Fabry disease is an X-linked recessive lysosomal storage disorder associated with a deficiency of α-galactosidase A resulting in accumulation of glycosphingolipids in the vascular endothelium, which leads to potentially fatal renal, cardiac, and cerebrovascular conditions (1). The incidence has been estimated to be 1:117,000 live births and 1:50,000 males (2). Registries of dialysis patients from the United States and Europe indicate that Fabry disease is the cause of end-stage renal disease (ESRD) for approximately 0.2% of patients on dialysis (3, 4). Among 250,352 patients who began renal replacement therapy in the United States between 1995 and 1998, only 42 patients were identified with Fabry disease as the cause of ESRD (5).
Chronic renal insufficiency may begin as early as the second decade with 53% developing ESRD by the age of 53 years (6). Before the availability of dialysis, ESRD was the most common cause of death for patients with Fabry disease, occurring at a mean age of 41 years (7). An examination of 95 patients with Fabry disease initiating dialysis between 1985 and 1998 found superior patient survival rates when compared with diabetics on dialysis, but lower compared with other causes of ESRD (5).
An earlier study that included 97 patients transplanted in the United States between 1988 and 1998 indicated that patients with Fabry disease had similar graft and patient survival rates when compared with age- and race-matched controls (8). Here, we reexamine outcomes in a cohort of 197 patients with Fabry disease, transplanted in the United States between 1987 and 2007.
All 233,280 kidney-only transplant recipients reported to the Organ Procurement Transplant Network (OPTN) from October 1987 to January 2007 were included in this study. Baseline characteristics, overall graft survival, patient survival, and death-censored graft survival were described and compared between recipients with ESRD because of Fabry versus those with ESRD secondary to other causes.
Patient demographic data were compared using chi-square test for categorical variables and rank-sum for continuous variables. Continuous variables such as age, cold ischemia time, duration of dialysis, and percent peak value of panel reactive antibodies (PRA) were categorized. Data were missing for panel-reactive antibodies in 4%, human leukocyte antigen mismatch 1%, body mass index (BMI) 6%, dialysis duration 7%, hypertension 29%, cardiovascular 35%, diabetes 31%, pulmonary 34%, cerebral vascular 35%, peripheral vascular 35%, and type of insurance for 26% of recipients.
The incidence of rejection during the initial hospital stay reflects a reported rejection episode or steroids or antibody given as anti-rejection treatment on the discharge follow-up record. The rejection rates after discharge were calculated based on reports of rejection or rejection treatment with steroids antibody on follow-up records 0 to 6 and 7–12 months posttransplant. Patients with inadequate reporting of rejection (11% at 6 months and 21% at 1 year) were excluded in rate calculations. Significance was tested with chi-square test.
Univariate survival curves were calculated using Kaplan-Meier with endpoints of graft failure, death-censored graft failure, and patient death. Cox proportional hazards were used to estimate the risk of graft loss, return to dialysis (death-censored graft failure), and death using stepwise regression to identify confounding factors. Factors considered in the multivariate models included year of transplant, recipient gender, age, race, BMI, previous transplant, peak panel-reactive antibody, level of human leukocyte antigen mismatch, type of insurance (public vs. private), dialysis duration, recipient history of hypertension, recipient history of diabetes, history of cardiovascular, cerebral vascular, peripheral vascular disease, donor gender, age, race, BMI, type (living, expanded, or standard criteria) cause of death, diabetes, cold ischemia time, cytomegalovirus serostatus in the recipient and donor, and whether tacrolimus, mycophenolate mofetil, sirolimus, or induction antibody was initially given for immunosuppression. Results with a two-tailed P value at or below 0.05 were considered statistically significant. Stata version 10 was used for all analyses.
In addition, a 10:1 matched control cohort was created consisting of kidney transplant recipients with other causes (non-Fabry) of ESRD matched for age decile, year of transplant (in 4-year bands), gender, and propensity score. Logistic regression analysis was used to identify factors associated with a diagnosis of Fabry disease and propensity scores were calculated based on the parameters with significant associations.
Baseline characteristics are outlined in Table 1. Most kidney transplant recipients with Fabry disease were male and White. None had a history of diabetes mellitus and fewer patients with Fabry had a history of hypertension compared with recipients with ESRD due to other causes. Recipients with Fabry disease reached ESRD earlier, yet had shorter dialysis time pretransplant compared with recipients with ESRD due to other causes. There were more living donor transplants and preemptive kidney transplants in those with Fabry disease. Fewer Fabry patients were sensitized (panel reactive antibodies >30%), with only 2% of Fabry patients having a prior transplant compared with 8% of those with other causes of ESRD.
The distribution of covariates for the 10:1 matched cohort is shown in the right column of Table 1. Gender, age, race, and BMI were similarly distributed compared with the Fabry cohort. A higher fraction of the matched cohort had hypertension or diabetes, but a lower fraction had peripheral vascular disease when compared with the Fabry cohort.
Factors Associated With Fabry Disease
Results for the logistic regression outlining factors associated with a Fabry diagnosis are summarized in Table 2. Transplant recipients aged 45 years or younger (compared with those 60 years or older), male, White, those who were nonobese, nondiabetic, and nonhypertensive were more likely to have a diagnosis of ESRD secondary to Fabry disease. In addition, recipients with a history of cerebrovascular disease were more likely to have a diagnosis of Fabry disease.
Table 3 outlines rejection rates for recipients with Fabry, all other causes of ESRD, and the matched cohort at three distinct periods: at discharge, during the first 6 months posttransplant and between 7 and 12 months. Recipients with Fabry disease had rejection rates of 11.2%, 10.1%, and 8.0% at discharge, by 6 months posttransplant, and between 7 months and 1 year posttransplant. At discharge, rejection rates were not significantly different between Fabry patients, those with other causes of ESRD, and the matched cohort. At 6 months, the Fabry cohort had a lower rate of rejection, but between 7 months and 1 year posttransplant there was no significant difference between the groups.
Graft and Patient Survival
Kaplan-Meier survival curves comparing overall graft survival, death-censored graft survival, and patient survival among transplant recipients with Fabry disease, all other recipients with other causes of ESRD, and the matched cohort are illustrated in Figure 1.
Overall graft survival up to 5 years posttransplant was superior among recipients with Fabry disease (5-year graft survival 74%) compared with those with ESRD due to other causes (5-year graft survival 69%). However, graft survival was similar among recipients with Fabry and those in the matched cohort (P=0.64).
Patient survival was not significantly different between recipients with Fabry versus other causes of ESRD (P=0.33). However, compared with the matched cohort (5-year patient survival 90%), recipients with Fabry had inferior patient survival up to 5 years posttransplant (5-year patient survival 81%).
When graft loss was censored for death with a functioning graft, recipients with Fabry disease had superior graft survival up to 5 years posttransplant (88%) compared with all other recipients (77%) with other causes of ESRD and the matched cohort (80%).
The results of Cox proportional hazards multivariate regression models describing the adjusted risk of graft loss, death-censored graft loss (return to dialysis), and patient death among recipients with Fabry disease are outlined in Table 4.
After adjusting for confounding factors, recipients with Fabry had a 62% higher risk of death compared with recipients with other causes of ESRD and more than double the risk of death compared with the matched cohort. The risk of returning to dialysis was approximately 40% less among Fabry patients compared with both other cohorts. The risk of overall graft loss was not significantly different for patients with and without Fabry. Other factors associated with graft loss and patient death among transplant recipients with Fabry disease include older age and expanded criteria donors (Table 5). The hazard ratios for graft loss and patient death were similar for Fabry recipients transplanted before and after 1999 when considering both matched and non-Fabry reference cohorts (Table 4).
Causes of Patient Death
The median age of death was 47 years compared with 56 years for other ESRD conditions (P<0.001). Graft failure was listed of the cause of death for only 1 (2.4%) of the 41 reported deaths in the Fabry cohort of our study. Other or unknown cause of death was reported for 56% of the Fabry cohort, 50% of the other ESRD (n=39,346), and 52% of the matched cohort (n=248). Reported causes of death were not statistically different among the three cohorts. The most common reported cause of death in the Fabry cohort was myocardial infarction (17% vs. 14% in other and 12% in the matched cohort). Infection was reported for 13% to 14% of the other and matched cohorts and 7% with Fabry. Cerebrovascular or stroke was the cause of death for 12% to 13% of the three cohorts.
Fabry disease is an uncommon cause of ESRD, making up 0.0167% of all causes of ESRD (5). However, Fabry disease may be an important cause of ESRD among young men who initiate dialysis at young age. Kidney transplantation is generally considered the renal replacement therapy for choice, particularly among younger patients. However, the baseline higher risk of premature cardiovascular and cerebrovascular disease among Fabry patients, and the further increased risk of these events attributed to ESRD and immunosuppressive medications posttransplant raises concerns that these patients are particularly high-risk candidates for transplantation (5). Several reports on outcomes postkidney transplant, mostly small single-center series, have been conflicting (5).
Should Patients With ESRD due to Fabry Disease Be Offered Kidney Transplantation?
Roughly 50% of patients with Fabry disease develop ESRD by the age of 50 years and without dialysis the mean age of death due to uremia is 40 years. Three-year patient survival for Fabry patients on dialysis has been reported for 70% (3, 4).
In our analysis of the OPTN/United Network for Organ Sharing database, we analyzed posttransplant outcomes of 197 transplant recipients with ESRD due to Fabry over a 20-year period.
Compared with all kidney transplant recipients with other causes of ESRD, recipients with Fabry disease did not differ in their risk of overall graft loss, but had a 40% lower risk of returning to dialysis and had roughly a 60% greater risk of death.
On the basis of these results, kidney transplantation seems to be at least as successful in Fabry patients as it is in patients with other causes of ESRD in graft survival posttransplant. This supports previous reports recommending transplantation as the treatment of choice for patients with ESRD due to Fabry disease (5, 8–10). However, to determine the exact benefit of transplantation in this group, the outcomes of transplanted Fabry patients should be compared with Fabry patients on dialysis.
Posttransplant Outcomes Compared with Matched Controls
To determine the incremental increase in the risk of graft failure and patient death among patients with Fabry disease, we performed a matched analysis. Compared with a 10:1 matched cohort (matched for age, gender, transplant year, and propensity score) of recipients with ESRD due to other causes, we found that Fabry patients had similar overall allograft survival, but inferior patient survival. On multivariate analysis, the risk of patient death for Fabry patients was more than double that of the matched cohort, but the risk of death-censored graft loss was nearly 40% lower among Fabry patients.
Our results differ from a prior retrospective analysis, using USRDS data, which reported that both graft and patient survival postkidney transplantation among 97 transplant recipients with Fabry disease were similar to age- and race-matched patients with other causes of ESRD (8). In that study, unadjusted 5-year graft and patient survival among Fabry patients was reported as 76% and 83%, respectively. Our analysis of 197 patients revealed similar unadjusted 5-year graft and patient survival rates of 74% and 80%, respectively.
However, there were notable differences in the patient survival rates for the matched cohorts between the two analyses. In the article by Ojo et al., matched cohorts had 82% 5-year patient survival rates whereas in our examination the matched cohort had an 89% 5-year survival. The propensity matching method not only matched patient race, age, and length of follow-up, but also covariates such as obesity and comorbid conditions. In addition, we performed a multivariate analysis that confirmed our unadjusted findings.
It is also possible that the overall management of cardiovascular risk factors among transplant recipients has improved over time, resulting in superior patient survival among all transplant recipients. This may incrementally benefit patients with traditional cardiovascular risk factors more so than Fabry patients and result in a greater increase in patient survival among non-Fabry patients. Interestingly, in comparing the causes of patient death among Fabry and non-Fabry patients posttransplant, there were no significant differences, with cardiovascular causes as the leading causes. However, given the overall low rate of death, it is difficult to make direct comparisons between causes of death.
Limitations inherent to a retrospective analysis must be considered when interpreting the results of this study. For instance, the criterion for inclusion to the Fabry cohort was dependent on the diagnosed cause of ESRD reported to the OPTN. It is possible that the actual prevalence is higher than currently reported.
In summary, kidney transplantation is a viable option for patients with ESRD due to Fabry disease. Recipients with Fabry disease have a lower risk of functional graft loss compared with all other recipients with other causes of ESRD. However, compared with a matched cohort of recipients with other causes of ESRD, Fabry patients had a higher risk of death, suggesting that these patients require further attention with regards to minimizing their risk of death due to cardiovascular disease.
1.Desnick RJ, Brady R, Barranger J, et al. Fabry disease, an under-recognized multisystemic disorder: Expert recommendations for diagnosis, management, and enzyme replacement therapy. Ann Int Med
2003; 138: 338.
2.Meikle PJ, Hopwood JJ, Clague AE, et al. Prevalence of lysosomal storage disorders. JAMA
1999; 281: 249.
3.Thadhani R, Wolf M, West ML, et al. Patients with Fabry disease on dialysis in the United States. Kidney Int
2002; 61: 249.
4.Tsakiris D, Simpson HK, Jones EH, et al. Report on management of renal failure in Europe, XXVI, 1995. Rare diseases in renal replacement therapy in the ERA-EDTA Registry. Nephrol Dial Transplant
1996; 11(suppl 7): 4.
5.Obrador GT, Ojo A, Thadhani R. End-stage renal disease in patients with Fabry disease. J Am Soc Nephrol 2002; 13(suppl 2): S144.
6.Branton MH, Schiffmann R, Sabnis SG, et al. Natural history of Fabry renal disease: Influence of alpha-galactosidase A activity and genetic mutations on clinical course. Medicine (Baltimore)
2002; 81: 122.
7.Branton M, Schiffmann R, Kopp JB. Natural history and treatment of renal involvement in Fabry disease. J Am Soc Nephrol
2002; 13(suppl 2): S139.
8.Ojo A, Meier-Kriesche HU, Friedman G, et al. Excellent outcome of renal transplantation in patients with Fabry’s disease. Transplantation
2000; 69: 2337.
9.Mignani R, Gerra D, Maldini L, et al. Long-term survival of patients with renal transplantation in Fabry’s disease. Contrib Nephrol
2001; 136: 229.
10.Kasiske BL, Cangro CB, Hariharan S, et al. The evaluation of renal transplantation candidates: Clinical practice guidelines. Am J Transplant
2001; 1(suppl 2): 3.
11.Kurth T, Walker AM, Glynn RJ, et al. Results of multivariable logistic regression, propensity matching, propensity adjustment, and propensity-based weighting under conditions of nonuniform effect. Am J Epidemiol
2006; 163: 262.
12.Baser O. Too much ado about propensity score models? Comparing methods of propensity score matching. Value Health
2006; 9: 377.