Primary cytomegalovirus infection and gastric ulcers in normal host.

A 42-year-old woman presented with epigastric pain and vomiting. Upper gastrointestinal endoscopy revealed three gastric ulcers. Histologic examination of biopsies from the ulcers showed cytomegalovirus inclusion bodies. The appearance of IgM antibodies to cytomegalovirus indicated a recent and primary infection. Stored serum from her last pregnancy 17 months previously contained no cytomegalovirus antibodies. A thorough evaluation of her immune system revealed no abnormality. We are aware of only two other cases where seroconversion was documented in normal hosts. Cytomegalovirus infections in the gastrointestinal tract of normal hosts are very unusual but a common cause of morbidity in immunocompromised hosts. We believe that cytomegalovirus may have a role in the pathogenesis of gastrointestinal lesions in nonimmunocompromised patients.

Sex difference in the sensitivity of the human pancreatic polypeptide cell to autonomic nervous stimulation in man.

We have examined the plasma human pancreatic polypeptide (hPP) response to beta-adrenergic (infusion of epinephrine-phentolamine), vagal cholinergic, (insulin-hypoglycemia), and extravagal cholinergic (secretion infusion) stimulation in healthy, nonobese, age-matched males and females. beta-Adrenergic stimulation caused a rise in plasma hPP concentration in males that was 6 times that of females, whereas there were no differences between the sexes in the mean responses of pulse rate, systolic and diastolic blood pressure, or in plasma concentrations of glucose, insulin, and glucagon. Extravagal cholinergic stimulation caused increases of hPP secretion that were similar in males and females. Insulin-induced hypoglycemia was the most efficacious stimulus, and caused a similar rise in hPP response in males and females. Thus, in males there is greater sensitivity of the hPP cell to beta-adrenergic stimulation than females. The plasma hPP responses appear to be a more sensitive, discriminating index of adrenergic autonomic function than cardiovascular and other pancreatic hormonal responses.

Gastrointestinal/pancreatic hormone concentrations in the portal venous system of nine patients with organic hyperinsulinism.

Percutaneous transhepatic sampling of blood in the portal venous system (TPVS) was used to; (1) localize hormone secreting tumors and help in differentiating tumors from diffuse disease (nesideoblastosis and hyperplasia with adenomata) in 9 patients with fasting hypoglycemia and hyperinsulinism, and (2) study the concentration an distribution of the immunoreactive peptides: insulin (IRI), gastrin (IG), glucagon (IRG), pancreatic polypeptide (hPP), and somatostatin (SRIF-LI), in the venous drainage of the uninvolved portion of the pancreas and GI tract. Localized elevations of IRI (64-920 microunits/ml) predicted tumor localization in 6 patients with single tumors that were not demonstrable angiographically. In one patient with nesideoblastosis and another with islet cell hyperplasia with adenoma, elevated IRI concentrations at multiple locations suggested a diffuse or multicentric process. Elevations of SRIF-LI in the same region as IRI elevations in one patient and of IRG in another patient suggested that these tumor produced two hormones. Some problems in the interpretation of portal venous insulin concentrations are discussed. The locations of maximum portal venous system plasma concentrations and portal-arterial gradients (mean +/- SE pg/ml) in five patients with small single insulinomas were: IG, gastrocolic trunk (126 +/- 27, 46 +/- 22); IRG, proximal splenic vein (130 +/- 30, 47 +/- 13) and gastrocolic trunk (131 +/- 23, 60 +/- 13); hPP, portal vein (164 +/- 48, 49 +/- 22); SRIF-LI, superior mesenteric vein (186 +/- 50, 57 +/- 20) and gastrocolic trunk (178 +/- 59, 55 +/- 21). It is concluded; (1) TPVS can be used successfully to localize single insulin-secreting tumors of the pancreas and to help distinguish them from diffuse disease but problems in such differentiation do occur, (2) circulating SRIF-LI and IRG are derived from both the pancreas and the gut, IG predominantly from the proximal gut and hPP from the head of the pancreas, and (3) The data provide new information for the interpretation of portal insulin concentrations in patients with organic hyperinsulinism and of hormone concentrations for localization of peptide-producing tumors of the pancreas other than insulinomas.

Truncal vagotomy abolishes the somatostatin response to insulin-induced hypoglycemia in man.

The mechanism whereby insulin-induced hypoglycemia stimulates release of immunoreactive somatostatin (SRIF-LI) into the peripheral circulation in man is unknown. We have measured the plasma SRIF-LI response to insulin-induced hypoglycemia in 16 healthy subjects and five subjects with prior truncal vagotomy. Mean nadir of plasma glucose was similar in the two groups (37 +/- 5 and 34 +/- 2 mg/dl in the control group). Hypoglycemia induced a brisk rise in plasma pancreatic polypeptide (hPP) concentrations in healthy subjects (maximum concentration 1563 +/- 245 pg/ml) whereas in none of the postvagotomy subjects was there a significant change in hPP concentrations, indicative of completeness of the truncal vagotomy. In healthy subjects SRIF-LI concentrations rose for a basal of 168 +/- 19 pg/ml to a maximum of 254 +/- 30 at 39 +/- 4 minutes (p less than 0.005) with an incremental area of 2.8 +/- 1 ng . min/ml above basel. In vagotomized subjects, the mean basal SRIF-LI concentration of 166 +/- 22 pg/ml was not significantly different from that in healthy subjects. After insulin injection, SRIF-LI concentration fell with a net decrement of -3.2 +/- 1 ng . min /ml below the basal. It is concluded that the SRIF-LI response to insulin-induced hypoglycemia is dependent upon vagal integrity. Section of the vagus unmasks a suppressive effect of insulin action or its metabolic or hormonal consequences on the concentration of SRIF-LI in plasma.