candesartan improves blood pressure control and reduces

Nephrol Dial Transplant (2010) 25: 967–976doi: 10.1093/ndt/gfp581Advance Access publication 3 November 2009

Candesartan improves blood pressure control and reduces proteinuriain renal transplant recipients: results from SECRET

Thomas Philipp1, Franck Martinez2, Helmut Geiger3, Bruno Moulin4, Georges Mourad5,Roland Schmieder6, Michel Lièvre7, Uwe Heemann8 and Christophe Legendre2

1Medizinische Klinik, Universitätsklinikum, Essen, Germany, 2Hôpital Necker & Université Paris Descartes, Paris, France,3Medizinische Klinik III-Nephrologie, Klinikum der Johann-Wolfgang-Goethe-Universität, Frankfurt, Germany, 4Service deNéphrologie-Transplantation Rénale, Hôpital Civil, Strasbourg, France, 5Department of Nephrology and Transplantation, HôpitalLapeyronie, Montpellier, France, 6Department of Nephrology and Hypertension, Friedrich-Alexander-University Erlangen-Nürnberg,Erlangen, Germany, 7Service de Pharmacologie Clinique-UMR5558, Université Lyon 1, Lyon, France, and 8Abteilung fürNephrologie der II.Medizinischen Klinik und Poliklinik, Klinikum rechts der Isar der TU München, München, Germany

Correspondence and offprint requests to: Thomas Philipp; E-mail: [email protected]

AbstractBackground. Hypertension is a risk factor for the twoleading causes of death in renal transplant recipients: car-diovascular disease (CVD) and graft failure. Despite this,the optimum medication for post-transplant hypertensionis unclear.Methods. The Study on Evaluation of Candesartan Cilex-etil after Renal Transplantation (SECRET) was an interna-tional multicentre, double-blind, randomized investigationof the angiotensin II type 1 receptor blocker (ARB) cande-sartan cilexetil versus placebo in renal allograft recipientsoriginally designed to study 700 patients for 3years. Thecandesartan dose was escalated from 4 to 16mg daily,followed by addition of co-medication, if needed, withthe aim of achieving a diastolic blood pressure (BP)<85mmHg. The primary efficacy variable was a compositeof all-cause mortality, cardiovascular morbidity and graftfailure.Results. SECRET was stopped prematurely as the prima-ry event rate was much lower than expected. At thatpoint, 502 patients were enrolled; 255 received candesar-tan and 247 placebo. Thirteen primary events had oc-curred in each group. Control of both systolic anddiastolic BP was better in the candesartan group. Urinaryprotein excretion and protein/creatinine ratio decreasedon candesartan but increased on placebo. Serum creati-nine and potassium were increased in candesartan pa-tients, but these changes were generally small.Conclusions. SECRET provides insights into the designand conduct of studies in this area and evidence for theutility of candesartan, which showed good safety and tol-erability, improved BP control and decreased proteinuriain renal transplant recipients.

Keywords: angiotensin II type 1 receptor blocker; blood pressure;candesartan; kidney transplantation; proteinuria

Introduction

Hypertension is a risk factor for the two leading causes ofdeath in renal transplant recipients: cardiovascular disease(CVD) and graft failure [1–3]. It is present in 60–85% ofpatients, contributing to a CVD incidence of 3.5–5.0% perannum. In spite of its importance, the optimum medicationfor post-transplant hypertension is not known. Nor hasthere been a long-term controlled trial of the effects ofany antihypertensive drug on morbidity and mortality inthis population.

The aetiology of hypertension after renal transplantationis multifactorial. Damage to the transplanted kidney withloss of function, activation of the renin–angiotensin–aldo-sterone system (RAAS) by microvascular lesions or renalartery stenosis and the blood pressure (BP)-raising proper-ties of some immunosuppressive drugs all contribute [1–4].Of these different mechanisms, activation of the RAAS is ofparticular importance. In addition to its effects on vascularfunction and fluid balance, angiotensin II also influencescellular processes in inflammation, tissue growth and fibro-sis, the effects of which can promote progression of chronicrenal disease [5,6].

Candesartan is a potent angiotensin II type 1 receptorblocker (ARB) that binds more tightly to the AT1-recep-tor and dissociates more slowly than earlier members ofthe class [7–9], providing effective 24-h BP control froma once-daily dose. In the only study of candesartan in re-nal transplant recipients published to date, an open-labeluncontrolled trial in 62 treated and 11 untreated patients,Omoto et al. [10] reported that candesartan lowered bothsystolic BP (SBP) and diastolic BP (DBP) by 15% andreduced proteinuria by 54%.

The Study on Evaluation of Candesartan Cilexetil afterRenal Transplantation (SECRET) is the first long-termprospective study of the effects of antihypertensive therapyon morbidity and mortality in renal transplant recipients(see Hiremath et al. [11]).

© The Author 2009. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.For Permissions, please e-mail: [email protected]

Downloaded from https://academic.oup.com/ndt/article-abstract/25/3/967/1913391by gueston 31 January 2018

Materials and methods

SECRET was a double-blind, randomized, controlled study conducted in47 centres in France, Germany, Czech Republic and Austria (see Appen-dix). Randomization commenced on 27 March 2000. The study was ap-proved by the appropriate ethics committees and performed in accordancewith the principles of Good Clinical Practice and the Declaration of Hel-sinki. All patients gave written informed consent.

Objectives

The primary objective was to determine whether candesartan reduces theincidence of a composite efficacy variable made up of three events: all-cause mortality, cardiovascular morbidity and all-cause graft failure (cre-atinine clearance <15mL/min or dialysis). Creatinine clearance <15mL/min or dialysis was selected as a cutoff in accordance with the NationalKidney Foundation Kidney Disease Outcomes Quality Initiative (NKF-KDOQI) criteria for stage 5 chronic kidney disease [12]. Secondary effi-cacy variables included cardiovascular mortality, creatinine clearance, 24-h urinary protein excretion (determined by the protein/creatinine ratiofrom spontaneous urine collection), protein/creatinine ratio, urinary albu-min (determined using Micraltest®—a semi-quantitative microalbuminur-ia detection test), occurrence of nephrotic syndrome (≥3.5g protein/24h confirmed on complete 24-h urine collection), blood lipids (triglycer-ides and total, low-density and high-density lipoprotein cholesterol) andhaemoglobin. Safety variables included adverse events, laboratory dataand physical examination. Creatinine clearance was estimated by theCockcroft and Gault formula [13].

Inclusion criteria

Inclusion criteriawere renal allograft transplantation within 1–10years, age30–69years, either sex, negative enalapril test [this test, which looks for noincrease in serum creatinine after a loading dose or a short treatment withenalapril, was performed in angiotensin converting enzyme (ACE) inhib-itor-naïve patients to exclude the possibility of renal artery stenosis], cre-atinine clearance ≥25mL/min (with variations <50% over the past 6months), either normotensive (sitting DBP <85mmHg and sitting SBP<135mmHg) or hypertensive (treated or untreated; sitting SBP <200mmHg and sitting DBP <105mmHg) and either non-diabetic or diabetic.Because of the known impact of diabetes on cardiovascular disease andlong-term transplant outcome, randomization was stratified on diabetes.

Exclusion criteria

These included treatment with an ACE inhibitor or ARB for a non-renalindication, pregnancy or the possibility of becoming so, breast feeding,sitting SBP ≥200mmHg, sitting DBP ≥105mmHg, aortic valve stenosis,proteinuria ≥2.0g/24h, proliferative diabetic retinopathy, hypotensiveepisodes, chronic liver disease, serum potassium ≥5.8mol/L, psycholog-ical problems, drug addiction, excessive alcohol consumption, participa-tion in another study within 30days, history of myocardial infarction,stroke or transient ischemic attacks, present or past renal artery stenosis,cancer (in the last 5years), wasting disease or hypersensitivity to an ARB.Treatment with an ACE inhibitor or ARB was stopped 4weeks beforerandomization.

Conduct of study

The target BP was sitting DBP <85mmHg. A total of 502 patients wererandomly allocated to candesartan cilexetil 4mg (n=255) or placebo (n =247), taken once daily in the morning, in addition to any existing antihy-pertensives. A pre-maintenance phase of 24weeks was followed by a main-tenance phase of 124weeks. During the former, the dose of study drug wasescalated up to 16mg daily, as long as it was tolerated and sitting DBP was≥85mmHg. Patients who did not achieve target BP were then given anti-hypertensive co-medication. The antihypertensive drugs used were se-quentially: diuretics, β-blockers, calcium channel blockers, centrallyacting sympatholytic drugs and postsynaptic α-blockers. Normotensivepatients without antihypertensive medication were kept on candesartan 4mg or placebo. All patients with a sitting DBP <95mmHg entered themaintenance phase.

During the maintenance phase, target sitting DBP remained <85mmHg. Patients originally hypertensive remained on the study medicationat the maximum tolerated dose. When improved control was necessary,antihypertensive co-medication was modified, as already described. If sit-ting DBP increased to ≥85mmHg in an originally normotensive patient,the study drug was escalated up to 16mg, followed by addition of co-medication, if needed.

Clinical procedures

BP was recorded in the morning after administration of study medication,using the arm with the highest median sitting diastolic pressure. It wasmeasured by an electronic oscillometric device (Pressolink™) five timesat 2-min intervals, with the patient seated and quiet, and the median valuewas calculated. Systolic BP was taken to be when the initial sound ap-peared and diastolic BP to be when it disappeared. In 80% of the patients,the BP values were taken ‘online’ and registered centrally.

Discontinuation

Patients were withdrawn at their request (withdrawal of consent) if sittingDBP was ≥95mmHg despite maximum permitted study drug and co-medication or for any of the following reasons: symptomatic hypotension(on the lowest dose of study drug, without co-medication), sitting SBP>220mmHg, sitting DBP >120mmHg, creatinine clearance <15mL/min,serum potassium >5.8mol/L, proteinuria >3.5g/24h, unexplained haemo-globin <8g/dL unresponsive to erythropoietin and treatment with an ARB,ACE inhibitor or non-steroidal anti-inflammatory drug (prolonged).

Sample size determination

Event rates of 39% in the placebo group and 29% in the candesartangroup had been expected during 3years of follow-up, based on the Collab-orative Transplant Database [2]. According to power tables for the F-testfor comparing two exponential survival distributions, 120 events were pre-dicted in the placebo group and 90 in the candesartan group. Accordingly,310 patients in each group were needed for the final analysis. Assuming anoverall drop-out rate of 13%, 350 per group were required.

Statistical analyses

Frequencies were compared by the log-rank test using Kaplan–Meier es-timates, Fisher's exact test, the chi-square test, and the Mantel–Haenszeltest. Continuous variables were compared by the non-parametric Mann–Whitney–Wilcoxon test. Time-to-event curves were constructed usingKaplan–Meier estimates and compared by the log-rank test. All random-ized patients given at least one dose of study drug were evaluated forefficacy and safety. The primary and secondary parameters were also ana-lysed in a per-protocol population (n=393; all patients who had been con-tinuously treated over their observation period). Patients with complianceof <75% or >125% were excluded from per-protocol analyses.

An interim analysis was predefined by the protocol. A decision to stopthe trial prematurely could be taken at that stage if the possibility of dem-onstrating a significant difference at the end of 3years was judged insuf-ficient (<60 events in total).

Results

A predefined interim analysis showed that there would betoo few events to permit any conclusions about the effectof treatment on the primary efficacy variable within theplanned time course of the study. The trial was thereforeterminated, and each patient made a final visit at the nextscheduled time point. Four hundred patients attended thelast visit (candesartan n=213, placebo n=187). The meanduration of follow-up was 21.4months on candesartan and19.5months on placebo. Forty-six (18.0%) patients on can-desartan and 63 (25.5%) on placebo had been withdrawn(Figure 1).

968 T. Philipp et al.


Baseline characteristics

There were no group differences in baseline characteris-tics (Table 1). Similar numbers in the two groups weretaking antihypertensive drugs (87%) [of these, four pa-tients (0.8%) were taking ACE inhibitors: three in thecandesartan group and one in the placebo group]. A highproportion (44% overall) were taking HMGCoA reduc-tase inhibitors.

Primary efficacy variable

When the study was stopped, there had been 13 occur-rences of the primary efficacy variable in each group (in-tention-to-treat population: 5.1% on candesartan, 5.3% onplacebo; P=0.763). Results for the three components ofthe primary variable appear in Table 2. In the per-proto-col population (n=393), five (2.39%) occurrences of theprimary variable occurred on candesartan and six (3.26%)on placebo (P=0.542).

Clinical secondary end points

In the intention-to-treat population, there were three deathson candesartan (none cardiovascular) and four on placebo(three cardiovascular). Nine patients on candesartan andfive on placebo suffered a non-fatal cardiovascular event.Graft failure was unequally distributed, occurring in one pa-tient on candesartan and in six on placebo. No group differ-ence was statistically significant. The per-protocol analysesalso did not demonstrate any significant differences.

From baseline to last visit, the reductions of sitting SBPand DBP were greater on candesartan and antihypertensiveco-medication than on placebo and antihypertensive co-medication (Table 3). Sitting DBP ≥95mmHg at two suc-cessive visits occurred in 20 candesartan and 44 placebopatients (P=0.0012). High BP (DBP ≥105mmHg and/orSBP ≥200mmHg) occurred in 11 patients on candesartan(4.3%) and 28 on placebo (11.3%) (P=0.004). At baseline,87% in both groups were taking antihypertensive co-medication. At Month 6, the figures were 77.3% and65.9% in the placebo and candesartan groups, respectively(P = 0.006). At the final visit, they were 89.1% and 69.8%(P<0.001). BP changes over time for the groups areshown in Figure 2.

Laboratory end points

Results were expressed as medians and median relativechanges (%) from baseline to last visit. Serum creatinineincreased by 10.8% on candesartan and by 4.3% on place-bo (P<0.001). It doubled in seven patients on candesartan(2.78%) and in four on placebo (1.63%). Creatinine clear-ance decreased by 9.8% and 4.4% in the candesartan andplacebo groups, respectively (P=0.002). It decreased by≥25% in 74 patients (29.7%) on candesartan and 51(21.5%) on placebo (P=0.048) and by ≥50% in six pa-tients on candesartan and three on placebo. Creatinineclearance was also used to define all-cause graft failure,using the Cockcroft formula. As identified above, thiswas unequally distributed with six failures in the placebogroup versus one in the candesartan group.

Fig. 1. Disposition of patients in the study.

Candesartan in renal transplant recipients 969


All indices of proteinuria decreased on candesartan, andeither increased or were unchanged on placebo (Table 4,Figure 3). Nephrotic syndrome (proteinuria >3.5g/24h)occurred in two patients in each group. The median rela-tive change from baseline to last visit for urinary protein/creatinine ratio was −15.0% on candesartan and 23.4% onplacebo (P<0.001 between groups) and for serum potassi-um concentration was 5.1% on candesartan and −1.9% onplacebo (P < 0.001 between groups). The mean (±SD) ab-solute change in serum potassium was 0.27±0.51 (range,−1.9 to 2.0) mmol/L on candesartan and −0.13±0.50(range, −2.5 to 1.4) mmol/L on placebo. There were nopatient withdrawals as a result of hyperkalaemia. In 5.9%of patients, an increase in potassium was reported as anadverse event. The median relative change in blood hae-moglobin concentration was 8.2% on candesartan and 0%on placebo (P < 0.001 between groups). The mean (±SD)absolute change in blood haemoglobin was −1.08±1.42(range, −5.3 to 6.7) g/dL on candesartan and 0.01±1.25(range, −4.2 to 4.8) g/dL on placebo. There were no statis-tically significant changes in serum lipids.

Safety results

Adverse events were noted in 93.7% of patients in the can-desartan group and 89.8% in the placebo group. The mostfrequent on candesartan were increased creatinine (18.8%),nasopharyngitis (17.3%), urinary tract infection (12.2%),anaemia (12.2%) and ankle oedema (11.8%). On placebo,they were oedema of the eyelids (16.6%), nasopharyngitis(15.7%), urinary tract infection (14.5%), increased creati-nine (13.7%) and headache (10.9%). The outcome was fa-tal in three patients on candesartan and in four on placebo,including one case of sudden death in each group.

Discussion

SECRET is the first long-term, controlled clinical study ofthe effects of an antihypertensive drug on morbidity and

Table 1. Baseline characteristics of the randomized patientsa

Candesartan n=255 Placebo n=247

Male/female 164/91 154/93Age (years)b 50.0 (11.6) 49.7 (10.9)Range 21–70 21–71

Number Caucasian 242 (95.7%) 231 (93.9%)Weight (kg)b 76.1 (15.1) 77.1 (15.9)Range 38–143 38–126

Body mass index (kg/m2)b 26.4 (5.2) 26.7 (5.0)Time since transplant (years)c 4.03 (2.01–7.40) 3.77 (1.90–7.03)Number on drugs for diabetes 26 (10.3%) 29 (11.8%)Type 1 diabetes 4 3Type 2 diabetes 21 26Diabetes type not recorded 1 0

Number hypertensive 247 (96.9%) 233 (94.7%)Years of antihypertensive medicationd 11.0 10.0Number on long-term treatment with ARB/ACE-Ie 106 (41.7%) 101 (41.1%)Number with CVDe 39 (15.4%) 37 (15.1%)

a

Data are missing/not reported for some patients, and percentages are determined using denominators with recorded data.b

Mean (SD).c

Median (inter-quartile).d

Mediane

ARB, angiotensin receptor blocker; ACE-I, angiotensin converting enzyme inhibitor; CVD, cardiovascular disease.

Table 2. Primary efficacy variable and its three components

Candesartan n=255 Placebo n=247

Total no. of compositeprimary events

13 (5.1%) 13 (5.3%)

All-cause mortality 3 (1.18%) 4 (1.62%)Cardiovascularmorbidity events

9 (3.53%) 5 (2.02%)

All-cause graft failure 1 (0.39%) 6 (2.43%)

Results are number of patients (%) in the intention-to-treat population. Nogroup difference was statistically significant (P>0.05). Two patients in theplacebo group had two end points.

Table 3. Changes in BP from baseline to last visit

Candesartann=254

Placebon=245 P

Systolic BP (mmHg)At baseline 137.92 (17.0) 137.76 (16.6) NSAbsolute change tolast visit

−6.78 (19.50) −0.60 (20.40) 0.0009

Percent change tolast visit

−3.98 (14.14) 0.49 (14.98) 0.0007

Diastolic BP (mmHg)At baseline 84.38 (10.13) 84.64 (9.25) NSAbsolute change tolast visit

−4.81 (12.05) −1.21 (12.03) 0.0019

Percent change tolast visit

−4.64 (14.45) −0.52 (14.95) 0.0022

Results are presented as mean (SD). NS, P>0.05.



mortality in kidney transplant recipients. As post-trans-plant hypertension is associated with high rates of cardio-vascular disease and graft failure, effective management ofBP in these patients is critical [1–4]. As the risk of graftfailure is increased by proteinuria, it is generally agreedthat efforts to normalize BP should be augmented when itis present and that preference be given to antihypertensivedrugs that have direct renal protective activity [2,14,15].There is good evidence that ARBs and ACE inhibitors havesuch activity in chronic renal disease secondary to hyper-

tension, diabetes mellitus and glomerulonephritis [14–20]. This can be attributed to a reduction of effects of an-giotensin II on inflammation, tissue growth and fibrosis inthe diseased kidney [5,6]. Preliminary studies of losartan[21–23], valsartan [24] and candesartan [10] have sug-gested that the direct renal protective activity of ARBs op-erates also in post-transplant nephropathy.

It was partly on account of the known protective activityof candesartan in other chronic nephritides that SECRET in-cluded normotensive as well as hypertensive patients. The

Fig. 2. Mean changes in sitting systolic (a) and diastolic (b) BP from baseline according to treatment group.



BP target chosen for the study was consistent with that re-commended for other patients with chronic kidney disease[25].

The primary efficacy variable was a composite of threeevents: all-cause mortality, cardiovascular morbidity andgraft failure. Thus, the study did not aim to provide infor-mation on the effects of candesartan on specific patholog-ical processes or clinical diagnoses but to assess its effect

on a broad measure of outcomes of practical value to thelong-term management of transplant recipients.

SECRET was stopped prematurely by the independentSteering Committee on the recommendation of the inde-pendent Data Safety Monitoring Board (unanimous vote)after an interim analysis had shown that the number of pri-mary events was much smaller than expected and wouldnot cross the statistical boundary required to establishthe superiority of one treatment over the other. The numberof primary events was 26, equally divided between cande-sartan and placebo. None of the group comparisons for thedifferent components of all-cause mortality, cardiovascularmorbidity and all-cause graft failure was statistically sig-nificant. Nor did the two groups differ significantly in car-diovascular mortality.

It is likely that the lower than expected event rate can beexplained by the close monitoring and intensive BP con-trol that the patients received under clinical trial condi-tions. All study patients were recruited in specializedtransplantation outpatient clinics, where BP managementis an important part of aftercare. Both treatment groupstherefore started at baseline with a relatively low BP(mean 137/84mmHg), and this was true even in those pa-tients whose RAAS-blocking drug was discontinued be-fore randomization.

The better BP control achieved in candesartan-treated pa-tients at the end of the study (131/80 versus 137/84mmHg,P<0.001) was probably related to the study protocol thatmandated investigators to adjust antihypertensive therapyin order to achieve a target DBP of 85mmHg. In addition,normotensive patients were not excluded and also receivedactive treatment with candesartan. The low BP prior to in-tervention and the even lower BP achieved after interven-tion are unique among published controlled renal studies.

Table 4. Urinary albumin and total protein concentrations, and 24-hurinary protein excretion rates in the candesartan and placebo groups

Candesartan Placebo P

Albumin concentrationBaseline concentration (mg/L) 16.40 16.70 NSAbsolute change (mg/L) −1.80 1.05 0.0001Relative change (%) −22.2 25.3 0.0001

Protein concentrationBaseline concentration (g/L) 0.11 0.11 NSAbsolute change (g/L) −0.01 0.00 0.003Relative change (%) −16.7 0.00 0.0002

Protein excretion rateBaseline rate (g/24h) 0.12 0.14 NSAbsolute change (g/24h) −0.01 0.03 <0.0001Relative change (%) −28.6 15.4 <0.0001

Protein/creatinine ratioBaseline (g/μmol) 0.01 0.02 NSAbsolute change (g/μmol) – –Relative change (%) −15.0 23.5 0.0003

All values are medians. Changes are from baseline to the last value undertreatment.

Fig. 3. Median changes in urinary protein excretion from baseline according to treatment group.



It is beyond question, however, that the results—particularlythe effects on proteinuria—were influenced by the diffe-rent levels of BP achieved in the two treatment arms.

It is nonetheless an important result that target BP va-lues as recommended by JNC 7 [or even better byAmerican Society of Nephrology (ASN) guidelines]were only achieved for the majority by inhibition ofthe RAAS. This was shown by the impressive differencein the withdrawal rates for lack of efficacy (17 in theplacebo group versus two in the candesartan group).Similarly, dropouts for ‘withdrawal of consent’ by thepatient were higher in the placebo group than in the can-desartan group (18 versus 9). These differences could beexplained if therapies that did not include an ARB wereless effective and more prone to side effects.

BP in both treatment groups, at the start and the end of theSECRET study, were lower than those usually reported inclinical trials of renal transplant recipients. Indeed, theywere lower than those frequently reported for patients withany renal disease or hypertension. This might be related tothe intensive monitoring of BP that was mandated in theprotocol.

The fact that the present patients could have been giventheir transplant up to 10years before the study may alsohave contributed to the lower than anticipated event rate.

Notwithstanding its statistical failure with regard to theprimary objective, SECRET provided other information ofvalue for patient management. As would be expected, thepercentage of patients who were given an antihypertensivedrug on top of the study drug was greater in the placebogroup than in the candesartan group. Of significance, how-ever, was the fact that BP was better controlled in thosetaking candesartan, despite the fact that efforts to achieveoptimum control were applied equally in both groups, us-ing the same types of drugs in the same sequence.

As significant as the improvement in BP control was thereduction in proteinuria produced by candesartan. Urinaryalbumin concentration, total protein concentration, 24-h to-tal protein excretion and protein/creatinine ratiowere all sig-nificantly reduced. While excretion rate decreased by28.6% from baseline to last visit on candesartan, it increasedby 15.4% in patients who received placebo. Thecorresponding values for protein/creatinine ratio were−15.0% and +23.5%. These findings support the suggestionprovided by retrospective analyses [26,27] and preliminaryclinical studies [21–24] that ARBs reduce proteinuria in re-nal transplant patients, as they do in other chronic nephriti-des [14–20].

It is now well recognized that angiotensin II has directeffects on inflammation, tissue growth and fibrosis thatwould be expected to promote progression of chronic kid-ney damage [5,6]. Experimental work has shown that this isat least partly a consequence of increased expression of sev-eral genes that participate in these processes. Szabo et al.[28] used the ACE inhibitor, enalapril, to inhibit angiotensinII production in rats with chronically rejecting kidney allo-grafts. Treated animals had less proteinuria, fewer histolog-ical signs of rejection including leukocyte infiltration andlower mRNA levels in kidney tissue for transforminggrowth factor-ß, platelet-derived growth factor, insulin-likegrowth factor, interleukin-1 and monocyte chemo-attractant

protein-1. More recently, Ma et al. [29] reported that ACEinhibition and AT1-receptor blockade induced regression ofexperimental glomerulosclerosis in rats and that this waslinked to decreased levels of plasminogen activator inhibi-tor-1 and tissue inhibitor of metalloproteinase-1, suggestingthat matrix modulation also plays an important role.

Figures 2 and 3 suggest, at first sight, that the group dif-ferences in BP and urinary protein excretion decrease withincreasing length of follow-up. These figures are very dif-ficult to interpret, however, because of the prematurestudy closure, the withdrawal of patients with uncon-trolled BP and the differential use of additional antihyper-tensive medication. The mean data at prolonged follow-uptherefore refer to very low and unbalanced numbers of pa-tients (approximately 10% of the initial population remainat 36months) that may be skewed compared with meanbaseline data for the 100% of patients initially enrolled.Strictly, the only relevant data are changes from inclusionto last visit for paired data.

The improvements in BP control and proteinuriaachieved with candesartan are clinically significant. Bothare risk factors for graft failure, and high BP is also a riskfactor for cardiovascular disease in renal transplant recipi-ents [1–3,15,16]. On the basis of our results, candesartanwould be expected to reduce cardiovascular morbidity, car-diovascular mortality and graft failure during the long-termmanagement of these patients. Indeed, while only one caseof graft failure occurred in the candesartan group, thereweresix in the placebo group. In this context, it is also of interestthat in a retrospective analysis of 1662 patients, Prema-sathian et al. [27] concluded that ACE inhibitors and ARBsare associated with greater rates of graft survival in patientswith elevated systolic BP and proteinuria than are otherantihypertensive drugs. By contrast, reduction of plasmalow-density lipoprotein concentration in kidney transplantrecipients, a risk factor for cardiovascular morbidity andmortality in this population, by treatment with fluvastatinhad no significant effects on the rate of graft failure [30].

It should be noted that the lower incidence of graft failurein the candesartan group versus placebo (one versus six) inour study was offset by an unexpectedly higher incidence ofcardiovascular morbidity events (nine versus five), thus ac-counting for the similar incidence of composite primaryevents in each group. Cardiovascular morbidity includedvarious events, such as acute coronary syndrome, myocar-dial infarction, angina, coronary revascularization andstroke. This numerical increase in cardiovascular morbidityin the candesartan group was, however, far from being sta-tistically significant and can be considered representative ofrandom fluctuation of very small numbers of events.

Recently published retrospective studies of renal trans-plant patients have shown conflicting results with respectto effects on RAAS blockade with ACE inhibitors andARBs. Heinze et al. [31] showed an improvement of ap-proximately 20% and 30% for patient and graft survival,respectively, at 5years in 2031 patients receiving ACE in-hibitors or ARBs. However, Opelz et al. [32] found no im-provement of graft or patient survival in those receivingACE inhibitors or ARBs in a retrospective study in 17209 renal transplant patients. There appears to be no clearreason to account for the sharply conflicting difference in



the results between these two retrospective studies, otherthan the fact that they were, by design, strongly subjectedto confusion bias [33].

Other differences between the candesartan and placebogroups accorded with the known effects of ARBs in chronicrenal disease. Increases in serum creatinine of 20–30% arecommon, owing to reductions of glomerular capillary pres-sure and filtration rate. The increase in serum potassium anddecrease in haemoglobin concentrations that occurred weresmall. Thus, candesartan has a good safety profile in renaltransplant recipients.

Appendix

Principal Investigators and Clinical Centres

France Prof Albert Fournier, Service de Néphrologie, Hô-pital Sud, Amiens; Prof Françoise Mignon, Service de Né-phrologie, Hôpital Bichat/Claude Bernard, Paris; Dr MichelDelahousse, Service de Néphrologie, Centre Medico-Chir-urgical Foch, Suresnes; Prof Paul Vialtel, Service de Né-phrologie, CHU-Hôpital Michallon, Grenoble; ProfMaryvonne Hourmant, Service de Néphrologie, CHU-Hôpital Hotel-Dieu, Nantes; Prof Georges Mourad, Servicede Néphrologie, CHU-Hôpital Lapeyronie, Montpellier;Prof Michel Godin, Service de Néphrologie, HôpitalBois-Guillaume, Rouen; Prof Bernard Bourbigot, Trans-plantation Rénale, Hôpital de la Cavale Blanche, Brest;Prof Claire Pouteil Noble, Service Adjoint Néphrologie,Hôpital Lyon Sud, Pierre-Benite; Prof Dominique Durand,Service de Néphrologie-Transplantation, CHU-Hôpital deRangueil, Toulouse; Prof Patrick le Pogamp, Service deNéphrologie, CHRU-Hôpital Pontchaillou, Rennes; ProfBernard Charpentier, Service de Néphrologie-Hémodia-lyse, Hôpital de Bicetre, Le Kremlin-Bicentre; Dr ElisabethCassuto-Viguier, Service de Néphrologie, CHU-HôpitalPasteur, Nice; Prof Jean-Marc Chalopin, Service de Né-phrologie, CHU-Hôpital Saint-Jacques, Besancon; ProfYvon Lebranchu, Transplantation-Immunologie Clinique,Hôpital Bretonneau, Tours; Dr Jean-François Subra, Ser-vice de Néphrologie, CHU-Hôpital D'Angers, Angers; DrChristian Noel, Service de Néphrologie, Hôpital Calmette,Lille; Prof Michèle Kessler, Service de Néphrologie,CHRU-Hôpital Brabois Adultes, Vandoeuvre-lès-Nancy;Prof Eric Rondeau, Service de Néphrologie A, Hôpital Ten-on, Paris; Prof Christophe Legendre, Service de Néphrolo-gie, Hôpital Saint-Louis, Paris; Prof Bruno Moulin, Servicede Néphrologie-Hémodialyse, Hôpital Civil, Strasbourg;Dr Olivier Toupance, Service de Néphrologie, CHU-Hôpi-tal Maison Blanche, Reims; Prof Denis Glotz, Service deNéphrologie-Immunologie, Hôpital Europeen G Pompi-dou, Paris; Prof Henri Kreiss, Service Transplantation Ré-animation, Hôpital Necker, Paris; Prof François Berthoux,Service de Néphrologie, Hôpital Nord, Saint Etienne. Ger-many Prof Dr med Thomas Philipp, UniversitätsklinikumEssen, Abteilung für Nieren-und-Hochdruckkrankheiten,Essen; Prof Dr med Helmut Geiger, Klinikum der Johann-Wolfgang-Goethe-Universität, Medizinische Klinik IV, Ne-phrologie, Frankfurt; Prof Dr med L Christian Rump, Uni-versitätsklinik Freiburg, Innere Medizin IV, Nephrologie,

Freiburg; Prof Dr med Teut Risler, Universitätsklinikum Tü-bingen, Medizinische Klinik-und-Poliklinik, Innere Medi-zin III, Tübingen; Prof Dr med Christoph Wanner,Klinikum der Bayerischen Julius-Maximilians-Universität,Würzburg; Prof Dr med Roland Schmieder, Universitätsk-linikum Erlangen-Nürnberg, Medizinische Klinik IV/Ne-phrologie, Nürnberg; Prof Dr med Günter Stein, Klinikumder Friedrich-Schiller-Universität, Klinik für Innere Medi-zin, Jena; Prof Dr med Hans-Helmut Neumayer, Medizi-nische Klinik Charité der Humboldt Universität, Berlin;Prof Dr. med Hermann Haller, Medizinische HochschuleHannover, Zentrum Innere Medizin, Hannover; Prof Drmed Bernhard Krämer, Klinikum der Universität Regens-burg II, Medizinische Klinik und Poliklinik, Regensburg;Prof Dr medManfredWeber, Kliniken der Stadt Köln,Med-izinische Klinik I, Köln; Prof Dr med Jürgen Floege, Uni-versitätsklinikum Aachen, Medizinische Klinik II derRWTH Aachen, Aachen. Austria Prof Dr med Erich Pohan-ka, Allgemeines Krankenhaus der Stadt Wien, Universi-tätsklinik für Innere Medizin III, Nephrologie undDialyse, Vienna. Czech Republic Dr. Stefan Vitko, IKEM,Dept of Nephrology, Prague.

Steering committee

Chairmen

Prof Dr T Philipp, Nephrologist, Essen, GermanyProf Dr C Legendre, Nephrologist, Paris, France

Investigators

Prof Dr H Geiger, Nephrologist, Frankfurt, GermanyProf Dr B Moulin, Nephrologist, Strasbourg, FranceProf Dr G Mourad, Nephrologist, Montpellier, FranceProf Dr R Schmieder, Nephrologist, Erlangen/Nürnberg,Germany

Sponsor

Dr D Girault (Laboratoires Takeda, Puteaux, France)Dr R Hübner (Takeda Pharma, Aachen, Germany)

End point committee

Central adjudication placed in LAUSANNE (Switzerland)

Prof Hans R Brunner (Lausanne) ChairmanProf M Burnier (Nephrologist, Lausanne)PD Dr G. Turini (Cardiologist, Lausanne)Dr G Devuyst (Neurologist, Lausanne)PD Dr D Hayoz (Angiologist, Lausanne)

Native speaking members(as appointed by the national steering boards)

Prof. Dr. Heemann, Nephrologist, MunichPD Dr Martinez, Nephrologist, Paris, France



Data management and safety board

Members

Prof E Ritz (Nephrologist, Heidelberg) (Chairman)Prof G Mayer (Nephrologist, Innsbruck)Prof B Moulin (Nephrologist, Strasbourg)Prof X Girerd (Nephrologist, Paris)Prof KH Jöckel, (Statistician Essen)

Consultants

Prof CG Brilla (Cardiologist, Marburg)Prof PG Steg (Cardiologist, Paris)

Independent statistical centre

Prof M Lièvre, Lyon (Chairman)

Acknowledgements. Data management was performed by ECRON,Frankfurt and SAM GmbH, Aachen, Germany.

Conflict of interest statement. This study was supported by LaboratoiresTakeda, France, and Takeda Pharma GmbH, Germany. H Geiger was aconsultant of Takeda Pharma concerning this study, M Lièvre's institutionreceived a grant for performing the interim and final statistical analysesand R Schmieder received grants from the sponsor. U Heemann, C Le-gendre, F Martinez, B Moulin, G Mourad and T Philipp have no conflictof interest to declare. The results presented in this paper have not beenpublished previously in whole or part, except in abstract format.

References

1. Weir MR. Blood pressure management in the kidney transplant recip-ient. Advan Chronic Kidney Dis 2004; 11: 172–183

2. Opelz G, Wujciak T, Ritz E. Association of chronic kidney graft fail-ure with recipient blood pressure. Collaborative transplant study. Kid-ney Int 1998; 53: 217–222

3. Ojo AO. Cardiovascular complications after renal transplantation andtheir prevention. Transplantation 2006; 82: 603–611

4. Sarnak MJ, Levey AS, Schoolwerth AC et al. Kidney disease as arisk factor for development of cardiovascular disease. A statementfrom the American Heart Association Councils on Kidney in Car-diovascular Disease, High Blood Pressure Research, Clinical Cardi-ology, and Epidemiology and Prevention. Circulation 2003; 108:2154–2169

5. Remuzzi GG, Perico N, Macia N et al. The role of the renin-angio-tensin-aldosterone system in the progression of chronic kidney dis-ease. Kidney Int Suppl 2005; S57–S65

6. Fogo AB. Mechanisms in nephrosclerosis and hypertension beyondhemodynamics. J Nephrol 2001; 14(Suppl 4): 563–569

7. Lacoucière Y, Asmar R. A comparison of the efficacy and duration ofaction of candesartan cilexetil and losartan, as assessed by clinic andambulatory blood pressure after a missed dose, in truly hypertensivepatients: a placebo-controlled forced titration study. Am J Hypertens1999; 12: 1181–1187

8. Fridman KU, Elmfeldt D, Wysocki M et al. Influence of AT1 receptorblockade on blood pressure, renal haemodynamics and hormonal re-sponses to intravenous angiotensin II infusion in hypertensive pa-tients. Blood Pressure 2002; 11: 244–252

9. Gradman AH, Lewin A, Bowling BT et al. Candesartan versus Lo-sartan Efficacy Comparison (CANDLE) Study Group. Comparativeeffects of candesartan cilexetil and losartan in patients with systemichypertension. Heart Disease 1999; 1: 52–57

10. Omoto K, Tanabe K, Tokumoto T et al. Use of candesartan cilexetildecreases proteinuria in renal transplant patients with chronic allo-graft dysfunction. Transplantation 2003; 76: 1170–1174

11. Hiremath S, Fergusson D, Doucette S et al. Renin angiotensin systemblockade in kidney transplantation: a systematic review of the evi-dence. Amer J Transplant 2007; 7: 2350–2360

12. K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease:Evaluation, classification, and stratification. Part 4. Definition andclassification of stages of chronic kidney disease. Am J Kid Dis2002; 39 (Suppl): S46–S75

13. Cockcroft DW, Gault MH. Prediction of creatinine clearance from se-rum creatinine. Nephron 1976; 16: 31–41

14. Maschio G, Alberti D, Janin G et al. Effect of the angiotensin-con-verting-enzyme inhibitor benazepril on the progression of chronic re-nal insufficiency. N Engl J Med 1996; 334: 939–945

15. Jafar TH, Stark PC, Schmid CH et al. Progression of chronic kidneydisease: the role of blood pressure control, proteinuria, and angioten-sin-converting enzyme inhibition. A patient-level meta-analysis. AnnIntern Med 2003; 139: 244–245

16. Mogensen CE, Neldam S, Tikkaanen I et al. Randomised control trialof dual blockade of renin-angiotensin system in patients with hyper-tension, microalbuminuria, and non-insulin dependent diabetes: thecandesartan and lisinopril microalbuminuria (CALM) study. BMJ2000; 321: 1440–1444

17. The GISEN group. Randomised controlled trial of the effect of rami-pril on decline in glomerular filtration rate and risk of terminal renalfailure in proteinuric non-diabetic nephropathy. Lancet 1997; 349:1857–1863

18. Lewis EJ, Hunsicker LG, Bain RP et al. The effect of angiotensin-converting-enzyme inhibition on diabetic nephropathy. N Engl JMed 1993; 329: 1456–1462

19. Brenner BM, Cooper MA, de Zeeuw D et al. Effects of losartan onrenal and cardiovascular outcomes in patients with type 2 diabetesand nephropathy. N Engl J Med 2001; 345: 861–869

20. Lewis EJ, Hunsicker LG, Clarke WR et al. Renoprotective effect ofthe angiotensin-receptor antagonist irbesartan in patients with ne-phropathy due to type 2 diabetes. N Engl J Med 2001; 345: 851–860

21. Calvino J, Lens XM, Romero R et al. Long-term antiproteinuric ef-fect of losartan in renal transplant recipients treated for hypertension.Nephrol Dial Transplant 2000; 15: 82–86

22. Del Castillo D, Campistol JM, Guirado L et al. Efficacy and safety oflosartan in the treatment of hypertension in renal transplant recipi-ents. Kidney Int Suppl 1998; 68: S135–S139

23. Inigo P, Campistol JP, Sarocho R et al. Renoprotective effects of lo-sartan in renal transplant recipients. Results of a retrospective study.Nephron Clin Pract 2003; 95: c84–c90

24. Muirhead N, House A, Hollomby DJ et al. Effect of valsartan on uri-nary protein excretion and renal function in patients with chronic re-nal allograft nephropathy. Transplant Proc 2003; 35: 2412–2414

25. Chobanian AV, Bakris GL, Black HR et al. Seventh Report of theJoint National Committee on Prevention, Detection, Evaluation andTreatment of High Blood Pressure, National Heart, Lung, and BloodInstitute. JAMA 2003; 289: 2560–2572

26. Holgado R, Anaya F, Del Castillo D. Angiotensin II type 1 (AT 1)receptor antagonists in the treatment of hypertension after renal trans-plantation. Nephrol Dial Transplant 2001; 16(Suppl 1): 117–120

27. Premasathian NC, Muehrer R, Brazy PC et al. Blood pressure controlin kidney transplantation: therapeutic implications. J Hum Hypertens2004; 18: 871–877

28. Szabo A, Lutz J, Schleimer K et al. Effect of angiotensin-convertingenzyme inhibition on growth factor mRNA in chronic renal allograftrejection in the rat. Kidney Int 2000; 57: 982–991

29. Ma LJ, Nakamura S, Aldigier JC et al. Regression of glomerulo-sclerosis with high dose angiotensin inhibition is linked to de-creased plasminogen activator inhibitor-1. J Am Soc Nephrol2005; 16: 840–842

30. Holdaas H, Fellström B, Jardine AG et al. Effect of fluvastatinon cardiac outcomes in renal transplant recipients: a multicentre,randomised, placebo-controlled trial. Lancet 2003; 361: 2024–2031



31. Heinze G, Mitterbauer C, Regele H et al. Angiotensin-converting en-zyme inhibitor or angiotensin II type I receptor antagonist therapy isassociated with prolonged patient and graft survival after renal trans-plantation. J Am Soc Nephrol 2006; 17: 889–899

32. Opelz G, Zeier M, Laux G et al. No improvement of patient or graftsurvival in transplant recipients treated with angiotensin-convertingenzyme inhibitors or angiotensin II type 1 receptor blockers: a Collab-

orative Transplant Study Report. J Am Soc Nephrol 2006; 17: 3257–3262

33. Smith GD, Phillips AN. Confounding in epidemiological studies:why “independent” effects may not be all they seem. BMJ 1992;305: 757–759

Received for publication: 5.2.09; Accepted in revised form: 9.10.09

Nephrol Dial Transplant (2010) 25: 976–984doi: 10.1093/ndt/gfp567Advance Access publication 28 October 2009

Effects of pioglitazone on subclinical atherosclerosis and insulinresistance in nondiabetic renal allograft recipients

Seung Jin Han1,2,*, Kyu Yeon Hur3,*, Yu Seun Kim4, Eun Seok Kang5,6, Soon Il Kim4,Myoung Soo Kim4, Jung Young Kwak7, Dae Jung Kim1, Sung Hee Choi8,Bong Soo Cha5,6 and Hyun Chul Lee2,5,6

1Department of Endocrinology and Metabolism, Ajou University School of Medicine, Suwon 443-721, Korea, 2Department ofMedicine, Graduate School, Yonsei University, Seoul, Korea, 3Department of Medicine, Sungkyunkwan University School ofMedicine, Seoul 135-710, Korea, 4Department of Surgery, Yonsei University College of Medicine, Seoul 120-752, Korea,5Department of Internal Medicine, Yonsei University College of Medicine, Seoul 120-752, Korea, 6Brain Korea 21 for MedicalScience, Yonsei University College of Medicine, Seoul 120-752, Korea, 7Severance Hospital Diabetes Center, Yonsei UniversityMedical Center, Seoul 120-752, Korea, and 8Department of Internal Medicine, Bundang Hospital, Seoul National University Collegeof Medicine, Seongnam, Korea

Correspondence and offprint requests to: Hyun Chul Lee; E-mail: [email protected]*Seung Jin Han and Kyu Yeon Hur contributed equally to this work

AbstractBackground. The aim of this study was to evaluate theeffect of pioglitazone treatment on the progression ofsubclinical atherosclerosis and insulin resistance in re-nal allograft recipients with no preoperative history ofdiabetes.Methods. Eighty-three patients without diabetes wererandomly assigned to either the pioglitazone group orthe control group. Carotid intima–media thickness(IMT), serum adiponectin level and lipid profile wereassessed before transplantation and at 12 months aftertransplantation. Insulin secretory function and insulin re-sistance were evaluated by the oral glucose tolerancetest.Results. The pioglitazone group showed a significant re-duction in the mean and maximum carotid IMT com-pared with the control group after 12 months (meancarotid IMT, 0.05 ± 0.04 vs −0.03 ± 0.07mm, P <0.001; maximum carotid IMT, 0.08 ± 0.05 vs −0.05 ±0.09mm, P < 0.001). Pioglitazone increased the adipo-nectin level, and the change in adiponectin was negative-ly correlated with carotid IMT changes. Pioglitazonetreatment increased the insulin sensitivity index compared

with the control group (−0.8 ± 3.1×10−2 vs +1.1 ±3.7×10−2, P = 0.036).Conclusions. These results suggest that pioglitazone treat-ment reduces the progression of carotid IMT and improvesinsulin resistance in renal allograft recipients without a his-tory of diabetes.

Trial Registration. Clinicaltrials.gov Identifier:NCT00598013

Keywords: carotid intima–media thickness; insulin resistance;pioglitazone; renal allograft

Introduction

Cardiovascular disease (CVD) is a common cause of mor-bidity and mortality after kidney transplantation [1]. Theannual risk of CVD is 3.5–5% in renal transplant recipi-ents, 50-fold higher than in the general population [2,3].The measurement of carotid intima–media thickness(IMT) by high-resolution B-mode ultrasonography is an

© The Author 2009. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.For Permissions, please e-mail: [email protected]

976 S.J. Han et al.


candesartan improves blood pressure control and reduces

Documents