Vascular permeability during the early and late phases of ocular anaphylaxis

The role of enhanced vascular permeability in a model of ocular anaphylaxis was investigated during both early- and late-phase reactions (EPR and LPR). Vascular permeability was assessed by measuring the extravascular retention of 125I-bovine serum albumin (125I-BSA) in ocular tissues. Ten groups of guinea pigs (n = 5-12 per group) were injected with dinitrophenylated (DNP) bovine gamma globulin emulsified in Freund's adjuvants and challenged after a 4-week interval by topical application of di-DNP-lysine to one eye and phosphate-buffered saline to the other eye. Thereafter, the eyes were examined and the animals were killed at different intervals after topical challenge. They were injected intravenously with 125I-BSA 0.5 hr before death. Retained radioactivity was measured separately in four tissues. The EPR (period between 0.5-1.5 hr after challenge) was characterized by enhanced retention of radioactivity in lids, conjunctiva, and orbital content. There was no significant retention of extravascular radioactivity in the globe and lacrimal gland. Thereafter (period between 2-3.5 hr after challenge), retained radioactivity was significantly diminished. The subsequent period, between 4.5-6.5 hr (LPR), was characterized by a smaller, although significant, increment of radioactivity retained in lids and conjunctiva but not in the other tissues examined. These findings indicate that enhanced microvascular permeability occurs during two phases in actively immunized guinea pigs challenged topically with di-DNP-lysine and that these phases correspond to the clinical signs that constitute the EPR and LPR of ocular anaphylaxis.

Leukotrienes, thromboxane A2, and prostaglandins during systemic anaphylaxis in sheep

We investigated the roles of eicosanoid mediators in acute systemic anaphylaxis in anesthetized sheep. Sheep were sensitized with dinitrophenylated Ascaris suum extract and were challenged with an intravenous injection of dinitrophenylated bovine serum albumin. During anaphylaxis, cyclooxygenase inhibitors eliminated the elevation of arterial plasma levels of thromboxane B2 and 6-ketoprostaglandin F 1 alpha but markedly elevated the levels of leukotriene E4 in lung lymph without significantly eliminating elevation of plasma levels of histamine. Most of the measured physiological abnormalities accompanying anaphylaxis were aggravated by cyclooxygenase blockade. Enhancement of this anaphylactic mediator response was associated with an accentuated and prolonged increase of airway pressure (P less than 0.05, compared with sensitized, antigen-challenged but otherwise untreated sheep), a more intense hypoxemia (P less than 0.0001), and leukopenia (P less than 0.001), changes that were largely eliminated by pretreating with the sulfidopeptide leukotriene (SPLT) antagonist FPL 55712, suggesting that the SPLTs were important mediators of these responses. In contrast, the prolonged, but less severe, systemic vascular collapse and the reduced pulmonary hypertension induced by cyclooxygenase inhibitors were not influenced by the SPLT antagonist. These results demonstrate that in sheep cyclooxygenase metabolites are mainly involved in the acute, but transient, systemic and pulmonary vascular response of systemic anaphylaxis, whereas SPLTs are primarily implicated in the airway and secondary cardiovascular response. SPLT may act either directly or by potentiating the release of and reactivity to histamine and other mediators. Our data therefore suggest that a combination of cyclooxygenase and lipoxygenase inhibition will be necessary to more effectively protect against the consequences of an anaphylactic reaction.

Expression of MHC class II (Ia) antigen by the neonatal enterocyte: the effect of treatment with interferon-gamma

Immunohistochemical techniques were used to probe the expression and inducibility of class II major histocompatibility complex (MHC) (Ia) antigens by the mouse enterocyte at various stages postpartum. Expression of Ia was related to both age and intestinal location. Ia antigen was not detected until at least 1 week post-weaning and was noted thereafter in both proximal and distal intestine. Both crypt and villus enterocytes were stained in the distal small intestine, but staining was restricted to the upper portion of the villus in the proximal small intestine. Moreover, the extent of staining and the intensity of staining were greater in the distal small intestine. The effect of a single injection of recombinant mouse interferon-gamma (IFN-gamma) on Ia expression by enterocytes of 16-day-old, suckling BDF1 mice was examined. Injection of distilled water (DW) or 1 to 2 x 10(4) U IFN-gamma did not induce enterocyte Ia expression. Doses of 4-10 x 10(4) U were effective inducers of Ia on crypt and, occasionally, on lower villus cells examined 24 hr later. Staining did not persist on the enterocyte beyond 48 hr. In conclusion, Ia is not normally expressed on small intestinal enterocytes of the mouse until after weaning; however, Ia expression can be induced earlier by treatment with IFN-gamma. It is not known whether failure to detect Ia expression prior to weaning reflects a lack of positive stimuli and/or the presence of inhibitory stimuli, possibly carried in the breast milk.

Schnitzler's syndrome: a broader clinical spectrum

Schnitzler's syndrome is characterized by chronic urticaria, recurrent fever, bone pain, and lymphadenopathy in conjunction with a serum IgM M component in a concentration that is usually less than 10,000 mg/L. Complement activation and cryoprecipitation do not appear to be involved. We report two additional patients who share many of the characteristics of this entity. These patients differ from patients previously reported because of the markedly elevated IgM M-component concentration in one patient and the severity of anemia in the second patient. An increased frequency of IgG autoantibodies to interleukin-1-alpha has been reported by other investigators; it has been suggested that an antibody-mediated prolongation of the half-life of interleukin-1-alpha might account for some of the symptoms and signs of this disorder. However, neither the mediators involved in the induction of nonpruritic urticaria nor the role of the IgM M component has been established.

Inhibition of immune complex-induced enteropathy by three different platelet-activating factor receptor antagonists

We previously showed that intravenous injection of rat anti-BSA-BSA complexes (IC) prepared in 5x antigen excess rapidly induced a striate pattern of serosal (to mucosal) hemorrhage and vascular congestion throughout the small intestine of the Sprague-Dawley rat. In this study, we tested the effect of three different platelet-activating factor (PAF) receptor antagonists on the development of lesions. L-652,731, a synthetic derivative of kadsurenone (at doses of 1.3-2.7 mg/kg), SRI 63-675, a substituted quinolinium compound (6.7-15 mg/kg), and WEB 2086, a thienotriazolodiazepine (5-25 mg/kg) were each capable of completely or partially inhibiting IC-induced enteropathy in the majority of animals tested. Pretreatment with WEB 2086 prevented IC-induced hemoconcentration but not neutropenia. The antagonists did not lower the level of blood complement nor interfere with the fall in complement induced by administration of IC. The ability of PAF receptor antagonists to completely or partially inhibit IC-induced small intestinal lesions suggests that endogenous PAF is a major mediator of IC-induced enteropathy.

Adverse reactions to heparin

Heparin is a medication that has gained widespread use in clinical medicine as the therapy of choice for acute anticoagulation in the prevention and treatment of thromboembolic disease. Therapy with heparin is associated with many potential adverse side effects. Heparin-induced skin necrosis is an uncommon complication of heparin therapy that is now believed to be a thrombotic complication of heparin-induced thrombocytopenia. The pathogenesis of this disorder is unknown, but it is presumed to be immunologically mediated. The diagnosis is frequently one of exclusion. Significant morbidity and mortality may arise from failure to recognize this adverse reaction.

Anaphylaxis during induction of general anesthesia: subsequent evaluation and management

Twenty-seven patients were referred for evaluation of anaphylaxis after induction of general anesthesia (GA) in which thiobarbiturates, muscle relaxants, or antibiotics were administered intravenously. Skin testing by the prick and intracutaneous methods was performed with dilutions of the thiobarbiturates and muscle relaxants; beta-lactam reagents were used in patients who had also received these drugs. No skin test reactivity was noted in 16 normal subjects. Skin tests were positive in 13 patients (thiobarbiturates in five, muscle relaxants in six, and antibiotics in two patients). Two patients were dermatographic and yielded indeterminate skin test results. Eleven of the 27 patients subsequently had GA; all patients received a premedication regimen of prednisone and diphenhydramine. Of three patients with negative skin tests, one experienced an arrhythmia, but no other signs attributable to anaphylaxis were noted. One patient with dermatographism had GA without a reaction. Positive skin tests implicated an agent that was avoided in seven patients; one of these patients experienced delayed urticaria/angioedema after the completion of GA. Thus, no patients developed anaphylaxis during subsequent GA for which agents producing positive skin tests were avoided, and a premedication regimen was used.

Early and late phases of ocular anaphylaxis in actively immunized guinea pigs

A model of ocular anaphylaxis with distinct early- and late-phase components was studied in actively immunized guinea pigs. Twenty guinea pigs were injected with dinitrophenylated (DNP) bovine gamma globulin emulsified in Freund's complete adjuvant and challenged topically with di-DNP-lysine. Clinical signs were monitored over a 48 h period. An early-phase reaction (EPR) characterized by conjunctival edema, conjunctival erythema, lid swelling, and lid redness was observed. This reaction peaked at 0.5 h after challenge and subsided to a low point at 3-4 h. Subsequently, a second episode of lid swelling and lid redness was observed at 4-8 h. All animals in both groups exhibited an EPR. In addition, 75% of the animals underwent an EPR and an LPR. No animals exhibited an isolated LPR. Of the animals that underwent a dual response, 47% were biphasic, 6% were prolonged and 47% were multiphasic. The development of an active model of ocular anaphylaxis exhibiting both EPR and LPR components will enable studies of mechanisms which regulate the frequency and magnitude of these ocular allergic responses.

Divalent hapten-induced intestinal anaphylaxis in the mouse enhances macromolecular uptake from the stomach

The capacity of the stomach to participate in anaphylaxis induced by the hapten N,N'-di-2,4,dinitrophenyllysine (di-DNP-lysine) was examined in BDF1 female mice immunized with dinitrophenylated Ascaris suum extract. Immunized animals underwent laparotomy and nontraumatic pyloric occlusion using a microvascular clamp. Following wound closure, animals were gavage-fed ovalbumin together with di-DNP-lysine. Other mice were subjected to systemic anaphylaxis by intravenous injection of di-DNP-lysine administered 1 min after gavage feeding of ovalbumin. The intravenous and intragastric administration of di-DNP-lysine led to a sixfold or greater increase in serum immunoreactive ovalbumin. Examination of 1-micron sections of gastric tissue from DNP-Asc-immunized and unimmunized mice showed an intact mucosal and submucosal architecture. A 75% increase in the number of mast cells below the muscularis mucosa was seen in immunized compared with unimmunized BDF1 mice. Gastric tissue sections from immunized mice challenged orally or intravenously with di-DNP-lysine showed compaction of erythrocytes in blood vessels, degranulation of mast cells, degenerative changes in the gastric epithelium, expulsion of mucus from gastric glands, and edema in the lamina propria. The present model may be useful for further defining the consequences of anaphylaxis on the development of immune responses to dietary antigens.

Monoclonal antibodies to rat Kupffer cells. Anti-KCA-1 distinguishes Kupffer cells from other macrophages

Two monoclonal antibodies, anti-KCA-1 and anti-KCA-2, directed against rat Kupffer cells (hepatic sinusoidal macrophages) were developed. Immunohistologic studies of the liver and analysis of isolated hepatic cells by immunofluorescence and flow cytometry showed that the reactivity of these antibodies was restricted to macrophages. Both KCA-1+ and KCA-2+ cells were located predominantly in the periportal region; in contrast, Ia+ sinusoidal cells were located primarily in the centrilobular region. Macrophagelike cells within the portal tracts expressed KCA-2 but not KCA-1. These findings indicate the presence of heterogeneity within the macrophage population of the liver. Anti-KCA-1 reactivity appeared to be almost entirely restricted to Kupffer cells; only a few macrophages in the thymus and a small number of cells in the bone marrow expressed KCA-1. In contrast, KCA-2 was more widely distributed; splenic, lymph node, and intestinal macrophages were intensely stained with anti-KCA-2. These studies indicate that KCA-1 is a marker of Kupffer cells.

Penetrating the conjunctival barrier. The role of molecular weight

Dinitrophenyl (DNP) derivatives of various molecular weights were tested for their ability to elicit ocular anaphylaxis after topical application to the eye of immunized animals. Adult male Sprague-Dawley rats were immunized by intraperitoneal injection of DNP-Ascaris suum extracts and alum and were then skin-tested with DNP-bovine serum albumin on day 13 post-immunization to assess their sensitivity to the DNP hapten. On day 14, animals were challenged topically with DNP derivatives in one eye; PBS was applied to the contralateral, control eye. Animals were evaluated clinically, and ocular tissues were processed for histologic evaluation. The compounds used for topical ocular challenge included the DNP derivative of egg albumin (MW 43,500 D), soybean trypsin inhibitor (MW 20,080 D), insulin (MW 5733 D), B-chain insulin (MW 3496 D), and lysine (MW 478 D). Only di-DNP-lysine elicited clinical signs of redness, edema, and tearing and histologic evidence of mast cell degranulation. None of the other compounds, tested in solutions of either equal numbers of milligram per milliliter or equimolar concentrations, elicited ocular anaphylaxis after topical application. A compound of low molecular weight, less than 3496, is needed to elicit ocular anaphylaxis when applied topically.

Determination of the interval during which one application of compound 48/80 to the rat conjunctiva influences the response to a second application

We studied the effect of one application of compound 48/80 to the conjunctiva of the rat on the response to a subsequent challenge. Rats treated once showed conjunctival edema and marked degranulation of conjunctival mast cells. A second exposure to compound 48/80, applied after an interval of 24 h, produced slight clinical effects but had marked effects on conjunctival mast cells. Approximately 90% of the mast cells could no longer be observed by light microscopy. As the interval between initial challenge and rechallenge was increased, the clinical response returned, mast cells were again observable in normal numbers, and the extent of degranulation returned to approximately 50%. After an interval of 7 days, rats responded clinically as they had to the first application of compound 48/80. Although the number of mast cells in the conjunctiva was normal and extensive degranulation occurred, exocytosis was not observed. Thus some effects of the first application of compound 48/80 persisted for at least 1 week. Whether the effects observed depend upon the continued presence of compound 48/80 in the mast cell or granule membrane remains to be determined.

Clinical patterns of ocular anaphylaxis in guinea pigs passively sensitized with IgG1 antibody

A model of topically induced ocular anaphylaxis in guinea pigs was developed. A guinea pig anti-dinitrophenyl (DNP) bovine gamma-globulin antiserum rich in IgG1 antibodies was injected subconjunctivally; various serum dilutions and latent periods were tested. The anaphylactic response was elicited with di-DNP-lysine applied topically. The response was characterized by rating 11 ocular signs. Although the 4-hour latent period is considered optimal for passive cutaneous anaphylaxis, a 14-hour latent period yielded the strongest ocular reaction and appeared to the most appropriate time for clinical assessment. Both an early-phase reaction (peaking in the 21st hour) and a late-phase reaction (peaking at 6-10 h) were observed. The two signs of the early-phase reaction that recurred in the late-phase reaction were periorbital swelling and lid redness. The anaphylactic reaction displayed different patterns in different animals: protracted, biphasic and multiphasic.

Effect of topical dexamethasone on the ocular allergic reaction in passively sensitized guinea pigs

We examined the effect of extended topical pretreatment with dexamethasone (Dex) on the ocular allergic response in passively sensitized guinea pigs. The passive IgG1 antibody dependent guinea pig model used in these studies demonstrates both early- and late-phase ocular anaphylactic reactions. Fourteen hours prior to challenge with the hapten di-DNP-lysine, three groups of guinea pigs were sensitized subconjunctivally with antiserum rich in IgG1 anti-DNP antibodies in both eyes (group 1, n = 11) or in one eye (group 2, n = 9; group 3, n = 10). The contralateral eyes of animals in groups 2 and 3 received normal guinea pig serum. At 1-hour intervals, beginning 6 h prior to challenge, guinea pigs in group 1 received 0.1% Dex in 0.5% hydroxymethylcellulose (MC) topically in one eye and 0.5% MC in their other eye. Animals in group 2 were treated on the same schedule with 0.5% MC only in both eyes. Animals in group 3 were not pretreated. Conjunctival edema, conjunctival erythema, periorbital swelling, and lid erythema were scored independently in the upper and lower lids of both eyes of all animals for 10 h. Comparisons between the scores of Dex-pretreated eyes in group 1 and MC-pretreated eyes in group 2 or untreated eyes in group 3 revealed no significant differences in any clinical sign at 0.5 h (early-phase reaction). However, these comparisons demonstrated significant reductions in all four clinical signs in animals of groups 1 from 1 to 8 h and marked reduction or complete absence of signs from 6 to 10 h (late-phase reaction).(ABSTRACT TRUNCATED AT 250 WORDS)

Histology of ocular late-phase reaction in guinea pigs passively sensitized with IgG1 antibodies

The ocular late-phase reaction (LPR) is a mast cell-dependent, delayed inflammatory reaction developing 4-12 h after the early-phase reaction (EPR). We developed a passive IgG1 antibody-dependent guinea pig model that clinically reproduced the biphasic reaction of ocular EPR and LPR. An EPR was observed in all animals; a biphasic, multiphasic or prolonged inflammation was observed in the animals maintained for 9 and 24 h. The substantia propria of eyes undergoing EPR (0.5 h) showed intense edema, mast cell degranulation (88%), and 4-fold increase in eosinophils. At 9 h, the neutrophils and eosinophils had increased 11- and 25-fold, respectively. The number of basophils and lymphocytes was significantly increased compared to the controls (p less than 0.05). Of the mast cells, 45% were degranulated compared with 19% for controls. Cellular reactions had subsided by 24 h. Conjunctival epithelium also accumulated inflammatory cells as did the stroma of the lid skin. These histologic changes in ocular tissues undergoing anaphylaxis demonstrated that mast cell degranulation was most severe in the early phase, whereas neutrophil, eosinophil, and basophil accumulation was most marked in the late phase.

Divalent hapten-induced intestinal anaphylaxis in the mouse: uptake and characterization of a bystander protein

We examined the mucosal barrier function during anaphylaxis induced by the hapten N,N'-di-2,4,dinitrophenyl-lysine (di-DNP-lysine) in BDF1 female mice immunized with dinitrophenylated Ascaris suum extract. Immunized mice were gavaged with 10 mg or 50 mg of ovalbumin (OVA) with or without N,N'-di-2,4,-DNP-lysine (di-DNP-lysine). Animals that received di-DNP-lysine underwent anaphylaxis and were observed to have significantly greater serum concentrations of immunoreactive OVA (iOVA) than control mice. The severity of anaphylaxis, which varied with the dose of di-DNP-lysine administered, influenced the uptake of OVA; greater amounts of iOVA were detected in serum of mice undergoing more severe anaphylaxis. On gel permeation of serum from both groups of mice, immunoreactive OVA was found to have a molecular size similar to native OVA. Di-DNP-lysine is a synthetic hapten that reliably induced anaphylaxis in sensitized animals challenged by gavage. Anaphylaxis resulted in the uptake into the circulation of greater quantities of an unrelated protein antigen present in the intestinal lumen. The protein antigen that was taken up into the circulation appeared to be intact and thus may have an influence on the development of the immune response, or lack thereof, to this bystander antigen.

Effect of fish-fat or beef-fat supplemented diet on immune complex-induced enteropathy in the rat

In contrast to animals on a beef fat-supplemented diet (BFD), animals maintained on a fish fat-supplemented diet (FFD) incorporate increased amounts of eicosapentaenoic acid (EPA) into membrane phospholipids. Generation of lipid mediators from such tissues favors the formation of compounds with less pro-inflammatory activity than are derived from tissues poor in EPA. Nevertheless, the FFD has not had a uniformly beneficial effect on animal models of inflammatory diseases. We previously showed that intravenous injection of rat anti-BSA-BSA complexes (IC) prepared in 5x antigen excess rapidly induced a striate pattern of serosal (to mucosal) hemorrhage and vascular congestion throughout the small intestine. In this study, we tested the effect of a BFD and FFD on immune complex-induced enteropathy. After six (Expt. 1) or eight weeks (Expt. 2) on the diet, rats were injected with IC and the severity of serosal hyperemia in the small intestine was scored. In some FFD, no lesions were seen under conditions which elicited moderate to severe lesions in BFD rats. In Expt. 1 involving 22 rats and in Expt. 2 involving 28 rats, those on the FFD had a significantly lower composite lesional score compared to those on the BFD, p less than 0.005 and p less than 0.005, respectively. These results indicate that the FFD had a beneficial effect on IC-induced enteropathy. It is suggested that this effect of the FFD may be mediated primarily by a reduction in availability of platelet-activating factor.

Effect of thermal injury in the rat on transfer of IgA protein into bile

Severe thermal injury is associated with bacterial sepsis; the intestine is considered a likely source of invasive organisms. Because IgA antibody in bile accounts for much of the specific immune defense of the upper intestinal tract in the rat, the effect of thermal injury on the quantity of IgA protein in bile was examined. Sprague-Dawley rats received a 20% to 30% body surface area burn under anesthesia. Eighteen hours later the common bile duct was cannulated and bile was collected for three hours. Total IgA protein in bile decreased 90% after thermal injury. The bile volume, the concentration of bile protein, and free secretory component did not change significantly. Although blood flow to the liver 18 hours after thermal injury was not changed, there was a significant reduction in total IgA concentration in the circulation; both monomeric (m-IgA) and polymeric IgA (p-IgA) were decreased. This finding may explain, in part, the reduced concentration of IgA protein in bile. Although not examined in this study, decreased local hepatic synthesis and/or transport of p-IgA across the hepatocyte may also contribute to the reduced IgA levels in bile.

Intestinal adaptation during lactation in the mouse. I. Enhanced intestinal uptake of dietary protein antigen

Small quantities of dietary protein antigens cross the intestinal epithelium of the lactating mouse, enter the circulation, are transferred across the mammary gland into the milk and reach the suckling neonate. In this study, we sought to determine whether intestinal uptake of ovalbumin (OVA) was enhanced in lactating compared to control mice. OVA was administered by gavage under ether anaesthesia. Blood was obtained at 15, 30, 60 and 120 min and immunoreactive OVA (iOVA) measured by enzyme immunoassay. At 30 and 60 min, a three- to four-fold higher concentration of iOVA was detected in lactating compared to control mice. Because this increase in concentration of iOVA might be explained by changes in plasma volume, rate of clearance of OVA from the circulation or altered uptake from the intestine, plasma volume was measured by isotope dilution after i.v. injection of 125I-bovine serum albumin (BSA) and clearance was assessed by measuring elimination of OVA from the circulation after i.v. injection of OVA. In comparison to controls, plasma volume of Day 7-10 lactating mice was increased two-fold and no difference in clearance rate was noted. Because the increase in concentration of iOVA in lactating mice is several-fold greater than in controls, we suggest that increased intestinal uptake of the protein occurs during lactation.

Lymphedema, lymphocytic myocarditis, and sarcoidlike granulomatosis. Manifestations of Whipple's disease

A patient with Whipple's disease presented with a long prodromal period characterized by granulomatous lymphadenitis and progressive lymphedema of the extremities. No gastrointestinal tract symptoms were present and a small bowel biopsy sample was normal. His clinical condition deteriorated with the onset of lymphocytic myocarditis. At autopsy, intestinal involvement with macrophages that stained positively with periodic acid-Schiff was limited primarily to the submucosa. Diffuse fibrous effacement of lymph nodes with afferent lymphangiectasia seemed to be the mechanism of diffuse lymphedema, protein-losing enteropathy, and hypoproteinemia. Whipple's disease, therefore, should be considered in the differential diagnosis of patients presenting with granulomatous disease, lymphocytic myocarditis, or unusual lymphedema.

Characteristics of the cellular response of the rat conjunctiva to topically applied leukotriene B4

Cellular events in the rat conjunctiva were studied at 1, 6 and 24 h following the topical application of leukotriene B4. A significant increase was noted in the number of eosinophils (p less than 0.01) and neutrophils (p less than 0.05) at 1 h and only in eosinophils at 6 h after leukotriene B4 application at the 2-micrograms dose. At the 200-ng dose, the eosinophil count was significantly elevated at 1 h only, and the number of neutrophils was not significantly changed at any time point. No significant increase in cell counts was observed at the 20-ng dose. In conclusion, certain doses of topically applied leukotriene B4 significantly increased the number of eosinophils and neutrophils in the rat conjunctiva.

Response of rat conjunctival mast cells to multiple versus single applications of compound 48/80

A single application of the mast cell secretagogue compound 48/80 to the surface of the rat eye induces significant histologic changes. Ocular anaphylaxis is usually the result of repeated, not single, exposures to allergenic substances. The response of conjunctival mast cells to repeated daily applications of compound 48/80 was, therefore, evaluated. Ninety rats received one dose of compound 48/80 or phosphate-buffered saline almost daily for 17 days. The frequency and degree of mast cell degranulation and the number of mast cells and other inflammatory cells in the subepithelial conjunctiva were determined histologically. The clinical response was most marked after one application of compound 48/80; repeated daily applications markedly reduced the clinical response. In eyes treated with multiple applications, 75% fewer mast cells were observable in the conjunctiva by light microscopy compared with phosphate-buffered saline treated eyes. Most mast cells were granulated; a few showed mild to moderate degranulation. Except for epithelial damage, no tissue injury was associated with multiple applications of compound 48/80. In contrast to conjunctivae subjected to a single application of compound 48/80, conjunctivae receiving multiple applications resembled that of phosphate-buffered saline controls.