HYPOTHERMIA AND CARDIAC SURGERY— SWAN 



405 



IS now routine. In addition, if there is evidence of postoperative bleeding, re- 

 exploration will, in the future, he more promptly done. 



Postoperative thromho-emholism was the cause of death in three patients, all 

 adults \vith repair of large atrial septal defects. We feel that the very large pul- 

 monary vascular tree associated with this disease may allow stagnation and intra- 

 vascular clotting when the blood flow through the lungs is drastically reduced by 

 repair of the lesion. For this reason, in such patients we are now giving postopera- 

 tive anticoagulants in an effort to forestall this complication. 



Cardiac arrhythmias, especially ventricular iibrillation, occurred with considerable 

 frequency in the early part of our series. Even th(jugh these hearts were usually 

 restored to a regular rhythm by resuscitative measures, the patients often died in 

 the postoperative period. Reducing the parameters of circulatory occlusion has been 

 one factor, we believe, in reducing the risk of this complication. 



In figure 1 is seen the relationship of the degree of hypothermia to mortality 

 rate. As can be seen, patients whose temperatures are lowered below 26° C. have a 

 sharp rise in their risk. 



In figure 2 is seen the relationship between the duration of total circulatory occlu- 

 sion and mortality rate. It is clear that maintaining circulatory arrest beyond eight 

 minutes also causes a marked rise in risk. 



It might be argued that the patients who were cooled below 25° C. were those 

 with the biggest, sickest hearts. We needed more time and, for this reason, sought 

 deeper hypothermia. Be that as it may, the fact remains that we did not achieve 

 a safe prolonged operative time in these patients by this means. 



For these reasons, we have come to believe that the safe parameters of open- 

 heart surgery under hypothermia as we now employ it are procedures which can 

 be done through a right cardiotomy, in less than eight minutes of occlusion time, 

 at temperatures above 26° C. 



Except for one patient with cerebral embolus, no brain damage was experienced 

 by any patient in this series. The degree of cooling appears adequate to protect the 



50-, 



40- 



30- 

 Number 



of 

 Procedures 20- 



20- 



I I PROCEDURES 



■ ■ DEATHS 



37-32 32-28 28-26 26-20 



Lowest Temperatures in Degrees Centigrade 



Fig. 1.— Mortality in relation to various temperature ranges (84 procedures j. 



