OBJECTIVE To evaluate the eosinophilic response in intestinal mucosa of horses with intestinal ischemia and reperfusion or with strangulation of the jejunum or colon. SAMPLE Mucosal samples from horses with naturally occurring strangulation (n = 24 horses) or distention (n = 6) of the jejunum or colon (11), with experimentally induced ischemia and reperfusion of the jejunum (6) or colon (15), or that were euthanized for reasons other than gastrointestinal tract disease (13). PROCEDURES Mucosal samples were collected and grouped by type of intestinal injury. Slides were stained with Luna eosinophil stain and histologically examined to determine eosinophil accumulation and distribution. Number of eosinophils per mm2 of mucosa was calculated as a measure of eosinophil accumulation. Additionally, mucosa was categorized into 5 regions; the percentage of eosinophils in each of the 5 regions, relative to the total eosinophil count in all regions, was determined. RESULTS Eosinophil migration toward and onto the luminal surface was evident in tissues after ischemia and reperfusion and after naturally occurring strangulating disease of the jejunum and colon, as indicated by a decrease in the number of eosinophils near the muscularis mucosa and an increase in the number of eosinophils on or near the luminal surface. Ischemia alone did not change eosinophil distribution in the jejunum or colon. CONCLUSIONS AND CLINICAL RELEVANCE Eosinophils responded to mucosal damage evoked by ischemia and reperfusion by migration toward and onto the luminal surface. This migration could represent an important component of the inflammatory response to injury in equine gastrointestinal mucosa.

Objective—To determine the effect of large colon ischemia and reperfusion on concentrations of the inflammatory neutrophilic protein calprotectin and other clinicopathologic variables in jugular and colonic venous blood in horses. Animals—6 healthy horses. Procedures—Horses were anesthetized, and ischemia was induced for 1 hour followed by 4 hours of reperfusion in a segment of the pelvic flexure of the large colon. Blood samples were obtained before anesthesia, before induction of ischemia, 1 hour after the start of ischemia, and 1, 2, and 4 hours after the start of reperfusion from jugular veins and veins of the segment of the large colon that underwent ischemia and reperfusion. A sandwich ELISA was developed for detection of equine calprotectin. Serum calprotectin concentrations and values of blood gas, hematologic, and biochemical analysis variables were determined. Results—Large colon ischemia caused metabolic acidosis, a significant increase in lactate and potassium concentrations and creatine kinase activities, and a nonsignificant decrease in glucose concentrations in colonic venous blood samples. Values of these variables after reperfusion were similar to values before ischemia. Ischemia and reperfusion induced activation of an inflammatory response characterized by an increase in neutrophil cell turnover rate in jugular and colonic venous blood samples and calprotectin concentrations in colonic venous blood samples. Conclusions and Clinical Relevance—Results of this study suggested that large colon ischemia and reperfusion caused local and systemic inflammation in horses. Serum calprotectin concentration may be useful as a marker of this inflammatory response.

Objective-To assess the effects of ischemia and reperfusion on indicators of oxidative stress, activation of eosinophils, and apoptosis in the large colonic mucosa of horses. Animals-40 horses. Procedures-In 1 or two 20-cm-long segments of the pelvic flexure, ischemia was induced for 1 or 2 hours followed by no reperfusion or 30 minutes and 18 hours of reperfusion in anesthetized horses. Mucosal specimens were collected before (controls; n = 20 horses) and after each period of ischemia, and full-thickness tissue samples were collected after each period of reperfusion. Sections of colonic tissues were stained for histomorphometric analysis or assessment of eosinophil accumulation. Nitrotyrosine was identified immunohistochemically, and severity of apoptosis was determined via the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling method. Results-Numbers of mucosal eosinophils were similar before induction of ischemia, after ischemia, and after ischemia-reperfusion. Eosinophil nitrotyrosine production increased significantly during ischemia and continued through 30 minutes of reperfusion; production was decreased at 18 hours of reperfusion but remained greater than that of the controls. In other leukocytes, nitrotyrosine generation peaked at 1 hour of ischemia and again at 18 hours of reperfusion. Compared with control findings, epithelial apoptosis increased gradually at 1 through 2 hours of ischemia with no further progression after reperfusion. Conclusions and Clinical Relevance-Results suggested that resident eosinophils in the large colon of horses react to mucosal injury from ischemia and reperfusion and may undergo oxidative stress under those conditions. Epithelial apoptosis could contribute to tissue damage.

Objective—To induce ischemia and reperfusion injury in the large colon mucosa of horses in vivo and evaluate the recovery and effects of components of an organ transplant solution on mucosal recovery in vitro. Animals—6 healthy horses. Procedures—Horses were anesthetized, and ischemia was induced for 60 minutes in the pelvic flexure, which was followed by reperfusion for 240 minutes. Ischemic (n = 4 horses), reperfused (6), and adjacent control (6) colonic mucosae were isolated for in vitro testing and histologic examinations. Tissues were mounted in Ussing chambers with plain Krebs Ringer bicarbonate (KRB), KRB with N-acetylcysteine (NAC), or KRB with a modified organ transplant solution (MOTS). Transepithelial electrical resistance (TER) and mannitol flux were used to assess mucosal integrity. Data were analyzed by use of ANOVA and Kruskal-Wallis tests. Results—The TER in reperfused tissues was similar to the TER in control tissues and greater than the TER in ischemic tissues, which was consistent with morphological evidence of recovery in reperfused tissues. Mannitol flux was greater in ischemic tissues than in reperfused tissues. The TER and mannitol flux were not significantly affected by incubation of mucosa with NAC or MOTS. Conclusions and Clinical Relevance—Ischemia induced during the brief period allowed rapid mucosal repair and complete recovery of tissue barrier properties during reperfusion. Therefore, reperfusion injury was not observed for this method of ischemic damage in equine colonic mucosa.

Thirty horses were randomly assigned to 1 of 5 groups. All horses were anesthetized and subjected to ventral midline celiotomy, then the large colon was exteriorized and instrumented. Colonic arterial blood flow was reduced to 20% of baseline (BL) and was maintained for 3 hours. Colonic blood flow was then restored, and the colon was reperfused for an additional 3 hours. One of 5 drug solutions was administered via the jugular vein 30 minutes prior to colonic reperfusion: group 1, 0.9% NaCl; group 2, dimethyl sulfoxide: 1 g/kg of body weight; group 3, allopurinol: 25 mg/kg; group 4, 21-aminosteroid U-74389G: 10 mg/kg; and group 5, manganese chloride (MnCl2): 10 mg/kg. Hemodynamic variables were monitored and recorded at 30-minutes intervals. Systemic arterial, systemic venous (SV), and colonic venous (CV) blood samples were collected for measurement of blood gas tensions, oximetry, lactate concentration, PCV, and plasma total protein concentration. The eicosanoids, 6-keto prostaglandin F1alpha, prostaglandin E2, and thromboxane B2, were measured in CV blood, and endotoxin was measured in CV and SV blood. Full-thickness biopsy specimens were harvested from the left ventral colon for histologic evaluation and determination of wet weight-to-dry weight ratios (WW:DW). Data were analyzed, using two-way ANOVA for repeated measures, and statistical significance was set at P < 0.05. Heart rate, mean arterial pressure, and cardiac output increased with MnCl2 infusion; heart rate and cardiac output remained increased throughout the study, but mean arterial pressure returned to BL values within 30 minutes after completion of MnCl2 infusion. Other drug-induced changes were not significant. There were significant increases in mean pulmonary artery and mean right atrial pressures at 2 and 2.5 hours in horses of all groups, but other changes across time or differences among groups were not observed. Mean pulmonary artery pressure remained increased through 6 hours in all groups, but mean right atrial pressure had returned to BL values at 3 hours. Mean colonic arterial pressure was significantly decreased at 30 minutes of ischemia and remained decreased through 6 hours; however, by 3.25 hours it was significantly higher than the value at 3 hours of ischemia. Colonic arterial resistance decreased during ischemia and remained decreased throughout reperfusion in all groups; there were no differences among groups for colonic arterial resistance. Colonic venous PO2, oxygen content, and pH decreased, and PCO2 and lactate concentration increased during ischemia but returned to BL values during reperfusion. Compared with BL values, colonic oxygen extraction ratio was increased from 0.5 to 3 hours. By 15 minutes of reperfusion, colonic oxygen extraction ratio had decreased from the BL value in all groups and either remained decreased or returned to values not different from BL through 6 hours. Colonic venous 6-keto prostaglandin F1alpha and prostaglandin E2 concentrations increased during ischemia, but returned to BL on reperfusion; there were no changes in thromboxane2 concentration among or within groups. Endotoxin was not detected in CV or SV blood after ischemia or reperfusion. There were no differences among or within groups for these variables. Low-flow ischemia and reperfusion (I-R) of the large colon caused mucosal injury, as evidenced by increases in percentage of surface mucosal disruption, percentage depth of mucosal loss, mucosal hemorrhage, mucosal edema, mucosal interstitial-to-crypt ratio, mucosal neutrophil index, submucosal venular neutrophil numbers, and mucosal cellular debris index. There was a trend (P = 0.06) toward greater percentage depth of mucosal loss at 6 hours in horses treated with dimethyl sulfoxide, compared with the vehicle control solution. There were no differences in the remainder of the histologic variables among groups. Full-thickness and mucosal WW:DW increased with colonic I-R, but there were no differences among groups. There was a trend (P = 0.09) toward neutrophil accumulation, as measured by myeloperoxidase activity, in the lungs after colonic I-R, but there were no differences among groups. There was no change in lung WW:DW after colonic I-R. There were no beneficial effects of drugs directed against oxygen-derived free radical-mediated damage on colonic mucosal injury associated with low-flow I-R. Deleterious drug-induced hemodynamic effects were not observed in this study.

Twenty-four horses were randomly allocated to 3 groups. All horses underwent a ventral midline celiotomy, and the large colon was exteriorized and instrumented. Group-1 horses served as sham-operated controls, group-2 horses underwent 6 hours of colonic ischemia, and group-3 horses were subjected to 3 hours of ischemia and 3 hours of reperfusion. Baseline blood samples were collected, then low-flow colonic ischemia was induced in horses of groups 2 and 3 by reducing colonic arterial blood flow to 20% of baseline. All horses were monitored for 6 hours. Citrated systemic venous (SV) blood samples were collected from the main pulmonary artery, and colonic venous (CV) samples were collected from the colonic vein draining the ventral colon. Samples were collected at 0, and 2, 3, 3.25, 4, and 6 hours for determination of one-stage prothrombin time, activated partial thromboplastin time, antithrombin III activity, and fibrinogen concentration. Data were analyzed statistically, using two-way ANOVA for repeated measures, and post-hoc comparisons were made by use of Student Newman Keul's test. Statistical significance was set at P < 0.05. There were significant decreases in all hemostatic variables by 2 hours in SV and SV samples from horses of all 3 groups, but there were no differences among the 3 groups for any of these variables. These hemostatic alterations could have been secondary to a hypercoagulable state or to fluid therapy-induced hemodilution. Colonic ischemia-reperfusion was not the cause of these alterations because these alterations also were observed in the sham-operated control horses. Significant temporal alterations existed even after accounting for the hemodilution. The most plausible explanation for these alterations is that hemostatic activation was incited by the celiotomy and manipulation of the colon during exteriorization and instrumentation. Comparison of paired SV and CV samples for each hemostatic variable revealed significant differences for the absolute values of one-stage prothrombin time and fibrinogen concentration, but not for activated partial thromboplastin time or antithrombin III activity. This indicates that monitoring SV hemostatic variables does not necessarily provide an accurate assessment of hemostatic function in regional vascular beds. Large-colon ischemia with or without reperfusion did not alter hemostatic function.

The role of platelet-activating factor (PAF) in mediating the colonic damage that develops after large-colon torsion was studied in 14 ponies. Morphologic changes in areas of the ascending colon and selected abdominal and thoracic viscera after 1 hour of large-colon torsion and 3 to 5 hours of reperfusion were determined, as well as the protective effects of systemic administration of a specific PAF antagonist (WEB 2086). Ponies were selected then allocated at random and in equal numbers to 2 groups that received 1 of 2 treatments prior to induction of large-colon torsion: group 1 - control (saline solution), and group 2 WEB 2086 (3 mg/kg of body weight loading dose and 3 mg/kg/h for the remainder of the study). In each pony, full-thickness tissue specimens from the gastrointestinal tract-cecum, pelvic flexure, left and right ventral colon, and right dorsal colon - heart, left lung, liver, left adrenal gland, spleen, and right kidney were collected and histologically evaluated. Edema, mucosal necrosis, and neutrophil infiltration in colonic sections were graded from 0 (normal) to 3 (most severe changes). Sections of liver and lung from 3 ponies in each group, and colon from 1 pony in each group, also were examined by transmission electron microscopy to determine the presence of ultrastructural alterations. Ischemia and reperfusion induced marked changes in all sections of colon in all ponies: moderate to severe submucosal edema, moderate necrosis of the superficial epithelium and lamina propria, and necrosis of the mucosal crypt epithelium. Extra- vascular neutrophil accumulation was evident in all sections of colon and cecum, but not in other tissues. Ultrastructural lesions were not present in hepato- cytes or pneumocytes, or in the endothelial cells of liver, lung, and colon. Bacteria were observed by electron microscopy in 5% of hepatic sinusoids. Administration of a specific PAF antagonist, WEB 2086, failed to reduce severity of the observed lesions, indicating that it was not cytoprotective at the dosage used in this model of ischemia-reperfusion injury.

Effects of low-flow ischemia and reperfusion of the large colon on systemic and colonic hemodynamic and metabolic variables were determined in horses. Twenty-four adult horses were randomly allocated to 3 groups: sham-operated (n = 6), 6 hours of ischemia (n = 9), and 3 hours of ischemia and 3 hours of reperfusion (n = 9). Low-flow ischemia was induced in groups 2 and 3 by reducing colonic arterial blood flow to 20% of baseline. Heart rate, arterial blood pressures, cardiac index, pulmonary artery pressure, right atrial pressure, and colonic blood flow were monitored. Arterial, mixed-venous, and colonic venous blood gas and oximetry analyses; PCV; and blood lactate and pyruvate and plasma total protein concentrations were measured. Data were recorded, and blood samples were collected at baseline and at 30-minute intervals for 6 hours; additionally, data were collected at 185, 190, and 195 minutes (corresponding to 5, 10, and 15 minutes of reperfusion in group-3 horses). There were no differences among groups at baseline or across time for any systemic hemodynamic or metabolic variable. Colonic blood flow did not change across time in group-1 horses. Colonic blood flow significantly (P < 0.05) decreased to 20% of baseline at induction of ischemia in horses of groups 2 and 3 and remained significantly decreased throughout the ischemic period in horses of groups 2 (6 hours) and 3 (3 hours). Colonic blood flow significantly (P < 0.05) increased above baseline by 5 minutes of reperfusion in group-3 horses. Colonic oxygen delivery and oxygen consumption, and colonic venous pH, PO2, percentage saturation of hemoglobin, and oxygen content were significantly (P < 0.05) decreased within 30 minutes after induction of ischemia in horses of groups 2 and 3; colonic venous PCO2, colonic oxygen extraction ratio, and lactate and pyruvate concentrations were significantly (P < 0.05) increased by 30 minutes of ischemia. These alterations continued throughout ischemia, but within 5 minutes of reperfusion in group-3 horses, these variables either returned to baseline (pH, PCO2, lactate, pyruvate), significantly (P < 0.05) increased above baseline (PO2, oxygen content, % saturation of hemoglobin), or significantly (P < 0.05) decreased below baseline (colonic oxygen extraction ratio). Colonic oxygen consumption remained decreased during reperfusion in group-3 horses. Colonic mucosal ischemia-reperfusion injury observed in this model of ischemia was associated with local colonic hemodynamic and metabolic alterations in the presence of systemic hemodynamic and metabolic stability. Reactive hyperemia was observed at restoration of colonic blood flow in group-3 horses and persisted during reperfusion. Colonic venous metabolic alterations were corrected at reperfusion, indicating adaptation of the colon to the return of blood flow and oxygen delivery with resultant decrease in anaerobic metabolism. The early alterations in these variables may simply represent a washout of metabolic by-products.

Intramural vascular patterns of the jejunum and colon were evaluated during ischemic strangulation obstruction (ISO, 70 minutes) and subsequent reperfusion (60 minutes) in 7 adult anesthetized horses. Microvasculature of experimental and control segments was described by comparison of results from microangiography, light microscopy, and scanning electron microscopy of vascular replicas. Experimental and control segments with isolated vascular arcades were removed either immediately after the experimental period or after 60 minutes of reperfusion. Blood was flushed from the vascular system by use of isotonic NaCl, and the segments were divided. Half of each segment was perfused with a modified radiopaque medium for microangiographic evaluation, and half was perfused with dilute methylmethacrylate to create a vascular replica to be studied by scanning electron microscopy. Microangiographic section also were evaluated for histologic changes. Microvasculature of jejunal control segments and all colon segments was similar to described normal microvasculature of the equine jejunum and ascending colon. In jejunal ISO segments, intramural perfusion was redistributed away from the mucosa. In the villi, the central arteriole was short and convoluted and the subepithelial capillaries were not filled. The submucosal vessels and crypt capillaries were congested, compared with those of controls, and the serosal vessels were not filled in the ischemic segments. Histologic grade II-III mucosal lesion was seen in jejunal ISO segments. Reperfused jejunal segments had a transmural hyperemic response, and previously unfilled capillaries were observed in all intestinal layers. After reperfusion, the mucosal lesion progressed to grade III-IV and a cellular infiltrate and edema formation were observed in the serosa. The intramural vasculature of the ischemic and reperfused colon remain unchanged. Minimal histologic damage was observed in the colon after 70 minutes of ISO or after 60 minutes of reperfusion.

Sheets of mucosa from the jejunum of healthy horses were mounted in incubation chambers and bathed with Krebs-ringer bicarbonate solution. Changes in tissue function and histologic appearance were compared after the following conditions: (1) control conditions for 30 minutes with 95% O2/5% CO2 in the gas phase; (2) same conditions as control, except incubation with superoxide dismutase (300 U/ml) during the last 18 minutes; (3) anoxia for 15 minutes with 95% N2/5% CO2, followed by reoxygenation for 15 minutes; (4) same conditions as 3, except incubation with superoxide dismutase during reoxygenation; and (5) anoxia for 30 minutes. Anoxia reduced the accumulation of radiolabeled L-alanine and caused cell swelling, as indicated by an increase in tissue water and tissue Na contents. Reoxygenation improved the tissue's ability to accumulate L-alanine, but tissue swelling continued after this treatment. Tissue Na content and L-alanine accumulation were restored to control values by reoxygenation with superoxide dismutase in the bathing medium. The grade of structural damage, as indicated by separation of eptihelial cells from villi, was equally severe after all, but control, conditions. Superoxide dismutase had no effect on the tissue control conditions. Results of this study suggest that superoxide radicals are involved in the pathogenesis of reperfusion injury in equine jejunal mucosa and that this may be of clinical importance in cases of small intestinal strangulation obstruction.