Study Guides, Notes, & Quizzes

First United Kingdom Heart and Renal Protection (UK-HARP-I) study: Biochemical efficacy and safety of simvastatin and safety of low-dose aspirin in chronic kidney disease

First United Kingdom Heart and Renal Protection (UK-HARP-I) study: Biochemical efficacy and safety of simvastatin and safety of low-dose aspirin in chronic kidney disease
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  First United Kingdom Heart and Renal Protection (UK-HARP-I)Study: Biochemical Efficacy and Safety of Simvastatin and Safetyof Low-DoseAspirin in Chronic Kidney Disease Colin Baigent, BMBCh, Martin Landray, MD, PhD, Craig Leaper, PhD, Paul Altmann, MD, FRCP,Jane Armitage, MBBS, MFPH, Alex Baxter, MA, Hugh S. Cairns, MD, Rory Collins, FMedSci,Robert N. Foley, MB, MSc, Valeria Frighi, MD, Karen Kourellias, Peter J. Ratcliffe, FRS,Mary Rogerson, MD, FRCP, John E. Scoble, MD, FRCP, Charles R.V. Tomson, DM,Graham Warwick, MD, FRCP, and David C. Wheeler, MD, FRCP ●  Background:   Patients with chronic kidney disease are at increased risk for cardiovascular disease, but the efficacy andsafety of simvastatin and aspirin are unknown in this patient group.  Methods:   Patients were randomly assigned in a 2  2factorial design to the administration of: (1) 20 mg of simvastatin daily versus matching placebo, and (2) 100 mg ofmodified-release aspirin daily versus matching placebo.  Results:   Overall, 448 patients with chronic kidney disease wererandomly assigned (242 predialysis patients with a creatinine level  >  1.7 mg/dL [ > 150   mol/L], 73 patients on dialysistherapy,and133patientswithafunctioningtransplant).Compliancewithstudytreatmentswas80%at12months.Allocationto treatment with 100 mg of aspirin daily was not associated with an excess of major bleeds (aspirin, 4 of 225 patients [2%]versusplacebo,6of223patients[3%]; P   notsignificant[NS]),althoughtherewasa3-foldexcessofminorbleeds(34of225[15%]versus12of223patients[5%]; P   0.001).Among those with predialysis renal failure or a functioning transplant atbaseline, aspirin did not increase the number of patients who progressed to dialysis therapy (7 of 187 [4%] versus 6 of188patients[3%]; P   NS)orexperiencedagreaterthan20%increaseincreatininelevel(63of187patients[34%]versus56 of 188 patients [30%];  P     NS). After 12 months of follow-up, allocation to 20 mg of simvastatin daily reducednonfasting total cholesterol levels by 18% (simvastatin, 163 mg/dL [4.22 mmol/L] versus placebo, 196 mg/dL [5.08mmol/L];  P   <  0.0001), directly measured low-density lipoprotein cholesterol levels by 24% (89 mg/dL [2.31 mmol/L]versus114mg/dL[2.96mmol/L]; P  < 0.0001),andtriglyceridelevelsby13%(166mg/dL[1.87mmol/L]versus186mg/dL[2.10mmol/L]; P  < 0.01),buttherewasnosignificanteffectonhigh-densitylipoproteincholesterollevels(2%increase; P   NS). Allocation to simvastatin therapy was not associated with excess risk for abnormal liver function test results orelevated creatine kinase levels.  Conclusion:   During a 1-year treatment period, simvastatin, 20 mg/d, produced asustained reduction of approximately one quarter in low-density lipoprotein cholesterol levels, with no evidence oftoxicity,andaspirin,100mg/d,didnotsubstantiallyincreasetheriskforamajorbleedingepisode.Muchlargertrialsarenowneededtoassesswhetherthesetreatmentscanpreventvascularevents. AmJKidneyDis  45:473-484. © 2005bytheNationalKidneyFoundation,Inc. INDEX WORDS: Statin; simvastatin; aspirin; cardiovascular disease; chronic kidney disease (CKD). Editorial, p. 607 I N PATIENTS WITH preexisting coronaryheart disease, large-scale randomized trialshave shown that decreasing low-density lipopro-tein (LDL) cholesterol levels with 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors(statins) reduces coronary mortality and morbid-ity in some types of high-risk patients. 1-4 Typi-cally, in those trials, an average reduction of approximately 40 mg/dL (1 mmol/L) maintainedfor approximately 5 years reduced the risk for acoronary event by approximately 25% to 30% 1 and the risk for ischemic stroke by a similaramount. 5,6 Patients with established chronic kid-ney disease (CKD) are at high risk for cardiovas-cular disease; therefore, the benefits of choles-terol-lowering therapy might be expected to besubstantial in this population. However, patients From the Clinical Trial Service Unit, Oxford University;Churchill Hospital, Oxford; Kings College Hospital; Guy’s Hospital; Royal Free Hospital, London; Hope Hospital,Salford; Southampton General Hospital, Southampton; and Southmead Hospital, Bristol, UK. Received July 29, 2004; accepted in revised form Novem-ber 9, 2004.Originally published online as doi:10.1053/j.ajkd.2004.11.015onJanuary25,2005.The HARP pilot studies and the Study of Heart and RenalProtection (SHARP) were funded by unrestricted grants from Merck&Co.C.B.,M.L.,J.A.,andR.C.eachreceivedreimburse-mentfromMerckfortravelexpensesinconnectionwithvariousspeaking engagements, but as members of the Clinical TrialService Unit, comply with the Unit’s policy of not acceptinghonoraria.SeeAppendixforUK-HARPPilotStudyInvestigators. Address reprint requests to Colin Baigent, BMBCh, Clini-cal Trial Service Unit, Radcliffe Infirmary, Oxford OX26HE, UK. E-mail:  ©  2005 by the National Kidney Foundation, Inc.0272-6386/05/4503-0004$30.00/0doi:10.1053/j.ajkd.2004.11.015 American Journal of Kidney Diseases,  Vol 45, No 3 (March), 2005: pp 473-484 473  with CKD have generally been excluded fromprevious trials, and there is currently no reliablerandomized evidence that decreasing cholesterollevels would be beneficial in these patients. 7,8 There are several reasons why the benefits of decreasing blood LDL cholesterol levels shownin other populations might not translate to pa-tients with CKD. First, observational studies of dialysispatientshavereportedanegativeassocia-tion between blood total cholesterol level andmortality. 9,10 Second, only approximately onequarter of cardiac mortality in such patientsappears to be definitely attributable to acutemyocardialinfarction 11 (generallycausedbyath-erosclerosis and thus potentially avoidable withcholesterol-lowering therapy), whereas the othercommon causes (eg, cardiac arrest, arrhythmia,and heart failure) are more likely attributable tocardiomyopathy and may not be as dependent oncholesterollevels. 12 Finally,althoughtheAssess-ment of Lescol in Renal Transplantation study 13 suggested that fluvastatin, 80 mg/d, was safe inrenal transplant recipients treated for 5 to 6years, most previous statin trials of predialysis ordialysis patients have been small and involvedonly a few months of treatment, 8 and there is aneed to establish the safety of more prolongedcholesterol-lowering therapy in such popula-tions.In patients at high risk for occlusive vasculardisease, a meta-analysis of randomized trialsshowed that a prolonged course of antiplatelettherapy reduced the risk for a serious vascularevent (myocardial infarction, stroke, or vasculardeath) by approximately one quarter. 14 Atotal of 2,632 hemodialysis patients were included in 14trials of antiplatelet therapy for the prevention of vascular access occlusion, and in these patients,antiplatelettherapywasassociatedwithastatisti-cally significant 41% reduction in risk for aserious vascular event. Because CKD is a causeof impaired hemostasis, 15,16 possible benefits of antiplatelet therapy might be offset by a highabsolute risk for bleeding. However, only 46major bleeds occurred in the hemodialysis pa-tients in this meta-analysis of randomized trials;therefore, it could not provide reliable informa-tion about a possible excess risk for major bleed-ing in association with antiplatelet therapy. 14 Therefore, there is a need for reliable directevidenceofwhethercholesterol-loweringtherapyor treatment with low-dose aspirin might be ableto prevent a worthwhile proportion of vascularevents (myocardial infarction, cardiac death,stroke, or revascularization procedures) withoutundue toxicity in patients with CKD. Becausesuch evidence could be provided by only a large-scale megatrial, we designed a randomized pilotstudy in preparation for such a trial. The princi-pal objectives of this pilot study (the first UKHeart and Renal Protection study [UK-HARP-I])are to: (1) assess the biochemical efficacy (interms of changes in lipid profile) and safety (interms of hepatic or muscle toxicity) of 1 year of 20 mg of simvastatin daily, (2) assess the safetyof 1 year of 100 mg of modified-release aspirindaily, and (3) establish the feasibility of a large-scale randomized trial with sufficient power todetermine effects on vascular events. METHODS  Eligibility Men or women 18 years or older were eligible if: (1) theywereapredialysispatientwiththemostrecentserumorplasmacreatinine level of 1.7 mg/dL or greater (  150   mol/L), ahemodialysisorperitonealdialysispatient,orhadafunctioningrenal transplant (with any creatinine level); and (2) their ownnephrologist and primary care physician did not consider therewasadefiniteindicationfor(orcontraindicationto)cholesterol-lowering therapy or aspirin. There was no upper limit to bloodcholesterol levels, but patients were not to be randomized if their own doctor considered that cholesterol-lowering therapyshould be prescribed. Patients also were to be excluded if therewas evidence of a recent history of acute uremia, history of chronicliverdisease,inflammatorymuscledisease(ie,dermato-myositis or polymyositis) or creatine kinase [CK] level greaterthan3timestheupperlimitofnormal(ULN),previousadversereaction to a statin or history of aspirin hypersensitivity (eg,aspirin-induced asthma or angioedema), concurrent treatmentwith a contraindicated drug (ie, nonstudy statin, fibrate, niacin,macrolide antibiotic; systemic azole antifungal, nefazodone, ororal anticoagulant therapy), high immediate risk for bleeding(eg, active peptic ulceration, recent injury, or hemophilia),child-bearing potential in the absence of a reliable method of contraception, a life-threatening condition other than CKD orvascular disease (eg, nonskin cancer or acquired immunodefi-ciency syndrome), frequent nonattendance at clinics or knownnoncompliance with drug treatments, or alcohol or substanceabuse.  Recruitment  Participating consultant nephrologists were asked to giveconsent for their hospital database to be screened for poten-tially eligible patients under their care. Selected patientswere sent a patient information leaflet and an invitation toattend a screening clinic. Patients attending a screening BAIGENT ET AL474  clinic were seen by a research nurse and asked about theirrenal and other medical history (including any history of vascular disease or diabetes mellitus), smoking habit, andalcohol consumption. Height and weight were measured forcalculation of body mass index. Seated blood pressure wasmeasured after 5 minutes of rest using an A&D UA-767Digital Blood Pressure Monitor (Tokyo, Japan).Individuals who were eligible and willing to considerparticipating were provided with a full explanation of thestudy, and written informed consent was obtained. Patientswho wished to have more time to consider participation,perhaps to allow discussion with their family or their owndoctors, were asked to return to a future clinic.Anonfastingblood sample was obtained for assessment of plasma lipids(total cholesterol, high-density lipoprotein [HDL] choles-terol,andtriglycerides),creatinine,CK,andalaninetransami-nase (ALT).Patients who agreed to take part were given a 1-month“run-in pack” of single-blind placebo medication (1 bottlecontaining placebo simvastatin tablets, and a second bottlecontaining placebo aspirin tablets) and instructed to take 1tablet from each bottle every evening. One key purpose of this run-in period is to help assess, before randomization, thelevel of compliance, because patients who took less than90% of the run-in medication or who dropped out for anyreason were not randomized. In addition, the run-in periodallowed time for the patient’s own nephrologist and primarycare physician to be informed of the patient’s clinical andbiochemical status (including lipid profile) and to considerwhether there was a definite indication for (or contraindica-tion to) statin or aspirin therapy, in which case the patientwas not randomized. In particular, this procedure ensuredthat no patient with a lipid profile considered to warranttreatment with a statin was randomized.At the end of the run-in period, patients who remainedeligible and willing to continue and had no abnormal livertransaminase or CK results were randomly allocated toadministration of 20 mg of simvastatin daily or matchingplacebo and, in a 2    2 factorial design, 100 mg of modified-releaseaspirindailyormatchingplacebo.Random-ization was by telephone to the Clinical Trial Service Unit,University of Oxford, UK. Minimized randomization 17 wasused to balance the treatment groups with respect to eligibil-ity criteria and other major prognostic factors. Follow-Up Patients were to be seen in study clinics for routinefollow-up checkups and blood safety monitoring at 1, 3, 6,and 9 months and 1 year. At each follow-up visit, patientswere asked about muscle pain, muscle weakness, or seriousadverse experiences occurring since the last visit. Regularnonstudy treatment was recorded, and, at the 6- and 12-month follow-up visits, compliance with the study treat-ments was assessed by means of tablet count (with patientsclassified as compliant if they had taken  80% of scheduledtablets). A decision about whether to continue with studytreatments was made, and any reason(s) for stopping wererecorded. At the 6-month visit, patients willing to continuereceived an additional treatment pack containing 6 months’supply of study drugs (with composition identical to therandomizationpack).Patientsstartingregulardialysistherapywere to continue taking the study treatments unless advisedotherwise by their doctor. In patients who received a renaltransplant after randomization, the study simvastatin ormatching placebo simvastatin was to be taken on alternatedays (with aspirin or matching placebo continued daily).A nonfasting blood sample was obtained at the 3- and12-month clinic visits for central laboratory assay of lipidprofile (total cholesterol, HDL cholesterol, directly measuredLDLcholesterol, 18,19 apolipoproteinA 1 , apolipoprotein B, andtriglycerides),ALT level to monitor liver function, CK level tomonitor possible muscle toxicity, creatinine, albumin, urate,andfullbloodcount.Allbloodsampleswerestoredintheclinicrefrigerator and transferred by refrigerated transport to thecoordinating center laboratory (Clinical Trial Service Unit,UniversityofOxford)forseparationandanalysiswithin2daysof venesection and subsequent long-term storage at  40°C foranysubsequentassaysthatmightberequired.Patients who became unable or unwilling to attend theclinics were to be contacted by telephone at the time of theirscheduled follow-up, but their allocated study simvastatin orplacebo tablets were to be stopped (because blood safetymonitoring could not be continued).At each follow-up visit,information was recorded on reports of all possible seriousadverseevents,includingsuspectedvascularevents(myocar-dialinfarction,stroke,hospitalizationforangina,hospitaliza-tionforfluidoverloadorcardiacfailure,coronaryrevascular-ization or other arterial procedure); renal events (start of dialysis therapy, change of dialysis modality, or renal trans-plantation); cancers; and other serious adverse events. Wealso examined potential adverse effects of aspirin on uricacid excretion by recording acute episodes of gout. Reportsof unexplained muscle pain, unexplained muscle weakness,and cancer diagnoses also were sought systematically.Possible adverse effects of low-dose aspirin therapy wereassessed in 2 different ways. First, we determined thenumbers of patients experiencing a decrease greater than 2g/dL(  20 g/L) in hemoglobin levels between any 2 visits orbetween randomization and a later visit. Second, we system-atically sought reports of all episodes of bleeding andcalculated the number of patients with at least 1 majorbleeding episode (ie, fatal or requiring hospitalization) andthe number with at least 1 minor bleeding episode (eg,epistaxis, ecchymosis, or bruising). Major bleeding wouldresult in the aspirin (or matching placebo) study treatmentbeing stopped permanently.Biochemical test result abnormalities requiring action wereto be reported promptly by the coordinating center to the studynurse, and patients were then recalled for further review. Astandard algorithm was used for the management of significantALT and CK level abnormalities. The coordinating centerobtained more detailed information about reports of all eventsrecordedonthefollow-upformbydirectcontactwiththestudynurse (or, if necessary, the patient’s nephrologist or primarycare physician) or the Office for National Statistics.All eventswere coded centrally according to a standard protocol. Thestudy was approved by the Anglia & Oxford MulticentreResearchEthicsCommittee. Statistical Analysis The prespecified primary analyses were intention-to-treatcomparisons of: (1) all patients allocated to simvastatin versus UNITED KINGDOM HEART AND RENAL PROTECTION I STUDY 475  all those allocated to placebo simvastatin, and (2) all patientsallocatedtoaspirinversusallthoseallocatedtoplaceboaspirin.Between-group differences in continuous lipid variables werecalculatedbyusinglinearregression,whichincludedatermforaspirintoallowforthefactorialdesign.Missingfollow-uplipidmeasurements were imputed with the value at randomization.Reported adverse events were analyzed in terms of the numberof patients who ever reported such an outcome. Such informa-tion was sought for all randomized patients irrespective of clinicattendance.Resultsforparticulargroupsareexpressedasmean    SD or number and percentage of individuals unlessstated otherwise. Differences between groups are expressed asmean  SEM.Two-sided  P  are used throughout, and  P  greaterthan0.05isconsiderednotsignificant(NS).The sample size was designed to be sufficient to detect a3-fold increase in annual risk for major bleeding with aspirinfrom 3.5% to 10.5%. It was estimated that a study of 600patients would have approximately 90% power to detect suchan increase at the 5% significance level. It also was calculatedthat a study of this size would provide reliable estimates of thesize of any mean changes in LDL cholesterol levels during 1year within each of the separate patient groups (predialysis,dialysis, and transplant patients) and help estimate the likelylong-termattenuationincholesterolleveldifferencesthatmightbeproducedbylossofcompliance.Thesedatawererequiredtoprovide reliable sample-size projections for a large-scale trialdesigned to assess the effects of cholesterol-lowering therapyonseriousvasculareventsinindividualswithCKD. RESULTS Study Population Theoriginaltargetsamplesizewas600patients,but recruitment was discontinued after an interimanalysisof448patientsshowedthattheannualrateof major bleeding events was less than anticipated(2.5%), suggesting that continuation would be un-likely to provide much additional information onaspirin-associated bleeding risks. Of 571 patientswho attended a screening appointment, 448 pa-tients (78%) were eventually randomized betweenOctober 1999 and March 2001. Of randomizedpatients, 242 patients (54%) were predialysis, 73patients (16%) were on dialysis therapy (39 pa-tients [9%], peritoneal dialysis; 34 patients [8%],hemodialysis), and 133 patients (30%) had a func-tioning renal transplant. The 4 randomized groupsgenerated by the 2  2 factorial design were wellbalanced with respect to the key prognostic vari-ablesassessedatbaseline(Table1).Approximately two thirds of participants were men, mean age was53years,meanbloodpressurewas143/84mmHg,mean body mass index was 27.4 kg/m 2 , 9% had ahistory of prior vascular disease, and 11% haddiabetesmellitus. Compliance With Allocated Treatment  Of 448 randomized patients, 71 patients (16%)stopped both treatments (that is, simvastatin [ormatching placebo] and aspirin [or matching pla-cebo]), 19 patients (4%) stopped aspirin (or match-ing placebo) only, and 11 patients (2%) stopped Table 1. Baseline Characteristics of Randomized Patients Baseline CharacteristicsSimvastatinAspirin(n  112)SimvastatinOnly(n  112)AspirinOnly(n  113)DoublePlacebo(n  111) Age (y) 54  14 52  15 52  16 54  15Men 78 (69.6) 79 (70.5) 81 (71.7) 76 (68.5)History of vascular disease* 9 (8.0) 10 (8.9) 10 (8.8) 10 (9.0)Body mass index (kg/m 2 ) 26.8  5.1 27.5  4.4 27.1  4.9 28.2  8.3Diabetes mellitus 12 (10.7) 12 (10.7) 13 (11.5) 11 (9.9)Ethnic srcinWhite 103 (92.0) 99 (88.4) 100 (88.5) 101 (91.0)Black 4 (3.6) 8 (7.1) 8 (7.1) 6 (5.4)Indian 4 (3.6) 3 (2.7) 4 (3.5) 4 (3.6)Other 1 (0.9) 2 (1.8) 1 (0.9) 0 (0.0)Systolic blood pressure (mm Hg) 145  20 141  20 144  20 142  21Diastolic blood pressure (mm Hg) 85  11 84  11 83  12 84  11Renal statusPredialysis 59 (52.7) 62 (55.4) 61 (54.0) 60 (54.1)Continuous ambulatory peritoneal dialysis 13 (11.6) 8 (7.1) 7 (6.2) 11 (9.9)Hemodialysis 11 (9.8) 6 (5.4) 7 (6.2) 10 (9.0)Transplant 29 (25.9) 36 (32.1) 38 (33.6) 30 (27.0)NOTE. Values expressed as mean  SD or number (percent).*Definedasahistoryofanginapectoris,myocardialinfarction,coronaryrevascularization,stroke,orperipheralarterialdisease.BAIGENT ET AL476  simvastatin (or matching placebo) only. There wasno excess of patients stopping active simvastatincompared with placebo simvastatin, either overall(42 patients [19%], simvastatin versus 40 patients[18%], placebo simvastatin; Table 2;  P  NS) orfor patients reporting adverse effects as the reasonfor stopping (7 [4%] versus 5 patients [2%];  P  NS). Although there was no excess of patientsstopping among those allocated to active aspirincompared with placebo aspirin overall (44 patients[20%], aspirin versus 46 patients [20%], placeboaspirin;  P  NS), allocation to aspirin therapy wasassociated with an excess of adverse effects result-ing in treatment discontinuation (20 [9%] versus 5patients [2%];  P    0.002), the most common of which was major or minor bleeding (15 [7%]versus 4 patients [2%];  P  0.01). Of 432 patientswho attended the 6-month follow-up visit or wereassessed by telephone (in which case, they hadstopped study simvastatin or matching placebo),the proportion of patients taking 80% or greater of tabletswassimilar(  88%)foreachoftherandom-izedtreatmentsandalsoforbothtreatments(  80%)among 442 patients with 12-month follow-up data.  Biochemical Efficacy and Safety of 20 mg of Simvastatin Daily Three months after randomization, allocation to20 mg of simvastatin daily reduced LDL choles-terol levels by 30%    2% (SE) ( P    0.0001),equivalent to a mean absolute reduction of 37mg/dL (0.95 mmol/L; Table 3). Total cholesterol level was reduced by 22%    1% ( P    0.0001);non-HDL cholesterol level, by 28%    2% ( P   0.0001); triglyceride level, by 15%    4% ( P   0.001); and apolipoprotein B level, by 23%  1%( P  0.0001). There was a marginally statisticallysignificantincreaseinHDLcholesterollevel(5%  2%;  P  0.05), but no significant effect on apoli-poprotein A 1  level (2%  1% increase;  P  NS).By 12 months after randomization, LDL choles-terol level was reduced by 24%  2%, equivalentto a mean absolute reduction of 29 mg/dL (0.75mmol/L;  P  0.0001); non-HDLcholesterol level,by 23%  2% ( P  0.0001); triglyceride level, by13%  5%( P  0.01);andapolipoproteinBlevel,by 19%  2% ( P  0.0001). At 12 months, therewasnosignificanteffectonHDLcholesterol(2%  2% increase) or apolipoprotein A 1  level (0.4%  1% increase; both  P  NS). Allocation to aspirintherapy did not modify the observed effects of allocation to simvastatin therapy on any compo-nent of the lipid profile at either time (data notshown).Mean baseline plasma LDL cholesterol levelsvaried among the 3 main categories of patientsstudied (predialysis, 123 mg/dL [3.18 mmol/L];dialysis, 111 mg/dL [2.86 mmol/L]; transplant,128 mg/dL [3.31 mmol/L]), but proportionalreductions in LDLcholesterol levels with simva-statin allocation were similar in each of thesecategories at both 3 months (predialysis, 33%  2% reduction; dialysis, 24%    5% reduction;transplant, 26%  3% reduction; chi-square testfor heterogeneity on 2  df   5.3;  P  NS) and 12months (27%  3%, 20%  6%, and 20%  4%reductions,respectively;chi-squaretestforheter-ogeneity on 2  df     3.5;  P    NS; Table 4). Likewise, the proportional effects of simvastatinallocation on total cholesterol, non-HDL choles-terol, HDL cholesterol, and triglyceride levelswere similar in each of these 3 disease categoriesat both 3 and 12 months (tests for heterogeneity, Table 2. Reasons for Stopping Allocated Study Treatment Reason for StoppingSimvastatin(n  224)Placebo Simvastatin (n  224)Aspirin(n  225)Placebo Aspirin(n  223) Patient request 17 (8) 14 (6) 13 (6) 18 (8)Abnormal liver function test results 2 (1) 1 (0.4) 0 (0) 0 (0)Contraindicated drug 1 (0.4) 6 (3) 3 (1) 7 (3)Adverse effectBleed 0 (0) 0 (0) 15 (7) 4 (2)Other 7 (3) 5 (2) 5 (2) 1 (0.4)Other reason 15 (7) 14 (6) 8 (4) 16 (7)Total 42 (19) 40 (18) 44 (20) 46 (20)NOTE. Values expressed as number of patients (percent).UNITED KINGDOM HEART AND RENAL PROTECTION I STUDY 477
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks