9 6 



Scientific Proceedings (8i). 



59 (1237) 



A consideration of the reduction of blood platelets in purpura. 

 By Alfred F. Hess, M.D. 



[From the Department of Health, New York City.] 



As is well known, typical purpura is characterized by a marked 

 diminution in the number of platelets; whereas the normal indi- 

 vidual has about 300,000 platelets per cubic centimeter of blood, 

 the purpuric individual has less than 100,000. Evidently this 

 abnormal condition may be due either to a lack of formation of 

 these cells, or to their increased destruction. A number of 

 studies have been undertaken to decide this question, but no 

 conclusive evidence has been brought forward. 



The coagulation time of the plasma centrifugalized for fifteen 

 minutes is normal in purpura or but slightly prolonged. If we 

 centrifugalize plasma for two hours we find that, whereas the 

 coagulation time in the normal case has been greatly prolonged — 

 for example — from 8 to 18 minutes (Case 1, Table I), in the 

 case of purpura, the centrifugation has frequently occasioned but 

 little or no delay in coagulation (see Table I). This distinction 

 may be interpreted as due to the fact that in the normal plasma 

 we have removed all the platelet cells, whereas in the purpuric 

 plasma, there is considerable platelet substance in solution which 

 could not be removed by means of centrifugation. 



This question can also be answered by means of employing a 

 dilute solution of hirudin, an antithrombic substance, which com- 

 bines with thromboplastic substances such as platelets. If we 

 remove practically all the suspended platelet cells by means of 

 prolonged centrifugation, we shall then be able to determine 

 whether or not there is platelet substance in solution by attempting 

 to neutralize it by the addition of hirudin. This test has been 

 carried out in many normal as well as purpura cases. It has been 

 found that whereas the addition of the hirudin solution to the 

 centrifugalized normal plasma brings about considerable delay 

 (in Case 3 of the table, from 8 to 30 minutes), in the purpuric 

 plasma on the contrary, a considerably less degree of retardation 

 is effected (for example, in Case 7 — from 11 to 16 minutes). We 



