Post-transplant hyperlipidemia affects the majority of solid organ transplant recipients (1,2). While hyperlipidemia has been shown to contribute to an increased cardiovascular risk in adult patients after renal transplantation, a similar link in pediatric recipients has not been established. Changes in serum lipid profiles reported after transplantation include an increase in total cholesterol, triglyceride, LDL cholesterol, and VLDL cholesterol (2–4). On the other hand, the effect of renal transplantation on HDL cholesterol is variable (5,6). Factors including age, body mass index (BMI), pretransplant lipid levels, presence of diabetes, use of antihypertensive medications, and immunosuppressive agents have been implicated as being contributory to posttransplantation hyperlipidemia (7). As part of our institutional protocol, all renal transplant recipients after 1999 have been managed preemptively with fixed-dose pravastatin for its potential benefits in reducing acute rejection (8,9), regardless of the recipients' serum lipid levels.
A previous study from our center evaluated the effect of pravastatin on lipid levels in a cohort of pediatric renal transplant recipients who were receiving maintenance steroids. The preemptive use of pravastatin reduced the incidence of hypercholesterolemia and also lowered LDL (10). An unexpected finding in our study was a substantial decrease in HDL levels over time in the cohort that was treated with pravastatin; data on HDL levels were not available in the control population. Multivariate analyses showed that the two predictors of low HDL were lower prednisone dose and lower creatinine clearance. However, an effect of pravastatin on reducing HDL could not be evaluated.
Since 2004, our center has adopted a novel steroid minimization immunosuppression protocol; even in this cohort, preemptive pravastatin was used as part of our protocol. This gave us the opportunity to analyze retrospective data on serum lipid levels among our pediatric recipients who were managed by steroid minimization, comparing them to patients who received maintenance steroids, to study the effect of steroid elimination on the prevalence and pattern of dyslipidemia, in the setting of preemptive pravastatin use. Data from time points when patients were receiving sirolimus were excluded from the analyses due to the known effect of sirolimus in contributing to dyslipidemia.
Patients and Methods
This was a retrospective longitudinal cohort study of children who received renal transplants at the University of California Davis Children's Hospital from January 2001 to January 2008. None of the recipients were known to have a genetic predisposition for dyslipidemia. No patient in either group was treated with diuretics or other medications known to contribute to dyslipidemia during the study period. As part of the protocol, all patients were treated with fixed-dose pravastatin from postoperative day 1. The dose of pravastatin was 10 mg once daily for children less than age of 10 years, and 20 mg once daily for older children. From 2001 to mid-2004, patients (group 1) were managed with steroid-based maintenance immunosuppression. After mid-2004, all patients were managed with a novel steroid minimization protocol (group 2). Table 1 depicts the protocols used during the study. The Institutional Review Board (IRB) at the University of California, Davis approved the study. Based on the practice at our center of using pravastatin preemptively in all patients and monitoring lipid profiles as part of our routine standard of care and since all data were analyzed retrospectively, we did not consider it necessary to obtain consent/assent from patients and the IRB gave us an exemption to that effect.
Materials and Methods
The clinical data collected for analysis included the age, sex, weight, height, BMI, ethnicity, donor source, mode of renal replacement, and antihypertensive medication requirements. Laboratory values including the pretransplant total cholesterol, and lipid profiles at 1, 3, 6, 9, and 12 months after transplantation were also abstracted; these were obtained in a fasting state. Pretransplant data on lipid levels other than the total cholesterol were not available.
BMI percentiles were based on the Centers for Disease Control and Prevention (CDC) 2000 growth charts and were calculated using a SAS programming code (11). Children were classified as dyslipidemic if any of the lipid values were abnormal relative to established values for either the 5th or 95th percentile for children—with the 5th percentile used for HDL and the 95th used for the others (12). We also analyzed two lipid ratios that have been shown to be predictors of adverse cardiovascular outcomes (13–15): (1) the cholesterol/HDL ratio (abnormal: >4.0) and (2) the LDL/HDL ratio (abnormal: >3.0).
Unadjusted comparisons between group 1 and group 2 were assessed for statistical significance using Fisher exact test for categorical data and t test for continuous measures. We used mixed-effects analysis of covariance models (a multivariate regression methodology for longitudinal data) to estimate adjusted between-group (steroid maintenance versus steroid minimization) mean differences on post-transplant lipid measures. For each of the lipid measures that we examined, these models specified that the mean level could be expressed as the sum of group-specific fixed effects for time since transplant (fixed slopes), group-type (fixed intercepts), group-specific effects in the acute phase of recovery (within 60 days of transplant), and fixed effects for gender, age at transplant, and BMI percentile (measured at 1 month post transplant). To account for patient-specific heterogeneity in baseline levels and over-time changes, random slopes and intercept effects were specified. For multivariate analyses, we used multiple imputation to account for intermittent missing data in follow-up outcome and dose measures. Version 9.2 of the SAS System was used for statistical computations. Statistical significance was achieved when the (two-sided) P value was <0.05. Because the hypotheses being addressed in this observational study concerned the effects of maintenance steroids on specific lipid measures, P values were not adjusted for multiple comparisons.
Twenty-nine children were transplanted during the study period, all of whom are included in the analysis; 16 of these were in group 1. Due to various reasons, including biopsy proven calcineurin inhibitor toxicity, hyperglycemia, and alopecia, five patients (four in group 1 and one in group 2) were started on sirolimus at some point in time during the study period. Demographic data by treatment groups are depicted in Table 2. There was no difference in the mean (±SD) and median age at the time of transplantation between the two groups (9.0 ± 6.8 years and 7.9 years for group 1 versus 10.5 ± 5.2 years and 11.2 years for group 2, respectively, P > 0.05). There were no significant differences between the two groups in any baseline characteristic, including the pretransplant total cholesterol, BMI, or estimated GFR (eGFR). Group 1 received oral steroids as part of the immunosuppression after renal transplantation. The initial dose of prednisone after transplantation was 2 mg/kg per day. At 1, 3, 6, 9, and 12 months, mean doses had been tapered to 0.69 ± 0.18 mg/kg per day, 0.20 ± 0.08 mg/kg per day, 0.16 ± 0.04 mg/kg per day, 0.14 ± 0.06 mg/kg per day, and 0.12 ± 0.07 mg/kg per day, respectively.
Unadjusted Comparisons of Lipid Levels
Data on unadjusted laboratory values of lipid levels and ratios for the two groups are depicted in Table 3.
Although there was no difference between the groups in their pretransplant total cholesterol, at the 1 month follow-up, statistically significant differences in both the cholesterol and HDL levels were noted between the groups. At 1 month, group 1 had a higher mean total cholesterol (200 ± 49 mg/dl in group 1 versus 135 ± 31 mg/dl in group 2, P < 0.05). Although the mean total cholesterol continued to be higher in group 1 throughout the study period, the difference was statistically significant only at 6 months posttransplant. Patients in group 2 had a significantly lower HDL level (70 ± 25 mg/dl in group 1 versus 35 ± 12 mg/dl in group 2, P < 0.05) at 1 month. The difference in HDL levels between the two groups remained significant at all time points, except at 9 months when the difference was of borderline statistical significance (P value = 0.06). During the first 30 days, over half of the difference in total cholesterol between the groups could be accounted for by the difference in the HDL levels. We also observed significantly higher LDL levels and a higher LDL/HDL ratio in group 1 compared with group 2 at 1 month.
There were two patients with nephrotic range proteinuria (urine protein to creatinine ratio >2.5) during at least one time point, one in each group. We did not see any correlation between dyslipidemia and nephrotic range proteinuria.
Point Prevalence of Dyslipidemia
Based on the cut off values for dyslipidemia, 25% of the patients at baseline had hypercholesterolemia, although the prevalence of elevated total cholesterol was not different between the two groups (27% and 23% in group 1 and group 2, respectively, P > 0.05). In spite of the fact that all patients were preemptively given pravastatin after transplantation, a large proportion of children in both groups remained dyslipidemic. Table 4 shows that 56% of patients in group 1 had abnormal total cholesterol; moreover 44% had high triglycerides at 1 month after transplantation. Although no patient in group 2 had hypercholesterolemia at the 1-month follow-up, 31% had abnormally high triglyceride levels. On the other hand, 46% of children in group 2 had abnormally low HDL. Compared with group 1, the higher prevalence of abnormally low HDL levels in group 2 was statistically significant at 1 and 6 months post transplant. We also observed higher prevalences of abnormal cholesterol/HDL and LDL/HDL ratios in group 2 throughout the follow-up period, but these were not statistically significant (P > 0.05, by Fisher exact test).
Mixed Effects Analysis of Covariance Results
Multivariate analyses were performed using mixed effects analysis of covariance models to estimate between-group differences on post-transplant lipid measures with adjustments for the fixed effects of steroid use, age at transplantation, BMI percentile, gender, and ethnicity on posttransplant lipid values as well as for random effects at the patient level, as described earlier. Estimated adjusted between-group differences on lipid measures are shown in Table 5. As can be seen from the table, steroid use had a significant effect on all lipid parameters, with the exception of serum trigycerides, at 1 month. The magnitude of the average effect of steroid use on total cholesterol was considerable at 1 month (66 mg/dl); although the effect was NS at subsequent time points, the size of the effect remained quite large. Similar observations were noted for LDL and for the two lipid ratios. For the HDL, a statistically significant effect of steroid use was noted at every single time point after transplantation, the greatest effect being in the early post-transplant period.
Dyslipidemia is a common problem after solid organ transplantation (7–9,16–18). Although diet may contribute to the dyslipidemia after renal transplantation, a recent study in children has disputed this association (19). The most important contributor to lipid abnormalities after transplantation is the choice of immunosuppressive agents (20). Corticosteroids, cyclosporine, and sirolimus have all been shown to have profound adverse effects on serum lipid levels (21–24), while tacrolimus and mycophenolate mofetil (MMF) appear to have minimal to no dyslipidemic effects (20–22,25).
The need to recognize and investigate dyslipidemia after transplantation is related to the well known observation that even in children, cardiovascular diseases are responsible for over 40% of all mortality in the setting of ESRD (26). Although the risk of cardiovascular mortality decreases significantly after transplantation, it continues to account for about 10% of all deaths in the post-transplant period (27). Among the many contributors to post-transplant cardiovascular mortality, is dyslipidemia. In the adult population, elevated total cholesterol, elevated LDL (28), and low HDL (29) have all been associated with an increased prevalence of cardiovascular events after transplantation.
The aim of our study was to investigate the impact of corticosteroids on lipid levels in pediatric renal allograft recipients. Corticosteroids enhance the activity of acetyl-CoA and free fatty acid synthetase and lead to elevations of total cholesterol, VLDL, and triglycerides. The induction of a hyperinsulinemic state by steroids increases the hepatic synthesis of LDL-receptor activity (30) and an increase in the rate-limiting enzymes involved in lipogenesis (31).
In addition to the potential benefit of lowering the risk of acute rejection (8), pravastatin has also been proven to be useful in reducing post-transplant hypercholesterolemia in both adults and children (10,32,33). Our previous pilot study showed that in a cohort of children immunosuppressed using a steroid based regimen, compared with controls, the use of pravastatin was associated with lower lipid levels at all time points after renal transplantation (10). Although no established guidelines exist on the use of statins in children less than 10 years of age, based on the long-term risk of cardiovascular morbidity and mortality after transplantation, we have felt that it is justifiable to use statins preemptively in pediatric renal allograft recipients, and a fixed-dose pravastatin has been part of our regimen after transplantation since 1999.
In this study, as shown in the results section, multivariable regression analyses suggest that steroids have an independent effect on cholesterol levels and hypercholesterolemia, most notably during the first month after renal transplantation. However, unlike the effect on total cholesterol levels, the effect of steroids on HDL differs (2). Corticosteroid therapy may raise the HDL level by increasing apolipoprotein A-1 synthetase and by decreasing the activity of cholesterol ester transfer protein (CETP) (34), as has been observed in patients with sarcoidosis (35). In a recent study, steroid withdrawal was associated with a 14% to 22% decrease in HDL in a transplanted cohort of adult recipients (36). In our pilot study in children, HDL levels significantly declined after transplantation, especially after 3 months, at a time when the dose of steroids had been reduced significantly (10). In the present study, we again observed significantly lower levels of HDL in our cohort that was not receiving maintenance immunosuppression with steroids (group 2). Regression analyses again confirmed the significant association of steroid use and HDL levels. Compared with group 2, group 1 had higher HDL levels at all time points, and the difference was significant even at the 12-month follow-up, when group 1 recipients were on very low dose prednisolone (0.12 ± 0.07 mg/kg per day).
The risk associated with this high in incidence of abnormally low HDL in our cohort with steroid minimization immunosuppressive therapy is unknown. However, there is growing evidence that low HDL levels may impart additional risk for future cardiovascular disease (37). Furthermore, although not statistically significant, we observed a trend for more children in group 2 having an abnormally high cholesterol/HDL and LDL/HDL ratios, which may further suggest a negative cardiovascular effect of excluding steroids after renal transplantation, as these ratios have been proven to be predictors of adverse cardiovascular outcomes.
The lack of a significant effect of corticosteroids on serum triglyceride levels may be related to the fact that many other factors, such as metabolic derangements, affect triglyceride levels and may have confounded the analyses, although none of our patients in this study were diabetic.
Hence, on one hand, the results of the present study do suggest an adverse effect of steroids on increasing the level of total cholesterol after renal transplantation in children, since the total cholesterol levels were higher at all time points in group 1, although the effect was transient and was only significant during the first month after transplantation, when steroid doses were relatively high. However, in children managed with a steroid minimization protocol, many more patients had lower HDL levels, when compared with the cohort that received maintenance steroids. The potential cardiovascular protective effect of eliminating steroids after renal transplantation, which had hitherto been promising, suddenly becomes a bit more questionable. Having stated that, we do need to recognize and emphasize that dyslipidemia is but one unwanted effect of chronic corticosteroid therapy, the others being weight gain, hyperglycemia, hypertension, growth failure, osteopenia, and many other cosmetic side effects, all of which need to be taken into consideration. Our study was too small and of too short a duration to investigate the effect of steroid minimization on the aforementioned morbidities.
Another variable that needs to be considered and merits further investigation is the role of pravastatin in preventing/treating dyslipidemia. At our center, pravastatin was included in our protocol for its immunomodulatory effects rather than for preventing lipid abnormalities; hence, pravastatin doses were not adjusted according to the individual patient lipid/HDL levels, nor was there a protocol to switch to alternative more-potent statins, which may be more effective in ameliorating lipid abnormalities in the steroid-based group (38), thereby reducing the incidence of hypercholesterolemia in this cohort. A complex dose effect for some of the statins on HDL levels has been shown in previous studies, and hence dose adjustments (in either direction) may be a strategy to consider when aberrations in HDL levels are noted (39). Additionally, the use of strategies, both pharmacologic and nonpharmacologic that may help in increasing HDL-levels, may help in making steroid-minimization even more attractive than it presently is (40).
Our study is limited by its retrospective nature and lack of data on pretransplant lipid parameters, other than the total cholesterol. We also acknowledge that the small sample size of our single institution study adversely affects the power of our analyses to find modest but clinically important associations and limits the generalizability of our findings. Nevertheless, we feel that our study is important, clinically relevant, and points to the need for prospective studies evaluating the long-term effect of steroid-minimization on dyslipidemia.
Our study suggests that in a pediatric cohort, in spite of using fixed-dose pravastatin, the use of maintenance steroids after renal transplantation transiently increases total cholesterol in a dose-related manner. A significant decline in HDL levels was observed in the steroid-minimization cohort, the underlying mechanism for which has yet to be clearly elucidated; moreover, its long-term impact on cardiovascular morbidity is unknown but is concerning enough to warrant further investigation. Having stated that, it is also important to recognize that the use of corticosteroids has other adverse effects that all need to be considered in a risk-benefit analysis before committing children to one or another regimen.
Published online ahead of print. Publication date available at www.cjasn.org.
1. Ong CS, Pollock CA, Caterson RJ, Mahony JF, Waugh DA, Ibels LS: Hyperlipidemia in renal transplant recipients: Natural history and response to treatment. Medicine (Baltimore) 73: 215–223, 1994
2. Kobashigawa JA, Kasiske BL: Hyperlipidemia in solid organ transplantation. Transplantation 63: 331–338, 1997
3. Andany MA, Kasiske BL: Dyslipidemia and its management after renal transplantation. J Nephrol 14: S81–S88, 2001
4. Pirsch JD, D'Alessandro AM, Sollinger HW, Knechtle SJ, Reed A, Kalayoglu M, Belzer FO: Hyperlipidemia and transplantation: Etiologic factors and therapy. J Am Soc Nephrol 2: S238–S242, 1992
5. Tse KC, Lam MF, Yip PS, Li FK, Lai KN, Chan TM: A long-term study on hyperlipidemia in stable renal transplant recipients. Clin Transplant 18: 274–280, 2004
6. Marubayashi S, Ohdan H, Tashiro H, Tokita D, Onoe T, Hayamizu K, Asahara T, Doi S, Okumoto S, Tanji C, Kawai T, Taniguchi Y, Fukuda Y, Dohi K: Studies on post-transplant dyslipidemia in kidney transplant patients. Hiroshima J Med Sci 54: 39–45, 2005
7. Kasiske BL: Cardiovascular disease after renal transplantation. Semin Nephrol 20: 176–187, 2000
8. Katznelson S, Wilkinson AH, Kobashigawa JA, Wang XM, Chia D, Ozawa M, Zhong HP, Hirata M, Cohen AH, Teraski PI, Danovitch GM: The effect of pravastatin on acute rejection after kidney transplantation—a pilot study. Transplantation 61: 1469–1474, 1996
9. Butani L, Pai MV, Makker SP: Pilot study describing the use of pravastatin in pediatric renal transplant recipients. Pediatr Transplant 7: 179–184, 2003
10. Butani L: Prospective monitoring of lipid profiles in children receiving pravastatin preemptively after renal transplantation. Pediatr Transplant 9: 746–753, 2005
11. Centers for Disease Control and Prevention: A SAS Program for the CDC Growth Charts. Available at http://www.cdc.gov/nccdphp/dnpa/growthcharts/resources/sas.htm
. Accessed November 20, 2009
12. US Department of Health and Human Services: Lipids Research Clinics Program. In: The Lipid Research Clinics Population Studies Data Book, Vol. 1. The Prevalence Study. Bethesda, MD, National Institutes of Health, 1527, 1980
13. Kinosian B, Glick H, Garland G: Cholesterol and coronary heart disease: Predicting risks by levels and ratios. Ann Intern Med 121: 641–647, 1994
14. Ballantyne CM, Hoogeveen RC: Role of lipid and lipoprotein profiles in risk assessment and therapy. Am Heart J 146: 227–233, 2003
15. Frontini MG, Srinivasan SR, Xu J, Tang R, Bond MG, Berenson GS: Usefulness of childhood non-high density lipoprotein cholesterol levels versus other lipoprotein measures in predicting adult subclinical atherosclerosis: The Bogalusa Heart Study. Pediatrics 121: 924–929, 2008
16. Penson MG, Fricker FJ, Thompson JR, Harker K, Williams BJ, Kahler DA, Schowengerdt KO: Safety and efficacy of pravastatin therapy for the prevention of hyperlipidemia in pediatric and adolescent cardiac transplant recipients. J Heart Lung Transplant 20: 611–618, 2001
17. Krmar RT, Ferraris JR, Ramirez JA, Sorroche P, Legal S, Cayssials A: Use of atorvastatin in hyperlipidemic hypertensive renal transplant recipients. Pediatr Nephrol 17: 540–543, 2002
18. Argent E, Kainer G, Aitken M, Rosenberg AR, Mackie FE: Atorvastatin treatment for hyperlipidemia in pediatric renal transplant recipients. Pediatr Transplant 7: 38–42, 2003
19. Siirtola A, Virtanen SM, Ala-Houhala M, Koivisto AM, Solakivi T, Lehtimäki T, Holmberg C, Antikainen M, Salo MK: Diet does not explain the high prevalence of dyslipidaemia in paediatric renal transplant recipients. Pediatr Nephrol 23: 297–305, 2008
20. Perrea DN, Moulakakis KG, Poulakou MV, Vlachos IS, Nikiteas N, Kostakis A: Correlation between lipid abnormalities and immunosuppressive therapy in renal transplant recipients with stable renal function. Int Urol Nephrol 40: 521–527, 2008
21. Mathis AS, Davé N, Knipp GT, Friedman GS: Drug-related dyslipidemia after renal transplantation. Am J Health Syst Pharm 61: 565–585, 2004
22. Hohage H, Arlt M, Brückner D, Dietl KH, Zidek W, Spieker C: Effects of cyclosporin A and FK 506 on lipid metabolism and fibrinogen in kidney transplant recipients. Clin Transplant 11: 225–230, 1997
23. Brattström C, Wilczek HE, Tydén G, Böttiger Y, Säwe J, Groth CG: Hypertriglyceridemia in renal transplant recipients treated with sirolimus. Transplant Proc 30: 3950–3951, 1998
24. Legendre C, Campistol JM, Squifflet JP, Burke JT: Sirolimus European Renal Transplant Study Group: Cardiovascular risk factors of sirolimus compared with cyclosporine: Early experience from two randomized trials in renal transplantation. Transplant Proc 35: S151–S153, 2003
25. Wlodarczyk Z, Vitko S, Salmela K, Czajkowski Z, Margreiter RTERRA Study Group: Lipid metabolism in renal transplant patients receiving tacrolimus/sirolimus combination therapy. Transplant Proc 37: 1871–1873, 2005
26. Johnstone LM, Jones CL, Grigg LE, Wilkinson JL, Walker RG, Powell HR: Left ventricular abnormalities in children, adolescents and young adults with renal disease. Kidney Int 50: 998–1006, 1996
27. Parekh RS, Carroll CE, Wolfe RA, Port FK: Cardiovascular mortality in children and young adults with end-stage kidney disease. J Pediatr 141: 191–197, 2002
28. Fazelzadeh A, Mehdizadeh AR, Ostovan MA, Raiss-Jalali GA: Predictors of cardiovascular events and associated mortality of kidney transplant recipients. Transplant Proc 38: 509–511, 2006
29. Kasiske BL, Guijarro C, Massy ZA, Wiederkehr MR, Ma JZ: Cardiovascular disease after renal transplantation. J Am Soc Nephrol 7: 158–165, 1996
30. Hirsch LJ, Mazzone T: Dexamethasone modulates lipoprotein metabolism in cultured human monocyte-derived macrophages. Stimulation of scavenger receptor activity. J Clin Invest 77: 485–490, 1986
31. Chan MK, Varghese Z, Moorhead JF: Lipid abnormalities in uremia, dialysis, and transplantation. Kidney Int 19: 625–637, 1981
32. Imamura R, Ichimaru N, Moriyama T, Shi Y, Namba Y, Nonomura N, Matsumiya K, Toki K, Takahara S, Okuyama A: Long term efficacy of simvastatin in renal transplant recipients treated with cyclosporine or tacrolimus. Clin Transplant 19: 616–621, 2005
33. Lisik W, Schoenberg L, Lasky RE, Kahan BD: Statins benefit outcomes of renal transplant recipients on a sirolimus-cyclosporine regimen. Transplant Proc 39: 3086–3092, 2007
34. Moulin P, Appel GB, Ginsberg HN, Tall AR: Increased concentration of plasma cholesteryl ester transfer protein in nephrotic syndrome: Role in dyslipidemia. J Lipid Res 33: 1817–1822, 1992
35. Salazar A, Mana J, Pinto X, Argimon JM, Hurtado I, Pujol R: Corticosteroid therapy increases HDL-cholesterol concentrations in patients with active sarcoidosis and hypoalphalipoproteinemia. Clin Chim Acta 320: 59–64, 2002
36. Lemieux I, Houde I, Pascot A, Lachance JG, Noël R, Radeau T, Després JP, Bergeron J: Effects of prednisone withdrawal on the new metabolic triad in cyclosporine-treated kidney transplant patients. Kidney Int 62: 1839–1847, 2002
37. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR: High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study. Am J Med 62: 707–714, 1977
38. Blumenthal RS: Statins: Effective antiatherosclerotic therapy. Am Heart 139: 577–583, 2000
39. Barter PJ, Brabdruo-Wogensen G, Palmer MK, Nicholls SJ: Effect of statins on HDL: A complex process unrelated to changes in LDL: Analysis of the VOYAGER database. J Lipid Res 51: 1546–1553, 2010
40. Brousseau ME, Schaefer EJ, Wolfe ML, Bloedon LT, Digenio AG, Clark RW, Mancuso JP, Rader DJ: Effects of an inhibitor of cholesteryl ester transfer protein on HDL cholesterol. N Engl J Med 350: 1505–1515, 2004