Transient inflammation and dysfunction of the eustachian tube secondary to multiple exposures of simulated gastroesophageal refluxant.

Gastroesophageal reflux is a common problem in the newborn and preschool periods. Recent research suggests that it may be related to eustachian tube dysfunction and otitis media with effusion. The purpose of this experiment was to investigate the relationship between simulated gastroesophageal reflux and eustachian tube dysfunction. Rat middle ears were repeatedly exposed (transtympanically) to pepsin in hydrochloric acid or to phosphate-buffered saline solution. Their eustachian tube function was evaluated by assessing passive opening and passive closing pressures, and active clearance of positive and negative pressure. The passive pressure function tests showed variable results. The rats exposed to pepsin in hydrochloric acid had an impaired ability to clear positive and negative pressure from the middle ear as compared to the rats exposed only to phosphate-buffered saline solution. The results demonstrate that multiple middle ear exposures to pepsin in hydrochloric acid leads to eustachian tube dysfunction in rats.

Late-phase allergy and eustachian tube dysfunction.

OBJECTIVE: This study investigated the role of late-phase allergy in the development of otitis media with effusion. METHODS: Brown Norway rats were sensitized to ovalbumin and later challenged transtympanically. Eustachian tube ventilatory function was assessed 2, 4, 8, 24, 28, and 32 hours postchallenge by measuring passive opening and closing pressures, active clearance of positive and negative middle ear pressure, and mucociliary clearance. RESULTS: The results demonstrate that exposure to transtympanic allergen induces eustachian tube dysfunction and subsequent formation of effusion. Allergic animals showed significant increases in passive and active opening pressures, as well as a decreased ability to actively clear middle ear pressure. Finally, the mucociliary was significantly impaired in all sensitized rats exposed to transtympanic allergen. CONCLUSION: These findings demonstrate that late-phase allergy leads to significant eustachian tube dysfunction and subsequent formation of effusion by impairing the ventilatory and clearance functions of the eustachian tube.

Function of the eustachian tube after weekly exposure to pepsin/hydrochloric acid.

OBJECTIVE: To determine the effects of repeated pepsin/hydrochloric acid (HCl) exposure on the eustachian tube (ET). STUDY DESIGN AND SETTING: ET function was studied in 22 rats. Group I (control) rats received transtympanic phosphate buffered saline solution; groups II (0.5 mg/ml) and III (2.0 mg/ml) received transtympanic pepsin/HCl. Test solutions were applied on day 0 with ET function evaluated on days 1, 2, 3, and 7 after exposure. Each 7-day period represents 1 cycle; all groups underwent 4 cycles. ET function was evaluated using passive opening and closing pressure, and active clearance of positive and negative pressure tests. RESULTS: Rats exposed to pepsin/HCl had elevated passive opening pressures and a decreased ability to clear positive and negative pressure. A temporal relationship exists. CONCLUSION: The results suggest middle ear exposure to pepsin/HCl leads to ET dysfunction in rats, and that this dysfunction is enhanced with repeated exposures. SIGNIFICANCE: Gastroesophageal reflux may induce ET dysfunction.

Action of histamine on eustachian tube function.

INTRODUCTION: The role of allergy in eustachian tube dysfunction is controversial. In this study, allergy was simulated by exposure to histamine, and eustachian tube function testing was performed in an experimental rat model. METHODS: Ventilatory function was assessed by measuring passive opening and closing pressures of the eustachian tube after challenge with either transtympanic or intranasal histamine. The mucociliary clearance time of the tubotympanum was assessed by observing dye transport from the middle ear to the nasopharynx after challenge with either transtympanic histamine or control solution. RESULTS: There was a statistically significant increase in passive opening and closing pressures with transtympanic histamine versus intranasal histamine. In addition, mucociliary clearance times of the tubotympanum after transtympanic histamine showed a statistically significant increase when compared with those after transtympanic control solution. CONCLUSIONS: Transtympanic histamine exposure causes eustachian tube dysfunction in the rat by increasing passive opening and closing pressures of the eustachian tube and impairing mucociliary clearance time.

An animal model for the study of genetic predisposition in the pathogenesis of middle ear inflammation.

OBJECTIVES: Chronic otitis media with effusion (COME) is the most prevalent inflammatory disease in children and is associated with numerous adverse long-term sequelae. Many factors have been associated with an increased risk of developing COME, one of which may be a genetic predisposition to the disease. To study the role that genetics play in the pathogenesis of COME, we used an animal model to compare the middle ear inflammatory responses in two different strains of rats (Lewis and Fisher). METHODS: In earlier studies, we demonstrated that exposure of the middle ear to endotoxin caused early extensive exudation and, later, goblet cell hyperplasia and mucin hypersecretion. In the present study, the animals were divided into six groups. In each group the animals were given transtympanic injection with gram-positive bacterial cell wall product (peptidoglycan-polysaccharide [PG-PS]). The middle ear bullae were studied at 1 week and 3 weeks after infection, and after systemic reinfection. Comparisons were made of the quantity of mucin exudate by enzyme-linked immunosorbent assay and by histological evaluation of the middle ear epithelial thickness. RESULTS: Our data demonstrate a statistically significant difference in middle ear inflammation and effusion formation between the two genetically different strains of rats. CONCLUSIONS: These data support the hypothesis that the middle ear response to PG-PS may be genetically determined and therefore suggest that genetic predisposition may play a role in the pathogenesis of COME.

Nitric oxide is a mediator of the late-phase response in an animal model of nasal allergy.

The presence of nitric oxide (NO) in the nose is well documented; however, the role of this molecule in nasal physiology is still poorly understood. Our laboratory has previously demonstrated that NO is a mediator of the immediate secretory response to an intranasal histamine challenge in a rat model of nasal allergy. Histamine challenge, however, does not elicit a late-phase response (LPR). To study the role of NO in the LPR, we developed a model of nasal allergy in which brown Norway rats are actively sensitized to the allergen ovalbumin and later challenged intranasally with either phosphate-buffered saline solution (vehicle), ovalbumin in vehicle, or ovalbumin and the NO synthase inhibitor N -nitro-l -arginine methyl ester. In each experiment, nasal lavage samples were collected 30, 120, 240, and 360 minutes after challenge. Lavage samples were analyzed for albumin content by ELISA, inflammatory cell concentration with a hemocytometer, and evidence of inflammation by light microscopy. Blocking NO synthesis with N -nitro-l -arginine methyl ester significantly inhibited both albumin exudation and inflammatory cell influx into the nasal cavity during the LPR. These data suggest that NO plays a role in the LPR of nasal allergy.

Cytotoxic changes in hair cells secondary to pneumococcal middle-ear infection.

To determine whether the bacterial toxins associated with otitis media could induce morphologic changes in the organ of Corti, we inoculated the middle-ear cavities of healthy guinea pigs with either Streptococcus pneumoniae or sterile saline and then examined the organ of Corti histologically at 1, 2, and 3 weeks postinoculation. We found that the outer hair cells (OHCs) in the infected ears underwent several changes that were dependent on both the length of time following inoculation and also the position of the OHCs in the cochlea. At 2 weeks postinoculation, 7.0 to 20% of the OHC nuclei from the infected animals became very swollen, with the most significant swelling occurring in the basal turn. At 3 weeks postinoculation, 2.5 to 3.5% of the OHCs were missing in the infected animals, with the most significant loss occurring in the basal and middle turns. These results suggest that bacterial otitis media can produce cytotoxic changes in the cochlea. These changes may be a clinically significant factor in the temporary and permanent sensorineural hearing loss that has been associated with bacterial otitis media.

Allergy increases susceptibility to otitis media with effusion in a rat model. Second place--Resident Clinical Science Award 1998.

To investigate the possible relationship between allergy and otitis media with effusion (OME), we investigated the hypothesis that allergen presentation to the middle ear causes functional disruption of the eustachian tube predisposing to the development of OME. Thirteen of 19 Brown-Norway rats were sensitized to ovalbumin, and the remaining 6 served as nonallergic controls. To mimic subclinical exposure to allergen, we transtympanically injected ovalbumin at a dose (0.01 mg) that produced no changes detectable by otologic examination. Next, both allergic and nonallergic rats were exposed to transtympanic injection of either low-dose (10 microg/mL) or high-dose (100 microg/mL) lipopolysaccharide to simulate bacterial exposure. The allergic rats were found to have larger middle ear effusions when exposed to high-dose lipopolysaccharide as compared with the nonallergic controls. This response could be inhibited by diphenhydramine. We conclude that allergen presentation to the middle ear of allergic rats causes eustachian tube dysfunction predisposing to OME.

Systemic reactivation of otitis media with effusion in a rat model.

OBJECTIVE: This study addresses the interaction of bacterial antigens, specifically peptidoglycan-polysaccharide (PG-PS) and lipopolysaccharide (LPS), in the induction and reactivation of mucoid middle ear effusions. METHODS: Twenty-seven rats underwent eustachian tube obstruction before inoculation of the middle ear bulla with PG-PS. Three weeks later, after resolution of all middle ear effusions, 6 rats were randomly selected and euthanized as the first control group (control I). The remaining 21 animals were randomly assigned to 3 groups that received intravenous injections of Krebs Ringer (control II), PG-PS, and LPS, respectively. These rats were euthanized 2 days after intravenous injection. Middle ear mucin production and histologic changes were measured in all animals. RESULTS: The mean concentrations of mucin were 0.94 +/- 0.52 mg/mL, 0.41 +/- 0.87 mg/mL, 16.33 +/- 3.67 mg/mL, and 1.15 +/- 0.41 mg/mL in the control I, control II, PG-PS, and LPS groups, respectively. Thus the mean concentration of mucin in the middle ear lavage samples was significantly greater in rats that were injected intravenously with PG-PS than in rats in other groups (P < 0.05). Histologic analyses demonstrated a greater degree of goblet cell hyperplasia in the PG-PS group than in other groups. CONCLUSIONS: This is the first animal model of recurring otitis media with effusion in which a systemic injection of PG-PS was used to reactivate a middle ear effusion in rats previously primed with a transtympanic injection of PG-PS. This study suggests that after otitis media with effusion has resolved, it may be reactivated by the presence of bacterial antigens and/or cytokines in the systemic circulation.

Mucin production in the middle ear in response to lipopolysaccharides.

OBJECTIVE: This study examined the response of middle ear tissue to establish the lowest dose of lipopolysaccharide to induce mucin production in a rat otitis media model. METHODS: Twenty-six male Sprague-Dawley rats' eustachian tubes were obstructed before transtympanic inoculation of the bulla tympanica with 35 microL of Krebs Ringer or 1, 10, 100, or 1000 microgram/mL lipopolysaccharide. After 7 days the effusion and a lavage were collected for mucin ELISA measurement, and tissue was collected for histologic evaluation. RESULTS: Mucin secretion was significantly increased in the 100 microgram/mL 51.20 +/- 13.6 microgram/mL (SE) and 1000 microgram/mL 69.42 +/- 8.57 microgram/mL groups when compared with the Krebs Ringer control group 1.84 +/- 0.28 microgram/mL (P < 0.05). Histologic evaluation shows goblet cell metaplasia and hyperplasia in the middle ear epithelium in the 1000 and 100 microgram/mL groups. CONCLUSIONS: The histology and ELISA results suggest that a middle ear effusion is generated with a dose of lipopolysaccharide as low as 100 microgram/mL.

Non-insulin-dependent diabetic microangiopathy in the inner ear.

UNLABELLED: Hearing loss has long been associated with diabetes mellitus. Microangiopathy, associated with thickening of the basement membranes of small vessels, has been implicated as a major source of multiple system organ disease. OBJECTIVE: This study was designed to evaluate changes in basement membrane thickness in the inner ear of laboratory animals suffering from non-insulin-dependent diabetes mellitus (NIDDM) with, and without, exposure to moderate intensity noise exposure in an attempt to extrapolate the same disease process in humans. DESIGN: Spontaneously hypertensive-corpulent non-insulin-dependent rats (SHR/N-cp) were selected as a genetic model for the above study. Both lean and obese rats were used in this study. A genetically similar control group of animals (LA/N-cp) were used as controls. These animals express both the lean and obese phenotypes, but they lack the NIDDM gene. Forty-eight animals in each group were sacrificed at the end of the study. The cochleas were dissected and fixed. The basement membrane of the stria vascularis was examined using transmission electron microscopy. SETTING: This study was a laboratory-based, standard animal study. MAIN OUTCOME: This study was designed to show microangiography of the inner ear as related to NIDDM with, and without, obesity and noise exposure. RESULTS/CONCLUSIONS: NIDDM alone does not cause statistically significant basement membrane thickening; however, NIDDM in combination with obesity and/or noise exposure did show significant thickening and the combination of all three showed the greatest thickening. NIDDM appeared to be the greatest contributing factor.

Role of nitric oxide in kainic acid-induced elevation of cochlear compound action potential thresholds.

Nitric oxide (NO) has specifically been found to mediate the effects of excitatory amino acids in the central nervous system (CNS). Excitatory amino acids are the primary neurotransmitters at the cochlear hair cell afferent nerve synapse. Recent studies in our laboratory demonstrate that NO synthase is an active enzyme in the spiral ganglion cells of the cochlea. Given our current understanding of neurotransmission in the cochlea, it is reasonable to postulate that the actions of NO in cochlear neuronal tissue are similar to the actions of NO in the CNS, and that NO acts as a neurotransmitter/neuromodulator in the cochlea. In addition, NO is implicated as a mediator of excitotoxicity in the CNS and may therefore play a similar role in excitotoxicity in the cochlea. To further elucidate the role of NO in cochlear excitotoxicity, this study investigated the effects of 7-nitroindazole (7-NI), a competitive inhibitor of neuronal nitric oxide synthase, with regard to kainic acid (KA)-induced elevation of compound action potential (CAP) thresholds. KA is a conformationally restricted analog of glutamate with well-known excitotoxic effects on SGC's and previously described inhibitory actions on cochlear CAP thresholds. In anesthetized gerbils, CAP thresholds were recorded before and after cochlear perfusions with control solutions of artificial perilymph solution and test solutions of KA. 7-NI was administered i.p. prior to KA perfusion in an effort to block its depolarizing and toxic effects. Results showed that cochlear perfusion with KA caused significant elevation (p < 0.05) of the mean CAP threshold. This threshold shift was significantly reduced (p < 0.05) in animals pretreated with 7-NI. These results indicate that NO is involved in the toxic effects on CAP thresholds elicited by KA in the cochlea.

Sodium nitroprusside/nitric oxide causes apoptosis in spiral ganglion cells.

OBJECTIVE: In the cochlea, excitatory amino acid receptor overstimulation induces toxicity in spiral ganglion neurons by an unknown mechanism. In the central nervous system, excitatory amino acid-induced toxicity is mediated by nitric oxide, which induces apoptosis in neurons. This study tested the hypothesis that cochlear nitric oxide-mediated toxicity is the result of induction of apoptosis in spiral ganglion neurons. METHODS: The cochleas of 15 gerbils randomly assigned to different groups were perfused for 30 minutes with a test solution of 1 mmol/L sodium nitroprusside, a nitric oxide donor, or a control solution of artificial perilymph. Animals were killed at varying times, including 2, 3, 4, 8, and 18 hours after perfusion. DNA fragmentation or in situ terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick end-labeling analysis was done on cochleas for detection of apoptosis. RESULTS: Analysis by both techniques demonstrated marked apoptotic cell changes in spiral ganglion neurons of sodium nitroprusside-treated cochleas evident 4 to 8 hours after perfusion, as compared with minimal to no evidence of apoptosis in spiral ganglion neurons of control specimens. CONCLUSIONS: Exposure to high levels of nitric oxide induces apoptosis in spiral ganglion neurons. Because apoptosis is a delayed, potentially reversible cell death pathway, this may present an opportunity for intervention to prevent or attenuate hearing damage induced by excitotoxic stimuli.

Inhibition of nitric oxide synthase causes elevation of hearing thresholds.

OBJECTIVE: Nitric oxide mediates the effects of excitatory amino acids in the central nervous system. The excitatory amino acids are thought to be the neurotransmitters at the cochlear hair cell-afferent nerve synapse. Nitric oxide synthase is present in spiral ganglion cells. This study investigated the role of nitric oxide in cochlear neurotransmission. METHODS: In gerbils, cochlear compound action potential thresholds were recorded before and after cochlear perfusions with control solutions of artificial perilymph solution and test solutions of S-methyl-L-thiocitrulline (MTC), a competitive inhibitor of nitric oxide synthase. Cochleas were also preperfused with L-arginine before perfusion with a mixture of MTC/L-arginine (to overcome competitive inhibition by MTC with L-arginine, the natural substrate of nitric oxide synthase). RESULTS: Cochlear perfusion with MTC caused significant elevations of compound action potential threshold of 51 dB as opposed to insignificant elevations of only 10 dB in control animals. An insignificant threshold shift of 9 dB was observed when L-arginine was coperfused with MTC. CONCLUSIONS: Nitric oxide is involved in neurotransmission/neuromodulation in the cochlea. Because nitric oxide is both a mediator of neurotoxicity and an initiator of apoptosis in the central nervous system, nitric oxide may play a role in these processes in the cochlea.

Interaction between hypertension and diabetes mellitus in the pathogenesis of sensorineural hearing loss.

The purpose of this study is to support the hypothesis that diabetic end-organ damage of the cochlea is augmented in the setting of hypertension. A historical perspective reviewing the effects of diabetes and hypertension as causative factors in the development of sensorineural hearing loss, as well as the basic epidemiology and pathophysiology of the renal and vascular effects of diabetes and hypertension, is presented. The results of audiologic findings in insulin-dependent diabetic patients, both normotensive and hypertensive, were analyzed and correlated with the results of animal studies to support the hypothesis that sensorineural hearing loss in patients and cochlear hair cell loss in animal studies result from the effects of hypertension in conjunction with insulin-dependent diabetes mellitus.

Role of tumor necrosis factor and interleukin-1 in endotoxin-induced middle ear effusions.

In a rat model, we investigated the role of tumor necrosis factor (TNF) and interleukin-1 (IL-1) in endotoxin-induced middle ear effusions (MEEs). After the eustachian tube was obstructed, the middle ear was transtympanically injected with 35 microL of either 1) 1 mg/ mL lipopolysaccharide (LPS); 2) LPS and 100 micrograms TNF binding protein (TNFbp); 3) LPS and 1 microgram IL-1 receptor antagonist (IL-1ra); or 4) LPS, TNFbp, and IL-1ra. Every 2 hours, the fluid within the middle ear was collected, and the quantity of albumin in the fluid, an index of vascular leakage, was determined by enzyme-linked immunosorbent assay. After 6 hours, the middle ear was fixed for histologic analysis. The TNFbp significantly attenuated vascular extravasation into the middle ear. The IL-1ra did not significantly alter effusion development. These results indicate that TNF, but not IL-1, is a mediator of LPS-induced MEE. Therefore, TNFbp may represent a novel approach to the treatment of otitis media with effusion.

The role of nitric oxide in the neural control of nasal fluid production.

The production of nasal fluids serves an important role in the protection of the upper respiratory system, but can also be a troublesome symptom of rhinitis. The chief sources of nasal fluids are serous and mucous glandular secretion, epithelial goblet cell exocytosis, and exudation from submucosal blood vessels. This study was designed to investigate the role of nitric oxide in neurogenically mediated nasal vascular exudation and mucus secretion. A rat model of the naso-nasal reflex was developed in which one nasal cavity was challenged with histamine while albumin and mucin production were measured in the continuously perfused contralateral side. Histamine challenge was associated with a significant rise in contralateral albumin and mucin content. Perfusion with a nitric oxide synthase inhibitor (L-NAME) in the nasal cavity contralateral to nasal challenge was found to block albumin leakage, but not mucin secretion, on that side. The inhibition of vascular exudation was overcome by the addition of L-arginine, the natural substrate of nitric oxide synthase, to the perfusate. Treatment of the ipsilateral nasal of the ipsilateral nasal cavity with L-NAME did not significantly after the contralateral response. A high correlation was observed between albumin and mucin concentration in the perfusate. These findings indicate that NO is a mediator of the effector arm of the naso-nasal reflex that increases vascular permeability, but is not involved in the sensory nerve afferent pathway or in reflex mucin release. Further elucidation of the role of NO in nasal physiology may lead to novel pharmacotherapeutic approaches to the treatment of allergic and non-allergic rhinitis.

Nitric oxide is a mediator of neurogenic vascular exudation in the nose.

Rhinorrhea is a troublesome symptom of rhinitis seen commonly by otolaryngologists. The sources of nasal fluid production are glandular secretions and exudation from submucosal blood vessels. This study was designed to investigate the role of nitric oxide in neurogenically mediated vascular exudation in the nose. A rat model of the nasonasal reflex was developed in which one nasal cavity was challenged with histamine while albumin exudation was measured on the contralateral side. Histamine challenge was associated with a significant rise in albumin leakage, indicating an increase in vascular permeability. Perfusion with a nitric oxide synthase inhibitor (N-nitro-L-arginine methyl ester (L-NAME)) in the nasal cavity contralateral to nasal challenge was found to block albumin exudation on that side. This inhibition was overcome by the addition of L-arginine, the natural substrate of nitric oxide synthase, to the perfusate. Treatment of the ipsilateral nasal cavity with L-NAME did not significantly decrease the contralateral response. These findings indicate that NO is an important mediator of the effector arm of the nasonasal reflex that increases vascular permeability but is not involved in the sensory nerve afferent pathway. Further elucidation of the role of NO in nasal physiology may lead to novel pharmacotherapeutic approaches to the treatment of allergic and nonallergic rhinorrhea.

Nitric oxide mediates mucin secretion in endotoxin-induced otitis media with effusion.

The mechanisms that regulate mucin release in chronic otitis media with effusion, a leading cause of hearing loss in children, remain largely unknown. We developed an animal model using Sprague-Dawley rats to determine the factors responsible for mucin production in chronic otitis media with effusion. N-nitro-L-arginine methyl ester (L-NAME), a competitive inhibitor of nitric oxide synthase, was used to investigate the role of nitric oxide in mucin secretion by the middle ear epithelium. All rats underwent eustachian tube obstruction. In the first set of rats, the middle ear was then injected transtympanically with 35 microl of either 300 mOsm Krebs-Ringer bicarbonate buffer (control group) or 1 mg/ml lipopolysaccharide in Krebs-Ringer (experimental group 1). In a second set of rats, the middle ear space was injected with lipopolysaccharide and then infused at a continuous rate for 7 days with either Krebs-Ringer (experimental group 2) or 1 mmol/L L-NAME in Krebs-Ringer (experimental group 3) through an osmotic infusion pump. After 7 days the volume of effusion and the quantity of mucin collected were significantly greater in lipopolysaccharide-exposed ears than in controls. In addition, antimucin immunostaining demonstrated mucous cell hyperplasia in response to lipopolysaccharide. The lipopolysaccharide-induced production of mucin and mucous cell hyperplasia was inhibited in ears treated with lipopolysaccharide and L-NAME. These results suggest that nitric oxide is a mediator in the pathway of mucin secretion in chronic otitis media with effusion.

The role of glucocorticoids in endotoxin-mediated otitis media with effusion.

OBJECTIVE: To understand the actions of glucocorticoids on the development and progression of endotoxin-mediated otitis media with effusion. METHODS AND DESIGN: The middle ears of 20 Sprague-Dawley rats were exposed to 35 microL of either 300-mOsm Krebs-Ringer solution (control; n = 5) or lipopolysaccharide, 1 mg/mL, dissolved in Krebs-Ringer solution (n = 15). Among the group that received lipopolysaccharide, 10 rats were randomly selected to receive dexamethasone (1 mg/kg intramuscularly), either 2 hours (n = 5) or 24 hours (n = 5) before the introduction of lipopolysaccharide. Middle ear fluid was sampled after 2, 4, and 6 hours of exposure. OUTCOME MEASURES: Middle ear fluid volume and albumin content were determined as measures of vascular extravasation. Histological sections of the middle ear mucosa were used to quantify the degree of leukocyte exudation. Data were analyzed by 1- or 2-way analysis of variance with the significance level set at P < .05. RESULTS: Lipopolysaccharide exposure caused a significant increase in the mean +/- SEM middle ear fluid volume from 29.9 +/- 0.99 to 45.8 +/- 1.4 microL between the 2- and 6-hour samplings. Lipopolysaccharide exposure caused a significant increase in the albumin content of middle ear fluid from 70.1 +/- 19.2 to 271.0 +/- 93.1 micrograms between the 2- and 6-hour samplings. Both increases were significant compared with controls. Lipopolysaccharide also caused a significant increase in leukocytes localized to the middle ear mucosa. Pretreatment of animals with dexamethasone for either 2 or 24 hours inhibited the lipopolysaccharide-induced changes. There were no differences between 2- and 24-hour pretreatment with dexamethasone. CONCLUSIONS: Dexamethasone inhibits the development of endotoxin-induced otitis media with effusion.