Microvascular changes in liver after ischemia-reperfusion injury. Protection with misoprostol

Morphological changes in the hepatic microvasculature were studied in experimentally induced ischemia-reperfusion injury in the rat using a vascular casting technique. Partial hepatic ischemia was induced for 90 min followed by 24 hr reperfusion. Microvascular casting was performed after 24 hr reperfusion by either intraarterial or intravenous infusion of acrylic resin (Mercox). After corrosion of the tissue, the cast was examined by scanning electron microscopy. Casts of normal livers showed good patency with no evidence of unfilled areas. The mean diameter of sinusoids was 14 +/- 3 microns with those in zone 1 slightly smaller than those in zone 3. Liver casts from rats subjected to ischemia and reperfusion resulted in gross disruption of normal architecture. The common characteristics seen in both prograde and retrograde casts were clusters of closed sinusoids around zones 2 and 3 of the liver acini, which resulted in cavities of various sizes. Varicosities were observed in some areas. The mean diameter of sinusoids in areas of patent microvascular structure (10 +/- 2 microns) was significantly smaller compared to those in normal livers (P < 0.001). Misoprostol given at 1 min before reperfusion markedly reduced the microvascular injury. The hepatic microvascular was generally intact with mild focal unfilled areas. The majority of the sinusoids were of normal size and no clusters of blind ending sinusoids were detected. The present study shows that hepatic ischemia-reperfusion results in extensive microvascular injury in the liver. The protective effects of misoprostol against this injury may occur at the vascular level.

Misoprostol protection against acetaminophen-induced hepatotoxicity in the rat

The hepatoprotective effects of misoprostol on acetaminophen (APAP)-induced toxicity were studied in the rat. Liver injury was evaluated at 36 hr after APAP administration by measuring serum ornithine carbamoyltransferase (OCT) and alanine aminotransferase (ALT) levels, by using tetranitroblue tetrazolium (TNBT) staining and by histological analysis. After APAP administration, peak serum levels of the drug were detected at 15 min. Liver GSH was depleted from control levels of 448 +/- 48 micrograms/g to 82 +/- 2 micrograms/g (P < 0.01) within 3 hr. Serum ALT levels increased significantly after 16 hr and H&E staining revealed significant hepatic necrosis after 12 hr. Rats treated with misoprostol before and after APAP administration showed reduced OCT and ALT levels at 36 hr of overdose (454 +/- 446 IU/liter and 2571 +/- 2944 IU/liter, respectively) compared to those without misoprostol treatment (1348 +/- 480 IU/liter and 6077 +/- 3025 IU/liter, respectively, P < 0.01). TNBT staining showed a reduced area of damage from 28.6 +/- 22.3% to 7.3 +/- 8.9% (P < 0.01), and H&E staining also showed less extensive hepatic necrosis in rats treated with misoprostol before and after the overdose. In a time sequence study, misoprostol treatment starting within 10 hr of overdose showed the same protective effect as when it was given before and after APAP ingestion. No protection was detected when the treatment was started during the development of hepatic injury. However, misoprostol given when injury was established seemed to be protective. Our results show that misoprostol protects the liver against APAP-induced injury if given within 10 hr of overdose.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of misoprostol on delayed ulcer healing induced by aspirin

Clinical studies have suggested that treatment with the prostaglandin E1 analog, misoprostol, leads to significant healing of ulcers in patients taking regular nonsteroidal antiinflammatory therapy. This study aimed to investigate mechanisms involved in this healing using a rat model. Gastric ulcers were induced by application of acetic acid using a standard technique. Rats were treated with 200 mg/kg aspirin, 100 micrograms/kg misoprostol, a combination of both treatments, or methylcellulose vehicle for up to two weeks, starting two days after ulcer induction. Ulcers were assessed by macroscopic measurements of area and by quantitative histological measurements. Aspirin delayed ulcer healing compared with controls, while misoprostol significantly reversed this effect. Quantitative histology revealed that misoprostol cotreatment significantly increased mucosal regeneration compared with aspirin treatment alone. However, misoprostol did not reverse the effects of aspirin on an index of wound contraction. We conclude that treatment with misoprostol significantly reverses the delayed healing effect of aspirin, and this may occur via an effect on epithelial regeneration.

Effect of nonsteroidal anti-inflammatory drugs on LFA-1 and ICAM-1 expression in gastric mucosa

Leukocyte adhesion to the endothelium appears to play an important role in gastric injury. This study aimed to develop immunohistochemical staining techniques to investigate the distribution and sequence of expression of both leukocyte [lymphocyte function associated antigen 1 (LFA-1)] and endothelial [intracellular adhesion molecule 1 (ICAM-1)] adhesion molecules in the mucosa after treatment with nonsteroidal anti-inflammatory drugs (NSAIDs). In control rats there were 803 +/- 72 LFA-1-stained cells/mm2 in the deep mucosa, 134 +/- 32 cells/mm2 in the superficial mucosa, and 6.4 +/- 1.2 ICAM-1-stained blood vessels/mm2 in the total mucosa. The number of ICAM-1-stained blood vessels in the mucosa increased significantly after 30 min of treatment with intragastric aspirin (30 mM; 25.2 +/- 7.2/mm2, P < 0.01) and indomethacin (20 mg/kg; 20.7 +/- 4.4/mm2, P < 0.01) before any appreciable mucosal damage was evident. This increase was reversed by treatment with misoprostol (100 micrograms/kg) in both aspirin- (7.6 +/- 1.7/mm2, P < 0.01) and indomethacin-treated animals (10.7 +/- 2.6/mm2, P < 0.05). There was no significant increase in LFA-1-positive cells until 60 min of NSAID treatment. We conclude that the adhesion molecules LFA-1 and ICAM-1 are expressed in the normal gastric mucosa and that the number of ICAM-1-stained blood vessels increase rapidly after NSAID treatment. This increase in ICAM-1 expression may be associated with an inhibition of prostaglandin synthesis by NSAIDs. These results provide further support for the role of early vascular changes in NSAID gastropathy.

Protection against gastric ischemia-reperfusion injury by nitric oxide generators

Nitric oxide appears to play an important role in maintaining gastric mucosal integrity. This study aimed to investigate whether a nitric oxide donor (sodium nitroprusside) or stimulation of endogenous nitric oxide synthesis (with acetylcholine) protects against gastric ischemia-reperfusion injury. Rats were subjected to 30 min of ischemia followed by 15 min of reperfusion. Injury was assessed by quantitative histology. Intravenous sodium nitroprusside (50-75 micrograms/kg) or acetylcholine (10-25 micrograms/kg), immediately before reperfusion, significantly reduced the percentage of mucosal injury compared with controls. Inhibition of nitric oxide synthesis by topical application of 12.5 mg/kg NG-methyl-L-arginine before acetylcholine treatment, abolished the effects of acetylcholine. The protective effects of acetylcholine and sodium nitroprusside did not appear to be related to local vasodilation since neither drug improved gastric blood flow and infusion of a non-nitric oxide vasodilator (papaverine, 1 mg/kg), had no protective effect on reperfusion injury. Sodium nitroprusside (50 micrograms/kg) and acetylcholine (25 micrograms/kg) significantly reduced polymorphonuclear leukocyte infiltration and extravasation into the mucosa compared with controls. NG-Methyl-L-arginine pretreatment before acetylcholine abolished these effects. We conclude that nitric oxide generators significantly reduce mucosal injury following ischemia-reperfusion and that this may occur via a reduction in polymorphonuclear leukocyte infiltration into the mucosa.

Polymorphonuclear leukocyte infiltration into gastric mucosa after ischemia-reperfusion

Polymorphonuclear leukocytes (PMN) have been implicated in the pathogenesis of ischemia-reperfusion injury. The objective of this study was to investigate PMN infiltration and distribution within the gastric mucosa of rats subjected to 30 min gastric ischemia followed by reperfusion. Sections of mucosa were stained for PMN using an immunoperoxidase technique, and injury was assessed by quantitative histology. In control animals, there were 4 +/- 2 PMN/mm2 in the superficial and 9 +/- 4 PMN/mm2 in the deep mucosa. This increased significantly to 67 +/- 9 PMN/mm2 (P < 0.05) and 160 +/- 53 PMN/mm2 (P < 0.01) respectively at 15 min of reperfusion. The percentage of these PMN which were extravasated was 83 +/- 4% in the superficial mucosa and 82 +/- 4% in the deep mucosa (P < 0.001 compared with control levels of 0% in superficial and 10% in deep mucosa). Significant PMN infiltration occurred before full expression of mucosal injury and treatment of rats with anti-PMN antisera blocked reperfusion injury (treated 10.7 +/- 1.4% mucosa damaged, controls 33.5 +/- 2.4%; P < 0.001). Treatment with allopurinol (100 mg/kg) significantly reduced the number of infiltrating PMN (superficial 7 +/- 1/mm2, deep 16 +/- 2/mm2; P < 0.01) and the percentage of extravasating PMN (superficial 40 +/- 10%, deep 30 +/- 15%; P < 0.01) while also significantly reducing tissue injury (21.9 +/- 1.9% mucosa damaged, P < 0.01 compared with controls). We conclude that the immunoperoxidase staining provides a simple means of identifying PMN in histological sections. Furthermore, our results support a role for PMN in gastric ischemia-reperfusion injury.

A multicenter trial for resuscitation of injured patients with 7.5% sodium chloride. The effect of added dextran 70. The Multicenter Group for the Study of Hypertonic Saline in Trauma Patients

OBJECTIVE

To evaluate the use of 250 mL of a 7.5% sodium chloride solution, both with and without added dextran 70, for the prehospital resuscitation of hypotensive trauma patients.

DESIGN

Double-blind randomized trial.

SETTING

Six trauma systems served by helicopter transport.

PATIENTS

Injured patients with systolic blood pressures less than 90 mm Hg at any time in the field or during helicopter transport.

INTERVENTIONS

Infusion of study solution, in the field or during transport, followed by conventional isotonic solutions as needed. Solutions studied in four cohorts were as follows: (1) lactated Ringer's; (2) 7.5% sodium chloride (hypertonic saline); (3) 7.5% sodium chloride combined with 6% dextran 70; and (4) 7.5% sodium chloride combined with 12% dextran 70.

MAIN OUTCOME MEASURES

Blood pressure response; survival to time of hospital discharge among the treatment groups; and survival compared with that predicted by norms from the Major Trauma Outcome Study (MTOS).

RESULTS

The mean (+/- SD) change in systolic blood pressure on arrival in the emergency department was significantly higher in the hypertonic saline solution group than that in the lactated Ringer's solution group (34 +/- 46 vs 11 +/- 49 mm Hg, P < .03). Overall survival in the four treatment groups was 49%, 60%, 56%, and 45% (not statistically significant). Survival in the hypertonic saline solution group, however, was significantly higher than that predicted by the MTOS norms (60% vs 48%, P < .001). Survival to hospital discharge in patients with baseline Glasgow Coma Scale scores of 8 or less was correlated with treatment group (P < .05 by logistic regression and P < .01 by Cox proportional-hazards analysis; with survival in the hypertonic saline solution group [34%] vs lactated Ringer's solution group [12%]).

CONCLUSIONS

Prehospital infusion of 250 mL of 7.5% sodium chloride is associated with an increase in blood pressure and an increase in survival to hospital discharge compared with survival predicted by the MTOS norms. Patients with low baseline Glasgow Coma Scale scores seem to benefit the most from 7.5% sodium chloride resuscitation. Hypertonic saline solution without added dextran 70 is as effective as the more expensive solutions that contain dextran 70.

The role of luminal factors in prostaglandin protection against ethanol-induced gastric mucosal injury

Prostaglandins (PG) protect the gastric mucosa against damage by several irritants, but the mechanisms remain unclear. A standard rat model of gastric injury induced by 50% ethanol was used to test the hypothesis that PG protection occurs either by increasing luminal fluid volume and hence diluting the irritant, or by production of protective factors within this fluid. Quantitative histology was used to assess microscopic mucosal damage. The increase in luminal fluid volume in prostaglandin E2 (PGE2)-treated animals was measured, and the ability of this increased fluid to protect saline-treated animals via dilution of the irritant was assessed. The transfer of protection by exchange of luminal fluid from PG to non-PG treated animals was also tested. Results showed that PGE2 induced a specific increase in luminal fluid volume of 45.6%. When given together with ethanol treatment the extra fluid volume was not protective. Removal of luminal fluid after PG and before ethanol treatment did not abolish protection and no protective factor was transferred with the luminal fluid. In conclusion, this study has shown that neither dilution of ethanol by accumulation of luminal fluid nor the presence of luminal factors is responsible for PG protection.

Sequence of gastric mucosal injury following ischemia and reperfusion. Role of reactive oxygen metabolites

The mechanisms of gastric mucosal injury following a period of ischemia remain unclear. The aim of this study was to determine the relative contributions of ischemia, reperfusion, and reactive oxygen metabolites to mucosal injury induced by temporary occlusion of the celiac artery. Rats were subjected to 30 min of gastric ischemia in the presence of 100 mM HCl. Reperfusion periods ranged from 1 min to 24 hr. Drug treatments included allopurinol (100 mg/kg) or a combination of superoxide dismutase (15,000 units/kg), catalase (90,000 units/kg), and desferrioxamine (50 mg/kg). Mucosal injury was assessed by quantitative histology and the extent of macroscopic hemorrhage. Approximately one third of the total injury to the volume of the mucosa (11.8 +/- 9.1%) was due to ischemia alone. Another third was blocked by allopurinol or superoxide dismutase, catalase, and desferrioxamine (22.1 +/- 6.9%, P less than 0.001; and 25.9 +/- 4.6%, P less than 0.01), respectively, compared with control (32.5 +/- 5.1%). In contrast, extensive surface mucosal injury (62.2 +/- 27.6%) occurred primarily during ischemia and was not affected by antioxidants. Macroscopic hemorrhage was halved by treatment with allopurinol (17.5 +/- 12.6%, P less than 0.01) or superoxide dismutase, catalase, and desferrioxamine (15.9 +/- 14.5%, P less than 0.01). We conclude that temporary celiac occlusion results in gastric mucosal damage that consists of both ischemic and reperfusion components. The majority of surface mucosal injury occurred during ischemia, whereas injury to the volume of the mucosa and the vasculature occurred equally during reperfusion and was associated with reactive oxygen metabolites.

Misoprostol hepatoprotection against ischemia-reperfusion-induced liver injury in the rat

The hepatoprotective effects of misoprostol, a PGE1 analog, against ischemia-reperfusion liver injury were studied using a rat partial liver ischemia model. Serum ornithine carbamoyltransferase (OCT) and alanine aminotransferase (ALT) levels were determined as biochemical indices of injury. Hepatic cell necrosis was assessed histologically using tetranitroblue tetrazolium (TNBT) and hematoxylin and eosin (H&E) staining. With placebo treatment, 90 min of partial hepatic ischemia followed by 24 hr of reperfusion resulted in increased levels of serum OCT (760 +/- 521 IU/liter) and ALT (4327 +/- 1982 IU/liter), while extensive hepatic necrosis was evident by TNBT and H&E staining. Treatment with two doses of 25 micrograms misoprostol/kg body weight at 1 min before ischemia and 1 min before reperfusion significantly reduced the serum levels of OCT and ALT (207 +/- 189 IU/liter, P less than 0.01 and 2075 +/- 1217 IU/liter, P less than 0.01, respectively) and hepatic necrosis. When a single dose of misoprostol was administered 1 min before reperfusion, similar protective effects were observed. However, when the treatment of misoprostol was delayed to 1 min after reperfusion, significantly less hepatoprotection was seen. Misoprostol exerted no hepatoprotection at all when it was administered at 5 min or later after reperfusion. These results demonstrate that misoprostol partially protects the liver against ischemia-reperfusion injury in the rat. The observation that the protective effect of misoprostol occurs only within the first minute of reperfusion suggests that its mechanism of action involves an early event in reperfusion injury, such as modifying the effects of reactive oxygen metabolites.

Codependency: a disorder separate from chemical dependency

This study examined the relationship between codependency (an excessive preoccupation with the lives, feelings, and problems of others), chemical dependency of a significant other, and depression. The Significant Others' Drug Use Survey (SODS) determined whether the subject was in a relationship with a significant other at risk of being chemically dependent. Beck's Depression Inventory (BDI) was used to assess depression. Two hypotheses were tested: first, that codependency exists independently of chemical dependency and, second, that codependent people tend to be more depressed than non-codependents. Results supported the first hypothesis, but not the second. A significant correlation between depression and having a significant other likely to be chemically dependent was observed. The usefulness of the concept of codependency is discussed with proposals for subsequent research.

Inability of cytoprotection to occur during a period of gastric ischemia

Prostaglandin E2 (PGE2), colloidal bismuth subcitrate (CBS), and sucralfate (SUC) are known to protect the gastric mucosa from ethanol injury. The proposed central role for the microcirculation in gastric mucosal defense and as a site for the expression of the protective effects of these agents was investigated in the rat stomach. Animals were pretreated with either PGE2, CBS, or SUC. Control rats were given normal saline. After allowing 15 min for expression of the pretreatment, ethanol was administered as a 10%, 25%, 50%, or 100% solution to groups of rats with normally perfused stomach and to other groups of rats in whom the stomach was made ischemic by cross-clamping the supracoeliac aorta immediately prior to the instillation of ethanol. The extent of gastric mucosal damage was measured using quantitative histological techniques and expressed as a percentage of surface area and volume of mucosa damaged. In the presence of ischemia, the extent of damage by ethanol was markedly increased, with total destruction of the mucosa by the 50% and 100% solutions. With 25% ethanol, the volume of mucosal damage was increased from 0.5% in the normally perfused stomach to 53.5% with ischemia. When 10% ethanol was instilled into the ischemic stomach, only 0.8% of the volume of the mucosa was damaged, which was not different from the volume of mucosa damaged after the ischemic stomach was exposed to normal saline alone (1.0%). Pretreatment with PGE2, CBS, or SUC did not significantly change the extent of damage seen with exposure of the ischemic stomach to 25% or 50% ethanol.(ABSTRACT TRUNCATED AT 250 WORDS)

Time as a factor in the expression of ethanol injury to the gastric mucosa

A technique of quantitative histology was used to assess the influence of time after injury on the histological expression of gastric mucosal damage. Rats, pretreated with either natural prostaglandin E2 or saline, were subjected to intragastric administration of either 50% or 100% ethanol. Fifteen minutes later the ethanol was removed from the stomach. Rats were sacrificed at either 30 min or 24 h after ethanol instillation. In rats pretreated with saline and subjected to 100% ethanol with or without prostaglandin pretreatment, the extent of deep mucosal damage was markedly underestimated by early evaluation. Only 5.4% of the volume of the gastric mucosa showed evidence of damage at 30 min after ethanol, compared with 57.3% of the volume of the mucosa, at 24 h after 100% ethanol exposure. Assessment of gastric mucosa 24 h after ethanol injury showed that PGE2 reduces the extent of surface area damaged and the volume of the mucosal damage. When 50% ethanol was used as the injurious agent, no difference was noted in the volume of the mucosa damaged when the stomach was assessed at either 30 min or 24 h after injury. These results indicate that full histological expression of injury is not present 30 min after 100% ethanol instillation, at least in part because of fixation of the gastric mucosa by 100% ethanol. Fifty per cent ethanol, which does not cause mucosal fixation, may be better as a test agent.

Sulindac inhibits the rate of growth and appearance of colon tumors in the rat

Sulindac (cis-5-fluoro-2-methyl-1-[p-(methylsulfinyl) benzylidene] indene-3-acetic acid), an inhibitor of prostaglandin synthesis, has been reported to cause regression of colon polyps in patients with familial polyposis coli and Gardner's syndrome. We examined the effect of sulindac on the growth of primary colon carcinomas in rats. Colon tumors were induced in 18 rats by repeated subcutaneous administration of dimethylhydrazine. The site and diameter of each tumor were measured via laparotomy and colonoscopy. Rats were randomized to receive either sulindac (10 mg/kg) twice daily or vehicle (0.5% methylcellulose). After 4 weeks of treatment, the site and size of tumors in the colon were again recorded. In eight rats receiving sulindac, no new tumors were identified, while in 10 control rats, 13 additional tumors were found after treatment. There was a significantly greater increase in size of the tumors in the control group (56.4 mm for 26 tumors) compared with the rats receiving sulindac (9.3 mm for 14 tumors). We report that sulindac inhibits the rate of development and the rate of growth of colon tumors in the rat.

Evaluation of putative cytoprotective properties of antiulcer drugs using quantitative histological techniques

The capacity for cytoprotection has been claimed for a number of drugs that may have a place in the treatment of peptic ulcer disease. In this study we have used quantitative histological criteria to evaluate the ability of these drugs to be cytoprotective and have compared their effects with that of natural prostaglandin E2 (PG). The standard rat model, with injury by instillation of 1 ml of absolute ethanol, has been used. Putative cytoprotective agents were administered 15 min prior to ethanol. Each animal was sacrificed 15 min after ethanol exposure. The stomach was removed and studied using an established quantitative histological technique. This technique provides a measure of the surface area of mucosa damaged and of the volume of mucosa damaged. Ethanol alone caused damage to 76% of the area of the rat stomach and 14% of the volume of the rat gastric mucosa. Pretreatment by PG (25 micrograms/ml) resulted in reduction of the area of damage to 45% and reduction of the percentage volume damage to 2.2%. The synthetic analog of PGE2, Enprostil (1 microgram/ml) achieved similar protective effects. With pretreatment with colloidal bismuth subcitrate (10 mg/kg) or sucralfate (25 mg/kg), no protection against the surface area damaged by ethanol was seen, but there was a marked reduction of the volume of mucosa damaged. Indomethacin pretreatment augmented the damage caused by ethanol. The protective effects of colloidal bismuth subcitrate and sucralfate were not blocked by pretreatment with indomethacin.(ABSTRACT TRUNCATED AT 250 WORDS)

Microvascular architecture of experimental colon tumors in the rat

Tumor cell proliferation is dependent upon concurrent growth of a supporting vasculature. This study aims to characterize the structural features of the microvasculature within a primary tumor model. There were 22 colon tumors induced in 16 rats by repeated administration of dimethylhydrazine. A cast of the microvessels was prepared by intraarterial administration of acrylic resin (Mercox). After corrosion of the tissue, the cast was examined by scanning electron microscopy. Tumors 2.6 to 12.0 mm in diameter were examined. Within polypoid carcinomas up to 5.7 mm in diameter, there were two distinct vascular zones, a luminal vascular zone continuous with the vasculature of normal mucosa and a central zone continuous with the normal submucosa and muscularis propria vessels. Within both vascular zones, the organization of microvessels had the same general pattern as in normal mucosa. However, in tumors with diameters greater than 5.7 mm, the vasculature was seen to be disorganized and of a greater density than normal. In the smallest tumors, few morphological changes were seen in the individual microvessels when compared to normal. However, with tumor growth, there was elongation and increased diameters of the microvessels within the tumor. Microvessels within the luminal zone of the tumors which could definitely be traced to veins had diameters of 50 to 100 microns (compared to 12 to 30 microns for normal venules). Individual microvessels varied in diameter along their course forming saccular dilations in places. Networks of frequently anastomosing microvessels were formed. Extravasation of resin occurred from some microvessels. Elongated vessels of uniform diameter which travel distances up to 2 mm without branching were seen and were probably arterioles. These appearances indicate that there are two distinct stages of development of the vasculature within primary tumors, an early phase where the tumor is supplied by the preexisting host microvessels, followed by a phase of proliferation of new vessels with abnormal morphological characteristics.

Gastric surgery for morbid obesity. The Adelaide Study

The efficacy of three gastric restriction operations were compared in a prospective randomized study of 310 morbidly obese subjects. The median patient age was 34 years (range, 18 to 62 years). They were predominantly female (13:1) and had a median pre-operative weight that was 198% of their ideal weight (range, 160% to 318%). There was an equitable dispersion of perceived risk factors between the groups under study and there were no deaths during the perioperative period. Compliance with follow-up at 3 years was 91%. When success was defined as a loss of more than 50% of excess weight or a current pregnancy, the success rates at 3 years were 17% for gastrogastrostomy, 48% for vertical gastroplasty, and 67% for Roux-en-Y gastric bypass (p less than 0.001). Although the gastric bypass operation took longer to perform, there were similar outcome patterns for the three groups during the postoperative period. We conclude that the Roux-en-Y gastric bypass is the preferred procedure for the surgical treatment of morbid obesity.

Distributions of neuropeptides in the human esophagus

The distributions of nerve cells and fibers with immunoreactivity for the peptides substance P, somatostatin, enkephalin, vasoactive intestinal peptide, gastrin-releasing peptide, and neuropeptide Y and the enzyme tyrosine hydroxylase were examined in 25 samples of human esophagus. These were compared with samples of stomach and intestine. In the smooth muscle of the muscularis externa, the muscularis mucosae, and beneath the epithelium, the most abundant nerve fibers contained vasoactive intestinal peptide and neuropeptide Y, in contrast to the scarcity of substance P, enkephalin, somatostatin, and gastrin-releasing peptide. Gastric and intestinal samples contained dense populations of fibers containing vasoactive intestinal peptide, neuropeptide Y, substance P, and enkephalin in the equivalent layers, but somatostatin- and gastrin-releasing peptide-immunoreactive fibers were scarce. Complete coexistence of vasoactive intestinal peptide and neuropeptide Y in nerve fibers within the muscle layers was demonstrated in the esophagus, but not in gastric and intestinal samples. The myenteric plexus along the length of the esophagus contained cell bodies and fibers reactive for vasoactive intestinal peptide, neuropeptide Y, enkephalin, and substance P. Somatostatin-immunoreactive cell bodies were very rare in the myenteric plexus, no gastrin-releasing peptide-immunoreactive cell bodies were seen, and both somatostatin and gastrin-releasing peptide-immunoreactive fibers were rare. In the upper esophagus, striated muscle bundles did not contain nerve fibers reactive for these peptides but immunoreactive fibers were seen in the muscularis mucosae and subepithelium. It is concluded that the esophagus has a different pattern of innervation by peptide-containing neurons than the stomach and intestines. Esophageal neurons can be classified into separate classes on the basis of their peptide content.

Management of diffuse peritonitis by prolonged postoperative peritoneal lavage

Prolonged postoperative peritoneal lavage has been used as a part of the management of 55 patients with diffuse peritonitis. The lavage technique consisted of 60 min cycles of instillation and drainage of a lavage fluid into the peritoneal cavity via a peritoneal dialysis catheter. The lavage fluid was a slightly hypertonic salt solution containing antibiotics, usually cefotaxime and metronidazole. The overall mortality rate in the series was 11% (6/55). Only one of these patients had residual abdominal sepsis present at post-mortem, the remaining deaths being due to a progression of the pre-existing disease. Five patients showed evidence of further intra-abdominal sepsis. In three of these patients this was associated with the presence of a previously well established abscess cavity. The overall results indicate that, for this group of patients recognized to be at high risk of mortality or further sepsis, the use of prolonged postoperative peritoneal lavage is associated with an improved outcome.