This section is from the book "Materia Medica And Therapeutics Inorganic Substances", by Charles D. F. Phillips. Also available from Amazon: Materia medica and therapeutics.
In 1852 Pravaz, of Lyons, excited the utmost interest by his discovery of the coagulating powers of ferric perchloride, its effects when injected into the vessels of animals, and its successful use in various forms of aneurism. His observations were confirmed and extended by Giraldes, Broca, and others, who formulated rules for securing a good and firm clot, and obviating the dangers of inflammation and embolism which were soon found to be involved. The greatest importance was attached to the purity and neutrality of the preparation, its due density and proportioned amount to the size of the aneurism, and to the securing of pressure on the vessel above and below the seat of operation. Five drops of an aqueous solution at 30° density (Beaume) = 1.261 sp. gr., or 10 drops at 20° (1.160), was the calculation for each cubic centimetre (15 gr.) of blood to be acted upon (Giraldes). Dieulafoy has calculated even less than this. If too strong a solution, as of 45° to 50° (Beaume) = sp. gr. 1.449 to 1.526, be used, the vascular coats may become inflamed or gangrenous, and, if compression be omitted, embolism certainly may occur; and it will be found that some fault in these respects would explain most of the serious and fatal results which excited the vehement opposition of Malgaigne and others to the new procedure.
I think that scarcely sufficient importance has been attached to some of the successful cases - notably to one of aneurismal tumor of the orbit- recorded by an American surgeon (Brainard: Lancet, ii., 1853). The ligature of one carotid had given only temporary relief, and the actual cautery still less, but a complete cure resulted from several injections of the lactate of iron (8 gr. to 1 dr.). Brainard considered this salt more suitable than the perchloride, as acting more slowly, and with less irritation or tendency to suppuration. Bribosia (Brussels), in a special treatise on the use of coagulant injections, considers them best adapted for such aneurisms as contain more liquid blood than fibrine, and are situated on the smaller arteries (e.g., those of the cranium), and not too near the trunk. It must be acknowledged, however, that the general opinion of modern surgeons is adverse to the use of the perchloride as a coagulant in aneurism: Mr. Hart points out that compression of the affected vessel above and below the sac is a sine qua non, and when this can be obtained usually safer methods of treatment may be employed ("Holmes' System," vol. iii., 2d Ed., 1870). Marsacci, in a recent work, came to the same conclusion; Gross and Erichsen also discourage it, though the latter speaks of curing with if a gluteal aneurism after some suppuration ("Science and Art of Surgery").