Host defense mechanisms in elective and emergency vascular surgery: predicting septic-related mortality.

This study examines host defense mechanisms in vascular surgery patients when admitted to the hospital to determine the frequency of abnormal host defense and the predictability of this information for septic-related mortality in affected patients. Sixty-four patients undergoing elective surgery (indications: 35 with aneurysms, 27 with aortoiliac disease, and two with renovascular disease) and nine patients undergoing emergency operation (indications: two with ruptured aneurysms, seven with aortoenteric fistulas) were skin tested when admitted to the hospital with five ubiquitous antigens (Candida, mumps, tuberculin, Trichophyton, and streptokinase-streptodornase [Varidase]) and read at 24 and 48 hours. Patients were anergic with none, relatively anergic with one, or reactive with two or more antigen responses (induration 5 mm). Eighty-nine percent of elective surgery patients were reactive, 5% were relatively anergic, and 6% were anergic. This compared with 64% reactive, 13% relatively anergic, and 17% anergic patients in a separate concurrent study of 565 surgical patients with gastrointestinal malignancies (p less than 0.001, chi 2). There were two septic-related deaths in the elective surgery group and both patients were anergic at admission, which resulted in a mortality rate of 50%. Two deaths not related to sepsis occurred in two reactive patients (4%). Logistic regression analysis predicted 3.9 deaths in the elective surgery group and four deaths were observed. Both patients with ruptured aneurysms were unreactive and both died of septic-related complications. Only one of seven patients admitted with an infected graft was reactive. All had surgery. Using the admission data of these seven patients, we predicted 2.3 deaths. Two deaths were observed 21 and 48 days after operation. Both patients were anergic and scored high probabilities of mortality (more than 0.6) at admission. The remaining emergency procedure patients scored lower probabilities of mortality (less than 0.4) at admission and none died.2+.