SAT-087 RELIABILITY OF NZ RISK OF DEATH SCORING SYSTEM IN KIDNEY TRANSPLANT RECIPIENTS

Full text

choice experiment to evaluate trade-offs between the outcomes of an
allocation policy and underlying values.
SAT-087
RELIABILITY OF NZ RISK OF DEATH SCORING
SYSTEM IN KIDNEY TRANSPLANT RECIPIENTS
Talreja, H
1
, Jackson, A*
1
, Cross, N
2
1
Counties Manukau Health, Renal Medicine, Auckland, New Zealand,
2
Canterbury DHB, Nephrology, Christchurch, New Zealand
Introduction: The Transplantation Society of Australia and New Zea-
land (TSANZ) Consensus Statement on Eligibility Criteria and Alloca-
tion Protocol states that deceased donor kidney transplant recipients
should have a reasonable post-operative life expectancy defined as
>80% likelihood of surviving 5 years after transplantation. In New
Zealand (NZ), an algorithm derived from published USRDS data has
been used to estimate survival probability for potential recipients since
2012. This represents rationing on the basis of probability of benefit,
due to the shortage of organs.
This study assessed the accuracy of survival predictions derived
from the algorithm applied retrospectively to a cohort of deceased
donor kidney transplant recipients (DDKTR) between 1 January 2002
and 31 December 2012; from two NZ tertiary care centres.
Methods: All DDKTR during the study period were included. Risk
assessments using the algorithm were performed retrospectively for
each person using prospectively collected predictor data (table 1) at
time of the transplant.
The date of first renal replacement therapy was date of first dialysis
or date of transplant for preemptive transplant recipients. The date of
listing was defined as date of inclusion on the DD waiting list. Biological
markers were taken nearest to transplant date.
Patients were followed 5 years post-transplant. The outcome of all-
cause mortality irrespective of persistence of transplant function was
captured.
Summary statistics were produced to describe living and DDKTR
separately. Within donor type strata, transplant recipients with >20%
5 year mortality risk were compared to lesser predicted risk recipients
to measure proportion living at 5 years. Differences were compared
with chi-squared statistics. Risk of death at 5 years was assessed using
logistic regression with predicted probability of death as independent
variable. Time to death was modelled using Cox proportional hazards
models with prediction of >20% 5 year mortality as independent
variable.
Results: 141 recipients were included in the study, 80 Centre 1 and 61
Centre 2.
Centre 1 patients were significantly younger (mean 53.1 vs 46.7
years, p=0.01), had more diabetic ESKD (16% vs 3%, p = 0.03), less
congestive heart failure (0% vs 7%) and were more likely of ‘other’
ethnicity (76% vs 25%, p <0.0001).
More Centre 1 patients had a predicted survival >80% (89% vs
72%, p = 0.02). Mean survival probability was higher at 90.6% in
Centre 1 versus 86.4% in Centre 2 patients (p=0.01). There were 14
deaths by five years (9.9%). A higher proportion of Centre 1 patients
were alive at 5 years (95% vs 84%, p 0.04).
Of the 26 patients with survival probability <80%, 13 had died at
5 years (50%). In comparison, only 1 of 115 patients with a survival
probability >80% had died at five years (p <0.0001) (Fig 1). For every
10% increase in predicted risk, the actual mortality increased by 2.6
(95%CI 1.01-6.57, p = 0.04).
Conclusions: A probability of survival of less than 80% at 5 years after
transplant based on algorithm was associated with poor survival post-
transplant in a New Zealand cohort. Risk stratification using this al-
gorithm is a valid way of estimating post-transplant survival.
Table 1
ISN WCN 2019 ABSTRACTS
Kidney International Reports (2019) 4, S1–S437 S41

SAT-088
CHARACTERISTICS OF TUBERCULOSIS IN
KIDNEY TRANSPLANT RECIPIENTS
Kafle, M*
1
, Poudyal, A
2
, Singh, D
3
1
Tribhuvan University- Institute of Medicine, Department of Nephrology
and Transplantation Medicine- Tribhuvan University Teaching Hospital,
Maharajgunj- Kathmandu., Nepal,
2
Tribhuvan University- Institute of Med-
icine, Department of Community Medicine and Public Health- Maharajgunj
Medical Campus, Maharajgunj- Kathmandu, Nepal,
3
Tribhuvan University-
Institute of Medicine, Department of Nephrology and Transplantation
Medicine- Tribhuvan University Teaching Hospital, Maharajgunj- Kath-
mandu, Nepal
Introduction: Tuberculosis (TB) is a major public health disease. Burden
of TB is higher in the underdeveloped countries. Community preva-
lence of active pulmonary TB is high in Nepal. Incidence of TB in
general population is estimated to be 154 per 100,000. Transplanted
patients are therapeutically immunosuppressed to avoid damage to the
allograft by body’s immune system. So, transplant population is at
higher risk of having TB. Moreover, disease itself, medications, pro-
cedures (e.g. use of intravenous contrast) used to diagnose TB may pose
risk to the graft kidney.
We have been doing living donor kidney transplants since 2008.
We looked at our hospital records to investigate whether TB poses a
threat to transplantation.
Methods: This is a retrospective analysis of data from patients trans-
planted in our hospital. We included all patients who underwent
kidney transplant at our centre. To ensure completion of any treatment
ongoing during transplantation, we included patients transplanted not
less than 6 months before the time of analysis. Total 517 transplants
were done till September 2018. We included first 485 patients who
would complete at least 6 months after transplant. Data from 9 patients
were missing.
Disease screening. In pretransplant workup, every donor and recipient
undergo routine chest X-ray and three daily sputum smear
examination.
Immunosuppression. Initially, we used daclizumab for induction.
Currently, we use rabbit antithymocyte globulin. Tacrolimus, myco-
phenolate and prednisolone are the maintenance drugs. We do not use
isoniazid prophylaxis.
To study the impact of TB by time at onset of disease, we classified
patients into 3 groups as follows:
Pre-transplant onset when the patient was on treatment at
transplant.
Early onset when disease occurred on or before 12 months after
transplant.
Late onset when disease occurried after 12 months of transplant.
We tested if onset of disease had any impact on patient survival.
Results: General character istics. Out of 485, 79.2% were male.
Mean age SD was 34.41 10.77 years. Number of patients on anti-
tubercular therapy (ATT) after transplant was 40. While 13 were on
ATT during transplantation, 27 developed the disease after
transplantation. One patient had history of completion of ATT
before transplant workup. Incidence of TB after transplantation was
5.5%. Median time since transplantation to diagnosis of TB was 32
months. Distribution of post transplant TB by organ system is shown
in Table 1.
Outcomes: Patient characteristics in each group are shown in Table 2.
All patients were treated with combination of anti-tubercular drugs.
Rifampicin was used in 92.6% cases. Two died during treatment and
one was on treatment during analysis. Median duration of ATT was 8
months.
There was no statistically significant survival difference between
early and late onset disease (p =0.3033) or pre and post transplant
onset disease (p=0.45) at 1 year after diagnosis. Patient survival at 3
year after diagnosis was significantly different between early and late
onset disease (p=0.035). Outcome variables measured are summarized
in Table 3.
Deaths. Four patients died. All of these acquired tuberculosis after
transplantation. One death was related to tuberculosis.
ISN WCN 2019 ABSTRACTS
S42 Kidney International Reports (2019) 4, S1–S437