Begin typing your search above and press return to search.
EPUB Before Print


Predictors of Improvement in Renal Function After Heart Transplant

Objectives: Moderate chronic renal insufficiency is often found in patients evaluated for heart transplant. Recovery of cardiac output after heart transplant might lead to improvement of renal function. In this study, our aim was to identify predictors of improve­ment of renal function after heart transplant.

Materials and Methods: Our study included a cohort of heart transplant patients treated from 2011 to 2016 whose main outcome was improved renal function, defined as glomerular filtration rate at 6 months after heart transplant of ≥ 10% compared with baseline (before transplant). Univariate and multivariate logistic regression was used to identify independent predictors.

Results: Our study included 83 patients, with 29% having improvement in renal function. Multivariate analyses identified baseline glomerular filtration rate (odds ratio of 0.95; 95% confidence interval, 0.93-0.98; P = .005), absence of hypertension (odds ratio of 4.94; 95% confidence interval, 1.37-17.8; P = .015), and elective heart transplant (odds ratio of 13.71; 95% confidence interval, 1.33-141; P = .028) as independent predictors. A scale developed with independent predictors showed good accuracy (area under the curve of 0.76). The probability for improvement in renal function was 7%, 23%, and 58% in patients with low, medium, and high scores, respectively (P < .001).

Conclusions: In patients with heart transplant, baseline glomerular filtration rate, absence of hypertension, and elective heart transplant were independent predictors of improvement in renal function after heart transplant.

Key words : Chronic renal insufficiency, Glomerular filtration rate, Heart transplantation, Prognosis, Renal insufficiency


Chronic renal insufficiency is a frequent comorbidity in patients with advanced heart failure,1 and severe disease is a contraindication for heart transplant (HT) unless a combined heart-kidney transplant is planned.2

Worsening renal function is a frequent com­plication in HT patients. The cause is multifactorial, but use of calcineurin inhibitors and other drugs is a primary cause.3 Thus, there is concern that patients with moderate chronic renal insufficiency will experience a deterioration of renal function soon after HT.

However, many patients develop moderate chronic renal insufficiency secondary to type 2 cardiorenal syndrome.1 Kidney function in these patients typically improves after HT, once cardiac output has been restored to normal.

The aim of this study was to identify predictors of kidney function improvement after HT.

Materials and Methods

We retrospectively evaluated all consecutive HT procedures performed between 2011 and 2016 at our institution. Clinical data recorded before surgery and at monthly follow-up until posttransplant month 6, month 12, and month 24 were reviewed. Pre-HT information included clinical characteristics of the patients, blood analysis results, and right heart catheterization data. Post-HT information included routine blood analysis results.

Patients who died before the 6-month follow-up or those who underwent a combined heart-kidney transplant were excluded from the study. This was an observational and retrospective study approved by the Ethics Committee of our hospital, and all procedures were in accordance with the Helsinki Declaration of 1975.

To assess kidney function, we estimated glomerular filtration rate (GFR) using the Chronic Kidney Disease Epidemiology Collaboration equation.4 “Baseline GFR” was defined as the average of GFRs calculated using the last 3 stable values of serum creatinine before HT. “Post-HT GFR” was defined as that estimated from the value of serum creatinine at the patient’s 6-month follow-up. Follow-up creatinine levels needed to be a stable value, defined as 2 consecutive values with less than 10% change and not related to an admission episode.

The primary outcome evaluated in this study was improvement in kidney function, defined as a post-HT GFR at the patient’s 6-month follow-up that was at least 10% greater than that at baseline.

All variables with biologic plausibility and those associated with kidney function were included as potential predictors of kidney function. Because baseline GFR is a potential confounder, it was included in the final analysis. Clinical variables that could be considered potential confounders were clinical baseline characteristics, drug treatment, pulmonary hemodynamic profile, and post-HT right ventricular dysfunction. The status of HT was defined as “emergency HT” if the recipient was on mechanical circulatory support as bridge to transplant; otherwise, the status was indicated as “elective HT.”

Statistical analyses
We evaluated our database to identify outliers or missing data. We used single imputation strategy when missing data were less than 20% of total data. We performed descriptive analyses with mean and standard deviation for quantitative variables and absolute frequencies and percentages for qualitative variables. We compared groups with and without improvement in kidney function with t test for independent groups or chi-square test. Univariate logistic regression was done for all variables that showed a trend for association with the outcome (P ≤ .300). After that, a stepwise backward selection was used to fit a multivariate logistic regression model: all variables with biological plausibility to be associated with renal function were included in the full model; variables with P > .05 were removed. Model accuracy was assessed with the area under the receiver operating characteristic curve. Internal validation of the statistical model was performed using bootstrap analyses with 500 replications. To use the results of the multivariate analyses, we developed a scale using the beta coefficients of multivariate logistic regression. Because pre-HT GFR was a quantitative variable, we transformed it into an ordinal variable (1 for every 10 mL/min/1.73 m2). To generate whole numbers, we multiplied the beta coefficients by 2. The scale allowed 3 points for the absence of hypertension, 5 points if the HT was elective, and -1 point for every 10 mL/min of baseline GFR.

Three groups were defined according to tertiles of the scale. Finally, a covariance analysis was used to predict GFR at month 6 according to the results of the scale groups: GFR at month 6 was the dependent variable, the scale group was the independent variable, and baseline GFR was a covariate. A two-tail P ≤ .05 was considered significant. We used Stata 14 (StataCorp LP, College Station, TX, USA) for statistical analyses.


Ninety-one HT procedures were performed from 2011 to 2016. Eight recipients were excluded from analysis: 6 died within 6 months of HT and 2 received a combined heart-kidney transplant. Therefore, 83 patients were included in the final analysis.

All patients completed at least 6 months of post-HT follow-up. Twenty-four recipients (29%) experi­enced a GFR improvement after HT. Comparisons of recipient characteristic between those with and without GFR improvement are shown in Table 1. The following variables showed trends toward statistical significance for predicting GFR improvement: elective HT; absence of systemic arterial hypertension; New York Heart Association class IV; use of furosemide, everolimus, or ganciclovir; baseline GFR; and days of hospitalization after HT. However, on multivariate analyses, the only independent predictors of renal function improvement were baseline GFR (odds ratio [OR] of 0.95; 95% confidence interval [CI], 0.93-0.98; P = .005), absence of systemic arterial hypertension (OR of 4.94; 95% CI, 1.37-17.8; P = .015), and elective HT (OR of 13.71; 95% CI, 1.33-141; P = .028) (Table 2).

Table 2 shows that lower pre-HT GFR predicts renal function improvement (OR of 0.95; 95% CI, 0.93-0.98), which means that, as GFR increases 1 mL/min/1.73 m2, the probability of GFR improvement decreases by 5%. Although this association seems paradoxical, it is in patients with moderate renal insufficiency where one would expect improvement in renal function and not in patients with normal renal function.

The scale developed with beta coefficients from multivariate logistic regression gave 3 points to absence of hypertension, 5 points for elective HT, and -1 point for every 10 mL/min in baseline GFR (Table 3). Figure 1 illustrates distribution of the scale. The scale accurately predicted improvement in renal function (area under the curve of 0.76). The probability for improvement of renal function was 7% in patients with -9 to -3 points, 23% in patients with -2 to -1 points, and 58% in patients with ≥ 0 points (P < .001; Figure 2). An analysis of covariance showed that predicted GFR 6 months after HT was clinically and statistically different between groups (Figure 3). Finally, patients with GFR improvement 6 months post-HT continued to experience better renal function such that their GFR was significantly higher at 12 and 24 months after HT than those who did not show improvement at 6 months post-HT. That is, in those with better versus worse renal function, GFR was 82 ± 22 versus 51 ± 21 mL/min/1.73 m2 at 6 months (P < .001), 69 ± 22 versus 54 ± 22 mL/min/1.73 m2 at 12 months (P = .010), and 68 ± 24 versus 54 ± 21 mL/min/1.73 m2 at 24 months (P = .048) (Figure 4).


In this study, baseline GFR, absence of systemic arterial hypertension, and elective HT were predictors of renal function improvement after HT. Importantly, renal function improvement was sustained over a long term (12 and 24 months) and GFR remained significantly higher in these patients than in patients without improvement in renal function at 6 months post-HT.

Glomerular filtration rate at 6 months post-HT was chosen as the main outcome for several reasons. First, many patients have complications early after HT that are associated with acute kidney injury; thus, renal function stabilizes only after the third or fourth month. Second, the estimated GFR 1 year after HT is likely to be more influenced by the chronic use of calcineurin inhibitors than a cardiorenal type 2 syndrome. Third, renal function typically worsens in the first 3 to 6 months after HT. Therefore, GFR 6 months after HT is a good estimate of expected long-term renal function.

Cardiorenal type 2 syndrome is characterized by chronic low cardiac output that causes decreased renal perfusion, leading to prerenal insufficiency. As previously described, when cardiac output improves, renal perfusion typically normalizes and GFR improves.1 Lindelow and associates5 evaluated patients for cardiorenal type 2 syndrome when assessing candidacy for HT; they administered dobutamine to patients with moderate renal insufficiency and assessed for GFR improvement, which would suggest a higher probability of GFR improvement after HT.

There is much known about predictors of developing severe chronic renal insufficiency after HT, including age, sex, baseline renal function, and the presence of diabetes or hypertension.6 Changes in immunosuppression have led to partial recovery of renal function.7-9 However, as far as we know, this is the first study where predictors of renal function improvement are reported. Pretransplant identi­fication of recipients with higher probabilities of kidney recovery after HT is a common and relevant clinical challenge. On one side, it is known that combined heart and kidney transplant shows better outcomes than HT alone in recipients with severe renal failure.10 On the other side, organ shortage and kidney transplant complications do not make it reasonable to pursue a double organ transplant in every HT candidate with kidney failure. The use of clinical predictors of post-HT renal recovery could help to distinguish recipients with irreversible renal dysfunction from those with reversible cardiorenal syndrome and therefore guide the need of a single or double organ transplant.

We believe that all predictors of renal improve­ment described in this study have biological plausibility. Undoubtedly, baseline GFR is associated with post-HT GFR. In this study, patients with higher baseline GFRs had lower probabilities of renal function improvement since they were less likely to have cardiorenal type 2 syndrome. On the other hand, the lower the GFR of the patient (note that all patients had baseline GFR > 30 mL/min/1.73 m2), the higher the probability of renal function improve­ment. Patients with moderate renal insufficiency include patients with cardiorenal type 2 syndrome. It is in these patients that we would expect improvement in renal function.

Sixteen percent of the HTs included in the study were performed in an emergency status; that is, patients needed circulatory support with short-term devices. The risk for complications is higher in this type of HT, and irreversible kidney damage can result from injury to the kidneys during surgery and during use of mechanical circulatory support.11,12 On the other hand, these injuries typically do not complicate elective HT, and thus these patients have a higher probability of renal function improvement.

Finally, hypertension is a known risk factor for chronic renal insufficiency in patients with and without HT.6 Not surprisingly, in our study, the absence of hypertension was a predictor of renal function improvement. Specifically, in the transplant population, the interaction of calcineurin inhibitors and antihypertensive treatment could account for the kidney injury related to this risk factor. There are many reasons for this association, but we suspect that the use of calcineurin inhibitors and other nephrotoxic drugs used after HT, when combined with hypertension, decreases the probability of renal function improvement.13

Posttransplant hemodynamic management and drug use were per protocol, but differences in inotrope or diuretic doses have not been studied and could affect kidney function. In addition, calcineurin inhibitor levels may have been different between the recipients, accounting for different kidney functions. Because calcineurin inhibitor levels are intended to be lower in renal failure patients, differences in kidney recover may be masked. We note that this was a one-center study with a relatively small sample size and is therefore susceptible to bias; however, a robust statistical analysis, which included bootstrap analysis, was carried out to try to avoid a type I error. In addition, the number of events per covariate was 8, which is an acceptable number to avoid overfitting of the model.14 We emphasize that the results should be interpreted cautiously. The scale described in this study is not intended to be used in other hospitals because it has not been validated externally. However, we believe that the predictors described in this study and use of the scale may guide clinicians in identifying the profiles of patients for which renal function might improve after HT. For instance, in a patient with moderate chronic renal insufficiency and no hypertension, an elective transplant might offer a good probability of improving renal function. On the other hand, a patient with hypertension on short-term mechanical circulatory support has a low probability that renal function will improve after HT.


In this study, baseline GFR, absence of hypertension, and elective HT were predictors of improvement in renal function after HT. These simple and accessible parameters could help to identify recipients with reversible kidney dysfunction who would not need a kidney transplant in addition to the HT.


  1. Ronco C, Haapio M, House AA, Anavekar N, Bellomo R. Cardiorenal syndrome. J Am Coll Cardiol. 2008;52(19):1527-1539.
    CrossRef - PubMed
  2. Mehra MR, Canter CE, Hannan MM, et al. The 2016 International Society for Heart Lung Transplantation listing criteria for heart transplantation: A 10-year update. J Heart Lung Transplant. 2016;35(1):1-23.
    CrossRef - PubMed
  3. Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med. 2003;349(10):931-940.
    CrossRef - PubMed
  4. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-612.
    CrossRef - PubMed
  5. Lindelöw B, Bergh CH, Herlitz H, Waagstein F. Predictors and evolution of renal function during 9 years following heart transplantation. J Am Soc Nephrol. 2000;11(5):951-957.
  6. Lachance K, White M, de Denus S. Risk factors for chronic renal insufficiency following cardiac transplantation. Ann Transplant. 2015;20:576-587.
    CrossRef - PubMed
  7. Andreassen AK, Andersson B, Gustafsson F, et al. Everolimus initiation and early calcineurin inhibitor withdrawal in heart transplant recipients: a randomized trial. Am J Transplant. 2014;14(8):1828-1838.
    CrossRef - PubMed
  8. Schweiger M, Stiegler P, Puntschart A, et al. Everolimus in different combinations as maintenance immunosuppressive therapy in heart transplant recipients. Exp Clin Transplant. 2012;10(3):273-277.
    CrossRef - PubMed
  9. Michel S, Bigdeli AK, Hagl C, Meiser B, Kaczmarek I. Renal recovery after conversion to an everolimus-based immunosuppression in early and late heart transplant recipients: a 12-month analysis. Exp Clin Transplant. 2013;11(5):429-434.
    CrossRef - PubMed
  10. Gill J, Shah T, Hristea I, et al. Outcomes of simultaneous heart-kidney transplant in the US: a retrospective analysis using OPTN/UNOS data. Am J Transplant. 2009;9(4):844-852.
    CrossRef - PubMed
  11. Chen Y-C, Tsai F-C, Chang C-H, et al. Prognosis of patients on extracorporeal membrane oxygenation: the impact of acute kidney injury on mortality. Ann Thorac Surg. 2011;91(1):137-142.
    CrossRef - PubMed
  12. Barge-Caballero E, Almenar-Bonet L, Gonzalez-Vilchez F, et al. Clinical outcomes of temporary mechanical circulatory support as a direct bridge to heart transplantation: a nationwide Spanish registry. Eur J Heart Fail. 2018;20(1):178-186.
    CrossRef - PubMed
  13. Azzadin A, Małyszko J, Małyszko JS, Tankiewicz A, Myśliwiec M, Buczko W. Effects of combination of cyclosporine with losartan or enalapril on kidney function in uremic rats. Pol J Pharmacol. 2002;54(5):469-473.
  14. Vittinghoff E, McCulloch CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol. 2007;165(6):710-718.
    CrossRef - PubMed

DOI : 10.6002/ect.2018.0035

PDF VIEW [275] KB.

From the 1Department of Heart Failure and Heart Transplant, Hospital Clinic de Barcelona, Barcelona, Spain; and the 2Department of Cardiology, Hospital de Cardiologia, Centro Medico Nacional Siglo XXI, Instituto Mexicano del Seguro Social, Mexico City, Mexico
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Juan Betuel Ivey-Miranda, Hospital de Cardiologia, Centro Medico Nacional Siglo XXI, Instituto Mexicano del Seguro Social, 330 Cuauhtemoc, Cuauhtemoc, Mexico City, Mexico ZC 06720
Phone: +52 555627 6900 (22007)