCK-8.

Somatostatin inhibits cholecystokinin-induced pancreatic protein secretion via cholinergic pathways.

Although somatostatin is a potent inhibitor of pancreatic exocrine secretion in vivo, its mechanism of action remains unclear. The influence of extrapancreatic nerves and intrapancreatic cholinergic activity on somatostatin-induced inhibition of pancreatic exocrine secretion was studied in conscious dogs. Chronic pancreatic fistulae were created in six mongrel dogs, and a second group of six dogs also underwent complete pancreatic denervation. The pancreatic responses to graded doses of cholecystokinin (12.5-200 ng/kg/h) and bethanechol (57-916 micrograms/kg/h), both alone and during background infusion of somatostatin-14 (800 pm/kg/h), were determined in all dogs. The cholecystokinin dose-response with a somatostatin-14 background was then repeated with the addition of atropine (10 micrograms/kg/h). In both groups of animals, cholecystokinin elicited a dose-dependent increase in pancreatic protein secretion that was inhibited significantly by somatostatin-14. Regardless of the status of extrapancreatic nerves, atropine further inhibited cholecystokinin-induced protein secretion beyond that evoked by somatostatin-14. In both innervated and denervated animals, cholinergic stimulation with bethanechol elicited a dose-dependent increase in pancreatic protein secretion that was unaffected by somatostatin-14. We conclude that extrapancreatic nerves do not mediate the inhibitory effects of somatostatin-14. Somatostatin-14 appears to inhibit cholecystokinin-induced pancreatic secretion by an intrapancreatic cholinergic mechanism.

Adrenergic pathways do not mediate peptide YY-induced inhibition of pancreatic exocrine secretion.

The influence of extrapancreatic nerves and intrapancreatic adrenergic activity on the inhibition of pancreatic exocrine secretion by peptide YY (PYY) was studied in conscious dogs. Chronic pancreatic fistulae were created in five mongrel dogs while a second group of five dogs also underwent complete pancreatic denervation. After recovery, a continuous infusion of secretin (62 ng/kg/h) and cholecystokinin (CCK; 50 ng/kg/h) was administered over 2 h. An infusion of PYY (400 pmol/kg/h) was then given randomly, during either the first or second experimental hour. The experiments were then replicated after establishing adrenergic blockade with continuous background infusions of either phentolamine (0.2 mg/kg/h), propranolol (0.5 mg/kg bolus) or a combination of phentolamine and propranolol. The secretin/cholecystokinin-induced bicarbonate and protein outputs were significantly inhibited by PYY in both the innervated and denervated animals. Adrenergic blockade failed to eliminate the inhibitory effects of PYY. We conclude that extrapancreatic neural pathways, including adrenergic mechanisms, do not mediate the inhibitory effects of PYY. The results suggest that PYY inhibits secretin/cholecystokinin-induced pancreatic response by an indirect nonadrenergic mechanism.