1. Concentric Circle Liver Distribution Models
Julie Heimbach, MD
Transplant surgeon, Mayo Clinic
UNOS Liver Committee

2. Current Distribution Unit 58 OPO/Donor Service Areas
PR & US VI DE MD MA RI HI
Arbitrary, with wide variability in size, and population
( million people.)

3. Background
During last year’s public comment period for broader sharing, multiple regions/individual public comments suggested looking instead at concentric circle model.
This has been the long-standing model used in thoracic transplantation, modified in 2006 for hearts.

4. HEART ALLOCATION STATUSES
Adult candidates
Pediatric candidates 1A
Mechanical circulatory support (MCS): VAD/30 days; TAH; IABP; ECMO;
MCS with complications;
Mechanical ventilation
Continuous single high-dose or multiple IV inotropes with continuous hemodynamic monitoring
Ventilator;
Mechanical assist device;
Balloon pump;
<6 months with reactive pulmonary hypertension >50% of systemic level;
Single high-dose or multiple inotropes 1B
MCS beyond 30 days;
Continuous IV inotropes
Single low-dose inotrope;
<6 months old;
Growth failure 2
All other active candidates
There are multiple criteria for Status 1A and Status 1B; these are defined separately by age group. In addition to the specified criteria, there is an exception process for candidates with a comparable medical urgency who do not meet the stated criteria.

5. HEART DISTRIBUTION UNITS
ZONE D:
1500-<2500 miles
ZONE C:
1000-<1500 miles
ZONE B: 500-<1000 miles
ZONE A: < 500 miles
LOCAL (DSA)
Thoracic organs are distributed based on Zones, rather than a regional system as is done for abdominal organs. The local unit represents the donation service area of the OPO where the donor hospital is located. Beyond the local unit, geographic zones are determined based on concentric circles from the donor hospital. For example, Zone A includes all candidates at transplant centers within 500 miles of the donor hospital but beyond the local DSA. Other zones are determined similarly based on increasingly larger concentric circles.
ZONE E:
>2500 miles

6. ALLOCATION ORDERING: Adult donors
Local Status 1A
Local Status 1B
Zone A Status 1A
Zone A Status 1B
Local Status 2
Zone B Status 1A
Zone B Status 1B
Zone A Status 2
Zone B Status 2
Zone C Status 1A
Zone C Status 1B
Zone C Status 2
Zone D Status 1A
Zone D Status 1B
Zone D Status 2
Zone E Status 1A
Zone E Status 1B
Zone E Status 2
Prior to the policy modification, a heart from an adult donor would be offered to all local candidates prior to being offered out to Zone A. Now only local Status 1A or 1B candidates receive offers before hearts are offered to Zone A 1A or 1B candidates. The changes on July 12, 2006, are shown in yellow.
The allocation ordering is a bit more complicated for pediatric donors, due to additional categories based on the candidate age and a recent change combining local and Zone A for some statuses. The current analysis focuses primarily on adult candidates and recipients so the pediatric allocation is not described in detail here. 6

7. FOR DEATHS PER 100 PATIENT-YEARS ON THE WAITING LIST: Adult Candidates
Status*
Waiting era: 7/12/04-7/11/06
Waiting era: 7/12/06-7/11/08
# patients ever waiting
# of deaths**
Patient years (PY) at risk
Deaths/
100 PY
STATUS 1A
2267
233
201.6
115.6
2713
205
270.5
75.8
STATUS 1B
3254
288
827.2
34.8
3748
217
857.5
25.3
STATUS 2
4936
268
4421.2
6.1
4359
222
3605.3
6.2
ALL COMBINED
7730
811
5549.5
14.6
7621
664
4811.9
13.8
Death rates per 100 patient-years on the waiting list are shown for all adult candidates on this slide, both the overall rate as well as stratified by medical urgency status. These rates are based on a 2-year waiting period prior to the policy implementation and a 2-year period following the implementation.
When looking at the overall (all statuses combined), the number of patients ever waiting during the two eras are very similar but the actual number of deaths is substantially lower, as is the number of patient-years waiting. When the deaths per 100 patient years is computed, you can see a decrease from 14.6 in the pre-policy era to 13.8 in the post-policy era.
Next let’s look at the death rates within each medical urgency status. As expected, the death rate decreases with decreasing medical urgency (Status 1A to 1B to 2) regardless of era.
It appears that there is a decline in mortality within Status 1A and Status 1B in the most recent era compared to the pre-policy era. The rate for Status 2 is almost identical in the two eras.
* Inactive (status 7) is combined with the previous active status
** Deaths include those reported to the OPTN or to SSDMF while on waiting list or within 7 days of non-transplant removal. 7

8. NUMBER OF HEART TRANSPLANTS PERFORMED
Transplant Era
Age group
TOTAL
Adult
Pediatric
Pre-policy: 7/12/05-7/11/06
1902
324
2226
Post-policy year 1: 7/12/06-7/11/07
1887
322
2209
Post-policy year 2: 7/12/07-7/11/08
1820
331
2151
The total number of heart transplants is similar over these three eras, but has decreased slightly in the two post-policy years.
The number of pediatric transplants is almost identical in the 3 eras. Any decrease that has occurred is in the adult recipients. But this may be a function of a more frequent use of LVADs (after approval of the HeartMate II) or other medical therapies, rather than as a result of the policy modification.

9. ISCHEMIA TIME BY ERA FOR ADULT RECIPIENTS
Number TXed
N with ischemia time
Median (minutes)
5th and 95th %-iles
Pre-policy: 7/12/05-7/11/06
1902
1786
192
94,
293
Post-policy year 1: 7/12/06-7/11/07
1887
1775
201
98,
302
Post-policy year 2: 7/12/07-7/11/08
1820
1755
199
100,
300
Post-policy partial year 3: 7/12/08-3/11/09
1154
1076
197
97,
308
In spite of an increased number of transplants performed outside of the local area (from 36% to 45% on prior slide), the median ischemia time has increased only slightly for adults: from 192 minutes to minutes in the post-policy eras. The 95th percentile has also increased slightly for adults, from 293 minutes pre-policy to 308 minutes in the most recent era. (Only 5 percent of transplants have a longer reported ischemia time).

10. SURVIVAL WITHIN 1 YEAR: Adult recipients: All statuses combined
We examined survival for a 20-month cohort transplanted pre-policy and a 20-month cohort post-policy. Post-transplant survival was almost identical in adult recipients in the two eras when all statuses were combined in spite of a greater percentage of transplants performed in Status 1A recipients. There was no statistical difference in these rates through 1 year. The survival rate at 1 year was 88% for both eras.
Please note that the survival scale has been expanded (as it is on the following slide).
NOTE: Scale of survival axis has been expanded.

11. SURVIVAL WITHIN 1 YEAR: Adult recipients: Status 1A at transplant
Though the survival rate for adult recipients transplanted in Status 1A is slightly higher in the post-policy era, it is not statistically different from the pre-policy survival rate. The 1-year survival rates was 85% for the earlier era and 87% for the later era. Though not displayed here, the survival rates for recipients in Status 1B and Status 2 were also not significantly different pre- and post-policy.
NOTE: Scale of survival axis has been expanded.

12. CONCLUSIONS The policy modification appears to be meeting its goals:
Waiting list mortality has declined for adult candidates.
A larger percentage of transplants are being performed in urgent candidates.
Post-transplant survival has not been adversely affected.
The results we have seen to date are consistent with the modeling predictions that were used in the policy development process.

13. Liver Concentric Circle Model:
Using LSAM, the SRTR modeled a system using concentric circles for distribution units for all adult and pediatric donors.
Additional runs were performed with sharing only for MELD > upper thresholds of 35, 32, 29, 25, and 22.

14. Methods Endpoints: deaths (pre and post) median distance traveled
Study Population
Data from listed candidates and available donor organs from 1/1/2006 to 12/31/2006 were included
Analytical Approach
LSAM was used to compare concentric circles for sharing above various MELD/PELD thresholds versus current. Results= average of 10 separate runs.
Endpoints:
deaths (pre and post)
median distance traveled

15. 1 nm=1.15 mi
= 250 nm

16. 1 nm=1.15 mi
= 500 nm

17. Concentric Circles for Adult Deceased Donor Livers–
T:\slides\Liver-Intestine Committee\2009\ \LI_req_6637_v2.ppt
Concentric Circles for Adult Deceased Donor Livers–
Current System
Regional – Status 1A/1B
Local – MELD/PELD  15
Regional – MELD/PELD  15
Local – MELD/PELD < 15
Regional – MELD/PELD < 15
National – Status 1A
National – Status 1B
National – MELD/PELD
Created By:_MKG_______________ Date: _07/15/09__
Contributing Analyst: _MKG_______ Date: _07/15/09__ 17

18. Concentric Circles for Adult Deceased Donor Livers– February 2010
T:\slides\Liver-Intestine Committee\2009\ \LI_req_6637_v2.ppt
Concentric Circles for Adult Deceased Donor Livers– February 2010
Concentric 250: 35 (29, 25, 22)
0-250 NM – Status 1A then 1B
0-250 NM – MELD/PELD  35 (29, 25, 22)
NM – Status 1A then 1B
0-250 NM – MELD/PELD (29, 25, 22)
NM– MELD/PELD  35 (29, 25, 22)
0-250 NM – MELD/PELD < 15
500-1,000 NM – Status 1A then 1B
NM – MELD/PELD (29, 25, 22)
500-1,000 NM – MELD/PELD  35 (29, 25, 22)
NM – MELD/PELD < 15
1,000+ NM – Status 1A/1B
500-1,000 NM – MELD/PELD < 35 (29, 25, 22)
1,000+ NM – MELD/PELD
Note: NM = nautical mile radii (w/ donor hospital at the center)
Created By:_MKG_______________ Date: _07/15/09__
Contributing Analyst: _MKG_______ Date: _07/15/09__ 18

19. Concentric Circles: March 2010
T:\slides\Liver-Intestine Committee\2009\ \LI_req_6637_v2.ppt
Concentric Circles: March 2010
Con (32, 29, 25, 22)
0-250 NM Status 1A
NM Status 1A
0-250 NM Status 1B
NM Status 1B
0-250 NM MELD/PELD  35 (32, 29, 25, 22)
NM MELD/PELD  35 (32, 29, 25, 22)
0-250 NM MELD/PELD (31, 28, 24, 21)
NM MELD/PELD (31, 28, 24, 21)
NM Status 1A
NM Status 1B
NM MELD/PELD  15
1000+ NM Status 1A
1000+ NM Status 1B
1000+ NM MELD/PELD  15
0-250 NM MELD/PELD < 15
NM MELD/PELD < 15
NM MELD/PELD < 15
1000+ NM MELD/PELD < 15
Note: NM = nautical mile radii (w/ donor hospital at the center)
Created By:_MKG_______________ Date: _07/15/09__
Contributing Analyst: _MKG_______ Date: _07/15/09__

20. Concentric Circles for Adult DD Livers: March
T:\slides\Liver-Intestine Committee\2009\ \LI_req_6637_v2.ppt
Concentric Circles for Adult DD Livers: March
February
March Con (32, 29, 25, 22)
0-250 NM – Status 1A, then 1B
0-250 NM – MELD/PELD  35 (29, 25, 22)
NM – Status 1A, then 1B
0-250 NM – MELD/PELD (29, 25, 22)
NM– MELD/PELD  35 (29, 25, 22)
0-250 NM – MELD/PELD < 15
500-1,000 NM – Status 1A
500-1,000 NM – Status 1B
NM – MELD/PELD (29, 25, 22)
500-1,000 NM – MELD/PELD  35 (29, 25, 22)
NM – MELD/PELD < 15
1,000+ NM – Status 1A
1,000+ NM – Status 1B
500-1,000 NM – MELD/PELD < 35 (29, 25, 22)
1,000+ NM – MELD/PELD
0-250 NM Status 1A
NM Status 1A
0-250 NM Status 1B
NM Status 1B
0-250 NM MELD/PELD  35 (32, 29, 25, 22)
NM MELD/PELD  35 (32, 29, 25, 22)
0-250 NM MELD/PELD (31, 28, 24, 21)
NM MELD/PELD (31, 28, 24, 21)
NM Status 1A
NM Status 1B
NM MELD/PELD  15
1000+ NM Status 1A
1000+ NM Status 1B
1000+ NM MELD/PELD  15
0-250 NM MELD/PELD < 15
NM MELD/PELD < 15
NM MELD/PELD < 15
1000+ NM MELD/PELD < 15
Created By:_MKG_______________ Date: _07/15/09__
Contributing Analyst: _MKG_______ Date: _07/15/09__ 20

21. Key differences Feb. to March:
small circle for 1A, then 1B, then for MELD> 35 (or other).
big circle for 1A, then 1B.
back to the small circle for 15-35, then big circle >35(other), then to small circle for MELD <15.
March:
small circle 1A, then big circle for 1A.
small circle 1B, then big circle for 1B.
small circle >35 (or other lower), big circle >35
0-250 for 15-35, then , then national before for MELD <15.

22. Median Distance Between Donor Hospital and Transplant Center: February

23. Median Distance Between Donor Hospital and Transplant Center: March
total increase mi

24. Decrease in Total Deaths (vs. Current) - February

25. Decrease in Total Deaths (vs. Current)-
March
Overall fewer deaths

26. Number of Transplants by MELD: March
*More 1A, MELD >20. Overall ~110 fewer transplants.

27. Number of Deaths by Type: March
*Overall ~ fewer deaths, slightly more post TX.

28. Number of WL Deaths by MELD- March
T:\slides\Liver-Intestine Committee\2009\ \LI_req_6637_v2.ppt
Number of WL Deaths by MELD- March
*similar small decrease in WL death in each MELD group
Created By:_MKG_______________ Date: _07/15/09__
Contributing Analyst: _MKG_______ Date: _07/15/09__

29. Summary
All of the two tiered concentric circle systems resulted in higher median distances traveled vs. current.
The distance increased as the MELD score for the upper threshold was reduced
All of the two tiered concentric circle systems resulted in fewer total deaths compared to current allocation rules in the simulations.
The decrease in the number of deaths ranged from 83 to 105.

30. Median Distance vs. Decrease in Total Deaths
Reg Shr Benefit
Five Zones
Concentric
250_1525
National
250_1532
250_1529
15/25
15/29
250_1522
15/22
15/35
250_1535
Concentric 35
15/32
Shr Pos Benefit
Reg Shr MELD
250_29
250_22
250_35
250_25
Shr35R,
National 15
Concentric
Current

31. Flight cost estimates: Life Source (MNOP) 2007-2009
=MNOP (MN, ND, SD)
Region 7= MN,SD, ND, WI, IL
2nd Tier =IA, NE, KS, MO, MI, IN
Local (MNOP) flight= $ 5,271 (n=520)
Import region 7 flight= $ 5,617 (n=52)
Import 2nd tier= $ 7,629 (n=22)
Import national= $ 13,273 (n=55)

32. Number of Transplants by MELD: Feb

33. Number of Deaths by Type: Feb.

34. Number of WL Deaths by MELD- Feb.
T:\slides\Liver-Intestine Committee\2009\ \LI_req_6637_v2.ppt
Number of WL Deaths by MELD- Feb.
Created By:_MKG_______________ Date: _07/15/09__
Contributing Analyst: _MKG_______ Date: _07/15/09__ 34