In the last decade, there has been a 2-fold increase in the number of patients in the United Kingdom who are suspended from the National Kidney Transplant Waiting List (NKTWL).1 An annual report on national transplant activity between 2017 and 2018 indicated that 1 in every 3 patients activated on the waitlist is suspended from the list at least once during their wait.1 Historically, the outcomes of these patients were never specifically included in the calculations that reported annual waitlist mortality.1
Some patients are being suspended as they are temporarily unavailable to receive a transplant due to nonclinical reasons.1 Others may be placed on the NKTWL and then be suspended almost immediately, allowing them to still collect waiting list prioritization points.1,2 Most patients, however, are suspended because their health has deteriorated, transiently or otherwise, and their unexpected clinical condition makes it unsafe for them to be considered for transplantation at that time.1
In the United Kingdom and elsewhere, we know alarmingly little about the outcomes of patients who have experienced a clinically induced suspension event while on the NKTWL. We could not identify any study that has specifically described the survival of suspended patients, either from the point of activation on the waiting list or following transplantation. Therefore, the true scale of the morbidity and mortality of this large cohort of patients remains unknown.
We performed a cohort-based study that assessed the effect of clinically induced suspension events on outcomes of patients on the NKTWL and used national data from the UK transplant registry, including all patients on the NKTWL in the period from 2000 to 2010, linked at patient level with administrative data of all deaths in the National Health Service (NHS). We used this fully representative cohort with highly complete follow-up to examine mortality and graft failure in patients who experienced a clinically induced suspension event after activation on the NKTWL. In our analysis, we built 2 models; in our first model, we explored the association of a suspension event with mortality from the point of registration on the NKTWL, and in our second model, we explored the association of suspension with mortality and graft failure from the point of transplantation.
MATERIALS AND METHODS
Linked UK Transplant Registry and Office for National Statistics Mortality Data
Since 1995, the UK transplant registry contains information about all kidney transplants performed in the 23 adult kidney transplant centers in the United Kingdom.3 The data set is managed by National Health Service Blood and Transplant. Since 1993, the mortality database managed by the Office for National Statistics (ONS) has provided a detailed source of information on deaths in England and Wales.4 It contains information on the place of death and the original underlying cause of death—taking into account information provided by medical practitioners and coroners.4
Linkage of ONS mortality data allowed the analysis of patient deaths in and outside the hospital for all patients in England and Wales who have a record in the UK transplant registry.4 The data linkage increased the information available on date and cause of death and involved matching patient identifiers: NHS number (unique patient identifier used in the NHS), gender, date of birth, and postcode. A deterministic linkage process was used in a hierarchical order according to the likelihood of the links being correct, and the process was performed by the Data Access and Request Service of NHS Digital.5
The study population included all first-time recipients of a deceased donor kidney who were aged 18 years or older and were activated on the NKTWL between January 1, 2000, and December 31, 2010 (Figure 1). Patients who were suspended for 30 days or more within the first 90 days were excluded. This was to avoid bias where patients are activated on the list and suspended soon after while the patient undergoes further assessment of their fitness for transplantation. Patients listed for multiorgan transplants or receiving living donor transplants were also excluded from the analysis. Information on recipient and donor characteristics, as well as cold ischemic time (CIT), and HLA mismatch for transplants was extracted, and cause of death was identified from both the UK transplant registry and the ONS mortality data. Two transplant surgeons considered all cause of death codes found in the ONS records and then grouped them according to cause of death criteria in the UK transplant registry.3
To analyze outcomes from the point of registration on the NKTWL, and from the point of transplantation, the study population was split into 2 cohorts. Cohort 1 was used to analyze outcomes from the point of registration and included 13 322 patients who were still listed for transplant at 1 year from first activation on the NKTWL. Patients were defined as having had a suspension event if they had been suspended for a period of 30 days or more in their first year on the NKTWL. The categorization of patients was based on suspension events after being activated on the NKTWL and to compare survival from listing we compared patients by whether they were suspended in their first year of activation. This method minimized the effect of potential survivor bias. To analyze differences in cause of death to which patients had been categorized, we split the cohort 1 into 4 separate groups (a) suspended, not transplanted; (b) suspended, transplanted; (c) not suspended, not transplanted; and (d) not suspended, transplanted. Cohort 2 included 12 238 patients and was used to analyze outcomes of patients who received a deceased donor kidney transplant. Patients in this cohort were defined as suspended if they had any suspension events of 30 days or more at any point before being transplanted.
Donor and recipients’ characteristics and cause of death were compared between those patients who had or had not experienced a suspension event. The χ2 test or Fisher exact test was used to compare categorical variables. CIT was defined as the duration between start of cold perfusion in the donor to start of blood flow through the organ in the recipient.6 Patient sensitization was defined as the HLA antibody calculated reaction frequency.7-9 This was calculated as the percentage of blood group identical donors in a pool of 10 000 recent organ donors with whom the patient is incompatible.7-9 A patient was defined as highly sensitized if they had a calculated reaction frequency of 85% or above.7-9
Kaplan-Meier estimates were used to compare posttransplant survival between suspended and nonsuspended patients. Patients were followed up until patient death or August 2018 for patients still alive. In cohort 1, patient survival was taken as the time from 1 year after activation on the NKTWL until patient death. Patients were followed up for a further 5 years (6 y from activation on the transplant list) or until death if earlier, regardless of whether the patient received a transplant or not. Patients were censored at 5 years or on removal from the transplant list where there was no death date. For those patients receiving a transplant within 5 years, patients were censored at last follow-up if earlier. In cohort 2, graft survival was taken as the time from transplantation to retransplant or return to dialysis, whichever was earlier, censoring for death with a functioning graft, at 5 years or at August 2018 if earlier. Patient survival was taken as the time from transplant until death, censoring at 5 years or at August 2018 if earlier. Unadjusted differences in patient and graft survival were assessed using the log-rank test.
Cox proportional hazards regression models were used to assess the risk-adjusted association of a suspension event on patient mortality or graft failure. Multivariable models were fitted using forward stepwise variable selection with the importance of all terms reassessed upon the addition of new factors. The importance of all factors was assessed in the multivariable models by comparing the log-likelihood ratio statistic of nested models. After identifying suitable risk factors, the importance of the suspension term was assessed.
Patients with missing data from any of the variables included in the final multivariable model were omitted or included as a categorical value where appropriate. In cohort 1, there were 6 patients omitted due to missing information on their ethnicity. In cohort 2, there were 8 patients who had no information on survival time, 18 patients in which it was unknown whether they required dialysis before transplantation and 7 and 10 patients whose respective sex or ethnicity was not recorded. In the analyses that assessed graft survival, 16 patients had no information on survival time, 141 patients whose CIT was unknown, 4 patients with no information on dialysis status, and 10 patients whose ethnicity had not been recorded.
SAS version 9.4 was used for all statistical analysis. A P of <0.05 was considered significant for each statistical analysis.
National Kidney Allocation Scheme
Before 2006, the National Kidney Allocation Scheme for donation following brain stem death was based on 3 tiers of patients defined by HLA mismatches.7 This involved the national allocation of well-matched kidneys in tiers 1 and 2, with regional allocation for less well-matched patients in tier 3.7 In each tier, a points score based on 6 factors—matchability, sensitization status, waitlist time, recipient age, difference in recipient and donor age, and transplant unit import/export balance—further determined specificity order for allocation.7 After 2006, a fully national allocation scheme involving a 5 tiered system was adopted and priority within each tier involved the calculation of points using additional factors that included blood group and a points allocated for a combined HLA mismatch and age variable.8,9 Kidneys donated after circulatory death were allocated to patients through local arrangements which are known to vary across transplant centers. Changes in the allocation system are unlikely to have affected the results of the study.
Cohort 1: Outcomes From the Point of Registration on the NKTWL
In the first year of registration, there were 2221 patients who experienced a suspension event and 11 101 patients who did not (Table 1). Patients who were suspended had an average of 1.64 suspension episodes while on the NKTWL and a median suspension duration of 246 days. They were also older and more likely to have diabetes or hypertension as their primary renal disease (Table 1). Patients who were suspended but not transplanted had the highest incidence of death, with 48% of the cohort dying within 5 years of activation on the NKTWL (Table 2). The most commonly observed cause of death in those who did not receive a transplant was cardiothoracic, including acute myocardial infarction and chronic ischemic heart disease.
Kaplan-Meier survival curves comparing outcomes from suspended and nonsuspended patients showed that 5-year patient survival is significantly worse for patients who were suspended in the first year on the NKTWL (65.7%; 95% confidence interval [CI], 63.6%-67.8%, Figure 2) compared with those who were not (79.9%; 95% CI, 79.1%-80.7%). The first Cox regression model, comparing suspended with nonsuspended patients and without adjustment for donor or recipient characteristics, found that a suspension event in the first year of being active on the NKTWL was associated with an almost 2-fold increase hazard of death (hazard ratio [HR], 1.93; CI, 1.77-2.10; P < 0.001; Table 3). Suspension in the first year was still associated with a higher hazard of death after case-mix adjustment (HR, 1.79; CI, 1.64-1.95; P < 0.001). A higher proportion of patients not suspended in the first year were transplanted (65% versus 45% in patients suspended, P < 0.001).
Cohort 2: Posttransplantation Outcomes
Of the patients receiving a transplant, there were 3292 (26.9%) patients who experienced a suspension event and 8964 who did not, before transplant (Table S1, SDC, http://links.lww.com/TP/B841). Suspended patients had an average of 2.2 suspensions, with a median suspension duration of 122 days. Compared with the nonsuspended patients, those who experienced a suspension event between 2000 and 2010 were more likely to be female, older, and from a nonwhite ethnic background. They were also more likely to be highly sensitized and to have diabetes and hypertension as their primary renal disease.
Kaplan-Meier estimates showed evidence of a difference in 5-year patient survival between suspended patients (82.7%; 95% CI, 81.2%-84.1%) and nonsuspended patients (88.2%; 95% CI, 87.5%-88.9%; P < 0.001, Figure 3A) but no difference in graft survival (84.7%; 95% CI, 83.3%-86.0% and 85.7%; 95% CI, 85.0%-86.4%; P = 0.08, respectively, Figure 3B). After adjusting for donor and recipient factors, the effects of a suspension before transplant are still evident in survival outcomes (Table 4). For example, patients who were suspended before receiving a transplant had a higher hazard of subsequent posttransplant mortality (HR, 1.21; 1.07-1.37; P = 0.003) and also graft failure (HR, 1.13; 1.00-1.27; P = 0.04).
Summary of Results
From 2000 to 2010, one-fifth of all patients had experienced a clinically induced suspension event within the first year of being activated on the NKTWL. We found almost one-half of the patients who had been suspended and then not gone onto receive a kidney transplant within 5 years died and most often from cardiothoracic-related causes. From the time of registration on the NKTWL, the rate of mortality of patients who had been suspended was nearly twice as high as those who had not been suspended. From the time of transplantation, graft and patient survival were also found to be significantly worse.
There are 2 main limitations to our analysis. The first limitation is that we did not have information on the cause of suspension for many patients. Therefore, our criteria for defining suspension events—removal from the list of 30 days or more—may on the one hand not have captured all patients who experienced a clinically induced suspension event and, on the other hand, included some patients who were suspended for reasons unrelated to their clinical condition. Those patients inadvertently included in the suspended group based on nonclinical criteria were more likely to have been physically fitter than those suspended due a deterioration of health and their potential misclassification could have led to an underestimation of the impact of clinically induced suspension events on mortality.
The second limitation is that in cohort 1 we deliberately categorized patients as being suspended only if they had experienced a suspension event in their first year of registration. Beyond the first year, it is probable that some patients in both the suspended and nonsuspended cohorts experienced other clinically induced suspension events and this approach will have masked the impact that these events may have had on mortality. However, by restricting the analysis in this way, we limited the survival bias in the suspended group but still demonstrated that the potentially detrimental effects of waiting to receive a transplant can occur very early following activation on the NKTWL.
Comparison With Other Studies
We could find no studies, in the United Kingdom or elsewhere, that had been specifically designed to assess the effect that a clinically induced suspension event had on waitlist mortality or posttransplantation outcomes. Instead, information relating to the outcome of patients on kidney transplant waiting lists appears to be largely restricted to national reports of annual kidney transplant activity.1,2 In the United Kingdom, the most recent report, also using data from the UK transplant registry, identified that from 2017 onward 6% (n = 884) of all patients registered on the NKTWL had been removed or had died while waiting for a deceased donor kidney transplant.1
In the United States, the last annual report of transplant activity, based on data from the Scientific Registry of Transplant Recipients and including kidney and kidney-pancreas patients listed for transplant in 2016, found that the number of patients inactivated from the waiting list had decreased and was now <10 000.2 This included decreases in the number of patients initially listed for transplant and then immediately suspended—a trend attributable to a new donor allocation system that no longer awards prioritization points to inactivated patients who are going through their pretransplantation work-up.2 In the same year, around 1 in 3 patients (n = 33 291) in the United States was completely removed from the kidney transplant waitlist of which one quarter were removed due to death or a deteriorating medical condition.2
Explanation of Findings
Explanations for the increasing numbers of patients experiencing a clinically induced period of suspension are likely to be multifactorial. Over the last decade, a relative shortage of donor organs could have created a cascade of events that starts with prolonged waiting times and ends with more patients being suspended or being removed from the NKTWL as the effects of sustained hemodialysis and chronic kidney disease are given more time to negatively impact their fitness for transplantation.1,2,10-13 We also know from other analysis that the proportion of patients on the NKTWL who have characteristics that make them susceptible to a suspension event—identified in our study as patients who are highly sensitized, older, of a nonwhite ethnicity, and who have diabetes or hypertension—has increased.1 Irrespective, our results suggest that for many listed patients, there is an optimal window of opportunity in which transplantation is feasible and safe and that, in addition to points awarded due to time on the waitlist, prioritization policies should incorporate a dynamic assessment of patients’ clinical condition.
Consistent increases in the number of patient suspended from the NKTWL—2305 in 2009 to 3203 in 2018—has occurred in the context of consistently low waitlist mortalities.1 Over time, significant increases in the number of kidney transplants performed, and in particular preemptive transplants (in 2018 in the United Kingdom, over one-third of living donor transplants and nearly one-fifth of deceased donor transplants were performed on patients who did not have any renal replacement therapy1) will have helped reduce the number of deaths on the NKTWL.1 However, it must also be acknowledged that the UK calculations of waitlist mortality have only ever included patients who, at the time of the analysis, were activated on the list.
We found that following a clinically induced suspension event, mortality increased from the time of registration and that, from the time of transplantation, mortality and graft failure increased. Also, almost half of all patients in our study who were suspended and never transplanted died, raising the possibility that earlier transplantation may have improved their outcome, especially as predominant causes of death in this cohort were from infection or cardiovascular-related conditions, both of which are attributable to sustained hemodialysis.13-15 However, we must also acknowledge that those who are suspended and then die, either on the waiting list or following transplantation, may represent a group of patients who have a different prognosis irrespective of whether they experience a suspension event10 and that suspension itself may be a proxy marker for poorer health that is not captured by our risk adjustment.
The patients suspended in the first year of activation represent a vulnerable cohort of patients whose outcomes are worse in comparison to other patients accepted for deceased donor kidney transplantation. The higher mortality of these patients, both before and after transplantation, should be recognized in future policies on organ allocation, and the outcomes of all suspended patients should be included separately in calculations of waitlist mortality.
This research should alert clinicians to the significance of a clinically induced suspension event. Our results indicate that these events represent the surrogate measure of a critical risk associated with increased mortality. Therefore, periods of clinically induced suspension from the NKTWL should now act as a proxy for patient fitness, be a catalyst for closer monitoring on the waiting list, and invoke a recognition that for these patients kidney transplantation may now be a life-saving rather than life-prolonging intervention. Also, in line with other forms of solid organ transplantation that are considered life-saving, consideration must be given to conducting research that predicts the outcomes of patients suspended from the NKTWL when the suspension event itself is included alongside other clinical criteria that are used to prioritize patients.16
In our analysis of waiting list mortality, we looked at early suspensions only. Considering that the average waiting time for deceased donor kidney transplantation is around 30 months—and some groups of patients may wait considerably longer—it is probable that clinically induced suspension events might involve a much higher number of patients.1 It is also possible that the impact of a suspension event on mortality differs according to both time spent on dialysis (dialysis vintage) and according to regional variations in access to transplantation.17 Therefore, to further define the role that suspension events have on the outcomes of patients, national databases need to better capture the causes of clinically induced suspension events and the characteristics of the patients who experience them.
Almost 1 in 5 patients experiences a clinically induced suspension event within the first year of activation on the NKTWL. A clinically induced suspension event is associated with worse survival both on the waiting list and following transplantation. Earlier prioritization of patients at risk of experiencing a suspension event may improve their outcomes.
The authors would like to thank all kidney transplant centers for providing data to the UK transplant registry. We would also like to thank all those involved in collecting and handling kidney transplant Service data at National Health Blood and Transplant (NHSBT). This report is independent research arising from a Doctoral Research Fellowship (DRF-2016-09-132) awarded to D.W. and supported by the National Institute for Health Research. The views expressed in this publication are those of the authors and not necessarily those of the NHS, the National Institute for Health Research, Health Education England, or the Department of Health.
2. Hart A, Smith JM, Skeans MA, et al. OPTN/SRTR annual data report: kidney. Am J Transplant. 2018; 18:18–113
3. NHS Blood and TransplantAccess data: information about how to access national and local data.Available at https://www.odt.nhs.uk/statistics-and-reports/access-data/
. Accessed February 18, 2019
5. NHS DigitalData Access Request Service (DARS).Available at https://digital.nhs.uk/services/data-access-request-service-dars
. Accessed August 12, 2018
6. Pan ET, Yoeli D, Galvan NTN, et al. Cold ischemia time is an important risk factor for post-liver transplant prolonged length of stay. Liver Transpl. 2018; 24:762–768
7. Fuggle SV, Johnson RJ, Bradley JA, et al.; Kidney Advisory Group of NHS Blood and TransplantImpact of the 1998 UK national allocation scheme for deceased heartbeating donor kidneys. Transplantation. 2010; 89:372–378
8. Johnson RJ, Fuggle SV, O’Neill J, et al.; Kidney Advisory Group of NHS Blood and TransplantFactors influencing outcome after deceased heart beating donor kidney transplantation in the United Kingdom: an evidence base for a new national kidney allocation policy. Transplantation. 2010; 89:379–386
9. Johnson RJ, Fuggle SV, Mumford L, et al.; Kidney Advisory Group of NHS Blood and TransplantA new UK 2006 national kidney allocation scheme for deceased heart-beating donor kidneys. Transplantation. 2010; 89:387–394
10. Stel VS, Kramar R, Leivestad T, et al. Time trend in access to the waiting list and renal transplantation: a comparison of four European countries. Nephrol Dial Transplant. 2012; 27:3621–3631
11. Gill JS, Tonelli M, Johnson N, et al. The impact of waiting time and comorbid conditions on the survival benefit of kidney transplantation. Kidney Int. 2005; 68:2345–2351
12. Sørensen VR, Heaf J, Wehberg S, et al. Survival benefit in renal transplantation despite high comorbidity. Transplantation. 2016; 100:2160–2167
13. Hernández D, Alonso-Titos J, Armas-Padrón AM, et al. Mortality in elderly waiting-list patients versus age-matched kidney transplant recipients: where is the risk? Kidney Blood Press Res. 2018; 43:256–275
14. Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med. 1999; 341:1725–1730
15. Hernández D, de la Nuez PC, Muriel A, et al. Clinical assessment of mortality risk in renal transplant candidates in Spain. Transplantation. 2014; 98:653–659
16. Barber K, Madden S, Allen J, et al.; United Kingdom Liver Transplant Selection and Allocation Working PartyElective liver transplant list mortality: development of a United Kingdom end-stage liver disease score. Transplantation. 2011; 92:469–476
17. Wu DA, Robb ML, Watson CJE, et al. Barriers to living donor kidney transplantation in the United Kingdom: a national observational study. Nephrol Dial Transplant. 2017; 32:890–900