Effect of the SRTR COVID-19 3-Month Carve-Out on Program- Specific Transplant Outcome Evaluations

Purpose: The SRTR January 2022 program evaluations (Jan 2022 program-specific reports [PSRs]) applied a COVID-19 carve-out where follow-up for transplants performed before March 13, 2020, ends on March 12, 2020;transplants performed from March 13, 2020, through June 12, 2020, are excluded;and transplants performed after June 12, 2020, are followed as usual. This study quantified the impact of the carve-out and investigated the effect of censoring COVID-19 deaths (in addition to the carve-out) on first-year posttransplant outcomes metrics in the Jan 2022 PSRs, with particular attention to investigating variation among OPTN regions. Method(s): The program-specific hazard ratios (HRs) for graft failure and patient death were estimated under 2 alternative scenarios and compared with the published HRs. In the first scenario, the COVID-19 carve-out was removed. In the second scenario, the COVID-19 carve-out was retained, but deaths due to COVID-19 infection that were not already carved out were censored. Result(s): Compared to the HRs from the Jan 22 PSRs as published with the COVID- 19 carve-out, adding censoring for the COVID-19 deaths that are not already removed by the carve-out results in very little change on average in the HRs (beta=1.0, r2=0.96). Removing the COVID-19 carve-out has a relatively larger impact on the estimated HRs (beta=0.89, r2=0.82) By geography, there were 2 slight yet statistically significant differences. When removing the carve-out, the average HR in the Northwest (OPTN Region 6) was 0.049 lower (95% CI: -0.087 to -0.011) than under the program evaluations with the carve-out. When censoring COVID-19 deaths in addition to the carve-out, the average HR in the Midwest (OPTN Regions 7, 8, and 10) was 0.009 lower (95% CI: -0.015 to -0.003) than under the program evaluations as published with only the COVID-19 carve-out. Conclusion(s): The HRs estimated by censoring COVID-19 deaths are highly correlated with those estimated with the carve-out alone. Removal of the carve-out resulted in greater variation in estimated HRs than the censoring scenario. Little variation by OPTN Region was observed, with the carve-out resulting in slightly higher HRs on average in OPTN Region 6. Censoring COVID-19 deaths imparted little regional variation, with HRs in the Midwest reduced on average by 0.009. The impact of the carve-out on program-specific evaluations will continue to be evaluated.

OPTN/SRTR 2020 Annual Data Report: Kidney.

The year 2020 presented significant challenges to the field of kidney transplantation. After increasing each year since 2015 and reaching the highest annual count to date in 2019, the total number of kidney trans- plants decreased slightly, to 23642, in 2020. The decrease in total kidney transplants was due to a decrease in living donor transplants; the number of deceased donor transplants rose in 2020. The number of patients waiting for a kidney transplant in the United States declined slightly in 2020, driven by a slight drop in the number of new candidates added in 2020 and an increase in patients removed from the waiting list owing to death-important patterns that correlated with the COVID-19 pandemic. The complexities of the pandemic were accompanied by other ongoing challenges. Nationwide, only about a quarter of waitlisted patients receive a deceased donor kidney transplant within 5 years, a proportion that varies dramatically by donation service area, from 14.8% to 73.0%. The nonutilization (discard) rate of recovered organs rose to its highest value, at 21.3%, despite a dramatic decline in the discard of organs from hepatitis C-positive donors. Nonutilization rates remain particularly high for Kidney Donor Profile Index ≥85% kidneys and kidneys from which a biopsy specimen was obtained. Due to pandemic-related disruption of living donation in spring 2020, the number of living donor transplants in 2020 declined below annual counts over the last decade. In this context, only a small proportion of the waiting list receives living donor transplants each year, and racial disparities in living donor transplant access persist. As both graft and patient survival continue to improve incrementally, the total number of living kidney transplant recipients with a functioning graft exceeded 250,000 in 2020. Pediatric transplant numbers seem to have been impacted by the COVID-19 pandemic. The total number of pediatric kidney transplants performed decreased to 715 in 2020, from a peak of 872 in 2009. Despite numerous efforts, living donor kidney transplant remains low among pediatric recipients, with continued racial disparities among recipients. Of concern, the rate of deceased donor transplant among pediatric waitlisted candidates continued to decrease, reaching its lowest point in 2020. While this may be partly explained by the COVID-19 pandemic, close attention to this trend is critically important. Congenital anomalies of the kidney and urinary tract remain the leading cause of kidney disease in the pediatric population. While most pediatric de- ceased donor recipients receive a kidney from a donor with KDPI less than 35%, most pediatric deceased donor recipients had four or more HLA mis- matches. Graft survival continues to improve, with superior survival for living donor recipients versus deceased donor recipients.

The effect of acuity circles on deceased donor transplant and offer rates across meld and exception statuses

Purpose: Acuity circles (AC) allocation was implemented on February 4, 2020. AC significantly changed the relative priority of candidates with allocation PELD/ MELD scores of 29 to 34. We therefore performed a difference-in-differences analysis for the effect of AC on adjusted deceased donor transplant and offer rates across PELD/MELD categories. Methods: The before-AC period was February 4, 2019 to February 3, 2020, and the after-AC period was February 4 to March 12, 2020, the day before the national declaration of emergency due to COVID-19. Deceased donor transplant rates used active candidate time on the waiting list during the study period. The deceased donor offer rate was the number of offers in the first 10 spots of match run a candidate received per person-year. Only offers before the final acceptance were included. Transplant and offer rates were adjusted for other candidate characteristics. Results: Candidates with PELD/MELD 29-32 and PELD/MELD 33-36 had larger differences in transplant rates before and after AC than candidates with PELD/MELD 15-28, while other PELD/MELD categories also had larger but non-significant dif Admin ferences compared to candidates with PELD/MELD 15-28 (Figure 1). In contrast, all candidates with PELD/MELD 29 and higher had dramatically larger offers rates before and after AC than candidates with PELD/MELD 15-28 (Figure 2). Conclusions: Taken together, the implementation of AC increased the relative access to deceased donor transplant for candidates with PELD/MELD of 29-36 without reducing access for candidates with higher allocation priority.

COVID-19 incidence was initially associated with post-transplant kidney graft failure

Purpose: To better understand the effect of COVID-19 on kidney posttransplant outcomes, we estimated the association of county-level COVID-19 incidence with kidney posttransplant graft failure. Methods: The study used a period-prevalent cohort of kidney recipients from March 13, 2019 to July 31, 2020 who received a transplant on or after January 1, 2000. The county-level incidence of COVID-19 for each kidney transplant program was determined from the New York Times database and aggregated into cases per 1,000,000 for each week before and after the national emergency declaration for COVID-19. Results: For each week, recipients were given the county-level incidence of the transplant program during the previous week. A two-dimensional spline estimated the effect of COVID-19 across calendar time and incidence. Conclusions: The effect of COVID-19 incidence had a nonlinear relationship with kidney graft failure, and the effect changed over the course of the pandemic. At the time of the national emergency declaration (March 13 to 19, 2020), the incidence of COVID-19 had a nonlinear effect (Figure 1, left panel): relatively flat up to an incidence of about 16, then the effect rapidly increased to a hazard ratio of about 3, for an incidence of 1024. This nonlinear effect attenuated during the weeks after the declaration of a national emergency. Roughly 10 weeks after the emergency declaration (May 22 to 28, 2020), the incidence of COVID-19 had a less dramatic effect on posttransplant graft failure rates (Figure 1, right panel). Thus, the emergence of COVID-19 coincided with a significantly higher rate of kidney graft failure, potentially from COVID-19 infection or patients not seeking for-cause medical care. However, after the initial disruption, kidney graft failure rates were less strongly associated with COVID-19 incidence, suggesting that kidney recipients and/or transplant programs may have adapted to the new conditions imposed by COVID-19. (Table Presented).

The initial impact of COVID-19 on reported graft failure rates and potential confounding of SRTR PSRS

Purpose: COVID-19 could bias the Scientific Registry of Transplant Recipients (SRTR) program-specific reports (PSRs), especially if its impact varied geographically. Methods: Recipients who received transplants from January 1, 2000 to April 30, 2020 and had graft function on March 13, 2019 were included. To assess the risk of confounding, we estimated the overall and donation service area (DSA)-specific differences in graft failure rates from March 13, 2019 to March 12, 2020 compared with rates from March 13 to April 30, 2020, after adjusting for recipient and donor characteristics. Results: Kidney, liver, and heart recipients had higher adjusted graft failure rates after COVID-19 than before (Figure 1). Graft failure rates for kidney and liver recipients who received a transplant in the New York City DSA were significantly higher after COVID-19 than before (Figures 2 and 3, respectively). Lung and heart transplant recipients had significantly less variability across DSAs. Conclusions: Taken together, these results suggest potential confounding of SRTR PSRs, especially for kidney and liver transplant programs in the New York City DSA. Thus, SRTR is censoring transplant follow-up after March 12, 2020 for PSRs released in January 2021 to minimize potential bias. However, further studies are required to identity long-term solutions for minimizing potential confounding of SRTR PSRs by COVID-19.

The effect of COVID-19 on offer acceptance rates by age and race groups

Purpose: COVID-19 causes more severe complications in older patients and disproportionately leads to poor outcomes in racial minorities. High offer acceptance rates indicate better access to transplant. Because access to transplant is critical for patients with end-stage organ failure, we investigated the effect of COVID-19 on offer acceptance rates by candidate age and race before and after the national emergency declaration on March 13, 2020 for kidney, liver, lung, and heart transplant. Methods: We used match run data from March 13, 2019 to August 31, 2020 and included offers that resulted in at least 1 acceptance. Logistic regressions estimated differences in offer acceptance by candidate age and race before and after COVID-19 (i.e., the effects for age and race interacted with an indicator of donors recovered after March 12, 2020), and the regressions adjusted for the location of the offer in the match run and other candidate and donor characteristics. Results: Overall, offer acceptance rates were lower for kidney, liver, and heart transplant after COVID-19 than before (Table 1). Differences in kidney offer acceptance across candidate age had a dose-response relationship: offer acceptance rates were higher in younger kidney candidates before and after COVID-19 than in older candidates. Offer acceptance rates for Black and Asian candidates decreased more before and after COVID-19 than for White candidates (Table 1). Offer acceptance rates for liver, lung, and heart candidates did not notably differ before and after COVID-19 by candidate age and race. Conclusions: Thus, COVID-19 inequitably affected kidney offer acceptance rates across candidate age and racial groups.