Angiotensin converting enzyme inhibitors and angiotensin


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Angiotensin converting enzyme inhibitors and angiotensin

  1. 1. Angiotensin-converting enzyme inhibitors and angiotensinreceptor blockers for adults with early (stage 1 to 3) non-diabetic chronic kidney disease (Review)Sharma P, Blackburn RC, Parke CL, McCullough K, Marks A, Black CThis is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library2011, Issue 10http://www.thecochranelibrary.comAngiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  2. 2. T A B L E O F C O N T E N T S1HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . .5BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1316ADDITIONAL SUMMARY OF FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . .18DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Analysis 1.1. Comparison 1 ACEi versus placebo, Outcome 1 Mortality. . . . . . . . . . . . . . . . . 44Analysis 1.2. Comparison 1 ACEi versus placebo, Outcome 2 Cardiovascular events. . . . . . . . . . . . 45Analysis 1.3. Comparison 1 ACEi versus placebo, Outcome 3 Doubling of creatinine. . . . . . . . . . . . 46Analysis 1.4. Comparison 1 ACEi versus placebo, Outcome 4 Doubling of creatinine by underlying renal pathologies. 47Analysis 1.5. Comparison 1 ACEi versus placebo, Outcome 5 Adverse events. . . . . . . . . . . . . . . 4848ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .54CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .54DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .55INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .iAngiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  3. 3. [Intervention Review]Angiotensin-converting enzyme inhibitors and angiotensinreceptor blockers for adults with early (stage 1 to 3) non-diabetic chronic kidney diseasePawana Sharma1, Rachel C Blackburn2, Claire L Parke2, Keith McCullough2, Angharad Marks2, Corri Black21Health Services Research Unit, University of Aberdeen, Aberdeen, UK. 2Institute of Applied Health Sciences, University of Aberdeen,Aberdeen, UKContact address: Corri Black, Institute of Applied Health Sciences, University of Aberdeen, Foresterhill, Aberdeen, Grampian, AB526NZ, UK. group: Cochrane Renal Group.Publication status and date: New, published in Issue 10, 2011.Review content assessed as up-to-date: 7 March 2010.Citation: Sharma P, Blackburn RC, Parke CL, McCullough K, Marks A, Black C. Angiotensin-converting enzyme inhibitors andangiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronic kidney disease. Cochrane Database of SystematicReviews 2011, Issue 10. Art. No.: CD007751. DOI: 10.1002/14651858.CD007751.pub2.Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.A B S T R A C TBackgroundChronic kidney disease (CKD) is a long term condition that occurs as a result of damage to the kidneys. Early recognition of CKDis becoming increasingly common due to widespread laboratory estimated glomerular filtration rate (eGFR) reporting, raised clinicalawareness, and international adoption of Kidney Disease Outcomes Quality Initiative (K/DOQI) classification. Early recognition andmanagement of CKD affords the opportunity not only to prepare for progressive kidney impairment and impending renal replacementtherapy, but also for intervening to reduce the risk of progression and cardiovascular disease. Angiotensin-converting enzyme inhibitors(ACEi) and angiotensin receptor blockers (ARB) are two classes of antihypertensive drugs that act on the renin-angiotensin-aldosteronesystem. Beneficial effects of ACEi and ARB on renal outcomes and survival in people with a wide range of severity of renal impairmenthave been reported; however, their effectiveness in the subgroup of people with early CKD (stage 1 to 3) is less certain.ObjectivesThis review aimed to evaluate the benefits and harms of ACEi and ARB or both in the management of people with early (stage 1 to 3)CKD who do not have diabetes mellitus.Search methodsIn March 2010 we searched The Cochrane Library, including The Cochrane Renal Group’s specialised register and The CochraneCentral Register of Controlled Trials (CENTRAL), MEDLINE and EMBASE. Reference lists of review articles and relevant studieswere also checked. The search was conducted using the optimally sensitive strategy developed by the Cochrane Collaboration for theidentification of randomised controlled trials (RCTs) with input from an expert in trial search strategy.Selection criteriaAll RCTs reporting the effect of ACEi or ARB in people with early (stage 1 to 3) CKD who did not have diabetes mellitus were selectedfor inclusion. Only studies of at least four weeks duration were selected. Authors, working in teams of two, independently assessed theretrieved titles and abstracts, and whenever necessary the full text of these studies were screened to determine which studies satisfiedthe inclusion criteria.1Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  4. 4. Data collection and analysisData extraction was carried out by two authors, independently, using a standard data extraction form and cross checked by two otherauthors. Methodological quality of included studies was assessed using the Cochrane risk of bias tool. Data entry was carried out by oneauthor and cross checked by another author. When more than one study reported similar outcomes, data were pooled using the random-effects model, but a fixed-effect model was also analysed to ensure the robustness of the model chosen and to check susceptibility tooutliers. Heterogeneity was analysed using a Chi² test on N-1 degrees of freedom, with an alpha of 0.05 used for statistical significanceand with the I² test. Where data permitted, subgroup analysis was used to explore possible sources of heterogeneity. The quality of theevidence was analysed.Main resultsFour RCTs enrolling 2177 participants met our inclusion criteria. Of these, three compared ACEi with placebo and one comparedACEi with ARB. Two studies had an overall low risk of bias, and the other two were considered to be at moderate to high risk of bias.Low to moderate quality of evidence (from two studies representing 1906 patients) suggested that ACEi had no impact on all-causemortality (RR 1.80, 95% CI 0.17 to 19.27, P = 0.63) or cardiovascular events (RR 0.87, 95% CI 0.66 to 1.14, P = 0.31) in peoplewith stage 3 CKD. For all-cause mortality, there was substantial heterogeneity in the results. One study (quality assessment: low riskof bias) reported no difference in the risk of end-stage kidney disease in those with an eGFR > 45 mL/min/1.74 m² treated with ACEiversus placebo (RR 1.00, 95% CI 0.09 to 1.11, P = 0.99). The (high risk of bias) study that compared ACEi with ARB reported littledifference in effect between the treatments when urinary protein, blood pressure or creatinine clearance were compared. No publishedstudies comparing ARB with placebo or ACEi and ARB with placebo were identified.Authors’ conclusionsOur review demonstrated that there is currently insufficient evidence to determine the effectiveness of ACEi or ARB in patients withstage 1 to 3 CKD who do not have diabetes mellitus. We have identified an area of significant uncertainty for a group of patients whoaccount for most of those labelled as having CKD.P L A I N L A N G U A G E S U M M A R YAngiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early chronic kidney disease whodo not have diabetesChronic kidney disease (CKD) is a long-term condition that occurs as a result of the kidneys being damaged. Progressive deteriorationof kidney function can lead to end-stage kidney disease (ESKD). People with ESKD cannot maintain healthy kidney function andneed kidney dialysis or transplant. In the early stages of CKD, patients may not have any outward symptoms or signs of illness, andmay only be detected following investigations such as urine or blood testing. Two types of drugs - angiotensin-converting enzymeinhibitors (ACEi) and angiotensin receptor blockers (ARB) - have been widely recommended in clinical guidelines for doctors to usein the management of CKD. This review identified four studies (enrolling 2177 people). Three studies compared ACEi to placebo orno treatment and one study compared ACEi to ARB. There is not enough evidence in the published literature at present to determinehow effective drugs in the ACEi or ARB families are for treating patients with early (stage 1 to 3) CKD who do not have diabetes.2Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  5. 5. SUMMARYOFFINDINGSFORTHEMAINCOMPARISON[Explanation]ACEicomparedtoplaceboforearly(stage1to3)non-diabeticchronickidneydiseasePatientorpopulation:patientswithearly(stage1to3)non-diabeticchronickidneydiseaseSettings:hospitalIntervention:ACEiComparison:placeboOutcomesIllustrativecomparativerisks*(95%CI)Relativeeffect(95%CI)NoofParticipants(studies)Qualityoftheevidence(GRADE)CommentsAssumedriskCorrespondingriskplaceboACEiMortality-CKDstage3Follow-up:36to42monthsStudypopulationRR1.8(0.17to19.27)1906(2studies)⊕⊕low1,2,3103per1000185per1000(18to1000)Mediumriskpopulation76per1000137per1000(13to1000)Cardiovascularevents-CKDstage3Follow-up:36to42monthsStudypopulationRR0.87(0.66to1.14)1906(2studies)⊕⊕⊕moderate2,3104per100090per1000(69to119)Mediumriskpopulation82per100071per1000(54to93)3Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  6. 6. *Thebasisfortheassumedrisk(e.g.themediancontrolgroupriskacrossstudies)isprovidedinfootnotes.Thecorrespondingrisk(andits95%confidenceinterval)isbasedontheassumedriskinthecomparisongroupandtherelativeeffectoftheintervention(andits95%CI).CI:Confidenceinterval;RR:Riskratio;GRADEWorkingGroupgradesofevidenceHighquality:Furtherresearchisveryunlikelytochangeourconfidenceintheestimateofeffect.Moderatequality:Furtherresearchislikelytohaveanimportantimpactonourconfidenceintheestimateofeffectandmaychangetheestimate.Lowquality:Furtherresearchisverylikelytohaveanimportantimpactonourconfidenceintheestimateofeffectandislikelytochangetheestimate.Verylowquality:Weareveryuncertainabouttheestimate.1Therewasasignificantheterogeneityof81%.OnestudyfavourscontrolwhileanotherfavoursACEi.2Thequalityofevidenceisimprecisebecausethetotalnumberofeventsislessthan300(athresholdrule-of-thumbvalue)(basedon:Muelleretal.AnnInternMed.2007;146:878-881<>)3Lessthan10studies,socreatingafunnelplotwasnotapplicable.4Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  7. 7. B A C K G R O U N DDescription of the conditionChronic kidney disease (CKD) is a long-term condition that oc-curs as a result of damage to the kidneys. In 2002, the US KidneyDisease Outcomes Quality Initiative (K/DOQI) proposed a clas-sification for CKD that has been widely adopted internationally(Levey 2003).K/DOQI Stages for CKDStage Description1 Normal GFR: GFR > 90 mL/min/1.73 m² with other evidence of chronic kidney damageª2 Mild impairment: GFR 60 to 89 mL/min/1.73 m² with other evidence of chronic kidney damageª3 Moderate impairment: GFR 30 to 59 mL/min/1.73 m²4 Severe impairment: GFR 15 to 29 mL/min/1.73 m²5 End stage kidney disease (ESKD): GFR < 15 mL/min/1.73 m²aOther evidence of CKD may be one of the following: persistentproteinuria; persistent haematuria (after exclusion of other causes,such as urological disease); structural abnormalities of the kidneysdemonstrated on ultrasound scanning or other radiological tests (e.g.polycystickidneydisease,refluxnephropathy); or biopsy-proven chronicglomerulonephritis (mostof thesepatients willhavemicroalbuminuriaor proteinuria, and/or haematuria).Classification is based on two markers: evidence of kidney damage(such as the presence of microalbuminuria, proteinuria or struc-tural abnormality); and the sustained impairment of glomerularfiltration rate (GFR) for at least three months. Normal GFR inyoung adults is around 100 to 120 mL/min/1.73 m².Early CKD has been described as stages 1 to 3 of the K/DOQIclassification.Atthese stages, apatientmayhave no outward symp-toms or signs of illness and only testing such as dipstick urinemeasurement for proteinuria/haematuria or blood test may detectthe presence of a renal abnormality.PrevalencePrevalence estimates for CKD vary substantially. Several large,high quality, population-based screening studies have reported theprevalence of CKD stage 3 to 5 disease to be around 3.8% to4.7%; more than 95% of people with an estimated GFR of lessthan 60 mL/min/1.73 m² have stage 3 disease (Coresh 2005; Drey2003; Hallan 2006). In most epidemiological studies, the GFR isestimated (eGFR) from serum creatinine (SCr) measurements us-ing an equation; several equations exist and this contributes to thevariation in prevalence reported in these studies. The prevalenceof stage 1 and 2 disease (based on microalbuminuria (albumin/creatinine ratio (ACR) of 17 to 250 mg/g for men or 25 to 355mg/g for women) or macroalbuminuria (ACR > 250 mg/g for menor > 355 mg/g for women) has been reported to be as high as 11%of the population (CDC 2007). The prevalence of CKD increaseswith age (Coresh 2005; Hallan 2006; Imai 2007; John 2004).An ageing population, escalating prevalence of diabetes mellitus(one of the major risk factors for CKD), and increasing recogni-tion are contributing to a reported increase in the prevalence ofearly CKD and its growing recognition as a major public healthproblem.Consequences of early CKDEarly CKD, whilst often asymptomatic, is an important healthissue and has implications for individuals and health services. Pro-gressive deterioration of kidney function can result in end-stagekidney disease (ESKD) and the need for renal replacement ther-5Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  8. 8. apy (RRT) in the forms of dialysis or transplantation. The rate ofprogression of CKD may be influenced by secondary factors suchas age, race, intraglomerular haemodynamic factors, hypertensionand proteinuria. ESKD has risen globally over the last two decadesat high cost to individuals, their carers and families, and healthservices. However, the proportion of people with early CKD whoprogress to ESKD is low (Daly 2007; Hallan 2006). A greaterrisk for people with early CKD is cardiovascular disease (CVD).One study that followed patients over a period of five years, re-ported that 3.1% of people with CKD progressed to requiringRRT, whereas 24.9% died before reaching dialysis, probably as aresult of CVD (Daly 2007).Early recognition and management of CKD affords the opportu-nitynotonlytoprepare forprogressive kidneyimpairmentand im-pending RRT (CHOICE Study 2001; Dogan 2005; Khan 2007;Kinchen 2002), but also for intervening to reduce the risk of pro-gression and CVD.Description of the interventionAngiotensin-converting enzyme inhibitors (ACEi) and an-giotensin receptor blockers (ARB) are two classes of antihyper-tensive drugs that act on the renin-angiotensin-aldosterone sys-tem (RAAS). Both drug classes have been widely recommendedin guidelines for the management of CKD, particularly in pa-tients with evidence of proteinuria, and have been reported to pro-vide both cardioprotective and renoprotective effects. Beneficialeffects of ACEi and ARB on renal outcomes and survival in peoplewith diabetic kidney disease (DKD) have been reported (Strippoli2006). For non-diabetic patients with moderate to severe CKD,there was evidence of benefit in terms of renal outcomes whetheror not proteinuria was present (Jafar 2003a). The evidence forcardioprotective effects, particularly in non-diabetic patients withCKD, is less consistent (ASCOT-BPLA Study 2005; HOPEStudy2000; Strippoli 2006).To date, reviews have combined evidence from study participantswith a wide range of severity of renal impairment but the subgroupof those with early CKD (stage 1 to 3) have not been presentedseparately.How the intervention might workAs kidneys become damaged and begin to lose nephrons, patientsexperience systemic hypertension, proteinuria and a progressivedecline in GFR (Metcalfe 2007). Common pathological changesare observed regardless of the underlying causes and include earlyrenal inflammation; tubulointerstitial injury, and glomeruloscle-rosis (Ruster 2006). The pathophysiology of progressive kidneyfunction loss involves complex haemodynamic, endocrine, and in-flammatory factors (Metcalfe 2007; Ruster 2006).ACEi and ARB both act to inhibit the RAAS endocrine system.ACEi mode of action includes blocking the conversion of inactiveangiotensin I to active angiotensin II at the level of the enzymeneeded for its conversion. ARB works at a later stage in the RAASsystem and selectively blocks the type I subtype which is a receptorfor angiotensin II (Kumar 2002). Once thought of as a systemicendocrine system important in mediating vascular tone, RAAS isnow understood to be complex, operating both systemically andlocally within the kidney. Products of the RAAS are understood toimpact on a wide range of renal, as well as haemodynamic, factorsthat contribute to the progression of CKD (Kshirsagar 2000a;Kumar 2002; Ruster 2006).The action of ACEi and ARB extends beyond simple blood pres-sure (BP) control and may reflect effects on the complex RAASpathways. The ability of both of these drugs to inhibit the RAASat different points means that they have the potential to moderatethe functional and structural changes that occur in progressive re-nal insufficiency (Giatras 1997a; Jafar 2003a; Kshirsagar 2000a).Why it is important to do this reviewEarly recognition of CKD is becoming increasingly common dueto widespread laboratory reporting of eGFR, raised clinical aware-ness, and international adoption of K/DOQI classification. Thehigh prevalence of early CKD means that many individuals andclinicians are faced with choices about management. AnotherCochrane review (Strippoli 2006) has reviewed the evidence ofeffectiveness in people with diabetic CKD and demonstrated thatACEi and ARB play a core role in the management and preven-tion of DKD where proteinuria is a key feature. For people with-out diabetes, particularly those with normal or mild to moderatekidney function impairment where proteinuria may or may notbe present, the role of ACEi and ARB is less certain. This reviewseeks to summarise the evidence in relation to benefits and harmsof ACEi and ARB for non-diabetic patients with early (stage 1 to3) CKD.O B J E C T I V E SThis review aimed to evaluate the benefits and harms of ACEi andARB or both in the management of people with early (stage 1 to3) non-diabetic CKD.M E T H O D SCriteria for considering studies for this reviewTypes of studies6Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  9. 9. All randomised controlled trials (RCTs) and quasi-RCTs (studiesin which allocation to treatment was obtained by alternation, useof alternate medical records, date of birth or other predictablemethods) looking at the effect of ACEi or ARB were included.The first period of randomised cross-over studies were consideredfor inclusion. Only studies of at least four weeks duration wereincluded.Types of participantsInclusion criteriaAll non-diabetic adults (18 years or over) with early CKD, withno restriction to gender or race, were considered. Early CKD wasdefined as K/DOQI stage 1 to 3. We included studies that mea-sured eGFR by any method (excretion of iohexol, inulin or simi-lar marker; estimated from 24 hour urine collection; or estimatedfrom SCr using a recognised equation).Studies defining CKD based on SCr or other thresholds of GFRwere included in the review where the results had been presentedseparately for those with K/DOQI stage 1 to 3. Studies in generalpopulations that included people with diabetes were kept in the re-view where the results were presented separately for those with andwithout diabetes. If the results were not presented separately, andless than 30% of the study population had diabetes mellitus, thenthe study was included and the effect on the outcomes assessed inthe sensitivity analysis. The same process was adopted for studypopulations that included people with specific renal pathologies(e.g. immunoglobulin A (IgA) nephropathy, lupus nephritis, poly-cystic kidney disease).Exclusion criteriaAny studies of patients with a diagnosis of diabetes mellitus (type Ior II) were excluded. Because the prognosis for people with specificrenal diagnoses cannot be generalised to the wider population withCKD, studies restricted to patients with a single specific renaldiagnosis (e.g. IgA nephropathy, lupus nephropathy, polycystickidney disease) were excluded.Types of interventionsAll ACEi and ARB or combinations were included as outlinedbelow:1. Treatment with ACEi versus placebo2. Treatment with ARB versus placebo3. Treatment with ACEi and ARB versus placebo4. Treatment with ACEi versus ARB.The ACEi class includes the drugs:• Benazepril• Captopril• Cilazapril• Delapril• Enalapril maleate• Fosinopril sodium• Imidapril hydrochloride• Lisinopril• Moexipril hydrochloride• Perindopril erbumine• Quinapril• Ramipril• Spirapril• Trandolapril• Zofenopril.The ARB class includes:• Candesartan• Eprosartan• Irbesartan• Losartan• Olmesartan• Telmisartan• Valsartan.Any dose and dosing regimen was included in the review. Com-bination preparations with medicines other than ACEi and ARBwere not included. Only oral preparations were included.As they are licensed, new drugs will be added to the review insubsequent updates.Types of outcome measuresPrimary outcomes1. All-cause mortality2. CVD morbidity and mortality (including myocardialinfarction, cerebrovascular accident, congestive heart failure)3. ESKD (including RRT)Secondary outcomes1. Quality of life (QoL) measured by the visual analogue scale,such as SF-36 and KDQoL2. Adverse events including but not limited to: allergicreactions, cough, headache, hyperkalaemia, hypotension,angioedema and acute kidney injury (AKI)3. Kidney failure progression defined by: GFR rate of decline,doubling of creatinine or progression of CKD stage4. Proteinuria and albuminuria including: progression ofmicroalbuminuria to macroalbuminuria, regression ofmacroalbuminuria to microalbuminuria, progression ofnormoabluminuria to microalbuminuria, regression ofmicroalbuminuria to normoabluminuria measured by protein/creatinine ratio (mg/mmol); urinary total protein excretion (g/247Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  10. 10. hours); urinary ACR (mg/mmol); urinary albumin excretion(µg/min)5. BP: systolic and diastolic BP (mm Hg) reported as meanchange from baseline or percentage reaching study specific target6. Costs: total health care costs7. Hospital admission rates.Search methods for identification of studiesWe searched the following resources:Electronic searches1. The Cochrane Renal Group’s specialised register (searched8 March 2010) and the Cochrane Central Register of ControlledTrials (CENTRAL) in The Cochrane Library, Issue 1, 2010).CENTRAL and the Renal Group’s specialised register containthe handsearched results of conference proceedings from generaland speciality meetings. This is an ongoing activity of theCochrane Collaboration and is both retrospective andprospective (Master List 2009). We did not therefore specificallysearch conference proceedings. The Cochrane Renal Group’sModule in The Cochrane Library provides an up-to-date list ofconference proceedings (Renal Group 2011).2. MEDLINE (OvidSP 1950 - 8 March 2010) search wasconducted using the optimally sensitive strategy developed forthe Cochrane Collaboration for the identification of RCTs(Lefebvre 2008) with a search strategy developed with inputfrom the Cochrane Renal Group’s Trial Search Co-ordinator.3. EMBASE (OvidSP 1980 - 8 March 2010) search wasconducted using a search strategy adapted from that developedfor the Cochrane Collaboration for the identification of RCTs(Lefebvre 2008) with a search strategy developed with inputfrom the Cochrane Renal Group’s Trial Search Co-ordinator.See Appendix 1 for search terms used.Searching other resources1. Reference lists of review articles and relevant studies.2. We planned to write letters seeking information aboutunpublished or incomplete studies to investigators where onlyabstracts were identified, but none of the included studiespresented this requirement.Data collection and analysisSelection of studiesThe review was undertaken by six authors (CB, PS, RB, CP,KMcC, AM). The search strategy described was used to obtain ti-tles and abstracts of studies potentially relevant to the review. Thetitles and abstracts were shared for screening by two teams of twoauthors who worked independently (CB, RB, CP and PS). Theauthors discarded the studies that were not applicable; however,studies and reviews that might include relevant data or informa-tion on trials were retained initially. Authors, working in teams oftwo as before, independently assessed the retrieved abstracts, andwhenever necessary the full text of these studies were screened todetermine which studies satisfied the inclusion criteria.Data extraction and managementData extraction was carried out by at least two authors, indepen-dently, using a standard data extraction form. Studies reportedin non-English language journals were to be translated before as-sessment, but none of the included studies required translation.Where more than one publication of one study existed, reportswere grouped together and only the publication with the mostcomplete data was included. Where relevant outcomes were onlypublished in earlier versions, these data were used. Any discrepancybetween published versions was highlighted. It was planned thatany further information required from the original investigatorwas to be requested by written correspondence and any relevantinformation obtained in this manner was to be included in thereview. Disagreements were resolved in consultation with KMcC.Assessment of risk of bias in included studiesThe following items were independently assessed by two authors(CP, RB) using the risk of bias assessment tool (Higgins 2008) (seeAppendix 2). Discrepancies were resolved by discussion with CB.• Was there adequate sequence generation?• Was allocation adequately concealed?• Was knowledge of the allocated interventions adequatelyprevented during the study?• Were incomplete outcome data adequately addressed?• Are reports of the study free of suggestion of selectiveoutcome reporting?• Was the study apparently free of other problems that couldput it at a risk of bias?Measures of treatment effectData entry was carried out by CB and cross checked by PS. Fordichotomous outcomes (such as death, cardiovascular morbidity,adverse events) results were expressed as risk ratios (RR) with 95%confidence intervals (CI). Where continuous scales of measure-ment were used to assess the effects of treatment (such as SCr),the mean difference (MD) was used. According to the protocol,the standardised mean difference (SMD) was to be used wheredifferent scales had been applied. In instances where change frombaseline data (change scores) were reported, the difference in meanchange scores was to be used. It was planned that if standard devi-ations for change scores were not available, missing data were not8Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  11. 11. to be imputed. Data were presented in tabular form and, whereappropriate, final score data and change from baseline were incor-porated into meta-analyses. If adjustment had been undertaken toaccount for baseline values, these data were to be reported. Timeto event data (such as survival, time to ESKD) were to be anal-ysed as a dichotomous variable where data were presented for allparticipants up to a specified time period. Alternatively, hazardratios (and 95% CI) were to be used, with application of the pro-portional hazard assumption, for the purpose of comparison andmeta-analysis (Higgins 2008).Unit of analysis issuesWe did not anticipate that there would be any non-standard de-signs, such as crossover studies and cluster-RCTs, but multiple armstudies could be identified. Here, all intervention groups relevantto the review were included. It was planned that if there were sev-eral relevant comparisons, all independent comparisons were tobe included. It was also planned that if a comparator group over-lapped (such as a single placebo arm), either comparison groupswould be combined (if appropriate), or only the most importantcomparison would be selected for meta-analysis.Dealing with missing dataIntention-to-treat was the primary analysis sought from studies.Missing outcomes data, and the implications were discussed butimputation of missing data was not undertaken.Assessment of heterogeneityHeterogeneity was analysed using a Chi² test on N-1 degrees offreedom, with an alpha of 0.05 used for statistical significance andwith the I² test (Higgins 2003). I² values of 25%, 50% and 75%correspond to low, medium and high levels of heterogeneity.Assessment of reporting biasesIt was planned that funnel plots were to be assessed for evidenceof publication bias, and to plot effect estimates against study sizewhere data permit.Data synthesisData were pooled using the random-effects model but the fixed-effects model was also analysed to ensure the robustness of themodel chosen and susceptibility to outliers.Subgroup analysis and investigation of heterogeneityWhere data were available subgroup analysis was undertaken toexplore possible sources of heterogeneity related to the followingcharacteristics:• CKD stage• Presence of microalbuminuria/proteinuria• Age and gender• Comorbidities (CVD, hypertension)• Distribution of underlying renal pathologies• Interventions (heterogeneity in treatments could be relatedto prior agent(s) used and the agent, dose and duration oftherapy)• Study quality.Adverse effects were tabulated and assessed with descriptive tech-niquesbecause theywere likelytobe differentforthe variousagentsused. Where possible, the risk difference (RD) with 95% CI wasto be calculated for each adverse effect, compared with either notreatment or another agent.It was planned that QoL measures would be tabulated and re-ported descriptively, or if data permitted, meta-analysis would beundertaken as described.Sensitivity analysisIt was planned to undertake sensitivity analysis to explore therobustness of findings to key decisions in the review process. Thesewere to be determined as the review process took place (Higgins2008).The authors aimed to determine the applicability of the results tothe individual with early CKD (stage 1 to 3).R E S U L T SDescription of studiesSee: Characteristicsof includedstudies; Characteristicsof excludedstudies.See Characteristics of included studiestables.Results of the searchElectronic searching retrieved a total of 2118 titles and abstracts,from which 126 potentially eligible studies were considered fur-ther. Analysis of the full text left us with four studies that metthe inclusion criteria (AIPRI Study 1996; Matsuda 2003; PEACEStudy 2006; REIN Stratum 1 1999). There was no disagreementamong the authors about the inclusion of studies. Further infor-mation is presented in the study flow diagram (Figure 1).9Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  12. 12. Figure 1. Study flow diagram.Included studiesTypes of studyAll four included studies were RCTs and were published in full.Types of participantsOf the four studies, three recruited patients with CKD using arange of definitions:1. Matsuda 2003: stage 1 to 3 CKD (SCr < 265 µmol/L or“creatinine clearance > 30 mL/min/1.73 m²” plus proteinuria >0.3 g/24 hours).2. AIPRI Study 1996: stage 3 CKD only (SCr 133 to 354µmol/L and 24 hour creatinine clearance (CrCl) 30 to 60 mL/min).3. REIN Stratum 1 1999: included those with eGFR 20 to 70mL/min/1.73 m² plus proteinuria ≥ 1 to 2.9 g/24 hours atbaseline but reported a relevant CKD group - eGFR > 45 mL/min/1.73 m²separately (stage 1 to 3a).4. PEACE Study 2006: included patients with stable coronaryartery diseases and those with reduced left ventricular function atbaseline. The study reported relevant CKD group with eGFR 45to 59.9 mL/min/1.73 m² (stage 3a) and eGFR < 45 mL/min/1.73 m² (minimum 27 mL/min/1.73 m²) (stage 3b).Therefore, in two studies the included participants were a sub-group of the originally randomised patients (PEACE Study 2006;REIN Stratum 1 1999).The proportion of participants in CKD stage 3a and 3b varied;39% (AIPRI Study 1996) to 88% (PEACE Study 2006) were instage 3a CKD, and 12% (PEACE Study 2006) to 61% (AIPRIStudy 1996) were in stage 3b CKD. Different definitions andmeasures of GFR were reported in each study.One study had fewer than 100 participants (Matsuda 2003); twohad between 100 and 1000 (AIPRI Study 1996; REIN Stratum 11999); and one had more than 1000 participants (PEACE Study2006).All studies included both male and female adult participants, al-though a higher proportion of males was common among all stud-ies. Two studies had an age range of 18 to 70 years (AIPRI Study1996; REIN Stratum 1 1999); one study included participantsaged over 50 years (PEACE Study 2006), and another indicatedthat at baseline, all participants were adults without specifying anage range (Matsuda 2003).There were two studies that included patients with diabetes mel-litus (AIPRI Study 1996; PEACE Study 2006), but these patientsrepresented less than 30% of the total participants at baseline. Onestudy included only people with stable coronary artery disease orleft ventricular failure (PEACE Study 2006).Hypertension was common among study participants: one studyincluded only patients with hypertension (Matsuda 2003); and inthe other three studies, more than 50% of participants were hyper-10Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  13. 13. tensive. Each study included multiple underlying kidney diseases:glomerular diseases accounted for 33% (AIPRI Study 1996) to98% (Matsuda 2003) of study participants; interstitial nephritisor polycystic disease was present in 7% (REIN Stratum 1 1999) to29% (AIPRI Study 1996); nephrosclerosis occurred 2% (Matsuda2003) to 16% (AIPRI Study 1996); and 18% (AIPRI Study 1996)to 47% (REIN Stratum 1 1999) of patients had unknown or otherrenal diagnosis. The results were not presented separately for dif-ferent disease subgroups.Two studies defined CKD based on CrCl level from 24 hour urinecollection at baseline (AIPRI Study 1996; Matsuda 2003); oneestimated GFR using iohexol clearance (REIN Stratum 1 1999);and another used eGFR (four-variable Modification of Diet inRenal Disease [MDRD] formula) to define relevant CKD groups(PEACE Study 2006). One study used SCr (133 to 354 µmol/L)in addition to CrCl level, thus excluding patients with the mostmild stage 3 CKD (AIPRI Study 1996).Two studies included patients with proteinuria at baseline (REINStratum 1 1999: ≥ 1 to 2.9 g/24 hours; Matsuda 2003: > 0.3g/24 hours); the remaining two studies did not report presenceor absence of proteinuria among participants at baseline (AIPRIStudy 1996; PEACE Study 2006).Types of interventionsThree of the four included studies compared three different ACEidrugs with placebo:• Ramipril (REIN Stratum 1 1999)• Benazepril (AIPRI Study 1996)• Trandolapril (PEACE Study 2006).None of the included studies compared ARB with placebo or ACEiand ARB with placebo.Only one study compared ACEi (perindopril or trandolapril) withARB (losartan or candesartan) (Matsuda 2003).Type of outcomesThree studies presented data on our listed primary outcomes:• All-cause mortality (AIPRI Study 1996; PEACE Study2006)• Cardiovascular morbidity/mortality (AIPRI Study 1996;PEACE Study 2006)• ESKD (REIN Stratum 1 1999).Of our listed secondary outcomes, no study reported findings forQoL, hospital admission rates or costs.Two studies reported their findings for stage 3a and 3b CKDseparately (AIPRI Study 1996; PEACE Study 2006) and one studyreported by low and high proteinuria level (Matsuda 2003).Excluded studiesA total of 93 studies were excluded. The main reason for exclusionwas that studies did not present outcomes separately for stage1 to 3 CKD. See Characteristics of excluded studies for furtherinformation.The COOPERATE Study 2003 was excluded after a letter ofretraction was published by the editors of The Lancet in October2009.The REIN study was split into two stratum based on baselineproteinuria. The low proteinuria study has been reported here.The high proteinuria study (> 3 g/24 hours) did not report find-ings separately by eGFR or CKD stage and was excluded (REINStratum 2 1997).Risk of bias in included studiesDetails of the risk of bias are given in the Risk of Bias Table. Twostudies had an overall low risk of bias (PEACE Study 2006; REINStratum 11999). One small studywascharacterisedbyparticularlyhigh risk of bias: there was little information reported about therandomisation method and no indication of blinding (Matsuda2003). Twostudiesreportedlongertermfollow-upbeyond the endof the treatment phase and were subject to substantial switchingbetween treatment groups (AIPRI Study 1996; PEACE Study2006). See Figure 2 and Figure 3 for the summary results for riskof bias.11Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  14. 14. Figure 2. Methodological quality summary: review authors’ judgements about each methodological qualityitem for each included study.12Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  15. 15. Figure 3. Methodological quality graph: review authors’ judgements about each methodological qualityitem presented as percentages across all included studies.Two studies reported data for relevant subgroups but the fullstudy population is not included here (PEACE Study 2006; REINStratum 1 1999).Using funnel plots to detect publication bias was not feasible be-cause of the small number of studies included in this review.AllocationThe method of sequence generation and allocation concealmentwas adequately reported in two studies (PEACE Study 2006;REIN Stratum 1 1999), unclear in one (AIPRI Study 1996), andthe method of randomisation was not described in the fourth study(Matsuda 2003).BlindingTwo studies were double blinded (AIPRI Study 1996; REINStratum 1 1999) and one study was blinded to treatment only(PEACE Study 2006). The study that compared two activetreatment groups did not report on blinding (Matsuda 2003).The PEACE Study 2006, which started randomisation in 1996,blindedstudytreatmentuntil 2002, whenthe study’ssteeringcom-mittee recommended open label treatment with ACEi of those pa-tients with diabetes, hypertension and microalbuminuria, or pro-teinuria. The REIN Stratum 1 1999 continued its treatment asopen label after 27 months of blinding.Incomplete outcome dataTwo studies reported no missing data (AIPRI Study 1996;Matsuda 2003) and the other two addressed missing data (PEACEStudy2006; REIN Stratum11999). Analysiswasperformedbasedon intention-to-treat analysis in all but one study (Matsuda 2003).Selective reportingThere was no evidence of the selective reporting of outcomes inthe included studies.Other potential sources of biasStudies presenting long term outcomes were based on open labelfollow-up (AIPRI Study 1996; PEACE Study 2006). The protein-uria subgroup of the REIN Stratum 1 1999 was stopped ahead ofschedule by the advisory panel because of significant benefits tothe treatment group. Matsuda 2003 provided insufficient infor-mation to determine if there were any other potential sources ofbias.Effects of interventionsSee: Summary of findings for the main comparison ACEicompared to placebo for early (stage 1 to 3) non-diabetic chronickidney disease; Summary of findings 2 ACEi compared toplacebo (single study outcomes)See Summary of findings for the main comparison; Summary offindings 2.ACEi versus placeboThree studies compared ACEi with placebo. Duration of treat-ment ranged from 36 (AIPRI Study 1996) to 63 months (REINStratum 1 1999). Follow-up extended to 6.6 years (AIPRI Study1996). None of the studies reported information for subgroupsbased on age, gender, proteinuria or comorbidities. Where sub-group data were available by disease stage and underlying kidney13Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  16. 16. disease, these are reported. Each of the studies used a differentACEi; therefore, it was not possible to explore differences in effectby agent, dose or duration of therapy. There were too few studiesto explore the effect of study quality on effect size. The summaryeffect was estimated in a meta-analysis of two studies (AIPRI Study1996; PEACE Study 2006). REIN Stratum 1 1999 did not reportthe number of participants with early stages of CKD who receivedACEi or placebo. A total of 1001 patients who were treated withACEi (benazepril and trandolapril) and 905 treated with placebowere included for analysis.Primary outcomesAll-cause mortalityTwo studies reported all-cause mortality for stage 3 CKD only.A total of 170 deaths occurred in 1906 patients (77/1001 in thetreated group versus 93/905 in the placebo group). In the AIPRIStudy 1996, a total of nine deaths occurred (8/300 in the treat-ment group versus 1/283 in the placebo group) during a treat-ment period of three years. In the PEACE Study 2006, 161 deathswere reported (69/701 in the treatment group versus 92/622 inthe placebo group) during a median follow-up period of 4.8 years.There was no statistically significant difference in the risk of deathamong those treated with ACEi compared with placebo. (Analysis1.1.1 (2 studies, 1906 participants): RR 1.80, 95% CI 0.17 to19.27). There was statistically significant heterogeneity (I² = 81%,P = 0.02); PEACE Study 2006 reported a statistically significantreduction in deaths in the treated group (RR 0.67, 95% CI 0.50to 0.89); AIPRI Study 1996 observed fewer deaths in the placebogroup.Long-term follow-upAIPRI Study 1996 reported an extended period of follow-up at theend of the treatment phase of the study to give a total median dura-tion of 6.6 years. During the extended follow-up, 64% of those pa-tients randomised to benazepril continued on an ACEi, and 61%in the placebo arm started treatment with an ACEi. At the end ofthis time no statistically significant differences were observed inthe number of deaths between the groups (based on their originaltreatment allocation); 25/300 patients who were originally treatedwith benazepril and 23/283 who received placebo had died (RR1.55; 95% CI 0.95 to 59.96).Cardiovascular mortality and morbidityTwo studies reported cardiovascular events (AIPRI Study 1996;PEACE Study 2006) for stage 3 CKD only. A total of 186 cardio-vascular events occurred in 1906 patients (92/1001 in the treat-ment group and 94/905 in the placebo group).There was no statistically significant difference in the risk of car-diovascular events in the ACEi group compared with the placebogroup(Analysis 1.2.1(2studies,1906participants): RR0.87, 95%CI 0.66 to 1.14). None of the included studies had statisticallysignificant point estimates. The estimate was dominated by onestudy (PEACE Study 2006) that contributed to 92% weight tothe summary estimate. There was no statistical evidence of het-erogeneity (I² = 0%, P = 0.49).Definition of cardiovascular events varied in two studies. PEACEStudy 2006 included cardiovascular death plus myocardial infarc-tion, and AIPRI Study 1996 included all cardiovascular events andstroke.ESKDOnly REIN Stratum 1 1999 reported kidney survival in a sub-group of patients whose baseline GFR was > 45 mL/min/1.73 m²(maximum GFR 70 mL/min/1.73 m²) . At 63 months, only 3%of patients in this subgroup progressed to RRT. No statistically sig-nificant difference in RR was reported between ACEi and placebo(RR 1.00, 95% CI 0.09 to 1.11, P = 0.99).Kidney impairment and progressionDoubling of creatinineAIPRI Study 1996 reported 88/583 participants experienced dou-bling of creatinine or the need for dialysis during a follow-up of 3years. (Of the 583 patients only two needed dialysis, one each inthe treatment group and the placebo group). There was a statis-tically significant reduction in the doubling of creatinine amongthe ACEi group versus placebo (Analysis 1.3.1: RR 0.51, 95% CI0.34 to 0.77).Change in GFRREIN Stratum 1 1999 reported the mean rate of change in GFR/month for patients whose baseline GFR was > 45 mL/min/1.73m² as 0.19 mL/min/1.73 m² (standard error (SE) 0.07) in theACEi group compared with 0.34 mL/min/1.73 m² (SE 0.09) inthe placebo group (P = 0.25).Adverse events and withdrawalsOnly AIPRI Study 1996 reported adverse events or withdrawalsfor relevant patient groups (See Table 1). A total of 71 adverseevents were reported in 583 patients (38/300 in the ACEi groupand 33/283 in the placebo group) (Analysis 1.5: RR 1.09, 95% CI0.70 to 1.68). Fewer than 1% of patients reported dry cough inboth the ACEi and the placebo groups. Few patients experienced14Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  17. 17. hyperkalaemia (2% ACEi versus 1% placebo). Hypertensive crisiswas not reported by any patient in the treatment group comparedwith 1.4% of patients in the placebo group.Adverse events caused withdrawal from the study by 12% of pa-tients in the ACEi group and 9% in the placebo group. With-drawals due to other events, such as loss to follow-up, protocol vi-olation or lack of cooperation, was 6% in the ACEi group and 8%in the placebo group. Further details on withdrawals and adverseevents are presented in Table 1.Twostudies(PEACE Study2006; REIN Stratum1 1999)reportedadverse events and withdrawals for those who were randomised atstudy level, but not for the relevant subgroups that were of interestfor this review.Secondary outcomesProteinuriaAIPRI Study 1996 reported on proteinuria observing a decreasein urinary protein of 29% in the ACEi group and an increase by9% in the placebo group at 36 months (statistical significance notreported).Blood pressureAIPRI Study 1996 reported change in mean BP from baselinevalues at 36 months. Mean systolic BP decreased by 4.5 to 8.0mm Hg in the ACEi group versus an increase of 1.0 to 3.7 mmHg in the placebo group at 6, 12, 24 and 36 months.Mean diastolic BP decreased by 3.5 to 5.0 mm Hg in the ACEigroup versus increases of 0.2 to 1.5 mm Hg in the placebo groupat 6, 12, 24 and 36 months. The study also reported that uncon-trolled hypertension decreased from 28% to 18% in the ACEigroup and there was an increase from 27% to 32% in the placebogroup.Subgroup analysesThere was insufficient data to pool estimates for subgroup analysisfor any of the outcomes. The analysis is based on single study data.CKD stagePEACE Study 2006, which reported results for all-cause mortalityand CVD events, stratified by stage 3a and 3b, found no statisticalsignificant effect of ACEi over placebo in either mortality (Analysis1.1.2, stage 3a: RR 0.73, 95% CI 0.52 to 1.01; Analysis 1.1.3,stage 3b: RR 0.64, 95% CI 0.34 to 1.20) or CVD events (Analysis1.2.2, stage 3a: RR 0.92, 95% CI 0.67 to 1.26; Analysis 1.2.3,stage 3b: RR 0.83, 95% CI 0.46 to 1.50).AIPRI Study 1996 reported results for doubling of creatinine byCKD stage. The effect of ACEi to reduce the number with dou-bling of creatinine varied according to stage of CKD. A 70% re-ductioninthe riskof doublingcreatinine was observedin the ACEigroup versus placebo for CKD stage 3a (Analysis 1.3.2: RR 0.30,95% CI 0.11 to 0.79) and 39% reduction for stage 3b (Analysis1.3.3: RR 0.61, 95% CI 0.39 to 0.94).Presence of proteinuriaREIN Stratum 11999 includedonlyparticipantswith proteinuria.AIPRI Study 1996 reported reduction in risk of the doubling ofcreatinine level (or ESKD) in people with stage 3 CKD stratifiedby baseline 24 hour urinary protein. For those with ≤ 1 g/L and> 1 g/L to < 3 g/L there was no statistically significant differencecompared with placebo (≤ 1 g: 31%, 95% CI -67 to 71; > 1 gto < 3 g: 53%, 95% CI -14 to 81) but a statistically significantreduction was observed for those with ≥ 3 g/L (66%, 95% CI 34to 82). A similar pattern was observed when adjusted for changein diastolic BP and change in urinary protein excretion. PEACEStudy 2006 did not report on proteinuria.Age and comorbiditiesPEACE Study 2006, which included older patients who had es-tablished CVD at baseline, reported a significant beneficial effectof ACEi on all-cause mortality, but not on cardiovascular or renaloutcomes.Most of the included study participants were hypertensive (PEACEStudy2006: 54%; AIPRI Study1996: 81%(ACEigroup),83% (placebo group); REIN Stratum 1 1999: 79% (ACEi group),85% (placebo group)). AIPRI Study 1996 reported the effect ofACEi on doubling of creatinine stratified by baseline diastolic BP.For those with a treated diastolic BP ≤ 90 mm Hg there was astatistically significant reduction in doubling of creatinine in thosetreated with ACEi compared with placebo (51%, 95% CI 13 to73). After adjusting for changes in diastolic BP and proteinuria,the difference was lost. For those with normal, untreated BP or atreated diastolic BP > 90 mm Hg at baseline, there was no statis-tically significant difference (58%, 95% CI -72 to 89; 50%, 95%CI -8 to 76 respectively).Distribution of underlying renal pathologiesAIPRI Study 1996 reported data for doubling of creatinine orESKDat36monthsbyunderlyingrenal pathology (Analysis 1.4)).The effect of ACEi in reducing doubling of creatinine varied ac-cording to the underlying renal pathologies. Among those withunderlying glomerular diseases, a statistically significant reductionin the doubling of creatinine was observed in the ACEi groupcompared with placebo (Analysis 1.4.1: RR 0.42, 95% CI 0.22to 0.81). In the patients with other renal diagnoses (diabetic kid-ney disease, polycystic kidney disease, nephrosclerosis, interstitialnephritis and unknown renal disorders), there was no statisticallysignificant difference observed.15Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  18. 18. None of the included studies reported ESKD events and changein GFR by subgroup.ARB versus placeboThere were no published studies identified that compared ARBwith placebo.ACEi plus ARB versus placeboThere were no published studies identified that compared ACEiplus ARB with placebo.ACEi versus ARBMatsuda 2003 compared ACEi with ARB. The study reported BP,CrCl and proteinuria stratified by baseline proteinuria level; thosewith mild (< 1 g/day), and moderate (> 1 g/day) proteinuria levels.The authors reported that no significant difference was observedbetween ACEi and ARB in terms of change of BP from baselineand at 48 weeks (no values reported). Twenty seven patients weretreated with ACEi (perindopril or trandolapril) and 25 patientswere treated with ARB (losartan or candesartan).Subgroup analysisPresence of proteinuriaNosignificantchange inurinaryproteinexcretionwasobservedforpatients with mild proteinuria who received either ACEi or ARB(data were reported graphically only). A statistically significantbenefit was observed among patients with moderate proteinuria at48 weeks: ACEi significantly reduced urinary protein excretion by54% (standard error of mean (SEM) 7%) (from 2.7 (SEM 0.5) g/day to 1.2 (0.2) g/day, P < 0.05). Similarly, a statistically significantreduction of 41% (SEM 6%) (from 2.7 (0.4) g/day to 1.6 (0.3)g/day, P < 0.05) was observed in the ARB group. For ARB versusACEi, there was a statistically significant reduction in the ARBgroup at 48 weeks (P < 0.05) but not at 12 weeks (P > 0.2).Among patients with mild proteinuria, a statistically significantreduction in systolic/diastolic BP was reported from baseline to48 weeks for both ACEi and ARB.• ACEi: difference of mean* 17/12 mm Hg; from 148/86(SEM 3/5) mm Hg to 131/74 (SEM 4/4) mm Hg, (P < 0.05)• ARB: difference of mean* 17/15 mm Hg; from 154/86(SEM 4/3) mm Hg to 137/71 (SEM 3/2) mmHg, P < 0.05).For those with moderate proteinuria, marked reductions in BP(systolic/diastolic) were observed at 48 weeks in the ACEi group(difference of mean* 28/12 mm Hg; from 152/90 (SEM 4/3) mmHg to 124 (SEM 3)(diastolic BP only reported in graph) mm Hg,P < 0.01) while only a modest reduction was observed in the ARBgroup (difference of mean* 13/10; from 150/89 (SEM 3/3) mmHg to 137/79 (SEM 4/3) mm Hg, P < 0.05).(*difference of mean baseline values and the values at 48 weeks).CrCl was stable (data were reported graphically only) in bothACEi and ARB treatment groups and for both mild and moderateproteinuria.16Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  19. 19. A D D I T I O N A L S U M M A R Y O F F I N D I N G S [Explanation]Outcomes Illustrative comparative risks* (95% CI) for studypopulationRelative effect(95% CI)No. of ParticipantsAssumed risk Corresponding riskPlacebo ACEiMortality - CKD stage 3aFollow-up: median 4.8yearsStudy population124 per 1000 91 per 1000(64 to 125)RR 0.73(0.52 to 1.01)1203Mortality - CKD stage 3bFollow-up: median 4.8yearsStudy population256 per 1000 164 per 1000(87 to 307)RR 0.64(0.34 to 1.2)157Cardiovascular events -CKD Stage 3aFollow-up: median 4.8yearsStudy population117 per 1000 108 per 1000(78 to 147)RR 0.92(0.67 to 1.26)1203Cardiovascular events -CKD Stage 3bStudy population244 per 1000 203 per 1000(112 to 366)RR 0.83(0.46 to 1.5)157Doubling of creatinine -CKD stage 3Follow-up: 36 monthsStudy population201 per 1000 103 per 1000(68 to 155)RR 0.51(0.34 to 0.77)583Doubling of creatinine -CKD Stage 3bFollow-up: 36 monthsStudy population239 per 1000 146 per 1000(93 to 225)RR 0.61(0.39 to 0.94)356Adverse eventsFollow-up: 36 monthsStudy population117 per 1000 128 per 1000(82 to 197)RR 1.09(0.7 to 1.68)583*The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effectof the intervention (and its 95% CI).CI: Confidence interval; RR: Risk ratio17Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  20. 20. D I S C U S S I O NSummary of main resultsDespite growing international emphasis on the early detectionand management of CKD, and that stage 1 to 3 CKD accountsfor most people with evidence of kidney impairment, we foundvery few studies that reported the effectiveness of ACEi or ARBin this population. Only three studies that compared ACEi withplacebo in people with stage 1 to 3 CKD were identified (AIPRIStudy 1996; PEACE Study 2006; REIN Stratum 1 1999), andone study that compared ACEi with ARB was found (Matsuda2003). Two studies were considered to be at moderate to highrisk of bias (Matsuda 2003; REIN Stratum 1 1999). The patientgroups included in the studies varied considerably, particularly interms of comorbidities and severity of proteinuria. This made itdifficult to draw any overall conclusions. Each of the four includedstudies differed in their definitions of CKD in terms of GFR cut-off levels and measures of GFR. These differences are particularlyimportant when identifying a cohort of people with stage 3 CKD.Invariably, participants in each study differed in regard to theirreported degrees of kidney impairment.The low to moderate quality evidence from two studies (AIPRIStudy 1996; PEACE Study 2006) suggested that ACEi had noimpact on all-cause mortality or cardiovascular events in peoplewith stage 1 to 3 CKD. There was substantial heterogeneity in theresults relating to all-cause mortality: one study reported a mod-est benefit and the other reported potential harm. The study thatshowed a small potential benefit included older patients (aged > 50years) with coronary artery disease (PEACE Study 2006). Approx-imately 20% of patients who participated in this study had diag-noses of diabetes at baseline. The study that reported potentiallyharmful outcomes (AIPRI Study 1996) included comparativelyyounger patients with no underlying CVD, but including smallnumbers of participants with hypertension and with diabetes mel-litus (4%). There was insufficient evidence to assess whether theseare true subgroups of the study population who had different re-sponses to ACEi treatment, or if these outcomes reflect chance orother variations in the studies, including the specific ACEi studydrug (Trandolapril was used in the PEACE Study 2006 and be-nazepril in the AIPRI Study 1996).In terms of renal outcomes, one study assessed as having a low riskof bias reported no difference in the risk of ESKD among thosewith an eGFR > 45 mL/min/1.73 m² treated with ACEi versusplacebo (REIN Stratum 1 1999). One study that featured a mod-erate risk of bias reported a 50% reduction in the risk of creatininedoubling in those with stage 3 CKD who were treated with ACEi(AIPRI Study 1996). This study also reported a reduction in pro-teinuria among people in the ACEi group. The greatest benefitwas observed in those people with stage 3a CKD.Only one study reported adverse events for the relevant study par-ticipants, and no statistically significant difference between pa-tients in the ACEi and placebo groups (AIPRI Study 1996). Veryfew people reported cough or hyperkalaemia.The one study with a high risk of bias that compared ACEi withARB (Matsuda 2003), reported little difference in effect betweentreatments when urinary protein, BP or CrCl were measured. Thisstudy did not report all-cause mortality, cardiovascular events,ESKD or adverse events.Overall completeness and applicability ofevidenceThis review highlights the striking lack of studies in people withstage 1 to 3 CKD, which is the population group that accountsfor most people identified as having CKD. Identification of CKDstages 1 and 2 requires evidence of kidney damage such as confir-mation of proteinuria, haematuria or structural abnormality. Thisrequirement may make identifying people to participate in studiesmore challenging, and even more so for investigators looking intosingle, specific renal diagnoses.The two major issues around applicability of the evidence to clin-ical practice and patients were:• definition of CKD and• the underlying pathologies that caused the kidney damage.Two of the four included studies that reported data for all-causemortalityandcardiovascularmorbidityandmortalitydefinedtheirCKD groups based on single eGFR assays (AIPRI Study 1996;PEACE Study 2006), and therefore were prone to classificationbias because of the chance of including people with AKI. Only onestudy based the definition of CKD on an eGFR using standardequations (PEACE Study 2006), which is the method most widelyadopted in clinical practice, but likely to underestimate GFR inthose with true GFR level of around 60 mL/min/1.73 m².The included patient population comprised mainly hypertensivepatients with varying renal pathologies, and 4% (AIPRI Study1996) to 19% (PEACE Study 2006) had diabetes mellitus at base-line. The high proportion of people with diabetes, and the in-clusion of only people with existing CVD in the PEACE Study2006, make it difficult to draw conclusions for the wider popula-tion with early CKD. The PEACE Study 2006 reported a reduc-tion in all-cause mortality consistent with other studies of ACEi inpopulations with CVD; however the investigators did not reportthe statistically significant benefits in kidney outcomes that wereobserved in other studies conducted among diabetic study partici-pants only. The proportions of patients with CVD and specificallyreported renal diagnoses varied from study to study.We identified a number of significant gaps in the evidence. Nostudies were identified that compared ARB with placebo or ACEicombined with ARB. Only some of the many preparations ofACEi and ARB drugs available were investigated in the studiesand none compared one preparation with another. Interventionfeatures such as compliance, timing, dosing or intensity could not18Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  21. 21. be elicited because of poor reporting in studies. Although most ofthe included studies reported on some measure of kidney functionover time, mortality and ESKD, none reported on quality of life,admissions to hospital or costs.Quality of the evidenceFigure 2 and Figure 3 summarise the quality of the included stud-ies. Although it was found that two studies were of low risk ofbias, the small number of studies, their relatively small size, andlow event rates for some of the primary outcomes meant that theevidence was of low to moderate quality or based on a single study.Findings should therefore be viewed with caution.Potential biases in the review processStrengths and limitationsThis review was conducted as per the protocol following pre-specified inclusion criteria and included comprehensive literaturesearches to find all relevant studies. Two authors screened all ti-tles, abstracts and full papers to avoid selection bias. There was noarbitration required during the selection process, data extractionor quality assessment that needed a third author.We found very few studies reporting outcomes for the group ofpatients of interest in this review (stage 1 to 3 CKD). We excludedstudies of single specific renal diagnoses, and it is acknowledgedthat some of these may include patients with evidence of isolatedproteinuria (stage 1 CKD). A large number of studies conductedin people with CKD were excluded because the authors did notreport their findings in subgroups that were relevant to this study.In addition, studies of the use of ACEi and ARB drugs in hyper-tensive patients may also include people with stage 1 to 3 CKD.Before asking authors to consider re-analysing data into specifiedsubgroups, we considered it appropriate to establish what datawere available in the published literature.Agreements and disagreements with otherstudies or reviewsWe did not find any other published reviews that matched ourinclusion criteria. Two systematic reviews reported that ACEi wasfound to have little benefit over placebo or other antihypertensivedrugs in reducing all-cause mortality among patients with CKDor hypertension (P = 0.12) (Jafar 2001) or those with diabetesmellitus (Strippoli 2006). In the meta-analysis (21 studies, 7295patients) of patients with diabetes and CKD, no survival benefitwas observed compared with placebo (RR 0.91, 95% CI 0.71 to1.17) except when treated with maximum tolerable doses (RR0.78, 95% CI 0.61 to 0.098) (Strippoli 2006).Jafar 2001 reported ACEi to be effective in reducing mean BP(mean decrease - systolic: 4.5 mm Hg; 95% CI 3.0 to 6.1 mm Hgand diastolic: 2.3 mm Hg; 95% CI 1.4 to 3.2 mm Hg); proteinexcretion (mean decrease: 0.46 g/d; 95% CI 0.33 to 0.59 g/d);and in reducing the risk of renal progression (doubling of SCr oronset of ESKD: RR 0.70 95% CI 0.55 to 0.88) compared withplacebo or other antihypertensive drugs. The 11 included studiesused different measures of CKD. Baseline mean creatinine was>155 µmol/L in all but two studies (where it was 124 µmol/Land 88 µmol/L). In general, these studies included participantswith greater kidney impairment than we have included here. Thefindings reported by Jafar 2001 were not stratified by CKD stage,and therefore, they could not be compared directly. The authorsdidnote thatthe benefitof ACEionreducingprogressiontoESKDwas modified by the presence of proteinuria at baseline (RR 0.09,95% CI 0.07 to 0.11) and was lower for those with proteinurialevels of 2.0 g/24 hours compared with 1.0 g/24 hours). Baselinecreatinine was not found to modify the effect of ACEi on ESKD.Similarly, in people with diabetes and CKD, ACEi produced a31% risk reduction of ESKD compared with placebo (RR 0.60,95%CI0.39to0.93). These outcomeswere foundto be similarforARB (RR 0.78, 95% CI 0.67 to 0.91) (Strippoli 2006). However,the findings were not reported separately by stage.A U T H O R S ’ C O N C L U S I O N SImplications for practiceWith increasedrecognitionof the importance and impactof CKD,the diagnosis and labelling of people with early (stage 1 to 3) CKDis now common. Drugs in the ACEi and ARB families are thecornerstone for management of CKD. This review highlights thatthere have been very few studies that report on the effectiveness ofACEi and ARB drugs for patients with early CKD, without thosestudies investigating single, specific renal diagnoses. Despite thatpeople with early CKD are at greater risk of all-cause mortalityand cardiovascular events than progression to ESKD (Daly 2007;Hallan 2006; Sharma 2010), we found very few studies that con-sidered the early management of CKD to reduce cardiovascularrisk, all-cause mortality or kidney disease progression.Our review demonstrated that there is currently insufficient evi-dence to determine the effectiveness of ACEi or ARB treatment forpatients with stage 1 to 3 CKD who do not have diabetes mellitus.Studies have been conducted among patients with single, specificrenal diagnoses but we have not included these here. Based onevidence from studies of specific renal conditions, current clinicalguidelines (Levey 2003; NICE Guideline 2008; SIGN Guideline2008) and practice support the use of ACEi and ARB drugs for19Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  22. 22. patients with proteinuric kidney disease. We have not found suf-ficient evidence to support any change in current practice.Implications for researchWe have identified an area of significant uncertainty for a group ofpatients who account for most of those labelled as having CKD.We identified more than 50 studies where relevant data may ex-ist, but the subgroups of patients that are necessary to derive sup-ported evidence were not presented in the original publications(see Characteristics of excluded studies). These studies may pro-vide relevant subgroup or individual patient data for analysis ofpatients with stage 1 to 3 CKD. In addition, studies of ACEi (suchas HOPE Study 2000; PART 2 (MacMahon 2000); PROGRESSCollaborative Group 2001; SCAT Study 2000) and ARB (suchas LIFE (Lindholm 2002); SCOPE (Lithell 2003)) conducted inpeople with hypertension or CVD may also include people whomeet the criteria of CKD stage 1 to 3 that would potentially enablesubgroup or individual patient data analysis.There ispotential needforfurtherRCTstobe conductedthatfocuson patients with stage 1 to 3 CKD. These studies should featuredesigns that are adequate in size and duration, include quality oflife outcomes and establish recruitment criteria to ensure that thestudy population can be generalised to the wider community.A C K N O W L E D G E M E N T SWe wish to thank the referees for their comments and feedbackduring the preparation of this review.R E F E R E N C E SReferences to studies included in this reviewAIPRI Study 1996 {published data only}Locatelli F, Carbarns IR, Maschio G, Mann JF, PonticelliC, Ritz E, et al.Long-term progression of chronicrenal insufficiency in the AIPRI Extension Study. TheAngiotensin-Converting-Enzyme Inhibition in ProgressiveRenal Insufficiency Study Group. Kidney International- Supplement 1997;52(Suppl 63):S63–6. [MEDLINE:9407424]Maschio G, Alberti D, Janin G, Locatelli F, Mann J,Motolese M, et al.The use of benazepril in the treatmentof progressive renal disease [abstract]. 13th InternationalCongress of Nephrology; 1995 Jul 2-6; Madrid, Spain.1995:65.∗Maschio G, Alberti D, Janin G, Locatelli F, Mann JF,Motolese M, et al.Effect of the angiotensin-converting-enzyme inhibitor benazepril on the progression of chronicrenal insufficiency. The Angiotensin-Converting-EnzymeInhibition in Progressive Renal Insufficiency Study Group.New England Journal of Medicine 1996;334(15):939–45.[MEDLINE: 8596594]Maschio G, Alberti D, Locatelli F, Mann JF, MotoleseM, Ponticelli C, et al.Angiotensin-converting enzymeinhibitors and kidney protection: the AIPRI trial. The ACEInhibition in Progressive Renal Insufficiency (AIPRI) StudyGroup. Journal of Cardiovascular Pharmacology 1999;33(Suppl 1):S16–20. [MEDLINE: 10028949]Matsuda 2003 {published data only}Matsuda H, Hayashi K, Homma K, Yoshioka K, KandaT, Takamatsu I, et al.Differing anti-proteinuric actionof candesartan and losartan in chronic renal disease.Hypertension Research: Official Journal of the JapaneseSociety of Hypertension 2003;26(11):875–80. [MEDLINE:14714578]∗Matsuda H, Hayashi K, Saruta T. Distinct time courses ofrenal protective action of angiotensin receptor antagonistsand ACE inhibitors in chronic renal disease. Journal ofHuman Hypertension 2003;17(4):271–6. [MEDLINE:12692572]PEACE Study 2006 {published data only}Solomon SD, Rice MM, Jablonski A, Jose P, DomanskiM, Sabatine M, et al.Renal function and effectiveness ofangiotensin-converting enzyme inhibitor therapy in patientswith chronic stable coronary disease in the Prevention ofEvents with ACE inhibition (PEACE) trial. Circulation2006;114(1):26–31. [MEDLINE: 16801465]REIN Stratum 1 1999 {published data only}Remuzzi G, Tognoni G, Ruggenenti P, Perna A, MaggiorreQ, Pizzarelli F, et al.A long term, randomized clinical trialto evaluate the effects of ramipril on the evolution of renalfunction in chronic nephropathies. Journal of Nephrology1991;4(3):193–202. [EMBASE: 1992157069]∗Ruggenenti P, Perna A, Gherardi G, Garini G, ZoccaliC, Salvadori M, et al.Renoprotective properties of ACE-inhibition in non-diabetic nephropathies with non-nephrotic proteinuria. Lancet 1999;354(9176):359–64.[MEDLINE: 10437863]Ruggenenti P, Perna A, Mosconi L, Matalone M, GaspariF, Garini G, et al.The angiotensin-converting-enzyme(ACE) inhibitor ramipril slows the rate of GFR decline(GFR) and the progression to end stage renal failure(ESRF) in proteinuric, non-diabetic chronic renal diseases[abstract]. Journal of the American Society of Nephrology1997;8(Program & Abstracts):147A.Ruggenenti P, Perna A, Remuzzi G, Gruppo Italiano diStudi Epidemiologici in Nefrologia. ACE inhibitors toprevent end-stage renal disease: when to start and whypossibly never to stop: a post hoc analysis of the REINtrial results. Ramipril Efficacy in Nephropathy. Journal ofthe American Society of Nephrology 2001;12(12):2832–7.20Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  23. 23. [MEDLINE: 11729254]References to studies excluded from this reviewAASK Pilot Study 1996 {published data only}Kusek JW, Lee JY, Smith DE, Milligan S, Faulkner M,Cornell CE, et al.Effect of blood pressure control andantihypertensive drug regimen on quality of life: the AfricanAmerican Study of Kidney Disease and Hypertension(AASK) Pilot Study. Controlled Clinical Trials 1996;17(Suppl 4):40S–6S. [MEDLINE: 8889353]Acone 2003 {published data only}Acone D, Cillo F, Giordano G. Long term effects ofcombined therapy with ace-inhibitor and angiotensinii receptor antagonist on non-diabetic nephropathies[abstract]. Nephrology Dialysis Transplantation 2003;18(Suppl 4):365.Appel 2008 {published data only}Appel LJ, Wright JT, Greene T, Kusek JW, Lewis JB,Wang X, et al.Long-term effects of renin-angiotensinsystem-blocking therapy and a low blood pressure goalon progression of hypertensive chronic kidney disease inAfrican Americans. Archives of Internal Medicine 2008;168(8):832–9. [MEDLINE: 18443258]Aranda 2005 {published data only}Aranda P, Segura J, Ruilope LM, Aranda FJ, Frutos MA,Lopez V, et al.Long-term renoprotective effects of standardversus high doses of telmisartan in hypertensive nondiabeticnephropathies. American Journal of Kidney Diseases 2005;46(6):1074–9. [MEDLINE: 16310573]Bakris 2000 {published data only}Bakris GL, Siomos M, Bolton WK, Hebert l, AgerwallR, Catanzaro D, et al.Differential effects of valsartan(v) and lisinopril (l) on potassium [k+] homeostasis inhypertensive patients with nephropathy [abstract]. Journalof the American Society of Nephrology 1999;10(Program &Abstracts):68A.∗Bakris GL, Siomos M, Richardson D, Janssen I, BoltonWK, Hebert L, et al.ACE inhibition or angiotensin receptorblockade: impact on potassium in renal failure. VAL-KStudy Group. Kidney International 2000;58(5):2084–92.[MEDLINE: 11044229]Bakris 2000a {published data only}Bakris GL, Weir MR. Angiotensin-converting enzymeinhibitor-associated elevations in serum creatinine: is this acause for concern?. Archives of Internal Medicine 2000;160(5):685–93. [MEDLINE: 10724055]Balamuthusamy 2008 {published data only}Balamuthusamy S, Srinivasan L, Verma M, AdigopulaS, Jalandara N, Hathiwala S, et al.Renin angiotensinsystem blockade and cardiovascular outcomes in patientswith chronic kidney disease and proteinuria: a meta-analysis. American Heart Journal 2008;155(5):791–805.[MEDLINE: 18440325]Bianchi 1992 {published data only}Bianchi S, Mazzoni A, Bigazzi R, Baldari G, CampeseVM. Effect of converting enzyme inhibitors (CEI) onrenal function in hypertensive patients with chronic renalfailure (CRF) [abstract]. Journal of the American Society ofNephrology 1992;3(3):280.Bonner 1993 {published data only}Bonner G, Lederle RM, Scholze J, Stumpe KO. Therapeuticsafety of perindopril in the treatment of mild hypertensionwith concomitant nephropathy. Arzneimittel Forschung1993;43(8):852–5. [MEDLINE: 8216441]Campbell 2003 {published data only}Campbell R, Sangalli F, Perticucci E, Aros C, ViscarraC, Perna A, et al.Effects of combined ACE inhibitor andangiotensin II antagonist treatment in human chronicnephropathies. Kidney International 2003;63(3):1094–103.[MEDLINE: 12631093]Carlsen 1989 {published data only}Carlsen JE, Hansen FM, Jensen HA. Efficacy and safety ofcilazapril in hypertensive patients with moderate to severerenal impairment. American Journal of Medicine 1989;87(6B):79S–82S. [EMBASE: 1990026790]Casas 2005 {published data only}Casas JP, Chua W, Loukogeorgakis S, Vallance P, SmeethL, Hingorani AD, et al.Effect of inhibitors of the renin-angiotensin system and other antihypertensive drugs onrenal outcomes: systematic review and meta-analysis. Lancet2005;366(9502):2026–33. [MEDLINE: 16338452]Cinotti 2001 {published data only}Cinotti G, Zucchelli P. Effect of lisinopril on the progressionof non-diabetic chronic renal failure [abstract]. NephrologyDialysis Transplantation 1997;12(9):A94.∗Cinotti GA, Zucchelli PC, Collaborative Study Group.Effect of Lisinopril on the progression of renal insufficiencyin mild proteinuric non-diabetic nephropathies. NephrologyDialysis Transplantation 2001;16(5):961–6. [MEDLINE:11328901]COOPERATE Study 2003 {published data only}Nakao N, Baba M, Seno H, Kasuga H, Toriyama T,Kawahara H, et al.Baseline b-adrenergic blockade-baseddifferences for the effect of renin-angiotensin blockadefor the treatment of non-diabetic hypertensive patientswith chronic kidney failure: a COOPERATE-Carvedilolsubstudy. [abstract no: F-PO621]. Journal of the AmericanSociety of Nephrology 2004;15(Oct):203A.Nakao N, Hasegawa H, Fujimori A, Fukagawa M. Baselineproteinuria level and an relative nephroprotective efficacyof combination therapy of angiotensin-converting-enzymeinhibitor and angiotensin-II receptor blocker: subgroupanalysis of the COOPERATE Trial [abstract no: F-PO191].Journal of the American Society of Nephrology 2006;17(Abstracts):377A.Nakao N, Seno H, Kasuga H, Toriyama T, KawaharaH, Fukagawa M. Effects of combination treatment withlosartan and trandolapril on office and ambulatory bloodpressures in non-diabetic renal disease: a COOPERATE-21Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  24. 24. ABP substudy. American Journal of Nephrology 2004;24(5):543–8. [MEDLINE: 15528874]Nakao N, Takada M, Nakagawa T, Sanaka T, Kayano T.Combination therapy of ace inhibitor and A-II receptorantagonist more powerfully retard progression of non-diabetic renal failure than monotherapy of each drug - amulticenter, three year, double-blind, randomized trialin Japan (COOPERATE) [abstract]. Nephrology DialysisTransplantation 2001;16(6):A90.Nakao N, Takeshi U, Yoshida A, Tadaka M, KimuraG. Is small-dose combination of angiotensin receptorblocker and angiotensin-converting enzyme inhibitor stillbeneficial in management of CRF?: A six-by-five factorialdesign prospective study [abstract]. Nephrology DialysisTransplantation 2003;18(Suppl 4):134–5.∗Nakao N, Yoshimura A, Morita H, Takada M, KayanoT, Ideura T. Combination treatment of angiotensin-II receptor blocker and angiotensin-converting-enzymeinhibitor in non-diabetic renal disease (COOPERATE):a randomised controlled trial. Lancet 2003;361(9352):117–24. [MEDLINE: 12531578]Coratelli 1988 {published data only}Coratelli P, Buongiorno E, Giannattasio M, Passavanti G.Antihypertensive efficacy of enalapril maleate on impairedrenal function. Kidney International - Supplement 1988;25:S204–6. [MEDLINE: 2846937]Crowe 2001 {published data only}Crowe A, Howse M, Kemp G, Vinjamuir S, Williams P.The antiproteinuric effect of losartan and its relationshipto systemic blood pressure and renal haemodynamics inpatients with chronic renal failure [abstract]. NephrologyDialysis Transplantation 2002;17(Suppl 1) Jong 1999 {published data only}de Jong PE, Navis G, de Zeeuw D. Renoprotective therapy:titration against urinary protein excretion. Lancet 1999;354(9176):352–3. [MEDLINE: 10437857]De Rosa 2001 {published data only}De Rosa ML, de Cristofaro A, Rossi M, Baiano A, CardaceP, Albanese L, et al.Irbesartan effects on renal function inpatients with renal impairment and hypertension: a drug-withdrawal study. Journal of Cardiovascular Pharmacology2001;38(3):482–9. [MEDLINE: 11486253]de Vinuesa 2006 {published data only}de Vinuesa SG, Goicoechea M, Kanter J, PuertaM, Cachofeiro V, Lahera V, et al.Insulin resistance,inflammatory biomarkers, and adipokines in patients withchronic kidney disease: effects of angiotensin II blockade.Journal of the American Society of Nephrology 2006;17(12Suppl 3):S206–12. [MEDLINE: 17130263]Deuber 2003 {published data only}Deuber HJ. Improvement of proteinuria by candesartan,eprosartan, irbesartan, losartan [abstract]. NephrologyDialysis Transplantation 2003;18(Suppl 4):365–6.Dyadyk 1997 {published data only}Dyadyk AI, Bagriy AE, Lebed IA, Yarovaya NF, SchukinaEV, Taradin GG. ACE inhibitors captopril and enalaprilinduce regression of left ventricular hypertrophy inhypertensive patients with chronic renal failure. NephrologyDialysis Transplantation 1997;12(5):945–51. [MEDLINE:9175047]Elung-Jensen 2003 {published data only}Elung-Jensen T, Heisterberg J, Kamper A, Sonne J,Strandgaard S. Renoprotective effect of high or low dosageof enalapril in progressive chronic nephropathy [abstract].Nephrology Dialysis Transplantation 2002;17(Suppl 1):239–40.∗Elung-Jensen T, Heisterberg J, Kamper AL, Sonne J,Strandgaard S. Blood pressure response to conventional andlow-dose enalapril in chronic renal failure. British Journal ofClinical Pharmacology 2003;55(2):139–46. [MEDLINE:12580985]Elung-Jensen T, Heisterberg J, Sonne J, StrandgaardS, Kamper AL. Enalapril dosage in progressive chronicnephropathy: a randomised, controlled trial. EuropeanJournal of Clinical Pharmacology 2005;61(2):87–96.[MEDLINE: 15761754]ESPRIT Study 2005 {published data only}Sharma AM, Hollander A, Köster J. Efficacy, Safety inPatients with Renal Impairment. treated with Telmisartan(ESPRIT) Study Group. Telmisartan in patients with mild/moderate hypertension and chronic kidney disease. ClinicalNephrology 2005;63(4):250–7. [MEDLINE: 15847251]Faulhaber 1997 {published data only}Faulhaber HD, Mann JFE, Stein G, Jansa U. Tolerabilityof valsartan and effects on renal function in treatedhypertensive patients with stable renal insufficiency[abstract]. Journal of the American Society of Nephrology1997;8(Program & Abstracts):315A.Fleischmann 2003 {published data only}Fleischmann EH, Klingbeil AU, Schneider MP, Delles C,Schmieder RE. The effect of treatment with ultra high dosecandesartan (64 mg) on proteinuria: Interim results of adouble-blind, randomized, multi-center study [abstract].Nephrology Dialysis Transplantation 2003;18(Suppl 4):133–4.Fliser 1995 {published data only}Fliser D, Rett K, Hubinger A, Ritz E. Influence of ACEinhibition on glucose tolerance in patients with stablechronic renal failure. Nephrology Dialysis Transplantation1995;10(5):643–7. [MEDLINE: 7566576]Garg 2005 {published data only}Garg JP, Ellis R, Elliott WJ, Hasabou N, Chua D,Chertow GM, et al.Angiotensin receptor blockade andarterial compliance in chronic kidney disease: a pilotstudy. American Journal of Nephrology 2005;25(4):393–9.[MEDLINE: 16088080]Giatras 1997b {published data only}Giatras I, Lau J, Levey AS. Effect of angiotensin-convertingenzyme inhibitors on the progression of nondiabetic renaldisease: a meta-analysis of randomized trials. Angiotensin-Converting-Enzyme Inhibition and Progressive Renal22Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
  25. 25. Disease Study Group. Annals of Internal Medicine 1997;127(5):337–45. [MEDLINE: 9273824]Greenbaum 2000 {published data only}Greenbaum R, Zucchelli P, Caspi A, Nouriel H, Paz R,Sclarovsky S, et al.Comparison of the pharmacokineticsof fosinoprilat with enalaprilat and lisinopril in patientswith congestive heart failure and chronic renal insufficiency.British Journal of Clinical Pharmacology 2000;49(1):23–31.[MEDLINE: 10606834]Hannedouche 2001 {published data only}Hannedouche T, Chanard J, Baumelou B, FrenchCollaborative Telmisartan Study Group. Evaluation of thesafety and efficacy of telmisartan and enalapril, with thepotential addition of frusemide, in moderate-renal failurepatients with mild-to-moderate hypertension. Journal of theRenin-Angiotensin-Aldosterone System 2001;2(4):246–54.[MEDLINE: 11881131]Haufe 1994 {published data only}Haufe CC, Sierakowski B, Jansa U, Stein G. Kidneyfunction in hypertensive patients with chronic renalfailure treated with the dual eliminated ACE-inhibitorspirapril [Renale Funktion bei hypertensiven Patienten mitchronischer Niereninsuffizienz unter Therapie mit demdual eliminierbaren ACE–Hemmer Spirapril]. MedizinischeKlinik 1994;89(8):416–20. [MEDLINE: 7968874]Henriksen 1985 {published data only}Henriksen IO, Hansen BL, Kemp E. Captopril inthe treatment of hypertension in patients with renalinsufficiency [Captopril til behandling af hypertension hospatienter med nyreinsufficiens]. Ugeskrift for Laeger 1985;147(31):2451–3. [MEDLINE: 3904113]Hogan 2002 {published data only}Hogan TJ, Elliott WJ, Seto AH, Bakris GL. Antihypertensivetreatment with and without benazepril in patients withchronic renal insufficiency: a US economic evaluation.Pharmacoeconomics 2002;20(1):37–47. [MEDLINE:11817991]Hollenberg 2003 {published data only}Hollenberg NK. Reduction of proteinuria; combined effectsof receptor blockade and low dose angiotensin-convertingenzyme inhibition. Current Hypertension Reports 2003;5(4):284–5. [MEDLINE: 12918528]HOPE Study 2001 {published data only}Mann JF, Gerstein HC, Pogue J, Bosch J, Yusuf S. Renalinsufficiency as a predictor of cardiovascular outcomes andthe impact of ramipril: the HOPE randomized trial. Annalsof Internal Medicine 2001;134(8):629–36. [MEDLINE:11304102]Hou 2006 {published data only}Hou FF, Zhang X, Zhang GH, Xie D, Chen PY, ZhangWR, et al.Efficacy and safety of benazepril for advancedchronic renal insufficiency. New England Journal of Medicine2006;354(2):131–40. [MEDLINE: 16407508]Ihle 1992 {published data only}Ihle B, Becker G, Whitworth J, Shahinfar S, Cnaan A,Kincaid-Smith P. The effect of angiotensin convertingenzyme inhibition (ACEi) on progression of chronic renalinsufficiency (CRI) [abstract]. Journal of the AmericanSociety of Nephrology 1992;3(3):284.Ihle 1996 {published data only}Ihle BU, Whitworth JA, Shahinfar S, Cnaan A, Kincaid-Smith PS, Becker GJ. Angiotensin-converting enzymeinhibition in nondiabetic progressive renal insufficiency: acontrolled double-blind trial. American Journal of KidneyDiseases 1996;27(4):489–95. [MEDLINE: 8678058]Ishimitsu 2005 {published data only}Ishimitsu T, Kameda T, Akashiba A, Takahashi T, AndoN, Ohta S, et al.Effects of valsartan on the progression ofchronic renal insufficiency in patients with nondiabetic renaldiseases. Hypertension Research - Clinical and Experimental2005;28(11):865–70. [MEDLINE: 16555574]Jafar 2001 {published data only}Jafar TH, Schmid CH, Landa M, Giatras I, Toto R,Remuzzi G, et al.Angiotensin-converting enzyme inhibitorsand progression of nondiabetic renal disease. A meta-analysis of patient-level data. Annals of Internal Medicine2001;135(2):73–87. [MEDLINE: 11453706]Jafar 2003b {published data only}Jafar TH, Stark PC, Schmid CH, Landa M, Maschio G, deJong PE, et al.Progression of chronic kidney disease: therole of blood pressure control, proteinuria, and angiotensin-converting enzyme inhibition: a patient-level meta-analysis. Annals of Internal Medicine 2003;139(4):244–52.[MEDLINE: 12965979]Kamper 1990 {published data only}Kamper AL, Thomsen HS, Nielsen SL, Strandgaard S.Initial effect of enalapril on kidney function in patients withmoderate to severe chronic nephropathy. ScandinavianJournal of Urology and Nephrology 1990;24(1):69–73.[MEDLINE: 2157277]Kamper 1992 {published data only}Kamper A, Strandgaard S, Leyssac PP. Late outcome of acontrolled trial of enalapril treatment in progressive chronicrenal failure [abstract]. Journal of the American Society ofNephrology 1993;4(Program & Abstracts):251.Kamper A, Strandgaard S, Leyssac PP. Randomisedcontrolled trial of enalapril in progressive chronic renalfailure [abstract]. Nephrology Dialysis Transplantation 1990;5(8):672–3.Kamper AL, Holstein-Rathlou NH, Leyssac PP, StrandgaardS. The influence of angiotensin-converting enzymeinhibition on renal tubular function in progressive chronicnephropathy. American Journal of Kidney Diseases 1996;28(6):822–31. [MEDLINE: 8957033]Kamper AL, Nielsen OJ. Effect of enalapril on haemoglobinand serum erythropoietin in patients with chronicnephropathy. Scandinavian Journal of Clinical andLaboratory Investigation 1990;50(6):611–8. [MEDLINE:2247767]∗Kamper AL, Strandgaard S, Leyssac PP. Effect of enalaprilon the progression of chronic renal failure. American Journal23Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers for adults with early (stage 1 to 3) non-diabetic chronickidney disease (Review)Copyright © 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.