Eosinophils in the Esophagus

This editorial is accompanied by an article. Please see: Liacouras CA et al. Primary eosinophilic esophagitis in children: successful treatment with oral corticosteroids. J Pediatr Gastroenterol Nutr 1998;26:380-385. The eosinophil is an enigmatic cell whose function continues to perplex clinicians and scientists. This bilobed, granulated leukocyte is present in both normal and diseased mucosal surfaces, yet its contributions to host protection and to tissue injury are not well defined. Does the eosinophil offer protection from toxins, microbes and other foreign antigens? Or does it harm the host through the initiation and perpetuation of inflammatory processes? Whatever the answer, the characterization of eosinophils within a variety of clinical states is gradually becoming clearer. In their study in this issue of Journal of Pediatric Gastroenterology and Nutrition, Liacouras et al. address an important and timely topic, the clinical characteristics and treatment of eosinophilic esophagitis, an emerging diagnostic consideration in the pediatric population(1). The authors hypothesize that a subset of patients with isolated esophageal eosinophilia would not respond to aggressive medical treatment for gastroesophageal reflux disease (GERD) but would improve with the use of oral corticosteroids. They prospectively examined 20 patients with chronic gastrointestinal symptoms and isolated esophageal eosinophilia, in whom symptoms and histopathologic abnormalities were refractory to aggressive and prolonged medical management of GERD. The control group consisted of patients with GERD whose diagnosis was based on history and clinical response to treatment of acid reflux disease. All index patients had normal pH probes, suggesting no acid reflux. Subsequently, these patients were treated with 4 weeks of oral corticosteroids in addition to treatment for acid reflux. At the end of treatment, all patients showed improvement in their clinical complaints and histologic abnormalities. In addition, 10 remained asymptomatic 1 year later. The other 10 patients required dietary restriction, corticosteroid therapy, or both for recurrent symptoms and 1 patient's symptoms persisted despite therapy. The investigators concluded that the use of corticosteroids in the treatment of this condition provides short and long-term benefits. This is the first and largest study in which the efficacy of oral steroid treatment in the treatment of eosinophilic esophagitis has been prospectively determined. In previous studies, the use of cromolyn, dietary restriction, amino acid-based formula, and steroids were examined, but neither in the numbers nor in such a well-defined pediatric population(2-4). The authors acknowledge their study was not placebo-controlled or double-blinded, but their results demonstrate a short-term benefit in all patients examined, and a long-term benefit in half. Their results are encouraging, because in no prior works has such a response from a pharmacologic agent been demonstrated. Unfortunately, they do not state whether complications from the systemic steroids were encountered in any of their patients. Short- and long-term use of steroids has been associated with hypertension, hyperglycemia, bone demineralization, psychiatric disorders, and growth disturbance, thus making this treatment potentially undesirable for children (5). The authors suggest an immunologic etiology and note allergic disorders were more common in patients with eosinophilic esophagitis than in those with GERD. In addition, they appropriately state that before referral for antireflux surgery, some patients with presumed GERD should undergo evaluation for eosinophilic esophagitis. Their use of the pH probe to distinguish patients with eosinophilic esophagitis from patients with GERD is of vital clinical importance. The author's experience with such a large number of affected pediatric patients and their documentation of clinical and histologic improvement provides clinicians and pathologists with important information to assist in caring for these patients, who can be difficult to diagnose and treat. This article provides valuable clinicopathologic information. The number of eosinophils infiltrating the esophagus is much greater in eosinophilic esophagitis when compared to the number in GERD. This astute observation stimulates a variety of questions regarding the role of eosinophils in this and other inflammatory diseases. Are eosinophils specific effector cells that are attracted in large numbers to sites of allergic inflammation as seen in food-allergic states? Or are they indiscriminate inflammatory cells that migrate into tissues affected with such inflammatory diseases as ulcerative colitis or Crohn's disease? Although these questions have not yet been explored in studies, investigators have determined mediators produced by eosinophils and have begun to dissect mechanisms by which eosinophils migrate into sites of allergic inflammation. Eosinophils contain preformed granule proteins and can synthesize such mediators as cytokines, platelet activating factor and leukotrienes. At least four eosinophil-derived granule proteins (EDGPs) are present within the eosinophil's cytoplasm. These highly charged granule-associated proteins(major basic protein, eosinophil cationic protein, eosinophil derived neurotoxin, and eosinophil peroxidase) are released after appropriate stimulation. These proteins are thought to participate in disease processes through their cytotoxic effects-i.e., their highly charged cationic nature increases cell permeability and causes the death of certain resident cells and parasitic organisms (6). Results of recent studies suggest that eosinophil proteins stimulate additional biologic effects. For example, eosinophil cationic protein shortens coagulation time, inhibits T-cell proliferation, and induces fibroblast DNA synthesis and the release of mucus into airways(7-9). Eosinophil-derived neurotoxin is associated with ataxia and destruction of Purkinje fibers and can inhibit T-cell proliferation (7,8). Eosinophil peroxidase can stimulate platelet aggregation, basophil, and mast cell histamine release and can inhibit T-cell proliferation (10,11). Lastly, major basic protein is unique in the eosinophil protein family in its ability to induce human neutrophil superoxide production, murine airway hyperresponsiveness, respiratory epithelial ion secretion, the expression of IL (interleukin)-6, IL-11, and IL-8 from human fibroblasts and tumor necrosis factor-α release from murine mast cells(12-17). In that each EDGP is structurally, and perhaps functionally, distinct, the determination of the individual EDGP's effects within the gastrointestinal tract will be critical in defining their role in the pathogenesis of inflammation in the gastrointestinal tract. The interstitial deposition of EDGPs has been identified in a variety of inflammatory disorders. For example, major basic protein deposition has been demonstrated in the skin, lung, and gut of patients with atopic dermatitis, asthma, and Crohn's disease respectively (18,19). Bronchial lavage fluids from asthmatic patients and stool effluents from patients with inflammatory bowel disease have increased levels of major basic protein when compared with levels in normal control subjects(20,21). In a recent study, electron microscopic analyses provided evidence of granule extrusion into the squamous mucosa of patients with esophagitis (22). These findings suggest that the mucosa is exposed to EDGPs, suggesting a role for the eosinophil in inflammatory diseases. In addition, eosinophils can generate and release a variety of newly synthesized mediators, including such multifunctional cytokines as IL-1α, -3, -5, -6, -8; tumor necrosis factor-α; transforming growth factor-α; transforming growth factor-β; granulocyte-macrophage colony-stimulating factor; and interferon-γ and lipid mediators, including leukotriene C4, prostaglandins, and platelet activating factor. These studies of the newly generated mediators remain descriptive, and the functional consequences of individual cytokines released by eosinophils remains to be determined. Although most chemoattractants affect multiple cell types, evidence suggests that some factors may preferentially and exclusively influence eosinophil migration. One such molecule is the chemokine, eotaxin. Human eotaxin, a member of the C-C chemokine family, is an eosinophil-specific chemotaxin and activator that is produced by the epithelium, the endothelium, and macrophages (23). This molecule is upregulated in a variety of animal models of allergic inflammation, and genetic disruption of the eotaxin gene results in a diminution of antigen-induced eosinophilia(24). Identification of this molecule in allergic gastrointestinal diseases will provide valuable information toward understanding the pathogenesis of eosinophil-related diseases. Pharmacologic treatments targeted specifically for eosinophils are not yet available. Instead, such treatments as corticosteroids, cromolyn, and elimination diets with broad spectrums of actions have been used to treat patients affected with eosinophil-related gastrointestinal disease. These therapies decrease the release of eosinophil mediators and reduce the number of eosinophils in the tissue spaces. Corticosteroids can inhibit the release of eosinophil growth factors IL-3, IL-5, and granulocyte-macrophage colony-stimulating factor, can reduce the numbers of circulating eosinophils, and can block the transition of eosinophils to its activated, hypodense state. Corticosteroids can be administered in a variety of ways, and the method of administration may be of clinical importance. A liquid corticosteroid preparation may provide topical treatment of the esophagus, as in the allergic condition atopic dermatitis, and thus may provide better therapeutic success than does the tablet form. In a recent abstract, it is suggested that the use of an inhaled corticosteroid preparation provides effective topical treatment for patients with eosinophilic esophagitis (25). Although the use of steroids has been shown to be effective, the long-term side effects of steroids in the pediatric population may be worse than the condition. Liacouras et al. and other investigators in their studies showed that the course of eosinophilic esophagitis can be chronic and can require repeated treatments. Some patients have only intermittent symptoms and long-term use of systemic steroids may not be justified. In addition to stabilizing mast cell granule release, cromolyn can also inhibit eosinophil mediator release and T-cell function. In case reports of patients with eosinophilic esophagitis and food allergic disorders, the use of this medication has met with limited success. By reducing the antigenic exposure that the esophagogastrointestinal tract encounters, dietary restriction alone, or in conjunction with an amino acid-based formula may ameliorate symptoms and reduce the inflammatory reaction. Exact dietary restrictions are difficult to identify, and the administration of a highly restricted diet or a specialized formula can be impractical for a pediatric patient. Although acid is seemingly not the primary culprit in the pathophysiologic course of eosinophilic esophagitis, acid blockade and prokinetic agents may assist in healing by reducing exposure of an already diseased distal esophagus to further insult. Patients in the study by Liacouras et al. received treatment with both acid-suppressive and prokinetic agents, as well as with corticosteroids, which is suggestive that the beneficial response relies on both the treatment of GERD and the immunosuppressive effects of corticosteroids. Further definition of clinicopathologic features and therapeutic options of this disorder will be important in assisting clinicians in the care of pediatric patients. Although the use of endoscopy to administer provocative testing with submucosal antigens has been described as a diagnostic and research tool in adult patients with food allergic disorders, the use of these techniques in pediatric patients has not been attempted and is unlikely to gain acceptance. Thus, examinations of biopsy samples from pediatric patients and use of in vivo models of allergy will be critical to defining mediators and chemotactic factors associated with this condition. Once these conditions are defined, a more targeted pharmacotherapy can be identified. Until then, clinical trials with well-defined control groups, using known treatments, with novel techniques of administration, will be important to perform. Because the exact cause and pathogenesis has not yet been determined, perhaps this clinicopathologic entity should not yet be termed allergic esophagitis. Although an allergic origin is likely, we still do not understand the specific antigen(s), chemoattractants, or mediators that promote eosinophil infiltration and esophageal dysfunction. Before adding the“allergic” description to this entity, with all of the stigmata that this word carries, we should first try to understand the basis for the disease. In addition, we do not know whether the esophageal squamous mucosa from affected patients differs from that in the remainder of the gastrointestinal tract, making it the primary organ involved. This condition has been characterized by isolated involvement of the esophagus, but Liacouras et al. describe diarrhea as one of part of the symptom complex. This symptom suggests involvement of the small or large intestine. Thus, we do not know whether the esophagus is the “primary” organ involved or whether other organs are involved in a more diffuse disorder. Whether the eosinophil is the marker for a systemic disorder or the central cell in a local process is not certain. With the knowledge that large numbers of eosinophils are present in this clinicopathologic condition, perhaps it is better termed idiopathic eosinophilic esophagitis until more definitive studies have been performed. Unraveling the riddles of the eosinophil in this disorder and in others involving the gastrointestinal tract will provide better understanding of inflammatory responses and better care for our patients.