Vascular leakage in the middle ear cavity was studied after i.v. administration of various substances in rats and determined by the Evans blue technique. Bradykinin, histamine, serotonin, acetylcholine, substance P (SP) and vasoactive intestinal polypeptide (VIP) resulted in extravasation of Evans blue. In the case of bradykinin and histamine, the leakage was dose dependent. Calcitonin gene-related peptide (CGRP) did not affect vessel permeability. In other experiments the effect of histamine antagonists was tested on production of middle ear effusion, caused by blowing air at 14 degrees C into the external auditory canal (EAC). The increase in vessel permeability in this otitis media model was inhibited by the H2-receptor antagonist cimetidine, at doses 0.1 and 1.0 mg/ml. Diphenhydramine, an H1-receptor antagonist, arrested only partly middle ear fluid accumulation. Our study demonstrated that various inflammatory mediators and neuropeptides are capable of inducing vascular leakage in the middle ear cavity. It was also concluded that H2-receptors are involved in the regulation of middle ear vascular permeability.

The production of middle ear effusion by mechanical stimulation was studied in sympathectomized and vagotomized animals. Mechanical stimulation was obtained by a 14 degree C airstream in the external auditory canal. In sympathectomized animals this procedure caused effusion production and vasoconstriction of pars flaccida vessels. Mechanical stimulation in the vagotomized animals did not cause any production of effusion material. Degranulated mast cells were observed in all animals, including controls. The study indicates that the vagal nerve, in contrast to the sympathetic nerves, is involved in the mechanisms causing the leakage of fluid into the middle ear cavity, when cold air is blown into the external auditory canal.

Previous studies have shown that arachidonic acid (AA) metabolites are important in the pathogenesis of otitis media with effusion. The AA metabolites in 4 different experimental models for otitis media were analyzed, and the effect of anti-inflammatory drugs was studied. Purulent otitis media was induced in rats by inoculation of Streptococcus pneumoniae in the tympanic bulla, serous otitis media by blocking the tympanal orifice of the eustachian tube, and mucoid otitis media by combining the two procedures. Middle ear effusion was also induced by stimulating the external auditory canal with cold air. Indomethacin and hydrocortisone were used to inhibit AA metabolism in the latter model. Lipoxygenase products dominated in the purulent and cold air otitis media models. Cyclooxygenase products dominated in the mucoid and serous models. Indomethacin inhibited accumulation of middle ear effusion in the cold air otitis media model, whereas hydrocortisone did not. Apart from AA metabolites, other mechanisms and mediators appear to be responsible for the increased vessel permeability observed in the cold air otitis media model, such as interactions between mast cells and nerves in the middle ear mucosa.

The site of leakage in middle ear vessels was determined and characterized in experimental otitis media in rats. Middle ear effusion was induced by intravenous administration or local application in the tympanic bulla of substance P (SP), acetylcholine, and histamine. In another experiment, a 14 degrees C airstream was blown into the external auditory canal. Colloidal carbon was used to trace leakage sites at the light and electron microscopic levels. All mediators tested and the 14 degrees C airstream resulted in an increased number of leaking vessels in the pars flaccida and the middle ear mucosa. The leakage sites were restricted to capillaries and postcapillary venules. Increased numbers of degranulated pars flaccida mast cells were observed for SP only. Interendothelial gaps formed in leakage vessels after administration of mediators and stimulation of the external auditory canal with a 14 degrees C airstream. Also, cytoplasmic vesicle-like structures within the endothelial cells increased in number following SP and histamine treatment, suggesting that an increased permeability in experimental otitis media does not occur exclusively through interendothelial gaps.

The distribution and origin of peptide-containing and sympathetic nerve fibers were studied in the tympanic membrane of rats with intact innervation, and in rats following sympathectomy, vagotomy, or capsaicin treatment. Nerve fibers showing substance-P-like immunoreactivity (SP-LI), calcitonin gene-related peptide (CGRP)-LI, vasoactive intestinal polypeptide (VIP)-LI, enkephalin-LI, neuropeptide Y (NPY)-LI and tyrosine hydroxylase (TH)-LI were detected along blood vessels in the pars flaccida, the external auditory canal and in the fibrocartilaginous ring of the pars tensa. In the pars flaccida there were numerous fibers demonstrating SP-LI and CGRP-LI, while there were few such fibers in the pars tensa. In both portions of the tympanic membrane these fibers were present within and beneath the keratinized stratified squamous epithelium. In the pars flaccida, nerve fibers showing SP-LI and CGRP-LI were also seen near mast cells. Sympathectomy led to a loss of nerve fibers showing TH-LI and NPY-LI, whereas the other peptide-containing nerve fibers remained unaffected. Vagotomy did not reduce the immunoreactivity for any of the neuropeptides studied. Capsaicin treatment caused a reduction in nerve fibers displaying SP-LI and CGRP-LI. The abundance of nerve fibers showing SP-LI and CGRP-LI in the keratinized squamous epithelium indicates that the tympanic membrane is richly supplied with sensory nerves. The localization of nerve fibers exhibiting these latter substances in the vicinity of mast cells in the pars flaccida suggests that this part of the tympanic membrane is a site where neurogenic inflammation occurs.