EVI DENCE iVIDENCE has accumulated which suggests that gram-negative endotoxin may act on elements of the nervous system to produce significant systemic pathology.1-7 Penner and Bernheim4 demonstrated that ulcerative lesions of the gastrointestinal tract, which are present in systemic canine endotoxin shock, appear after the introduction of very small doses of Shigella endotoxin into the third ventricle. Such doses would be completely ineffective if given systemically.4 Palmerio et al5 have confirmed these observations in the dog and rabbit with Escherichia coli endotoxin. Kass et al7 observed that the lethal dose of E coli endotoxin is smaller when injected into the carotid artery of rats and that endotoxin lethality can be eliminated by posterior hypothalamic ablation.We have previously demonstrated that small doses of E coli endotoxin placed into the cisterna magna or lateral ventricle of dogs can produce a variety of systemic pathologic lesions associated with respiratory and circulatory failure.8,9The marked systemic pathology seen in meningitis1012 may also be induced by the action of bacte¬ rial products on crucial central nervous centers. The present experiments were de¬ signed to determine the effect of sterile meningococcal endotoxin placed within the cerebrospinal fluid of the ventricular system of dogs.
Material and MethodsThe principles of laboratory animal care pro¬ mulgated by the National Society for Medical Research were observed. Mongrel dogs weigh¬ ing 10 to 30 kg (22 to 66 lb) were used through¬ out this study. Endotoxin was administered to dogs anesthetized with 30 mg/kg intravenous sodium pentobarbital. The endotoxin itself was introduced into the lateral cerebral ventricle or into the third ventricle. An incision was made from the midsaggital plane vertically anterior to the ear, the muscles split, and a small disk of parietal bone removed with a trephine. An 18gauge spinal needle was passed stereotactically into the brain until ventricular fluid was en¬ countered, several milliliters of fluid were al¬ lowed to escape, and the endotoxin was in¬ jected slowly. A few milliliters of saline was then used to wash through the endotoxin, after which the intraventricular pressure was meas¬ ured. The spinal needle was then withdrawn slowly and the tissues were approximated by suture. Control dogs were injected with equiva¬ lent volumes of isotonic saline into the ventric¬ ular system or with endotoxin intravenously. Immediately after endotoxin administration, the dogs were returned to their cages where they were allowed to regain consciousness. At intervals thereafter, clinical and neurologic examinations were performed.Complete autopsies were performed as soon after death as possible. Postmortem cerebrospinal fluid and heart blood bactériologie cultures were done immediately in four of seven of the animals dying from intraventricular endotoxin.Normal control dogs were killed by large intra¬ venous injections of sodium pentobarbital. All organs were weighed, examined grossly, and the tissues fixed ...