The role of arachidonate metabolites in ocular inflammation.

PGs and leukotrienes are said to be mediators of various aspects of inflammation. In the rabbit eye, PGs are primarily involved in vasodilation and at least in rabbits, disruption of the blood-aqueous barrier. These effects vary qualitatively in different species. In the responses of the eye to chemical, mechanical, or neurogenic injury, PG involvement is limited, as indicated by the fact that cyclooxygenase inhibitors do not totally abrogate the responses. Inflammation, whether in the eye or in other tissues, is more complex than the injury responses. Complement pathway, arachidonate metabolites, cytokines, and monokines are major components in inflammation, but not necessarily in the response of tissues to injury. Topical or intracameral administration of leukotriene B4 causes PMN chemotaxis without affecting the blood-aqueous barrier. Peptidoleukotrienes such as leukotriene C4 and D4, on the other hand, are not chemotactic, but they do increase the permeability of the conjunctival microvasculature. Cyclooxygenase and 5-lipoxygenase products have been detected in the anterior chamber during the course of experimentally induced ocular inflammation and in humans with uveitis, but whether their presence is incidental or causal is yet to be established. In spite of numerous investigations, the precise role and the extent of involvement of arachidonic acid metabolites in ocular and non-ocular inflammatory diseases remain controversial. Pharmacologic studies on the effects of CO inhibitors like indomethacin in experimental models of inflammation have provided evidence that PGs participate in vascular reactions but not in leukocyte infiltration. PGs are also capable of interacting with other vasoactive agents to enhance the responses. In humans, such roles for PGs in inflammatory diseases are probably incidental. In patients with rheumatoid arthritis, treatment with CO inhibitors provides symptomatic relief without attacking the underlying cause of the disease. Post-operative vascular leakage is reportedly inhibited by some cyclooxygenase inhibitors (Mishima et al., 1989), but it remains to be seen whether they effectively arrest the processes that in some cases lead to the progression of the disease. Most of the 5-lipoxygenase inhibitors examined to date in experimental animals have been found to be non-selective. Only recently have highly selective 5-LO inhibitors become available, and these are currently being examined. Thus, evidence from animal studies with 5-LO inhibitors suggests the involvement of LTB4 in leukocyte infiltration and of peptidoleukotrienes in bronchospasm as well as in increased microvascular permeability of the conjunctiva and other