Risk Factors for Cardiovascular Disease in Children and Young Adults after Renal Transplantation

Abstract

Despite good outcomes in pediatric renal transplantation, life expectancy is reduced, mostly as a result of accelerated atherosclerosis. A comprehensive evaluation of cardiac status and risk factors for cardiovascular disease was performed in 60 patients after renal transplantation (age 3 to 29 yr; mean 15.8). Posttransplantation diabetes was diagnosed in 7%. Half of the patients did not engage in any physical activity, and this was associated with increased body mass index. Uncontrolled hypertension was found in 13% of patient, and 53% were on antihypertensive medications. BP index was associated with left ventricular mass index (LVMI). Dyslipidemia was relatively uncommon, with hypercholesterolemia found in 15% and elevated LDL cholesterol found in 10% of patients. Hyperhomocysteinemia was frequent (58%); in most patients, it was not due to folate or B₁₂ deficiency. Lipid and homocysteine abnormalities were associated with cyclosporine therapy. Echocardiography demonstrated normal LVMI in 93% of patients, although LVMI was higher than in healthy control subjects. Cardiac troponin I was normal in all patients, but N-terminal pro-brain natriuretic peptide was elevated in 35% and was associated with LVMI and renal function. Although present cardiac status is relatively normal in pediatric renal transplantation patients, cardiac risk factors are common, and strategies to prevent cardiovascular disease need to be developed.

The long-term outcome of pediatric renal transplantation recipients has improved dramatically in the past 3 decades, as a result of the use of more potent immunosuppressive agents and a decline in mortality from infections. However, the expected life span is shortened, compared with an age-matched population, mostly as a result of accelerated cardiovascular disease (CVD). Heart disease is the second most common cause of death in children after infection, and is the leading cause of death in young adults who have undergone renal transplantation. These children in many cases can lead a full and normal life, but they may have a 10-fold increased risk for cardiac death compared with their peers (^1–3). Although the cardiovascular morbidity and mortality in renal transplant recipients are much lower than in dialysis patients, they still remain unacceptably high.

Research efforts and resources now are being invested in studying the causes and treatment strategies of CVD in adult transplant recipients (⁴). Some of the traditional risk factors for cardiac disease and death, such as hypertension, obesity, left ventricular hypertrophy (LVH), diabetes, hyperlipidemia, and hyperhomocysteinemia, are more common in renal transplant recipients (^1,2,4–6). This may be due to persistence of a problem that predated transplantation or may result from renal insufficiency or the use of immunosuppressive medications (⁷). Other potential risk factors for heart disease are emerging, such as evidence of inflammation (elevated C-reactive protein [CRP] and fibrinogen) (^8,9); cardiac markers (troponins, B-type natriuretic protein [BNP], and N-terminal pro-BNP) (^9–14); and other factors, such as lipoprotein (a) (¹⁵), elevated calcium × phosphate product, and cardiac calcifications (¹⁶). The prevalence of these risk factors has not been studied in the pediatric renal transplantation population.

The aim of this study was three-fold: (1) To perform a comprehensive assessment of traditional, as well as newer, cardiovascular risk factors in children and young adults who have undergone renal transplantation; (2) to look for correlations between patient characteristics, graft function, immunosuppressive medications, and the prevalence of these risk factors; and (3) to evaluate cardiac structure and function and assess the relationship between the various risk factors or markers and cardiac status in these patients.

Materials and Methods

This cross-sectional study included children and young adults up to the age of 29 yr who were followed at our pediatric nephrology clinic and had a functioning renal transplant. Informed consent was obtained from the patient or parents, if the patient was a minor. Exclusion criteria were acute illness or pregnancy at the time of study or estimated GFR (eGFR) <20 ml/min per 1.73 m².

Medical history was taken, including, age, gender, underlying renal disease, dialysis mode and duration, age at transplantation, source of kidney transplant (living or deceased donor), number of previous transplantations, immunosuppression including current prednisone dose per weight, use of pulse corticosteroid for rejection, and other medications. Other data collected included whether the patient had structural heart disease, LVH, or congestive heart failure before transplantation or smoked or was exposed to passive smoking by parents or spouse. Patients were asked about family history of coronary artery disease or stroke, and whether they engaged in regular physical activity.

On the day of the evaluation, weight and height were recorded and body mass index (BMI) was calculated and plotted on age-specific percentiles. BP was measured in accordance with the recommendations of the Fourth National Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents (¹⁷), and mean systolic and diastolic measurements were calculated from the values that were obtained on the day of evaluation and the two previous clinic visits. These values then were divided by the 95th percentile BP measurements for age, gender, and height percentile to produce the BP index (BPI). In adults aged ≥18 yr, the value for age 17 yr was used. An individual with systolic (SBPI) or diastolic BPI (DBPI) >1 was considered to be hypertensive.

Blood and urine samples were collected after an overnight fast, other than water and regular medications. Serum electrolytes, glucose, blood urea nitrogen, creatinine, calcium, phosphorus, magnesium, uric acid, total cholesterol, triglycerides, and urine creatinine were measured on the Vitros analyzer (Ortho Clinical Diagnostics, Johnson & Johnson, Raritan, NJ). Creatinine measurement by this method is traceable to the HPLC method. eGFR in children who were younger than 18 yr was calculated using the Schwartz equation and compared with normal values for age and gender. For adults who were older than 18 yr, eGFR was calculated using the Modification of Diet in Renal Disease 2 (MDRD2) formula (¹⁸). Complete blood count was performed on the Cell-Dyn 4000 analyzer, homocysteine was performed on the Axsym analyzer (both Abbott Laboratories, Santa Clara, CA), and fibrinogen was measured on the ACL Futura plus analyzer (Instrumentation Laboratory, Lexington, MA). Parathyroid hormone was determined by Immulite (Diagnostic Products Corp., Los Angeles, CA); glycated hemoglobin and HDL cholesterol were measured on the Cobas Mira analyzer (Roche, Basel, Switzerland); and vitamin B₁₂, red blood cell folate concentration, and cardiac troponin I were determined on the Access Immunoassay system (Beckman, Fullerton, CA). CRP was determined by nephelometry on the BN2 analyzer (Dade Behring, Liederbach, Germany), lipoprotein (a) (Lp(a)) analysis was performed by the immunoturbidimetric method on Cobas Integra (Roche), and N-terminal pro-B type natriuretic peptide (nt-pro-BNP) was determined using the Elecsys proBNP system (Roche). Urine total protein was measured using the pyrogallol red direct colorimetric method (Sentinel Diagnostics, Milano, Italy).

Cardiac Structure and Function

Echocardiography was performed by a senior pediatric cardiologist using the HP Sonos 4500 machine. Left ventricular mass (LVM) was calculated using measurements by two-dimensional echocardiography. LVM index (LVMI) was calculated by dividing by height in meters to the power of 2.7 to correlate with body size. LVMI >2 SD from the mean for age was used to define LVH (¹⁹). Relative wall thickness (RWT) was measured to define left ventricular geometry. A combination of LVH and elevated RWT (≥0.41) is defined as concentric LVH, whereas if RWT is normal (<0.41), then LVH is considered to be eccentric. Concentric remodeling is defined as normal LVMI with elevated RWT (^6,20). Shortening fraction was measured using M-mode echocardiography, and diastolic function was assessed by Doppler measuring the mitral inflow e/a ratio. A control group consisted of 42 children and young adults (aged 3 to 27 yr; mean 11.5 ± 4.6), who underwent echocardiography for evaluation of an innocent murmur, with no abnormal structural findings found on the echocardiogram.

Each patient performed the 6-min walk test, in accordance with American Thoracic Society recommendations (²¹). This test records the distance walked in 6 min. Minimal normal values are 580 m for young men and 500 m for young women; there are no normal values for children.

Statistical Analyses

Variables are presented as mean ± SD. For comparison of continuous variables, the independent sample t test was used when two groups were compared, and the ANOVA procedure was applied when more than two groups were compared. Association between two continuous variables was assessed by calculating Pearson correlation coefficient. The Fisher exact test was used for assessing the association between two categorical variables. For simultaneous testing of the effect of more than one independent variable on a continuous dependent variable and for correction for possible confounders, the multivariate analysis of covariance model was applied. Multiple linear regression analysis was used to determine the effect of more than one continuous variable on a continuous dependent variable. All tests performed were two-tailed, and a P value of 5% or less was considered as statistically significant.

Results

Patient Characteristics

Sixty renal transplant recipients who were aged 3 to 29 yr were found to be eligible to participate in the study, and all underwent evaluation. Thirty-four were children who were younger than 18 yr, and 26 were young adults. Renal dysplasia and genetic kidney diseases composed the majority of the causes of renal failure. Six patients had undergone preemptive renal transplantation, and 25 had been on dialysis for <1 yr. Five patients had received a second renal transplant. Patient characteristics are summarized in Table 1. Immunosuppressive medications are shown in Table 2.

A family history of premature coronary artery disease or stroke was present in 16 (27%) patients. Two (3%) patients smoked up to 1 pack a day of cigarettes, and 20 (33%) reported passive smoking at home. Four (7%) children developed persistent posttransplantation diabetes (Table 3). One of them had acute-onset insulin-dependent disease, as part of the Wolfram syndrome. The remaining three patients initially were found to have slightly elevated glycated hemoglobin levels with normal fasting glucose concentrations, and subsequent glucose loading revealed diabetes. They were treated with dietary modification, with marked improvement. Fasting serum glucose and glycated hemoglobin concentrations were normal in all other patients.

Blood Pressure

Thirty-two (53%) patients were treated with antihypertensive agents, 10 of whom required two or more medications. Twenty-five (42%) patients were administered calcium channel blockers; 19 (32%) patients were administered β blockers; 11 patients received a combination of both; and two patients each were treated with α blockers, diuretics, and angiotensin-converting enzyme inhibitors. Eight (13%) patients were found to have uncontrolled hypertension, as defined by a BPI >1 (seven with elevated SBPI and one with a combination of elevated SBPI and DBPI). Mean SBPI was 0.93 ± 0.06, and mean DBPI was 0.80 ± 0.1, suggesting a tendency toward elevated SBP. BP was not affected by choice of immunosuppressive medications. DBP but not SBP was weakly correlated with GFR (r = −0.274, P = 0.043). For patients who are at high risk for CVD, it has been suggested that target BP measurements should be below the 90th, percentile or <120/80 mmHg, in adults (^17,22). When we indexed the patients’ mean BP measurements to these values, we found that 38 and 12% of patients had increased SBPI and DBPI, respectively.

Renal Function

The mean eGFR in the study group was 76.4 ml/min per 1.73 m². One third of the patients had an eGFR >90 ml/min per 1.73 m² with no evidence of kidney damage, such as hyperfiltration or proteinuria. Twenty-four patients had stage 2 chronic kidney disease (CKD), 13 fell into the category of stage 3 CKD, and three were stage 4 (Table 3). The eGFR values were higher in the pediatric age group (<18 yr; 86.7 ± 27.2 ml/min per 1.73 m²) compared with the young adults (64.6 ± 30 ml/min per 1.73 m²; P = 0.04). This may be due partly to the different methods used to calculate renal function. However, there was a trend toward a longer time from transplantation in the older group (4.5 ± 3.2 yr in the pediatric group versus 6.1 ± 3.5 yr in the young adults; P = 0.07).

Physical Activity and BMI

Patients were defined as physically active when they engaged in moderate physical activity three times weekly or more, mildly active when they engaged in mild to moderate physical activity up to three times weekly, or sedentary when they had no regular exercise. Half of the study group reported that they did not engage in any regular physical activity. Fifteen (25%) patients were physically active, and 14 (25%) were mildly active.

Mean BMI percentile for age was 49 ± 31% (range 2 to 96%), and 11 (18%) patients were overweight. Five children were found to be overweight (BMI >85th percentile for age), and five were underweight (<5th percentile of BMI for age). In the group of young adults, the mean BMI was 22.1 (range 16.2 to 30.4); six patients were overweight (BMI >25), and five were underweight (BMI <18.5). Individuals who led a sedentary lifestyle were more likely to be overweight than those who were physically active (nine [30%] of 30 versus two [6.7%] of 30; P = 0.04).

Physical conditioning was assessed by administration of the 6-min walk test (²¹). Fourteen patients did not perform the test for the following reasons: age <5 yr, poor vision, significant developmental delay, avascular necrosis of the hip, or declined to participate in this part of the study. Eighty percent of those who performed the test had normal results, and an additional five (10%) younger children had values within 100 m of the normal adult values.

Anemia, Calcium, and Phosphorus

Anemia was defined as a hemoglobin concentration at least 2 SD below the mean normal value for age and gender or those who were treated with erythropoietin. Twenty-three (38%) patients were anemic, 10 of whom were treated with recombinant erythropoietin. The frequency of anemia was identical in the pediatric and young adult groups and was associated with a lower eGFR.

Calcium and phosphorus serum concentrations were within normal limits for age, other than one patient with mild hypophosphatemia 3 mo after transplantation, which later resolved. Calcium × phosphorus product was <55 in all patients, other than one 3-yr-old patient with a product of 65.9. Mean intact parathyroid hormone concentrations were elevated at 108 pg/ml; 31 (52%) patients had values that exceeded the laboratory upper limit of 60 pg/ml, 16 of whom had values >100 pg/ml.

Dyslipidemia

None of the patients was receiving lipid-lowering medications before this study. Hypercholesterolemia (>200 mg/dl) was found in nine (15%) patients, with elevated LDL cholesterol (>130 mg/dl) in six (10%) patients (Table 4). Applying the more stringent cutoff, as is recommended for high-risk populations, yielded 18 (30%) patients with LDL cholesterol values that exceeded 100 mg/dl. Decreased HDL cholesterol concentrations (<40 mg/dl) were found in nine (15%) patients. Serum triglycerides were high (>150 mg/dl) in 18 (30%) patients. LDL cholesterol levels were correlated with age (P = 0.008) but not with eGFR. Patients who were treated with cyclosporine had higher total cholesterol and LDL cholesterol levels than those with tacrolimus-based immunosuppression (Table 5). Analysis of covariance showed that this association was not due to differences in age, years from transplantation, or eGFR between the two groups.

Serum triglycerides were not affected by calcineurin inhibitor treatment but were higher in patients who were treated with mycophenolate mofetil compared with patients who were taking azathioprine (135 versus 104 mg/dl; P = 0.04). This disparity was maintained after correction for age, time from transplantation, and eGFR.

Serum Lp(a) concentration was elevated (>30 mg/dl) in 26% of patients, with a range of up to 2.5 times the upper limit of normal (Table 4). Patients with elevated Lp(a) were more likely to have hypercholesterolemia and high LDL cholesterol values than those with normal Lp(a) values. Lp(a) values were not correlated with GFR or influenced by choice of immunosuppression.

Markers of Inflammation

Laboratory testing was performed only when acute illness was not present. CRP was elevated (>0.5 mg/dl) in 11 (18%) patients, and fibrinogen concentration was increased (>450 mg/dl) in eight (13%) patients. Five patients had abnormal results of both markers, two of whom carried a diagnosis of familial Mediterranean fever (Table 4). No association was observed between markers of inflammation and renal function or anemia.

Homocysteine and Folic Acid

Fasting plasma homocysteine levels were elevated in 35 (58%) patients, using the cutoff value of 12.5 μmol/L (Table 4). Twelve patients had values that exceeded 20 μmol/L, three of whom were found to be homozygous for the C677T mutation of the MTHFR gene. A correlation was found between homocysteine levels and eGFR (r = −0.383, P = 0.003). Vitamin B₁₂ deficiency was found in 11 (18%) patients, 10 of whom had elevated homocysteine. Red blood cell folate levels were slightly below normal in three (5%) patients who had hyperhomocysteinemia. However, 24 of 35 patients with hyperhomocysteinemia had no abnormalities of vitamin B₁₂ or folate. None of the patients had received vitamin supplementation at the time of study or in the preceding year. In cases in which vitamin deficiency or hyperhomocysteinemia was found during the study, subsequent treatment was initiated.

Cyclosporine-based immunosuppression was associated with higher plasma homocysteine levels (Table 5). Patients who were treated with cyclosporine were older and had a lower eGFR than those who received tacrolimus-based immunosuppression; however, after correction for GFR, the difference remained significant (P = 0.03).

Echocardiography

Five patients had previously diagnosed cardiac anomalies, including dilated cardiomyopathy, Ebstein anomaly, tetralogy of Fallot, constrictive pericarditis, and discrete subaortic stenosis; the last three had undergone surgical correction. These patients were excluded from analysis of echocardiographic parameters.

The mean LVMI was 30.9 ± 7.4 g/m^2.7 and was elevated in only four of 55 renal transplant recipients in our study. The RWT was normal in the four patients with elevated LVMI, which classifies them as eccentric LVH. All four patients had hypertension; for three of them, hypertension was uncontrolled. An additional six patients had concentric remodeling (elevated RWT with normal LVMI), two of them had controlled hypertension, and the others had normal BP. Although the mean LVMI was well within normal limits, it was significantly higher than in the control group (n = 42; mean LVMI 23.4 ± 5.1; P < 0.001; Figure 1) and was not correlated with age in either group. Left ventricular end diastolic volume and RWT were not significantly different in patients and control subjects. LVMI correlated with SBPI (r = 0.469, P = 0.001) and with DBPI (r = 0.370, P = 0.009) but not with the number of antihypertensive medications prescribed, time on dialysis before transplantation, or time since transplantation. Patients with a patent arteriovenous fistula had a higher LVMI than those who did not have an arteriovenous fistula (35.9 ± 7.6 versus 28.6 ± 7; P = 0.037). There was a tendency toward increased LVMI in patients with lower eGFR and in patients who were treated with cyclosporine; however, this did not reach statistical significance. No association was seen between anemia and increased LVMI.

Shortening fraction was normal in all patients, other than one young man who had a mild decrease of shortening fraction with no other abnormalities. The e/a ratio, which is a parameter of left ventricular diastolic performance, was abnormal in only two patients and was not correlated to BPI.

Two boys, aged 12 and 15 yr, were found to have small calcifications on the left ventricular wall. Both had well-functioning renal grafts and normal serum calcium and phosphorus levels. Intact parathyroid hormone concentration was normal in one patient and slightly elevated in the other (80 pg/ml), who had experienced severe hyperparathyroidism before transplantation. The calcifications had not been noted on earlier echocardiograms.

Cardiac Markers

Serum troponin I was below the level of detection (0.08 μg/L) in all patients. Mean serum nt-pro-BNP concentrations were elevated at 231.7 ± 283.7 pg/ml, and 19 (35%) patients, excluding those with structural heart disease, had values that exceeded 200. No association was found between nt-pro-BNP levels and BPI or the presence of an arteriovenous fistula. Multiple regression analysis demonstrated that both decreased eGFR and increased LVMI were independently associated with high nt-pro-BNP concentrations (Figures 2 and 3).

Combination of Risk Factors

The number of risk factors for CVD was counted in each patient. We included hypertension, diabetes, smoking (including passive smoking), positive family history, lack of regular exercise, dyslipidemia, and overweight in the traditional risk factors and elevated CRP, fibrinogen, homocysteine, Lp(a) or nt-pro-BNP, and LVH and anemia in the “novel” risk factors. The mean number of traditional risk factors was 2 ± 1.3 and the number of novel risk factors was 1.9 ± 1.3 per patient. Six patients had no traditional risk factors, 17 had only one, 20 had two, 10 had three, and seven patients had more than three traditional risk factors for CVD.

Discussion

Many studies have shown that risk factors for CVD are common in adults with renal failure and after transplantation (^3,4,7). Although cardiovascular mortality is relatively high in pediatric renal transplant recipients, it does not necessarily follow that the same risk factors are to be found in children and young adults. Most adults with renal failure have established atherosclerosis or clinical cardiac abnormalities before reaching ESRD, and in some, diabetes or chronic hypertension is the cause of renal impairment (⁴). The clinical characteristics of childhood-onset renal insufficiency are very different from those of adult-onset disease; therefore, risk factors and markers for CVD in children and young adults who are living with a kidney transplant need to be assessed. In this study, we attempted to examine the prevalence of these factors in a comprehensive manner, although our ability to draw meaningful conclusions is limited because of the relatively small number of patients and the absence of concrete end points. The study design is cross-sectional, which describes the population at a single point, precluding patient follow-up.

Traditional Risk Factors

We evaluated 60 patients with renal failure that started in childhood, in the setting of a pediatric renal transplant clinic. Hypertension was a common finding in our population, with more than half of the patients taking antihypertensive medications and 13% with uncontrolled hypertension, which mostly was systolic. These frequencies are slightly lower than those noted in previous studies of pediatric renal transplant recipients, possibly as a result of the lower prevalence of obesity and that all patients in our group are white (^6,22,23). It has been suggested that lowering BP goals below the 90th percentile, or below 120/80, in adults should be recommended in patients with CKD to decrease risk for CVD, as well as to slow the decline of renal graft function (^17,22). With this goal in mind, 38% of patients had elevated SBP and 12% had elevated DBP. In patients with chronic renal disease, office BP measurements may underestimate the prevalence of hypertension. Ambulatory BP monitoring is more sensitive and also may correlate better with LVMI in children (^24,25).

Surprising, we did not find a high prevalence of obesity in our patients; in fact, only one individual fell into that category, although 18% of patients were defined as overweight. Physical conditioning, assessed by the 6-min walk, was acceptable in most patients, but lack of regular physical activity was associated with a higher BMI. Of note, 17% of patients were underweight, either children with a BMI <5th percentile or young adults with a BMI of <18.5, which also may be associated with a higher risk for all-cause and cardiovascular mortality (²⁶). It could be of interest to evaluate diet in this population in a future study.

Hyperlipidemia in pediatric renal transplantation recipients is well described. Previous studies have shown a prevalence of 50%, even in studies that were performed as recently as 2000 (⁵). In contrast, we report a much lower frequency of lipid abnormalities. Even when applying the limit of 100 mg/dl LDL cholesterol concentrations (2.59 mmol/L), 70% of patients had normal values. Significantly higher serum total and LDL cholesterol concentrations were seen in patients who were treated with cyclosporine compared with those who were taking tacrolimus-based immunosuppression. This has been noted in studies that demonstrated improved lipid profiles after changing immunosuppression from cyclosporine to tacrolimus (^27,28). In previous studies that examined dyslipidemia in transplant recipients, most or all of the patients were treated with cyclosporine, whereas the majority of our patients are receiving tacrolimus (⁵). This, as well as relatively normal BMI in our patients and low-dosage alternate-day corticosteroid treatment, may account partly for the differences in blood lipids compared with previous studies. Diet also may play a role; although eating habits in Israel among Jews are closer to the Western rather than Mediterranean diet, saturated fat and cholesterol intake is lower and monounsaturated fat intake is higher among Arabs (^29,30). Looking separately at these two ethnic populations, there was a trend toward slightly lower LDL cholesterol levels in Arab patients, which did not reach statistical significance.

Emerging Risk Factors for CVD

Hyperhomocysteinemia is common in renal insufficiency and after renal transplantation, with an inverse relationship between serum homocysteine concentration and GFR (³¹). Although it is clear that elevated homocysteine concentration in adults with renal disease is associated with CVD, whether this just may be a surrogate marker for the degree of renal compromise or an independent risk factor remains to be established. Folic acid and vitamin B₁₂ supplementation have been shown to decrease homocysteine levels in adults with chronic renal insufficiency and after renal transplantation. We previously showed that folic acid supplementation can normalize homocysteine levels in children who are on hemodialysis (³¹). However, it remains unclear whether treatment with folic acid to decrease homocysteine concentrations in patients with renal disease can modify the risk for heart disease. Hyperhomocysteinemia was a common finding in young renal transplant recipients in this study, despite normal vitamin B₁₂ and folate concentrations in the majority of patients; the relationship to GFR is shown clearly.

Cyclosporine treatment was associated with elevated total and LDL cholesterol, hyperhomocysteinemia, but not BPI. Conversely, the patients who developed diabetes all were on tacrolimus therapy. These possibly deleterious effects on cardiovascular health need to be considered when choosing immunosuppressive therapy (²⁷).

Chronic anemia in renal transplant recipients may be overlooked as a cause of LVH that contributes to cardiac morbidity and mortality (^32,33). Treatment with recombinant erythropoietin of anemic patients who have ESRD has become the standard of care in Western countries. Posttransplantation anemia, however, is a surprisingly frequent finding, even in patients with reasonable graft function, and occurs in children and young adults, as we demonstrate in our study. Other causes of anemia need to be explored, such as iron deficiency or anemia as a result of treatment with antimetabolites. However, many renal transplant patients require erythropoietin therapy to restore hemoglobin to normal or near-normal levels. Although we did not find a link between anemia and LVH in our patients, this may be due to our relatively aggressive evaluation and treatment practice of anemia in renal transplant recipients.

Atherosclerosis is an inflammatory disease, and markers of inflammation such as CRP have been shown to predict the relative risk for future cardiovascular events in healthy individuals and in adults with ESRD (^8,9). Elevated serum CRP is associated with common carotid artery structure and function abnormalities, which are early markers of atherosclerosis, both in healthy children and in children with diabetes (^34,35). Although no specific treatment consistently decreases CRP levels, elevated CRP possibly should prompt interventions that are aimed at treating other modifiable risk factors.

Despite the presence of several risk factors for CVD in our young patients, the echocardiographic findings were somewhat reassuring. Only 7% of patients had LVH, using the 95th percentile of LVMI as a reference (¹⁹). The patients exhibited a significantly higher mean LVMI than did control subjects. Our findings are in contrast to Mitsnefes et al. (⁶), who found a high prevalence of LVH in pediatric renal transplant recipients, which might be explained by a higher rate of uncontrolled hypertension in their population. We found that LVMI correlated with BPI and the presence of an arteriovenous fistula. Despite our relatively benign findings on cardiac echocardiogram, this does not exclude the presence of occult atherosclerosis and other vascular abnormalities. Further studies, using more sensitive vascular intermediate function measures, such as carotid artery intimal-medial thickness and brachial artery flow–mediated dilation, could provide information on the presence of early-stage CVD.

The use of nt-pro-BNP in patients with compromised renal function is problematic because of renal clearance. A recent study in adults suggested that prediction of cardiac dysfunction could be made if different cutoff levels were used for patients with decreased renal function (¹³). Nt-proBNP was found to be significantly predictive of death in adults with end-stage renal failure and had a better predictive value than other cardiac biomarkers such as troponin and CRP (³⁶). In a large, community-based study that included adults with no known heart or kidney disease, McKie et al. (³⁷) found a correlation between LVH and nt-pro-BNP levels. No data exist regarding nt-pro-BNP levels and LVM in children with renal disease; however, in children with congenital heart disease, higher nt-pro-BNP levels were reported to correlate with worse postoperative course (³⁸). In our study, we demonstrate a correlation between nt-pro-BNP and LVMI in children and young adults. Patients with ESRD also may have elevated cardiac troponin I levels, and this has been shown to be a predictor of all-cause mortality (³⁶); however, none of our pediatric renal transplant recipients had detectable levels of troponin I.

We did not address the role of corticosteroid therapy in connection with risk factors for CVD; 97% of our patients received low-dosage oral prednisone as part of their immunosuppressive protocol. Corticosteroids are known to be associated with hypertension, dyslipidemia, and diabetes, and studies have shown an improvement in cardiac risk factor profile in renal transplant recipients after corticosteroid withdrawal (³⁹).

According to the US Renal Data System, the risk for CVD in a 25-yr-old renal transplant recipient is similar to that of a 55-yr-old person without renal disease (³). It was shown recently that the risk for myocardial infarction in young adults (aged 18 to 30 yr) is 2.7% in the first 3 yr after renal transplantation (⁴⁰). Although echocardiography revealed normal cardiac anatomy and function in the large majority of our patients, we need to be mindful of identifying and, if possible, modifying cardiovascular risk factors in these young people, who have a lifetime ahead of them. Patient education regarding healthful eating and regular exercise, as well as the increased risk of smoking, should be incorporated into regular clinic visits. Attention to BP control is mandatory, with goals of therapy perhaps being lowered to the 90th percentile, as in other high-risk groups. It is important to monitor periodically a fasting lipid profile and consider treatment in adolescents and young adults, with treatment goals equivalent to those of high-risk populations (^41,42). Studies that follow the effect of treatment for hyperlipidemia in children need to be performed. Screening for other risk factors such as diabetes, anemia, chronic inflammation, and hyperhomocysteinemia, as well as judicious use of echocardiography, should be part of the treatment of these young patients. It is reasonable to assume that reducing risk factors for CVD in young renal transplant patients will decrease morbidity and mortality; however, long-term prospective studies are needed to validate this assumption.

Figure 1: Left ventricular mass index (LVMI) in renal transplant recipients and healthy control subjects. Mean LVMI in transplant recipients (excluding five patients with structural heart disease): 30.9 ± 7.5 versus 23.4 ± 5.1 g/m^2.7 in control subjects (P < 0.001).

Figure 2: Correlation between estimated GFR and N-terminal pro-B-type natriuretic protein (nt-pro-BNP) in patients (r = −0.532, P < 0.001).

Figure 3: Correlation between LVMI and nt-pro-BNP in patients. (r = 0.32, P = 0.04).

Table 1: Baseline characteristics

Table 2: Immunosuppression

Table 3: Risk factors for cardiovascular disease, per history and physical examinationa

Table 4: Risk factors for cardiovascular disease in young renal transplant recipients: Laboratory findingsa

Table 5: Lipids and homocysteine, by treatment groupa

This research was supported in part by a clinical research grant from the Mirsky Foundation.

Portions of this article were presented in abstract form at the annual meeting of the American Society of Nephrology; November 8 through 13, 2005; Philadelphia, PA.

The authors wish to thank Hanna Amsalem and the Shaare Zedek endocrinology laboratory staff for their superb technical support.