The general similarity of the functional disturbances seen in the cases of tropical blackwater fever, hemoglobinuria from other causes and the experimental hemoglobinuria produced in dogs by transfusion with their own laked blood, indicate that both clinical symptoms and pathological lesions are essentially referable to intravascular hemolysis. This view is supported by pathological evidence secured from post mortem examinations of human cases (7, 3) and of the dogs used in these experiments.
Anuria and lesser grades of renal functional impairment seem to be referable primarily to occlusion of the renal tubules with coagulated blood pigment debris. Actual cellular injury appears to play a minor part. The functional disturbances which attend the initial phase of the disease, vomiting and shock, provide conditions peculiarly unfavorable for the secretion of urine and, therefore, may play an important secondary rôle in the production of anuria or oliguria. The shock is seen only in severe cases with massive hemolysis and may well be quite similar in its origin to that following hemorrhage. The logical therapeutic procedure would seem to be early blood transfusion, where this is not precluded, as it was in case A-5680, by the appearance of abnormal blood agglutinins. Transfusions have already been recommended and utilized with some success (9, 28, 29).
The initial vomiting, like the chill and temperature, resembles the symptoms that follow intravenous injections of any foreign protein. The cause of the subsequent vomiting is not clear. The result of emesis is, however, inevitably to deplete the salt and water stores of the body and thereby to diminish the materials required for the prdouction of urine. The rational way to meet this situation would seem to be the parenteral administration of glucose and saline solutions. Such measures may, however, fail to achieve their purpose if the vomiting is permitted to continue. Oral administration of fluids, in the face of vomiting, may only wash more salt from the body. The withdrawal of all fluids by mouth and the substitution of parenteral saline and glucose solutions may, in these circumstances, put an end to vomiting, restore the body fluids to their proper composition and, by providing conditions more favorable to urine formation, accelerate the return of normal renal function and recovery.
The actual effects of the intravascular hemolysis appear to be extremely transitory. Even without the intervention of the kidneys, in cases with the most massive hemolysis, both the pigment and the protein radicles of the released hemoglobin are removed from the blood with the greatest rapidity. This affords a further reason for believing that early treatment by transfusion and infusion, might, by preventing suppression of urine and minimizing impairment of renal function, eliminate the later secondary effects of the disease.
No support has been found for the theory that acidosis occurs in blackwater fever to offer an indication for alkaline therapy. The administration of bicarbonate in the cases described led to the development of alkalosis. In the more severe cases the alkali reserve was already somewhat above normal, before bicarbonate was given, presumably as a result of vomiting. Baker and Dodds (3), in experimental studies of hemoglobinuria, in rabbits, find that in an alkaline urine hemoglobin remains unprecipitated and unchanged, while in a highly acid urine (pH < 5.8) it is converted to methemoglobin and is precipitated if the salt concentration in the urine is excessively high. These observations have been cited as new support for alkaline therapy (28, 30) and do afford a more rational basis for this mode of treatment than the acidosis theory did. They also seem to offer an argument against the use of excessive amounts of salt solution. It must be borne in mind, however, that the patient with extreme base depletion from vomiting will excrete an acid urine, even when the serum bicarbonate is extremely elevated and that the administration of sodium chloride will, in this case, render the urine alkaline (31, 32). Furthermore, the urine after excessive vomiting contains minimal amounts of sodium chloride and the urinary concentration of this salt is not greatly increased by the administration of enough sodium chloride to restore the normal concentration in the serum. Finally dilution of the hemoglobin in the urine by the provision of a large urine volume may more than offset the effects of both reaction and salt concentration.