Effects of the 2006 U.S. thoracic organ allocation change: Analysis of local impact on organ procurement and heart transplantation

Methods 

Results 

Eighty patients were transplanted in the pre-APC era. Their mean age was 43 ± 19 years, 22 (27%) were female, and 12 (15%) were <18 years of age. Seventy patients were transplanted in the post-APC era. The mean age was 46 ± 18 years, 12 (17%) were female, and 7 (10%) were <18 years of age. A comparison of the UNOS status at transplantation between eras is shown in Figure 1. After the APC, there was an increase in the number of patients transplanted in Status 1A (24% to 43%, p = 0.015) and Status 1B (20% to 33%, p = 0.093). There was a significant decrease in the number of Status 2 patients undergoing heart transplantation in the post-APC era (56% to 24%, p = 0.001).

View Large Image Download to PowerPoint

Figure 1. UNOS status at transplantation before and after thoracic organ allocation policy change (based on Organ Procurement and Transplantation Network data).

Since the APC we have seen a significant increase in the mean waiting time for Status 1A patients (5 days to 11 days, p = 0.002), in the mean graft ischemic time (196 minutes to 223 minutes, p = 0.022), in the proportion of patients with graft ischemic time >4 hours (36% to 51%, p = 0.044), and in the proportion of patients on ventricular assist devices (VADs) at the time of transplantation (17% to 31%, p = 0.036) (Table 2). Although the difference did not reach statistical significance, after the APC there was an absolute increase in the number of donor hearts imported from outside of our local procurement area (54% to 66%) and in the median procurement travel distance (306 to 355 miles). A cost analysis comparing the two eras shows that the transportation cost as well as the total procurement cost per imported donor organ both increased significantly after July 2006 (Figure 2).

Table 2. Recipient, Donor, and Procurement Characteristics in the U.T.A.H. Cardiac Transplant Program Before and After the Change in Thoracic Organ Allocation Policy

		Before July 2006 (n = 80)	After July 2006 (n = 70)	p-valuea
	1. Waiting time in last status, mean days (range)
	UNOS Status 1A	5(0–16)	11(1–29)	0.002
	UNOS Status 1B	24(0–64)	46(1–604)	NS
	UNOS Status 2	166(6–684)	210(5–896)	NS
	2. Graft ischemic time, mean minutes (range)	196(60–450)	223(54–420)	0.022
	3. Patients with graft ischemic time more than 4 hours, n (percentage)	29(36)	36(51)	0.044
	4. Patients with ventricular assist devices, n (percentage)	14(17)	22(31)	0.036
	5. Length of hospital stay, median days (range)	10(0–98)	11(2–172)	NS
	6. Imported donor hearts, n (percentage)	43(54)	46(66)	NS
	7. Procurement travel distance, median miles (range)	306(0–799)	355(0–799)	NS

Based on the Organ Procurement and Transplantation Network data and Intermountain Donor Services data. U.T.A.H., Utah Transplant Affiliated Hospitals; UNOS, United Network for Organ Sharing.

a p < 0.05 considered statistically significant (NS, not statistically significant).

View Large Image Download to PowerPoint

Figure 2. Total procurement and transportation cost per imported donor organ before and after July 2006 (based on Intermountain Donor Services data).

The waiting-list mortality did not change significantly between the two eras (p = 0.75; Figure 3). In the pre-APC era, a total of 94 patients were on the cardiac transplant waiting list and 6 of them died within 6 months of listing. In the post-APC era, 95 patients were on the waiting list and 5 of them died within 6 months of listing. Short-term post-transplant survival has also not changed significantly. Four patients died within 30 days of transplant in the pre-APC era as compared with 2 patients in the post-APC era (p = 0.5; Figure 4).

View Large Image Download to PowerPoint

Figure 3. Survival on heart transplant waiting list before and after thoracic organ allocation change (based on Organ Procurement and Transplantation Network data).

View Large Image Download to PowerPoint

Figure 4. Thirty-day survival after heart transplant, before and after the thoracic organ allocation policy change (based on Organ Procurement and Transplantation Network data).

Discussion 

The intent of the new thoracic organ allocation policy was to decrease mortality on the cardiac waiting list by expanding regional sharing of organs for the sickest patients. Although the statistical modeling that led to the change was meticulous, the real-life results of any such change are always awaited with some degree of apprehension. The waiting list is not static and it is therefore possible that, with time, the impact of an allocation intervention can change. In addition, the allocation modeling was done on national data, but significant regional differences exist in both the distribution of patients with different urgency status on the waiting list, as well as in the proportion of patients transplanted in the different categories of urgency status in the different regions.9 Therefore, it can be expected that the changes to be seen at different transplant centers and OPOs may vary as well. Our analysis focused on assessment of the impact of the APC on a 4-hospital transplant program and a local OPO. We have not seen a change in mortality on the waiting list. There was a significant increase in the number of heart transplant candidates transplanted in a higher urgency Status 1A and a noticeable decrease in the number of organs allocated to the more stable candidates in the Status 2 category. These changes were accompanied by a significant increase in average graft ischemic time, and procurement cost.

What are the implications of these findings? Our study was not powered to demonstrate changes in waiting-list mortality and this will have to be addressed by analyses of the national OPTN data. We have seen, however, significant changes that did affect the processes of care of patients awaiting heart transplantation. The change resulted in greater allocation of hearts to patients with higher urgency status. This trend observed after the APC is consistent with the national data from the OPTN/Scientific Registry of Transplant Recipients, which showed an increase in patients transplanted in Status 1A (40% in 2005, 42% in 2006, 50% in 2007) and a decrease in patients transplanted in Status 2 (25% in 2005, 20% in 2006, 14% in 2007).10, 11 This tendency clearly appears to be a positive development as statistical models by Stevenson et al5 and Krakauer et al6 demonstrated that the highest survival benefit of heart transplantation is realized by transplanting patients in the higher urgency status. Lietz et al showed that 1-year mortality of candidates in Status 1 who did not undergo transplantation approached 60%, as compared with 10% in Status 2 candidates.12 In fact, it has been questioned whether all Status 2 patients benefit from transplantation or whether it would be of benefit to define sub-groups of Status 2 patients more likely to derive benefit.13, 14, 15, 16, 17 Although some Status 2 patients remain on the waiting list in relatively stable condition for prolonged periods of time, a significant proportion of these patients will decompensate and be upgraded to Status 1A or 1B.18

The increase in the proportion of patients transplanted in Status 1A was, however, associated with additional changes that may be of concern. The median waiting time to transplant in a Status 1A patient has increased significantly. In our clinical practice, we can no longer plan on receiving a suitable organ for a patient upgraded to Status 1A within a few days. We believe this is one of the reasons we have also seen a significant increase in the number of patients being bridged to transplantation with ventricular assist devices. Arguably, compared with transplantation alone, this approach results in excess morbidity, mortality, and markedly higher cost.19 In addition, we have observed a significant increase in the mean graft ischemic time; in half of our patients, graft ischemic time surpassed 4 hours after the APC was implemented. Although we have not seen a change in short-term mortality, the longer graft ischemic time could still impact long-term survival of the patients. Further, the fact that an increased number of grafts are now being transported over longer distances by air brings two additional considerations. One is the increase in transportation and overall procurement cost that we have documented. A second relates more to logistics. These changes are happening at a time when corporate travel is in rapid decline and, with that, a decrease in the number of corporate jets available for hire for organ procurement. In our experience, it has been more difficult to secure predictable, continuous availability of aircraft for organ procurement. At times, we have had to request air service from other regions, resulting in time delays and further increases in cost.

To what degree the experience of other regions will mirror ours remains to be seen. It will also be important to examine whether the effect of the new allocation algorithm remains constant, or whether it may be altered by the very consequences of its implementation.

Conclusion 

In conclusion, in a single-area study, the 2006 change in UNOS thoracic allocation policy has resulted in an increased proportion of patients transplanted in Status 1A. In addition, there has been an increase in the time to transplant for Status 1A patients and an increase in mean graft ischemic time and in procurement costs. We did not observe a change in mortality of patients on the waiting list. The long-term impact of the allocation change on heart transplant recipient survival should be evaluated carefully.