Abstract
Objectives:
This study examined whether African American race was associated with an elevated risk of chronic kidney disease (CKD) post-cardiac transplantation.
Background:
CKD often occurs after cardiac transplantation and may require renal replacement therapy (RRT) or renal transplant. African American patients have a higher risk for kidney disease as well as worse post-cardiac transplant morbidity and mortality. It is unclear, however, if there is a propensity for African Americans to develop CKD after cardiac transplant.
Methods:
The Institutional Review Board of Columbia University Medical Center approved the retrospective study of 151 adults (57 African American and 94 non-African American) who underwent single-organ heart transplant from 2013-2016. The primary outcome was a decrease in estimated glomerular filtration rate (eGFR), development of CKD, and end stage renal disease (ESRD) requiring RRT after 2 years.
Results:
African American patients had a significant decline in eGFR post-cardiac transplant compared to non-African American patients (−34 ± 6 vs. −20 ± 4 mL/min/1.73 m2, P<0.0006). African American patients were more likely to develop CKD stage 2 or worse (eGFR <90 mL/min/1.73 m2) than non- African American patients (81% vs. 59%, p<0.0005).
Conclusions:
This is the first study to report that African American patients are at a significantly higher risk for eGFR decline and CKD at 2 years post-cardiac transplant. Future investigation into risk reduction is necessary for this patient population.
INTRODUCTION:
Cardiac transplantation is a lifesaving operation for patients with end stage heart failure and carries a higher post-transplant median survival than other solid organ transplants [1, 2]. Despite improved patient longevity and quality of life in the modern era, there are still certain demographic and clinical risk factors associated with increased morbidity and mortality after transplant. Elucidating such risk factors may allow for improved monitoring, early intervention, and prevention of poor outcomes in vulnerable cardiac transplant patients.
Chronic kidney disease (CKD) occurs frequently after heart transplantation, with reported rates as high 12.9% amongst surviving patients at 5 years post-transplant[3]. The mortality associated with CKD in the general population is well known, with one longitudinal study documenting 5 year mortality rates of 19.5%, 24.3%, and 45.7% amongst patients with CKD stages 2, 3, and 4 respectively[4]. In the transplant population, patients who develop CKD have over double the mortality risk compared to patients with normal post-cardiac transplant renal function, and patients requiring renal replacement therapy (RRT) have four times the risk of mortality [5, 6].
The stages of chronic kidney disease include: Normal kidney function, Stage 1, CKD GFR >90 mL/min/1.73m2; Mild loss of kidney function, Stage 2, GFR 60-89 mL/min/1.73m2; Mild to moderate, Stage 3a, GFR 45-59 mL/min/1.73m2; Moderate to severe, Stage 3b, GFR 30-44 mL/min/1.73m2; Severe loss of kidney function, Stage 4, GFR 15-29 mL/min/1.73m2; and kidney failure, Stage 5, GFR <15 mL/min/1.73m2 [1,2,3]. In a 20 year-long Danish retrospective study, consisting of 471 heart transplant patients, a decrease in renal clearance to eGFR <60 mL/min/1.73m2 at just 1-year post-transplant was independently associated with a significant increase in death at 5 years post-transplant. In fact, an eGFR <60 mL/min/1.73m2 at 1 year post-transplant was a better predictor of need for renal replacement therapy and death 5 years post-transplant than eGFR <60 at the time of transplant as well as acute kidney injury (AKI) requiring perioperative renal replacement therapy [7]. Given the mortality associated with the development of CKD, especially in the post-transplant population, it is crucial to identify which of these patients may have higher risk of developing this comorbidity.
African American race is associated with increased rates of CKD in the general population independent of established risk factors such as hypertension, diabetes, socioeconomic risk factors, age, and sex [8–10]. Patients of African descent are also at increased risk for poor outcomes after liver, kidney, and heart transplantation compared to other racial groups, while controlling for socioeconomic status, diabetes, and race mismatch between donor and recipient [11, 12]. Despite this dual risk for both CKD and adverse allograft outcomes, no study to date has demonstrated higher rates of post-cardiac transplant CKD in adult African American patients— in large part because there have been a scarcity of studies that consider race as a risk factor in the development of CKD following cardiac transplant. In fact, one study that determined rates of kidney disease among solid organ transplants found that rates in Caucasian and African American patients were similar, but kidney disease was only defined as eGFR <30 mL/min/1.73m2, and the study did not analyze renal clearance outcomes in single organ cardiac transplant patients [1].
There are several known risk factors for CKD post-cardiac transplant, and they can be categorized temporally into risk factors present prior to transplant, events that occur in the perioperative period, and risk factors that occur after transplant. Pre-transplant risk factors include previously diagnosed hypertension, diabetes, older age, female sex, and pre-transplant CKD [1, 13, 14]. In the perioperative period, acute kidney injury, particularly injury leading to either a decrease in GFR >25% of the baseline GFR or a GFR<30 mL/min/1.73m2, is associated with increased risk for end stage renal disease (ESRD), mortality, and graft failure at just 1 year post-transplant [15–19]. Post cardiac transplant, frequent hospitalizations and rejection episodes are associated with kidney injury and worse mortality outcomes [5]. Long term calcineurin inhibitor (CNI) use is also associated with CKD, even though it is essential to prevent rejection in cardiac transplant recipients [20].
In our study, we investigated whether African Americans after cardiac transplantation had an elevated risk of CKD, as measured by decline in eGFR new diagnosis of CKD, or worsening CKD from baseline. We hypothesized that the African American patients race was associated with increased CKD after transplant despite controlling for established risk factors such as pre-cardiac transplant hypertension and diabetes, perioperative AKI, post-cardiac transplant rejection, and CNI use.
METHODS:
STUDY AIMS
This retrospective, single center cohort study aimed to determine if African American race was an independent risk factor for post heart-transplant CKD (Figure 1). The primary outcome was a composite categorical variable of decrease in eGFR >50, as calculated by Modification of Diet in Renal Disease formula (MDRD), and/or increase in CKD severity per the National Kidney Foundation Classification staging system for CKD from the time of heart transplant and at 2 years post-transplant. The secondary outcomes were new development of stage 2 or greater chronic kidney injury and ESRD requiring RRT.
Figure 1:
Patients included in study
STUDY POPULATION
There were 171 total adult cardiac transplant patients between 2013 and 2016 (Figure 1). 20 patients were excluded from analysis: 6 received dual organ transplants, 6 died before 2 years follow up, and 8 received their follow up transplant care at other centers and data was not available for analysis at 2 years post-transplant. The remaining 151 adult (age 18 and older) patients who received a single organ heart transplant between 2013-2016 at Columbia University Medical Center, New York, NY were studied. Patients who identified as black or of African descent were placed in the “African American” category. All other patients were placed in the “non-African American” category. Data was collected from the patient’s original records including age, sex, and significant perioperative and post-transplant information. Our protocol was approved by The Columbia University Institutional Review Board.
To stratify CKD at time of transplant, at 1 year, and 2 years follow up, the National Kidney Foundation Classification staging system was used, which divides CKD severity into 5 stages [21]: Stage 1: Normal kidney clearance is an eGFR ≥ 90 mL/min/1.73m2; Stage 2: eGFR ≥ 60-89 mL/min/1.73m2; Stage 3: eGFR ≥ 30-59 mL/min/1.73m2; Stage 4: eGFR ≥15-29 mL/min/1.73m2; and Stage 5: eGFR <15 mL/min/1.73m2, at which point dialysis or kidney transplant is generally needed. Known risk factors for CKD were divided into categories based on temporal risk factors: pre-cardiac transplant, perioperative transplant period (defined as during the same hospital admission in which a patient receives his or her heart transplant), and post-cardiac transplant.
STATISTICAL ANALYSIS
Data is presented as a mean ± standard deviation of the mean for continuous variables and as a percent for categorical variables. Continuous variables were compared with Student’s unpaired t-test and categorical variables were compared using the χ2 test. Simple linear regression analysis was used to determine each individual variable’s effect on kidney dysfunction. In order to adjust for baseline eGFR and the impact this may have on the magnitude of eGFR changes, individual variables were tested for association with a categorical variable defined as a decrease in eGFR of greater than 50 and/or worsened CKD stage. Simple linear regression was also used to further assess the relationship between tacrolimus dosing and the primary outcome according to quartiles within our data set as defined below, as well as the relationship between African American race and the primary outcome within each quartile. All statistical tests were 2-tailed, and P values that were <0.05 were considered statistically significant. Correlation between single variables with P <0.05 was determined using multivariate linear regression. Analyses were performed using Stata/IC version 15.1 software (StataCorp, TX, 2017).
PRE-CARDIAC TRANSPLANT VARIABLES
Pre-transplant variables were chosen based prior studies that had demonstrated an association with the development of CKD post-cardiac transplant [15–19]. Patients were diagnosed with hypertension if they had a chart diagnosis and were on anti-hypertensive medications prior to the development of heart failure, or if they had noted systolic blood pressure values ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg during 2 separate clinic visits prior to transplant. A diabetes diagnosis was based on at least one pre-transplant hemoglobin A1C (in %) reading of 6.4 or greater, or if they had a chart diagnosis of diabetes. The etiology of each patient’s heart failure was based on cardiac imaging prior to cardiac transplant and cardiac explant pathology. Operative notes both prior to and at time of transplant were used to determine the need for ventricular assist device (VAD) as bridge to transplant (BTT).
PERIOPERATIVE TRANSPLANT PERIOD VARIABLES
AKI was determined to be significant if it met at least Stage 1 kidney injury defined by the Kidney Improving Global Outcome (KDIGO) criteria AKI: creatinine increase >25.4 μmol/l within 48 h or 1.5-1.9 times baseline within 7 days [22]. This was based on creatinine measured at any point during the hospitalization in which they were transplanted. Operative notes, medication administration records, and clinic notes were used to determine surgical variables, transfusion data, length of time on vasopressors and inotropes, and length of time spent in the intensive care unit (ICU) and inpatient hospitalization.
POST-CARDIAC TRANSPLANT VARIABLES
Echocardiographic data
Every transthoracic echocardiogram (TTE) performed after the perioperative period (at least 30 days post-transplant) was analyzed to determine right ventricular dysfunction and tricuspid regurgitation. Severity of right ventricular dysfunction in particular has been linked to higher rates of CKD in patients because of a decrease in cardiac output as well as maintaining venous pressure [23]. Right ventricular dysfunction was graded as none/trace, mild, mild/moderate, moderate, or moderate/severe. If dysfunction was moderate or greater, it was considered significant. Since tricuspid dysfunction is a common finding post-transplant and associated with functional RV dysfunction, an identical gradation system was used for tricuspid regurgitation that was assessed via color doppler and continuous wave doppler. If tricuspid regurgitation was measures as more than trace or mild, it was considered significant.
Tacrolimus data
Tacrolimus trough levels were analyzed daily in the acute post-operative period and every few weeks to months post discharge. In particular, 3 months post-transplant, tacrolimus levels are often the highest, the most likely to vary, and are more likely to correlate with post-cardiac transplant rejection risk and kidney injury [24, 25]. Tacrolimus dose at 2 years post-cardiac transplant was also recorded for each patient as a marker of patient’s long-term maintenance dosing. Patients were divided into quartiles based on maintenance tacrolimus dose, with quartile 1 consisting of patients with maintenance doses < 3mg/day, quartile 2 with doses 3–6 mg/day, quartile 3 with doses 6–8 mg/day, and quartile 4 with doses >8 mg/day. Linear correlation was then assessed for patients within each quartile with the primary outcome, as well as for African Americans specifically within each quartile.
Change in estimated glomerular filtration rate (eGFR)
eGFR was calculated at 2 years post-cardiac transplant again using the MDRD formula and was subtracted from the baseline eGFR at the time of cardiac transplant. If a patient required RRT by 2 years post-cardiac transplant, their eGFR was set at 15 mL/min/1.73m2. The difference in eGFR was compared between each patient.
RESULTS:
BASELINE CHARACTERISTICS
The majority of the patient population studied were Caucasian (42%) and African American patients (38%). The remaining patients were Latino (12%), Asian (7%), or other (1%). There was no difference between African American and non-African American patients in terms of age, sex, mechanical support as bridge to transplant, pre-cardiac transplant hypertension, or pre-cardiac transplant diabetes (Table 1). African American patients had a higher baseline eGFR compared to non-African American patients (92 mL/min/1.73m2 vs. 77 mL/min/1.73m2. p <0.001), and African American patients had a lower rate of CKD stage 2 or lower, or eGFR <90, at the time of transplant (46% vs. 74%, p <0.001).
Table 1:
Baseline pre-transplant patient characteristics of heart transplant recipients at Columbia between 2013 and 2016.
| Characteristics | Non-African American (N = 94) | African American (N = 57) | Total (N = 151) | P value |
|---|---|---|---|---|
| Patient factors | 55 ± 11 | 53 ± 11 | 54 ± 10 | 0.18 |
| Age (year ± SD) | 68 (72%) | 36 (63%) | 104 (69%) | 0.31 |
| Male (N, %) | 33 (35%) | 24 (42%) | 57 (38%) | 0.39 |
| Pre-transplant (N, %) | 77 (82%) | 45 (79%) | 122 (81%) | 0.38 |
| Diabetes mellitus | 39 (41%) | 11(19%) | 50 (33%) | 0.02 |
| Hypertension | 37 (39%) | 35 (61%) | 72 (48%) | 0.01 |
| Etiology of Heart Failure (N, %) | 7 (7%) | 0 | 7 (4.6%) | 0.04 |
| Ischemic Cardiomyopathy | 5 (5%) | 7 (12%) | 12(9%) | 0.07 |
| Dilated Cardiomyopathy | 11(12%) | 11 (19%) | 22 (15%) | 0.20 |
| Congenital Heart Disease | 2 (2%) | 0 | 2 (1.3%) | 0.53 |
| Hypertrophic Cardiomyopathy | 35 (35%) | 13 (22.4%) | 48 (32%) | 0.06 |
| Other* | 31 (31%) | 12 (20%) | 43 (28%) | 0.12 |
| 2nd Cardiac Transplant | 60 (64%) | 40 (70%) | 100 (66%) | 0.42 |
| Obesity (BMIa, kg/m2) | 8 (9%) | 5 (9%) | 13 (9%) | 0.96 |
| Pre-transplant BMI > 30 | 26 (27%) | 12 (21%) | 38 (25%) | 0.36 |
| 2 years Post-transplant BMI > 30 | 77 ± 25 | 92 ± 27 | 82 ± 27 | 0.0009 |
| 70 (74%) | 26 (46%) | 96 (64%) | 0.0001 | |
| Implant Devices (N, %) | 25 (28%) | 8 (18%) | 33 (22%) | 0.07 |
| LVADb as BTTc | ||||
| BiVADd or RVADe device as BTT | ||||
| No device prior to transplant | ||||
| Mean eGFRf at transplant | ||||
| Baseline eGFR < 90 mL/min/1.76m2 | ||||
| Baseline eGFR < 60 mL/min/1.76m2 |
Continuous variables presented as mean +/− SD. Categorical variables presented as n (%)
Other indicates that patients received transplant for conditions other than ischemic, dilated, hypertrophic, or congenital cardiomyopathy. This included infiltrative diseases such as amyloid and sarcoid as well as primary valvular disease.
BMI = Body Mass Index;
LVAD = Left Ventricular Assist Device;
BTT = Bridge to Transplant;
BiVAD = Biventricular Assist Device;
RVAD = Right Ventricular Assist Device;
eGFR = estimated glomerular filtration rate
PRE-OPERATIVE RISK FACTORS
Non-African Americans were twice as likely to suffer from ischemic cardiomyopathy than African Americans, who were at 1.5 fold higher risk of dilated cardiomyopathy. Non-African American patients had higher rates of obesity and CKD III (eGFR<60) than African American patients, though not statistically significantly higher than African American patients.
PERIOPERATIVE AND POST-OPERATIVE RISK FACTORS
Perioperative risk factors for CKD among African American and non-African Americans were similar (Table 2), although non-African American patients had longer hospitalizations compared to African American patients. Post-cardiac transplant risk factors for CKD in African American and non-African American patients were also similar (Table 3), although mean maintenance tacrolimus dose was higher in African American patients than non-African American patients.
Table 2:
Perioperative statistics and association with patient race.
| Characteristics | Non-African American (N = 94) | African American (N = 57) | Total (N = 151) | P value |
|---|---|---|---|---|
| Vasopressor use (days) | 5.0 ± 4.2 | 4 ± 2.7 | 4.6 ± 3.7 | 0.85 |
| Primary graft failure requiring mechanical support (N, %) | 14 (15%) | 12 (21%) | 26 (17%) | 0.33 |
| Days on dobutamine | 6.2 ± 5.2 | 7.3 ± 6.9 | 6.6 ± 5.9 | 0.28 |
| Days on milrinone | 7.4 ± 4.0 | 7.4 ± 5.5 | 7.4 ± 5.0 | 0.99 |
| Days to extubation | 3.0 ± 3.3 | 3.1 ± 3.0 | 3.0 ± 3.2 | 0.86 |
| Units of PRBCsa needed | 5.1 ± 3.8 | 4.5 ± 2.9 | 4.9 ± 3.5 | 0.32 |
| Average ICUb length of stay (days) | 8.5 ± 5.7 | 8.3 ± 3.0 | 8.4 ± 5.8 | 0.44 |
| Average hospital length of stay (days) | 61 ± 63 | 42 ± 44 | 58 ± 57 | 0.03 |
| Ischemic time (minutes) | 175 ± 55 | 177 ± 53 | 176 ± 54 | 0.86 |
| Cardiac bypass time (minutes) | 158 ± 46 | 159 ± 46 | 159 ± 46 | 0.91 |
| Significant AKIc peri-transplant | 57 (61%) | 43 (75%) | 100 (66%) | 0.09 |
Continuous variables presented as mean +/− SD. Categorical variables presented as n (%). Data was considered statistically significant if p-value < 0.05. Significant AKI defined as 50% decrease in eGFR in perioperative period.
PRBCs = Packed Red Blood Cells;
ICU = Intensive Care Unit;
AKI = Acute Kidney Injury
Table 3:
Distribution of post-transplant risk factors for chronic kidney disease at 2 years post-transplant.
| Characteristics | Non-African American (N = 94) | African American (N = 57) | Total (N = 151) | P value |
|---|---|---|---|---|
| Mean tacrolimus dose | 5.2 ± 2.9 | 7.6 ± 4.1 | 6.1 ± 3.6 | 0.003 |
| Patients with TTEa evidence of right ventricular dysfunction | 12 (13%) | 12 (21%) | 24 (16%) | 0.18 |
| Patients with TTE evidence of tricuspid regurgitation | 19 (20%) | 11 (19%) | 30 (20%) | 0.89 |
| Average number of hospitalizations at 2 years post-transplant | 1.8 ± 1.7 | 1.6 ± 1.5 | 1.8 ± 1.7 | 0.9 |
| Significant AKI (50% decrease in eGFR or greater) after discharge | 27 (29%) | 19 (33%) | 46 (30%) | 0.79 |
| Tacrolimus blood level <8 ng/mL 5 days post-transplant | 9 (10%) | 5 (9%) | 14 (9%) | 0.86 |
| Tacrolimus blood level >15 ng/mL measured any time post-transplant | 7 (7%) | 2 (4%) | 9 (6%) | 0.98 |
| Patients who had at least 1 episode of rejection over 2 years requiring medication adjustment | 21 (22%) | 14 (26%) | 35 (23%) | 0.75 |
Continuous variables presented as mean +/− SD. Categorical variables presented as n (%). Data was considered statistically significant if p-value < 0.05.
TTE = transthoracic echocardiography.
CHANGE IN ESTIMATED GLOMERULAR FILTRATION RATE AND CKD SEVERITY
We further analyzed kidney clearance by examining the mean decrease in eGFR in African American patients and found it to be a significantly greater decrease compared to non-African American patients at 2 years post-transplant, 34mL/min/1.73m2 vs. 20 mL/min/1.73m2; P < 0.001 (Figure 2A). African Americans had significantly greater decreases in eGFR when compared separately to Caucasians and all other races in our analysis as well (Figure 2B). We analyzed the eGFR of African Americans and non-African Americans at heart transplant by grouping into eGFR >= 90, eGFR >= 60-89, eGFR >45-59, eGFR >30-44, eGFR >=15-29 and eGFR <15 and found that there was similar heterogeneity in both groups at heart transplant and at 2 years post-transplant (Figure 3A and 3B).There were 6 patients who developed ESRD requiring RRT after transplant (4 were African American), which did not reach significance. Although patients with CKD prior to transplant had less absolute change in eGFR, these patients were more likely to develop ESRD. As such, most of the patients who developed ESRD had an eGFR <60 prior to transplant (4 out of 6 patients). The differences in kidney clearance between African American and non-African American patients at baseline and at 2 years post-transplant are shown in Figure 3C.
Figure 2:
Mean change in GFR in (A) African American and non-African American patients at 2 years post-transplant and (B) African American, Caucasian and other races at 2 years post-transplant
Figure 3:
Kidney clearance in African American and non-African American patients (A) pretransplant and (B) post-transplant (C) number of patients who remained without CKD at transplant and 1- and 2-years post-transplant.
RISK FACTOR REGRESSION ANALYSIS
We performed simple regression analysis data related to decline in eGFR of greater than 50 or increase in severity of CKD at 2 years post-transplant and found that African American race, perioperative AKI and higher maintenance tacrolimus dose were associated with significant decline in eGFR (Table 4). To further investigate tacrolimus dose and its association with race and eGFR, African American and non-African American patients were divided into quartiles based on doses in the following ranges: less than or equal to 3.5 mg/day, 3.5-6 mg/day, 6-8 mg/day, or greater than 8 mg/day (Figure 4). Patients in quartile 1, corresponding to the lowest maintenance doses of tacrolimus, were significantly more likely to be non-African American, while those requiring the highest tacrolimus doses, quartile 4, were significantly more likely to be African American (Table 5).
Table 4:
Univariate and Multivariate logistic regression analysis for change in eGFR >50 and/or increase in CKD rank at 2 years post-transplant
| Characteristics | Univariate logistic regression | Multivariate logistic regression | ||||
|---|---|---|---|---|---|---|
| P-value | HR | 95% CI | P-value | HR | 95% CI | |
| African American race | 0.001 | 3.2 | 1.6 - 6.4 | 0.047 | 2.14 | 1.01 - 4.5 |
| Tacrolimus dose | 0.002 | 1.2 | 1.1 - 1.4 | 0.010 | 0.029 | 1.05 - 1.3 |
| Perioperative AKI | 0.012 | 2.4 | 1.2 - 4.9 | 0.065 | 2.0 | 0.96 - 4.2 |
| Age at transplant | 0.56 | 0.99 | 0.96 - 1.0 | |||
| ICU length of stay | 0.57 | 1.0 | 0.99 - 1.0 | |||
| Hospital length of stay | 0.59 | 1.0 | 0.99 - 1.0 | |||
| Heart failure secondary to ischemic cardiomyopathy | 0.06 | 0.52 | 0.26 - 1.0 | |||
| Heart failure secondary to dilated nonischemic cardiomyopathy | 0.06 | 1.9 | 0.98 - 3.6 | |||
| Days on dobutamine post-transplant | 0.82 | 0.99 | 0.93 - 1.0 | |||
| Number of PRBCs given perioperative | 0.32 | 0.95 | 0.87 - 1.1 | |||
| Days on milrinone post-transplant | 0.35 | 0.97 | 0.90 - 1.0 | |||
| Ischemic time | 0.77 | 1.0 | 0.99 - 1.0 | |||
| Cardiopulmonary bypass time | 0.33 | 0.99 | 0.99 - 1.0 | |||
| Time to extubation | 0.24 | 0.94 | 0.85 - 1.0 | |||
| LVAD as BTT | 0.34 | 0.95 | 0.85 - 1.1 | |||
| RVAD as BTT | 0.61 | 0.74 | 0.24 - 2.3 | |||
| Right ventricular dysfunction | 0.51 | 1.3 | 0.62 - 2.7 | |||
| Tricuspid regurgitation | 0.96 | 0.98 | 0.51 - 1.9 | |||
| Pre-transplant diabetes | 0.45 | 0.82 | 0.50 - 1.4 | |||
| Univariate logistic regression | ||||||
| Characteristics | ||||||
|
| ||||||
| P-value | HR | 95% CI | ||||
| Pre-transplant hypertension | 0.55 | 0.68 | 0.19 - 2.4 | |||
| Rejection episode requiring medication adjustment | 0.17 | 1.7 | 0.79 - 3.7 | |||
| Elevated tacrolimus concentration | 0.11 | 0.51 | 0.22 - 1.2 | |||
| Mechanical support needed perioperative | 0.61 | 0.74 | 0.24 - 2.3 | |||
| Female sex | 0.90 | 1.0 | 0.52 – 2.1 | |||
Data presented as p-value, Hazard Ratio (HR), and 95% Confidence Interval (CI). Individual variables were considered statistically significant if p-value <0.05. Individual variables found to be statistically significant were included in multivariate analysis.
Figure 4:
Scatterplot showing change in eGFR based on maintenance tacrolimus dose quartile at 2 years post-transplant in non-African Americans vs. African Americans.
Table 5:
| Characteristics | Tacrolimus Quartile 1 (≤3 mg/day) |
Tacrolimus Quartile 2 (3-6 mg/day) |
Tacrolimus Quartile 3 (6-8 mg/day) |
Tacrolimus Quartile 4 (≥8 mg/day) |
|---|---|---|---|---|
| CKD 2 or greater, eGFR≤90 (no CKD vs. CKD) | 10 vs. 31, p=0.061 | 15 vs. 23, p=0.652 | 15 vs. 24, p=0.759 | 20 vs. 24, p=0.139 |
| Race (non-AA vs. AA) | 35 vs. 6, p=0.0003 | 22 vs. 16, p=0.96 | 23 vs. 16, p=0.63 | 14 vs. 19, p=0.006 |
| Change in eGFR or increase in severity of CKD | P = 0.0001 | P = 0.96. | P = 0.047 | P = 0.006 |
| EC = −0.36 | EC = −0.0047 | EC = 0.18 | EC = 0.25, | |
| SE = 0.082 | SE = 0.094 | SE = 0.092 | SE = 0.008 | |
| African American race and relationship with Change in eGFR and/or increase in severity of CKD | P = 0.018 | P = 0.001 | P = 0.001 | P = 0.006 |
| EC = 0.20 | EC = 0.028 | EC = 0.027 | EC = 0.023 | |
| SE = .08 | SE 0.082 | SE 0.081 | SE 0.083 |
Categorical variables shown within quartiles as N vs. n with associated p-value. Data from regression analyses shown as p-value, estimated correlation coefficient (EC), and standard error (SE).
AA = African American;
non-AA = non-African American
Low tacrolimus dosing was also associated with less of a risk of worsening renal clearance. This was seen across both racial groups. However, African American race remained significantly associated with increased risk of worse kidney dysfunction in each of the tacrolimus dosing quartiles (Table 5).
Multivariate regression analysis was performed in order to further evaluate each variable that reached significance in univariate analysis for the composite outcome of decline in eGFR >50 or increase in CKD rank. African American ethnicity and tacrolimus dose remained significant whereas perioperative AKI did not. These variables also demonstrated no significant correlation indicating that they are each unique predictors of kidney injury post-transplant.
DISCUSSION
Retrospective pediatric studies have described African American race as a risk factor of both CKD and ESRD post-transplant, usually after at least 10 years follow-up. Furthermore, rates of ESRD are generally reported to be 3-10% at 7-10 years follow-up[19, 26, 27]. Our study demonstrates that rates of CKD and decrease in eGFR in adult African American patients are elevated compared to non-African American patients just 2 years after transplant. In our study, the incidence of ESRD was 7% in African American patients and 2% in non-African American patients (4% total) at just 2 years post-transplant. Our finding of a significant difference in rates of post-transplant renal failure by race may reflect greater poster to detect such a difference secondary to the larger proportion of African American patients in our sample size (38%), compared to databases used in post-transplant outcome studies [11, 28].
African American patients also had significantly higher baseline eGFRs compared to non-African American patients. In particular, non-African American patients were more likely to have CKD stage II (eGFR <90 mL/min/1.76m2). Interestingly, more severe CKD (stage III or worse) was associated with less decline in eGFR but higher risk of ESRD — 4 out of 6 of the patients who developed ESRD had a baseline eGFR of <60 mL/min/1.73m2. The association of more advanced CKD with less subsequent decline in eGFR may be due to the high proportion of non-African Americans with CKD at baseline, since non-African Americans carried lower risk for disease progression. Supporting this is the fact that African American patients developed greater declines in eGFR and higher rates of CKD post-cardiac transplant independent of baseline eGFR. Also, since patients with CKD have a lower baseline eGFR, increasing kidney injury will be associated with less of a change in eGFR even though kidney injury in this group may be more significant clinically.
African American patients also required higher tacrolimus maintenance doses compared to non-African American patients, which is consistent with dose requirements noted in prior studies [29, 30]. Higher tacrolimus levels were also associated with an increased risk of a lower eGFR post-transplant. Of note, a large retrospective study in 2011 demonstrated higher morbidity and mortality in African American patients, which was largely attributed to higher rates of rejection. These findings sparked increased diligence to preventing rejection in African American patients, including maintenance of tacrolimus levels despite need for increased dosing [31].
POSSIBLE RISK FACTORS PLACING AFRICAN AMERICAN PATIENTS AT INCREASED RISK OF CHRONIC KIDNEY DISEASE POST-CARDIAC TRANSPLANT
African-Americans are more likely to express CYP3A5 enzymes than people of Caucasian, Indian, and Chinese descent, which is linked to increased metabolism of tacrolimus [32]. The higher tacrolimus dose requirement in African American patients and its effect on kidney clearance is not well understood with different studies reaching opposing conclusions. In one study involving 1078 renal transplant patients, African Americans required higher tacrolimus doses to achieve therapeutic levels at the increased risk of developing interstitial fibrosis/tubular atrophy [29]. However, in another study involving 1609 patients after non-renal solid organ transplant in which total drug levels of tacrolimus and cyclosporine over 10 years were compared between African Americans and non-African Americans, it was demonstrated that African American patients were not at increased risk of decreased renal clearance based on calcineurin exposure compared to patients of other ethnic backgrounds. This was despite a general decrease in eGFR associated with higher levels of calcineurin levels in the blood [30].
In our cohort, African American patients were on average maintained on higher tacrolimus doses than non-African American patients. Consistent with prior data, African American tacrolimus blood levels did not differ significantly from non-African American patients despite higher tacrolimus dosing overall [33]. They also did not have higher rates of rejection compared to non-African American patients. Higher tacrolimus dosing was associated with larger decreases in eGFR and CKD severity in all racial groups, and African Americans were more likely to be placed on higher tacrolimus doses than non-African Americans. Despite this, African American race was associated with decreased kidney clearance even when controlling for tacrolimus dosing. Due to the discrepancy of tacrolimus dosing requirements between patients and the difficulty of maintaining levels in a narrow therapeutic window, other mechanisms of tacrolimus maintenance have developed such as measuring biomarkers for T-cell function [34]. New biomarkers such as calcineurin phosphatase activity and the quantitative analysis of inflammatory markers such as IL-2 and INF-γ will hopefully decrease episodes of rejection and decrease risk for kidney injury in post-cardiac transplant patients.
In the general population, multiple studies have demonstrated that African American patients are at increased risk of chronic kidney disease compared to non-African American patients. This is thought to be, in part, due to higher rates of chronic diseases such as hypertension and diabetes [35, 36]. In fact, these conditions were often associated with CKD post-transplant at follow up 5-10 years post-transplant [1, 7]. In our cohort, however, the rates of hypertension and diabetes did not differ between African American and non-African patients, and they were not associated with CKD at 2 years post-transplant. This is likely because these conditions chronically evolve and their effect on the kidneys may take many years to manifest. Likewise, many variables that were associated with CKD in prior studies did not reach significance in this study (Table 4).
A final possible explanation for African Americans’ enhanced CKD risk is allelic variations in kidney protein expression. Several studies have shown that patients of African descent are up to 3 times more likely to carry the APOL1 G1 allele, and having two copies of this allele is associated with renal failure requiring RRT up to 10 years earlier compared to patients who did not despite lack of underlying risk factors for kidney disease such as diabetes [37–39]. It would be important for future studies to determine whether any treatment aimed at slowing either kidney disease progression or the underlying disease process will have a difference in efficacy in individuals with or without APOL 1 risk alleles [40]. Although our study did not incorporate genetic influence on kidney outcomes after cardiac transplantation, this may be an interesting future area of investigation. In addition, given the potential impact of tacrolimus dosing on the development of CKD in African Americans post-transplant, providers should balance the potential susceptibility to renal toxicity with prevention of allograft rejection as it relates to tacrolimus trough targets in this specific population.
LIMITATIONS
This was a retrospective single center analysis, which by its nature increases the risk of selection bias. The relatively small sample size also increases the risk of a type I statistical error. The rates of post-cardiac transplant CKD in this cohort were larger than those reported in prior cardiac transplant studies, as was the incidence of pre-cardiac transplant comorbidities including obesity and diabetes. Longer observation time would further solidify trends in CKD. One variable that has been found in previous studies to be a potential risk factor for the development of CKD, specifically in African American patients that is notably absent from our study is socioeconomic status. While previous studies have used different data as a surrogate marker for socioeconomic status, i.e. income, education level, it is important to note that when these variables were adjusted for in studies looking at incidence of CKD in African-Americans vs. non-African American patients, differences in socioeconomic status were not able to fully explain the increased propensity of African-Americans for the development of CKD [35, 41–43]
There are implications to understanding health disparities. An additional limitation is that ethnic or different racial groups are poorly defined categories. Thus, there is a need to understand the genetic and environmental bases of these disparities. APOL1 has been confirmed to occur at higher rates in populations of African descent as compared to Caucasians, but its prevalence in the general population and its effect on post-transplant eGFR decline has not been studied. APOL1 may help explain racial differences in post-transplant GFR decline observed in this study and may serve as a better predictor of GFR decline for each individual patient than generalizations based on self-reported ethnicity.
Another limitation is that our study uses an estimated GFR (eGFR), instead of a measured GFR (mGFR). In general practice, eGFR is an estimated value for glomerular filtration rate calculated by the MDRD equation. GFR estimates based on creatinine-based formula may overestimate measured kidney clearance in cardiac transplant recipients, especially in the perioperative period. This is due to associated low muscle mass and chronic illness at the time of transplant, which leads to lower creatinine levels [19, 44].
CONCLUSIONS
African American patients had greater loss of renal clearance post-cardiac transplant compared to non-African American patients despite having lower baseline CKD and the absence of statistically significant differences other transplant CKD risk factors such as hypertension, diabetes, primary graft dysfunction, and perioperative mechanical support. Increased tacrolimus dosing was associated with worsening kidney clearance, and African American patients were placed on higher doses of tacrolimus. However, despite controlling for tacrolimus dosing, African American race remained significantly associated with worsening CKD and a decrease in eGFR post-transplant.
Acknowledgments
FUNDING: E.Y.W. was supported by NIH K08HL122, NIH R01HL152236, the Louis V. Gerstner, Jr. Scholars Program, Lewis Katz Prize, the Esther Aboodi Endowed Professorship at Columbia University, the M. Iréne Ferrer Scholar Award from the Foundation of Gender Specific Medicine and gift from Howard and Patricia Johnson.
Abbreviations
- AKI
acute kidney injury
- BTT
bridge to transplant
- CKD
chronic kidney disease
- ESRD
end stage renal disease
- ICU
Intensive Care Unit
- RRT
renal replacement therapy
- TTE
transthoracic echocardiogram
- VAD
Ventricular Assist Device
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
DISCLOSURES: None of the listed authors have any relevant disclosures.
REFERENCES:
- 1.Ojo AO, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med, 2003. 349(10): p. 931–40. [DOI] [PubMed] [Google Scholar]
- 2.Rana A, et al. Survival benefit of solid-organ transplant in the United States. JAMA Surg, 2015. 150(3): p. 252–9. [DOI] [PubMed] [Google Scholar]
- 3.Thomas HL, et al. Incidence, Determinants, and Outcome of Chronic Kidney Disease After Adult Heart Transplantation in the United Kingdom. Transplantation, 2012. 93(11): p. 1151–1157. [DOI] [PubMed] [Google Scholar]
- 4.Keith DS, et al. Longitudinal Follow-up and Outcomes Among a Population With Chronic Kidney Disease in a Large Managed Care Organization. Archives of Internal Medicine, 2004. 164(6): p. 659–663. [DOI] [PubMed] [Google Scholar]
- 5.Hamour IM, et al. Chronic kidney disease after heart transplantation. Nephrology Dialysis Transplantation, 2009. 24(5): p. 1655–1662. [DOI] [PubMed] [Google Scholar]
- 6.Al Aly Z, et al. The natural history of renal function following orthotopic heart transplant. Clin Transplant, 2005. 19(5): p. 683–9. [DOI] [PubMed] [Google Scholar]
- 7.Fortrie G, et al. Renal function at 1 year after cardiac transplantation rather than acute kidney injury is highly associated with long-term patient survival and loss of renal function - a retrospective cohort study. Transpl Int, 2017. 30(8): p. 788–798. [DOI] [PubMed] [Google Scholar]
- 8.Xue JL, et al. Longitudinal study of racial and ethnic differences in developing end-stage renal disease among aged medicare beneficiaries. J Am Soc Nephrol, 2007. 18(4): p. 1299–306. [DOI] [PubMed] [Google Scholar]
- 9.Li S, et al. Differences between blacks and whites in the incidence of end-stage renal disease and associated risk factors. Adv Ren Replace Ther, 2004. 11(1): p. 5–13. [DOI] [PubMed] [Google Scholar]
- 10.Burrows NR, Li Y, and Williams DE, Racial and ethnic differences in trends of end-stage renal disease: United States, 1995 to 2005. Adv Chronic Kidney Dis, 2008. 15(2): p. 147–52. [DOI] [PubMed] [Google Scholar]
- 11.Liu V, et al. Persistent racial disparities in survival after heart transplantation. Circulation, 2011. 123(15): p. 1642–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mahle WT, Kanter KR, and Vincent RN, Disparities in outcome for black patients after pediatric heart transplantation. J Pediatr, 2005. 147(6): p. 739–43. [DOI] [PubMed] [Google Scholar]
- 13.DePasquale EC, et al. Influence of Pre-Transplant Chronic Kidney Disease (CKD) on Adult Heart Transplant (HT) Outcomes. Journal of Heart and Lung Transplantation, 2013. 32(4): p. S166–S166. [Google Scholar]
- 14.Delgado JF, et al. Risk factors associated with moderate-to-severe renal dysfunction among heart transplant patients: results from the CAPRI study. Clin Transplant, 2010. 24(5): p. E194–200. [DOI] [PubMed] [Google Scholar]
- 15.De Santo LS, et al. Implications of acute kidney injury after heart transplantation: what a surgeon should know. Eur J Cardiothorac Surg, 2011. 40(6): p. 1355–61; discussion 1361. [DOI] [PubMed] [Google Scholar]
- 16.Wang L, et al. Severe Acute Kidney Injury after Cardiac Transplantation is a Marker of Poor Early Outcome and Impacts on Late Renal Function - A UK Cohort Study. The Journal of Heart and Lung Transplantation, 2019. 38(4): p. S143–S144. [Google Scholar]
- 17.Gude E, et al. Acute renal failure early after heart transplantation: risk factors and clinical consequences. Clin Transplant, 2010. 24(6): p. E207–13. [DOI] [PubMed] [Google Scholar]
- 18.Boyle JM, et al. Risks and outcomes of acute kidney injury requiring dialysis after cardiac transplantation. Am J Kidney Dis, 2006. 48(5): p. 787–96. [DOI] [PubMed] [Google Scholar]
- 19.Kolsrud O, et al. Renal function and outcome after heart transplantation. J Thorac Cardiovasc Surg, 2018. 155(4): p. 1593–1604 e1. [DOI] [PubMed] [Google Scholar]
- 20.Starzl TE, et al. Kidney transplantation under FK 506. JAMA, 1990. 264(1): p. 63–7. [PMC free article] [PubMed] [Google Scholar]
- 21.Levey AS, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int, 2005. 67(6): p. 2089–100. [DOI] [PubMed] [Google Scholar]
- 22.Machado MN, Nakazone MA, and Maia LN, Acute kidney injury based on KDIGO (Kidney Disease Improving Global Outcomes) criteria in patients with elevated baseline serum creatinine undergoing cardiac surgery. Rev Bras Cir Cardiovasc, 2014. 29(3): p. 299–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dini FL, et al. Right ventricular dysfunction is associated with chronic kidney disease and predicts survival in patients with chronic systolic heart failure. Eur J Heart Fail, 2012. 14(3): p. 287–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Liu BY, et al. The Effects of CYP3A5 Genetic Polymorphisms on Serum Tacrolimus Dose-Adjusted Concentrations and Long-Term Prognosis in Chinese Heart Transplantation Recipients. Eur J Drug Metab Pharmacokinet, 2019. [DOI] [PubMed] [Google Scholar]
- 25.Tholking G, et al. The tacrolimus metabolism rate influences renal function after kidney transplantation. PLoS One, 2014. 9(10): p. e111128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Choudhry S, et al. End-stage renal disease after pediatric heart transplantation: A 25-year national cohort study. J Heart Lung Transplant, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lee CK, et al. Pre-transplant risk factors for chronic renal dysfunction after pediatric heart transplantation: a 10-year national cohort study. J Heart Lung Transplant, 2007. 26(5): p. 458–65. [DOI] [PubMed] [Google Scholar]
- 28.Lui C, et al. Racial Disparities in Patients Bridged to Heart Transplantation With Left Ventricular Assist Devices. Ann Thorac Surg, 2019. 108(4): p. 1122–1126. [DOI] [PubMed] [Google Scholar]
- 29.Taber DJ, et al. African-American race modifies the influence of tacrolimus concentrations on acute rejection and toxicity in kidney transplant recipients. Pharmacotherapy, 2015. 35(6): p. 569–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Yessayan L, et al. Race, Calcineurin Inhibitor Exposure, and Renal Function After Solid Organ Transplantation. Transplant Proc, 2015. 47(10): p. 2968–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Morris AA, et al. Racial and ethnic disparities in outcomes after heart transplantation: A systematic review of contributing factors and future directions to close the outcomes gap. J Heart Lung Transplant, 2016. 35(8): p. 953–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chen L and Prasad GVR, CYP3A5 polymorphisms in renal transplant recipients: influence on tacrolimus treatment. Pharmgenomics Pers Med, 2018. 11: p. 23–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Beermann KJ, et al. Tacrolimus dose requirements in African-American and Caucasian kidney transplant recipients on mycophenolate and prednisone. Clin Transplant, 2014. 28(7): p. 762–7. [DOI] [PubMed] [Google Scholar]
- 34.Brunet M, et al. Therapeutic Drug Monitoring of Tacrolimus-Personalized Therapy: Second Consensus Report. Ther Drug Monit, 2019. 41(3): p. 261–307. [DOI] [PubMed] [Google Scholar]
- 35.Tarver-Carr ME, et al. Excess Risk of Chronic Kidney Disease among African-American <em>versus</em> White Subjects in the United States: A Population-Based Study of Potential Explanatory Factors. Journal of the American Society of Nephrology, 2002. 13(9): p. 2363–2370. [DOI] [PubMed] [Google Scholar]
- 36.Brancati FL, et al. The excess incidence of diabetic end-stage renal disease among blacks. A population-based study of potential explanatory factors. JAMA, 1992. 268(21): p. 3079–84. [PubMed] [Google Scholar]
- 37.Tzur S, et al. APOL1 allelic variants are associated with lower age of dialysis initiation and thereby increased dialysis vintage in African and Hispanic Americans with non-diabetic end-stage kidney disease. Nephrol Dial Transplant, 2012. 27(4): p. 1498–505. [DOI] [PubMed] [Google Scholar]
- 38.Dummer PD, et al. APOL1 Kidney Disease Risk Variants: An Evolving Landscape. Semin Nephrol, 2015. 35(3): p. 222–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Zhang JY, et al. UBD modifies APOL1-induced kidney disease risk. Proc Natl Acad Sci U S A, 2018. 115(13): p. 3446–3451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Grams ME, et al. Race, APOL1 Risk, and eGFR Decline in the General Population. J Am Soc Nephrol, 2016. 27(9): p. 2842–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Perneger TV, Whelton PK, and Klag MJ, Race and end-stage renal disease. Socioeconomic status and access to health care as mediating factors. Arch Intern Med, 1995. 155(11): p. 1201–8. [PubMed] [Google Scholar]
- 42.Barreto SM, et al. Chronic kidney disease among adult participants of the ELSA-Brasil cohort: association with race and socioeconomic position. J Epidemiol Community Health, 2016. 70(4): p. 380–9. [DOI] [PubMed] [Google Scholar]
- 43.Parsa A, et al. APOL1 risk variants, race, and progression of chronic kidney disease. N Engl J Med, 2013. 369(23): p. 2183–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Borrows R and Cockwell P, Measuring renal function in solid organ transplant recipients. Transplantation, 2007. 83(5): p. 529–31. [DOI] [PubMed] [Google Scholar]