## E6

The incidence of death secondary to sepsis was also significantly higher in re-transplanted patients (60.7% of deaths in retransplants vs. 29% in primary recipients). Among those retransplanted patients in whom sepsis was the primary cause of death, there was a significantly increased incidence of fungal infection (16 of 34 patients in the retransplanted group vs. 1 of 9 patients in the primary group.) The total average hospital and ICU stay was significantly longer in both retransplanted patients as well as in pediatric patients (irrespective of whether retransplantation was performed during the same or different admission.) Both adult and pediatric retransplantation patients accrued higher total charges.

A mathematical model derived from the UCLA data can be applied to predict relative outcome based on characteristic donor and recipient variables. 12 This system can be employed to identify a sub-group of patients in whom the expected outcome is too poor to justify retransplantation. The regression equation that estimates 1-year survival in retransplanted patients is shown: Estimated survival = 0.611exp(R-1.6856)

Where 0.611is the mean 1 year survival for the patient group and R is the patient risk score calculated by: R=0.726 x ischemia + 0.561 x vent. + 0.0292 x tbil + 0.202 x Cr + 0.526 x age group. In this equation the three categorical variables are defined as follows: ischemia = 1 if >/= 12 hrs and 0 if < 12 hrs., vent.= 1 if pre-op ventilation is required and 0 if not, age group = 1 if adult and 0 if child. The two continuous variables T bil and Cr are represented as preoperative serum values in mg/dl. The mean overall risk score for the group is 1.6856. From the above equation it was determined that a risk score of > 2.3 corresponds to an expected 1 year survival of <40%. We have arbitrarily suggested that an expected survival of less than one-half of that expected from a primary transplant is an unacceptable use of a valuable organ.

Slightly cumbersome in nature, the above equation was used as the basis for a simplified five point scoring approach, in which all five variables were binary. Each covariate was assigned an equal weight of 1 point, (i.e., 1 point was awarded for adults, organ ischemia > 12h, pre-op ventilator requirement, tbili >/=13, and Cr>/ 1.6). Patients were then grouped into 1 of 6 risk classes based on the sum total of

0 0