Patients with advanced heart failure (HF) typically have impaired renal function due to reduced renal blood flow, use of diuretics and renin-angiotensin-aldosterone system inhibitors, and underlying primary renal diseases such as atheromatous, hypertensive, and diabetic renal disease. In a large meta-analysis of more than 80,000 patients with HF, moderate-to-severe impairment of renal function (approximating to estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m2) was found in 29% of patients (1). Chronic kidney disease (CKD) has superseded chronic renal failure as the term that now covers both decreases in renal function, as evidenced by a low GFR, and also markers of earlier damage such as albuminuria (2).
Renal failure in the general population is associated with an increased risk of all-cause morbidity and mortality not adequately explained by the classical risk factors such as age, diabetes, ethnicity, deprivation, and other comorbidity (3, 4), and in immunocompromised patients, the chance of infection may further increase these risks.
Heart transplantation is an effective treatment for advanced HF with 1-year posttransplantation survival of 83% in the United Kingdom (5), but prolonged exposure to immunosuppressive agents can adversely affect renal function. Calcineurin inhibitors play a crucial role in posttransplantation immunological management, enabling high survival rates, but they can also cause renal failure due to their nephrotoxicity (6–8).
Posttransplantation renal failure is therefore a commonly occurring late complication of heart transplantation (9, 10), with an associated impact on mortality and it is important that clinicians can quantify and, where possible, minimize the potential risks for patients. National cohort studies have described the risk in the United States (11, 12) and Canada (13), and a cross-sectional study from Spain (14, 15) has recently been published, but these are the only large, multi-center reports.
In this study, we have linked two UK databases with comprehensive long-term follow-up data to (a) determine the incidence of CKD in the UK heart transplantation population; (b) identify risk factors for the development of CKD; and (c) assess the impact of CKD on subsequent survival.
There were 1815 adult heart transplantations performed in the UK at eight centers during this 11-year period, and after excluding 83 transplantations in accordance with the criteria specified, 1732 transplantations were available for analysis with a median follow-up of 5 years. Tables 1 and 2 present the recipient, donor, and transplantation characteristics for the cohort. Figure 1 presents the proportion of patients at National Kidney Foundation (NKF) CKD stage 3, 4, or 5, or receiving dialysis, by time posttransplantation. The number of patients available for analysis decreased over time due to insufficient follow-up and deaths posttransplantation, so results are presented as percentages. Immediately before transplantation, 50% of patients had CKD stage 3 or higher, consistent with previous observations regarding poor renal function of patients with advanced HF. After transplantation, this proportion increased rapidly to 77% within 1 year, then remained fairly constant throughout the posttransplantation period. However, the proportion of patients with more advanced CKD (investigated here) continued to increase from 3% at transplantation, to 11% at 1-year and more than 15% at 6 years posttransplantation and beyond.
Two hundred eighty-one patients received temporary dialysis or hemofiltration in the immediate posttransplantation period, but this was not included in the definition of CKD. However, 54 patients (3.1% of the whole cohort) received long-term renal replacement therapy (RRT); 33 hemodialysis, 16 peritoneal dialysis, 3 unknown dialysis modality, and 2 preemptive kidney transplantations. Median time from heart transplantation to commencing RRT was 6.0 years (range 0–9.9 years). One-year patient survival on chronic dialysis was 88.5% (95% confidence interval [CI] 76.1%–94.6%), with no significant difference between the hemodialysis and peritoneal dialysis groups (P=0.12). A total of 21 patients were listed for a kidney transplant and 10 patients received a kidney transplant. At 1-year after kidney transplantation, all patients were alive.
All factors in Tables 1 and 2 were considered as candidate risk factors in the Cox model for time to CKD, and the resulting model for the 1495 patients who survived at least 3 months is shown in Table 3. The model found that earlier transplantation date, shorter ischemia time (IT), female, older, hepatitis C virus (HCV) positive, diabetic recipients, and those with a higher CKD stage at registration or transplantation were at increased risk of developing CKD after heart transplantation. There were also significant differences between transplantation centers. There was an increased risk for those with a greater deterioration in renal function in the first 3 months after transplantation and those needing postoperative hemofiltration were more likely to have CKD at follow-up. However, a significant interaction term identified that patients who received hemofiltration in the earliest period of this study (1996–1999) were not at higher risk of CKD. Immunosuppression regimen and use of induction therapy were not identified as risk factors, but data are only available for the peri-operative period (Table 1).
The long-term survival model, conditional on 3-month survival, is shown in Table 4. After adjusting for other risk factors, patients surviving to a given time and at CKD stage 4 or 5 (excluding dialysis), or on dialysis, all have a significantly increased risk of death relative to otherwise identical patients at CKD stage 3 or lower at that same time. The hazard ratio is 1.66 (95% CI 1.22–2.26) for patients at stage 4, 8.54 (95% CI 4.60–15.87) for patients at stage 5 (excluding dialysis), and 4.07 (95% CI 2.36–7.02) for patients on dialysis. In a separate model, eGFR was fitted as a time-dependent variable and after adjusting for other risk factors, a patient with an eGFR 5 mL/min/1.73 m2 lower than an otherwise identical patient at the same time had an increased risk of death of 1.04 (95% CI 1.01–1.08, P=0.03).
For patients who died beyond 3 months posttransplantation, there were statistically significant differences in the causes of death between the CKD and non-CKD groups (P<0.001, data not shown). For 16% of the CKD group, the cause of death was given as renal failure compared with 2% of the non-CKD group. Patients in the non-CKD group were more likely to die of cardiac causes or infection.
Our UK data highlight an important incidence of renal dysfunction among heart transplant recipients. Preoperative renal function was impaired for many patients, and deteriorated further within the first year after transplantation. In the long-term, approximately 80% of recipients had NKF CKD stage 3 or higher and approximately 15% developed CKD as defined in this study (NKF CKD stage 4 or 5 or preemptive kidney transplantation). By 5 years after transplantation, 12.9% of surviving patients had CKD, which is slightly higher than the 10.9% rate reported for a large US cohort (11), and the 9.2% rate reported in a cross-sectional Spanish study with mean follow-up of 6.7 years (14), but important differences in the methods used and the definition of CKD mean direct comparisons cannot be made.
In our study, 3.1% of patients received long-term RRT during the follow-up period, which is comparable with published rates of 3.9% (13) and 2.1% (12) in two other national studies, and 2.7% to 7.9% in a number of single-center studies (16–22), although the differential lengths of follow-up confound direct comparison. Here, 1-year survival on dialysis was 88.5% (95% CI 76.1%–94.6%), which compares favorably with other studies which report rates of 60% to 82% (16, 21–23), although a French study of 16 patients reported a rate of 100% (19). There have been few studies of dialysis survival for heart transplant recipients compared with other populations. In our study, the 88.5% 1-year survival rate was comparable with the UK national data for prevalent dialysis patients in 2008 aged younger than 65 (91.8%, 95% CI 91.3%–92.3%) (24). This contrasts with a Canadian study that reported significantly inferior 5-year survival on dialysis for heart transplant patients compared with matched patients receiving dialysis for other causes (13).
In common with previous studies, increasing recipient age (11, 15, 18, 25, 26), female sex (11, 15, 25), diabetic status (11, 15, 25, 27), and requirement for temporary dialysis or hemofiltration perioperatively (11, 25) were all found to be associated with risk of developing CKD. However, we found that hemofiltration was not associated with CKD in the subgroup from the earliest era (1996–1999), although this observation is currently unexplained. Positive HCV status was rare (n=14) but was associated with a significantly increased risk consistent with the results of a large US study (11).
As in other studies (11, 15, 26–29), impaired renal function immediately before transplantation was associated with developing CKD. We also identified renal function at time of listing as an independent risk factor, suggesting that persistently impaired renal function before transplantation adds additional risk. The inclusion of CKD stage before transplantation could be considered a “self-fulfilling prophecy,” but a sensitivity analysis found that all other risk factors, except HCV status, remained significant when these factors were omitted from the risk model. In addition, the pathogenesis of the cardiorenal syndrome (CRS) in HF appears distinct from that of posttransplantation CKD. CRS may be reversible if cardiac function improves and this was often seen in the precyclosporine era of heart transplantation (7). Other factors can impair renal function after transplantation including; cardiopulmonary bypass (as in other cardiac surgery ) and blood transfusions as part of the transplantation operation, immunosuppression, and primary cardiac allograft dysfunction (9). Despite the different mechanisms involved, preoperative renal dysfunction was significantly associated with posttransplantation CKD. We also found that patients whose renal function deteriorated rapidly in the first 3 months posttransplantation were at increased risk, reflecting other studies which found that a fall in renal function in the first year had an impact on later CKD (17, 18, 21, 26).
Patients transplanted in the more recent years of the study were at decreased risk of CKD, compared with those transplanted in 1996 to 1999 (hazard ratios of 0.57 and 0.40 for transplants in 1999 to 2003 and 2003 to 2007, respectively, for patients who did not receive hemofiltration). This may relate to evolving renal sparing immunosuppressive strategies, but there are no data available to assess this. Longer IT was also protective against CKD, which seems counterintuitive and we did not find any significant interaction terms which may explain this finding. IT was not identified as a risk factor in any of the other studies reviewed, so this result requires corroboration. Significant differences observed between transplantation centers are also unexplained. They may partly reflect differences in long-term immunosuppression, but this cannot be established because immunosuppressive data collected by the UK Cardiothoracic Transplant Audit (UKCTA) are limited to the perioperative period. Transplantation center remains significant when looking at 5-year survivors only, and when modeling time to CKD within the first 3 years, suggesting the differences cannot be explained by variation in early or late mortality between centers. Case selection and factors that are not in the risk-adjustment model may play a role.
Our analysis demonstrates that CKD has an adverse impact on subsequent mortality, and describes the effect associated with each CKD stage. Patients with CKD stage 4, 5, or on dialysis were all at increased risk of mortality, but the highest risk was associated with having CKD stage 5 without receiving dialysis. Implementation of RRT reduced this risk but not to the risk at CKD stage 4. This is plausible because the patient’s effective renal function on dialysis is still lower than normal and there are complications caused by, or associated with, the treatment, in particular cardiovascular disease and infection. The hazard ratio of 4.07 for mortality on dialysis was similar to that in other studies (16, 19, 22). It is clear that renal dysfunction is a serious comorbidity after transplantation that impacts survival and so efforts to avoid CKD and manage it effectively are important.
Our study benefits from the linkage of renal and cardiac databases to report on the incidence of CKD in this UK-wide cohort of more than 1700 patients with no selection bias and comprehensive long-term follow-up. However, our study has limitations. Although the dataset is large and comprehensively populated, the number of risk factors available for analysis was relatively small and, in some cases, inadequately detailed (e.g., immunosuppression data from the perioperative period only, and without dosages). The number of patients receiving dialysis or kidney transplantation was small limiting the analysis that could be performed in these subgroups. Finally, we acknowledge that there is a small risk of CKD misclassification as there is only a single creatinine value in any 12-month period to calculate eGFR. This is unlikely to change the overall findings with respect to prevalence and impact on mortality as fluctuation of creatinine occurs in both directions and only affects a small proportion of nonhospitalized patients (31).
In conclusion, we have confirmed that several factors associated with posttransplantation renal failure identified in other studies are also relevant in the UK heart transplantation population. We have also identified some novel risk factors: the patient’s renal function at time of listing; shorter IT and a center-effect that require further investigation; and a decreasing risk of developing CKD for more recent transplantations. We have described the impact of CKD on subsequent survival according to stage; identifying that the institution of dialysis reduces the mortality risk, but not to the level of CKD stage 4.
The pursuit of novel immunosuppressive strategies (32) that minimize kidney damage, while maintaining the high rate of patient and allograft survival that has been achieved in the cyclosporine era, remains important. However, our findings suggest that efforts to reverse the CRS before transplantation and changes in peritransplantation management to minimize the risk of acute renal injury will also be important.
MATERIALS AND METHODS
Data were obtained from the UKCTA and the UK Renal Registry (UKRR), which have been described previously (24, 33) and have high levels of data completeness; 1-year follow-up data have been returned to the UKCTA for more than 99% of transplantations (24), and dialysis start date is 97.6% complete in the UKRR (personal correspondence). We analyzed all adult (≥18 years) heart transplantations performed in the UK between April 1996 and March 2007, and matched these with data held by the UKRR on all patients receiving RRT before December 2007 using six data fields held by both registries. Multi-organ, repeat and heterotopic transplantations and recipients with no serum creatinine data were excluded from the cohort. Approval for use of patient identifiable data is obtained under the provisions of Section 251 of the NHS Act 2006 as regulated by National Information Governance Board (http://www.nigb.nhs.uk/). Transplantation data were initially collected on the basis of presumed consent, but more recently informed consent has been sought when registering the patient for transplantation. In the UK, audit projects do not require separate research ethics committee approval (34).
The UKCTA collects recipient serum creatinine at registration, transplantation, 3 and 12 months posttransplantation, and then annually until patient death. The GFR for each patient at each time point was estimated using the four-variable Modification of Diet in Renal Disease equation (35), categorized into NKF CKD stages (2), and summarized by time posttransplantation. CKD stage 3 was split into stage 3(i) (40–59 mL/min/1.73 m2) and stage 3(ii) (30–39 mL/min/1.73 m2) to examine the less than 40 mL/min/1.73 m2 threshold, recommended by the International Society for Heart and Lung Transplantation as a relative contraindication to transplantation (36). Serum creatinine data were only included in this analysis if reported within a 2-month window spanning the relevant follow-up period. Serum creatinine data were deleted from all analyses if reported after dialysis commenced, or a subsequent kidney or second heart transplantation performed. For patients who received long-term RRT after heart transplantation, survival on dialysis, and after kidney transplantation was estimated using the Kaplan-Meier method.
CKD was defined as NKF CKD stage 4 or 5 (eGFR<30 mL/min/1.73 m2, or dialysis) or receipt of preemptive kidney transplant. The time from heart transplantation to the development of CKD was modeled using Cox proportional hazards regression on the subset of patients who survived at least 3 months posttransplantation, so that the impact of early changes in renal function on long-term function could be evaluated. Missing data were less than 3% for all but three risk factors and were imputed by allocation to the modal category in these cases. HCV status was missing in 17% of cases, and was imputed randomly according to the proportions observed in the complete case data. Missing IT was imputed using the median local or imported IT for that center. Missing eGFR at time of transplantation was imputed based on registration eGFR and waiting time, and missing eGFR at 3 months was imputed based on transplantation eGFR.
The impact of renal failure on mortality was assessed using a nonproportional hazards Cox regression model with time-dependent covariates (37) to signify the different stages of renal failure. This model was fitted to the same subset of patients who had survived at least 3 months after heart transplantation. This Cox model with time-dependent variables was used to describe the increased hazard of death for a patient with a particular stage of renal failure relative to an identical patient without advanced renal failure at the same time point. The time-dependent variables were added to a model containing previously identified factors associated with 1-year survival after heart transplantation in the UK (38). Missing data were imputed using the methods specified in the development of this existing model. All statistical analyses were undertaken using SAS (for Windows) version 9 (SAS Institute Inc., Cary, NC), and a P value less than 0.05 was considered statistically significant.
This work was undertaken on behalf of the Steering Group of the UKCTA: Dr. N.R. Banner (chairman) and Dr. André Simon, Harefield Hospital, Harefield; Prof. R.S. Bonser, Queen Elizabeth Hospital, Birmingham; Prof. P. Corris, Freeman Hospital, Newcastle; Mr. P. Braidley, Northern General Hospital, Sheffield; Mr. S. Tsui and Dr. J. Parameshwar, Papworth Hospital, Papworth Everard; Prof. N. Yonan, Wythenshawe Hospital, Manchester; Dr. M. Burch, Great Ormond Street Hospital, London; Mr. U. Nkere, Golden Jubilee National Hospital, Glasgow; Dr. J. van der Meulen, Royal College of Surgeons of England; Prof. D. Collett, NHS Blood and Transplant; and Dr. I. Stephens, NSCT. The authors thank the transplant coordinators and renal center staff for collecting the audit data.
1. Smith GL, Lichtman JH, Bracken MB, et al.. Renal impairment and outcomes in heart failure: Systematic review and meta-analysis. J Am Coll Cardiol 2006; 47: 1987.
2. Levey AS, Eckardt KU, Tsukamoto Y, et al.. Definition and classification of chronic kidney disease
: A position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005; 67: 2089.
3. Matsushita K, van der Velde M, Astor BC, et al.; Chronic Kidney Disease
Prognosis Consortium. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: A collaborative meta-analysis. Lancet 2010; 375: 2073.
4. Quinn MP, Cardwell CR, Kee F, et al.. The finding of reduced estimated glomerular filtration rate is associated with increased mortality in a large UK population. Nephrol Dial Transplant 2010. doi:10.1093/ndt/gfq505.
5. Transplant Activity in the UK—Activity Report 2009/10. NHS Blood and Transplant 2010.
6. Kahan BD. Cyclosporine. N Engl J Med 1989; 321: 1725.
7. Myers BD, Ross J, Newton L, et al.. Cyclosporine associated chronic nephropathy. N Engl J Med 1984; 311: 699.
8. Myers BD, Sibley R, Newton L, et al.. The long-term course of cyclosporine-associated chronic nephropathy. Kidney Int 1988; 33: 590.
9. Bloom R, Doyle A. Kidney disease after heart and lung transplantation. Am J Transplant 2006; 6: 671.
10. Wilkinson AH, Cohen DJ. Renal failure
in the recipients of nonrenal solid organ transplants. J Am Soc Nephrol 1999; 10: 1136.
11. Ojo AO, Held PJ, Port FK, et al.. Chronic renal failure
after transplantation of a nonrenal organ. N Engl J Med 2003; 349: 931.
12. Hornberger J, Best J, Geppert J, et al.. Risks and costs of end-stage renal disease after heart transplantation
. Transplantation 1998; 66: 1763.
13. Alam A, Badovinac K, Ivis F, et al.. The outcome of heart transplant recipients following the development of end-stage renal disease: Analysis of the Canadian Organ Replacement Register (CORR). Am J Transplant 2007; 7: 461.
14. Crespo-Leiro MG, Delgado JF, Paniagua MJ, et al.. Prevalence and severity of renal dysfunction among 1062 heart transplant patients according to criteria based on serum creatinine and estimated glomerular filtration rate: Results from the CAPRI study. Clin Transplant 2010; 24: E88.
15. Delgado JF, Crespo-Leiro MG, Gómez-Sánchez MA, et al.. Risk factors associated with moderate-to-severe renal dysfunction among heart transplant patients: Results from the CAPRI study. Clin Transplant 2010; 24: E194.
16. Satchithananda DK, Parameshwar JP, Sharples L, et al.. The incidence of end-stage renal failure
in 17 years of heart transplantation
: A single center experience. J Heart Lung Transplant 2002; 21: 651.
17. Cantarovich M, Hirsh A, Alam A, et al.. The clinical impact of an early decline in kidney function in patients following heart transplantation
. Am J Transplant 2009; 9: 348.
18. Lindelöw B, Bergh C, Herlitz H, et al.. Predictors and evolution of renal function during 9 years following heart transplant. J Am Soc Nephrol 2000; 11: 951.
19. Frimat L, Villemot J, Cormier L, et al.. Treatment of end-stage renal failure
after heart transplantation
. Nephrol Dial Transplant 1998; 13: 2905.
20. Sénéchal M, Dorent R, Tézénas du Montcel S, et al.. End-stage renal failure
and cardiac mortality after heart transplantation
. Clin Transplant 2004; 18: 1.
21. van Gelder T, Balk AH, Zietse R, et al.. Renal insufficiency after heart transplantation
: A case-control study. Nephrol Dial Transplant 1998; 13: 2322.
22. Goldstein DJ, Zuech N, Sehal V, et al.. Cyclosporine-associated end-stage nephropathy after cardiac transplantation: Incidence and progression. Transplantation 1997; 63: 664.
23. Villar E, Boissonnat P, Sebbag L, et al.. Poor prognosis of heart transplant patients with end-stage renal failure
. Nephrol Dial Transplant 2007; 22: 1383.
24. Ansell D, Feehally J, Fogarty D, et al.. UK Renal Registry 2009 12th Annual Report of the Renal Association. Nephron Clin Pract 2010; 115; (suppl 1): c1.
25. Hamour IM, Omar F, Lyster HS, et al.. Chronic kidney disease
after heart transplantation
. Nephrol Dial Transplant 2009; 24: 1655.
26. Jokinen JJ, Tikkanen J, Kukkonen S, et al.. Natural course and risk factors for impaired renal function during the first year after heart transplantation
. J Heart Lung Transplant 2010; 6: 633.
27. Rubel JR, Milford EL, McKay DB, et al.. Renal insufficiency and end-stage renal disease in the heart transplant population. J Heart Lung Transplant 2004; 23: 289.
28. Vossler MR, Ni H, Toy W, et al.. Pre-operative renal function predicts development of chronic renal insufficiency after orthotopic heart transplantation
. J Heart Lung Transplant 2002; 21: 874.
29. Odim J, Wheat J, Laks H, et al.. Peri-operative renal function and outcome after orthotopic heart transplantation
. J Heart Lung Transplant 2006; 25: 162.
30. Kumar AB, Suneja M. Cardiopulmonary bypass-associated acute kidney injury. Anesthesiology. 2011; 114: 964.
31. de Lusignan S, Tomson C, Harris K, et al.. Creatinine fluctuation has a greater effect than the formula to estimate glomerular filtration rate on the prevalence of chronic kidney disease
. Nephron Clin Pract 2011; 117: C213.
32. The International Society of Heart and Lung Transplantation, Guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2010; 29: 914.
33. Banner NR, Thomas HL, Curnow E, et al.. The importance of cold and warm cardiac ischemia for survival after heart transplantation
. Transplantation 2008; 86: 542.
35. Levey AS, Greene T, Kusek JW, et al.. A simplified equation to predict glomerular filtration rate from serum creatinine J Am Soc Nephrol 2000; 11: A0828.
36. Mehra MR, Kobashigawa J, Starling R, et al.. Listing criteria for heart transplantation
: International Society for Heart and Lung Transplantation guidelines for the care of cardiac transplant candidates—2006. J Heart Lung Transplant 2006; 25: 1024.
37. 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.
38. Rogers CA, Ganesh JS, Banner NR, et al.. UK Cardiothoracic Transplant Audit. End of year report from the audit Steering Group to the National Specialist Commissioning Advisory Group London: The Royal College of Surgeons of England. 2004.