TRANSFUSION OF BLOOD 833 



results as in benzol poisoning. Cases of acute idiopatkic purpura nemor- 

 rhagica best illustrate this. Here repeated transfusion checks hemorrhages 

 and supplies red corpuscles, and in so doing the transfused corpuscles may 

 keep the individual alive until the unknown cause diminishes so that the 

 platelets can return to normal as sometimes occurs. In cases of secondary 

 purpura hemorrhagica, and other hemorrhagic states, where the cause can- 

 not be removed, no real benefit can be anticipated from repeated 

 transfusion. 



Transfusion also finds valuable use in improving the condition of the 

 patient with anemia before operation is undertaken, even though the 

 anemia is not great. Ottenberg and Libman, among others, have com- 

 mented on the value of transfusion preparatory to operative procedures. 



Transfusion has been .used to combat sepsis and toxemias such as 

 eclampsia, but no definite beneficial results have been obtained. 



From time to time transfusions have been reported for the cure of 

 carbon monoxid poisoning, but there is almost no evidence forthcoming 

 to show that transfusion is beneficial in this condition. Crile and Lenhart 

 found that transfusion was the most efficient therapy in the restoration 

 of dogs overcome by carbon monoxid gas, but clinical results have not met 

 with the same success. Henderson has summarized our present knowledge 

 concerning the effects of carbon monoxid as follows: It is a physio- 

 logically harmless gas except in its affinity for hemoglobin, and its toxic 

 effects are entirely due to the inability of the blood combined with carbon 

 monoxid to transport oxygen. Hemoglobin has a very great affinity for 

 carbon monoxid, but the combination is not a permanent one and is rapidly 

 broken up in the presence of oxygen or pure air. Injury resulting from 

 this gas occurs during the time in which the patient breathed carbon 

 monoxid. When placed in an atmosphere of pure air almost all of the 

 carbon monoxid is eliminated from the body within a period of one to 

 three hours, if recovery is to occur. Transfusion cannot repair the injury 

 caused by this gas. The treatment consists mainly in fresh air and symp- 

 tomatic measures. However, in some instances transfusion may be very 

 beneficial, as suggested by Lindeman's case. 



In other conditions, such as nitrobenzene poisoning, there occur other 

 forms of altered hemoglobin than CO-hemoglobin, namely, methemoglobin 

 and NO-hemoglobin, which prevent oxygen from being transported. The 

 amount of these abnormal forms of hemoglobin may be so great that ex- 

 treme cyanosis is present and less than 30 per cent of oxyhemoglobin re- 

 mains. Under such conditions transfusion may be required. Usually 

 with the formation of altered hemoglobin the patient's condition is not 

 severe enough to require transfusion. Cases of nitrobenzene poisoning 

 show a surprising tendency toward spontaneous recovery when the source 

 of the poisoning is removed, as is the case in CO poisoning. However, 

 we have seen death occur from the effects of this substance and others, 



