In the treatmentofpatientswithadvancedchronickidneydisease (CKD) (which can be considered present when the estimated glomerular filtration rate [GFR] falls to <20 mL/min/ 1.73m2 and individuals become eligible to accrue time on the kidney transplant waiting list in the United States), a paramount goal is preventing or retarding progression to endstage renal disease and the requirement of dialysis. At earlier stages of CKD, blockade of the renin-angiotensinaldosteronesystem(RAAS) is amainstay therapy; angiotensinconverting enzyme inhibitors (ACEIs) or angiotensin II receptor blockers (ARBs) clearly slow renal function decline, especially among thosewith proteinuria. However, the use of ACEIs or ARBs in advanced CKD remains uncertain. This important clinical question is the subject of a new study by Hsu and colleagues.1 Physiciansmayhesitate to prescribeACEIs or ARBs in advanced CKD for several reasons. Angiotensin-converting enzymeinhibitorsandARBs increaseserumpotassiumlevels, and their use in advanced CKDmay lead to dangerous hyperkalemia. InitiationofRAASblockade isoftenassociatedwithshorttermreductions inGFR,asmuchas30%.This reduction is generally believed to be due to reductions in intraglomerular pressure, hemodynamically reversible and self-limited (unless there are unusual coexisting conditions, such as severe renal artery stenosis).2,3 However, in advancedCKD, a 30% difference in GFR may determine whether the patient needs immediate initiation of dialysis. There have also been reports of substantial increase in GFR (more than 30%) after cessation of ACEI or ARB therapy in advanced CKD, resulting in postponement of dialysis.4 Other investigatorshaveemphasized that theuseofACEIs andARBsheightens the riskof acute-on-chronic renal failure.5 PatientswithCKDareat increased risk for acutekidney injury,6 andnumerouspatientswithmoderateoradvancedCKDdonot recover fromanepisodeof dialysis-requiring acute kidney injury but instead become dependent on long-term dialysis. Therefore, a superimposed episodeof severe acute kidney injury can immediately precipitate end-stage renal disease in some CKD patients.7 Thus, although somedata from randomized clinical trials suggest that RAAS blockade can be implemented safely and is efficacious in advanced CKD,8,9 much uncertainty remains about use of ACEIs andARBs in non–clinical trial participants with advancedCKD,particularly the elderly. The studybyHsu andcolleagues1usesacomprehensiveTaiwanesedatabase that captures health care utilization data of more than 95% of the hospitals in Taiwan and99%of the country’s population of 23 million. The investigators studied 28497patientswith stable, hypertensive, advanced CKD receiving erythropoiesisstimulatingagent (ESA) therapywhohaveserumcreatinine levels greater than 6 mg/dL (to convert to micromoles per liter, multiplyby88.4).Of these, 14 117usedACEIs/ARBs, and14380 werenonusers.UseofACEIs/ARBswasassociatedwitha lower hazard ratio for long-term dialysis (0.94 [95% CI, 0.91-0.97]) andthecompositeoutcomeof long-termdialysisordeath (0.94 [0.92-0.97]). Similar results were seen in several sensitivity analyses. The strengths of this study include its large size and its being nationally representative. The large proportion of patients with diabetes mellitus enhances generalizability because these patientswere excluded fromprior studies.8,9Hsu and colleagues1 provide reassuring information complementing the data from randomized clinical trials that RAAS blockadecanbesafely implementedeven inadvancedCKD.8,9Based on the age and sex distribution provided (more than half the patientswere older than65years and female), all patients had an estimated GFR of 12 mL/min/1.73m2 or less. A large, randomized clinical trial seems unlikely to be undertaken in this subset of patients in the foreseeable future. However, several caveats must be kept in mind when interpreting these data. No information regarding blood pressure,proteinuria,orother importantpatient-level clinical characteristics was provided. Themajor weakness of this study is that the actual levels of renal function were unknown because the database did not contain serum creatinine values. Instead, the level of renal functionwas inferred based onprescription of an ESA. The authors’ claim that the 28 497 patients studied had serum creatinine levels of greater than 6 mg/dL (andhematocrit levels of <28%) is basedonTaiwanNational Health Insurance reimbursement regulations bywhich patients meeting these criteria “could receive ESA treatment tomaintainahematocrit levelnot exceeding36%.”1Thus, confoundingbydifferences in the level ofGFRbetweenusers and nonusers of ACEIs/ARBs is a concern. The strength of the association between the level of GFR and the need for dialysis is very strong,10 at anorder ofmagnitude stronger than the association between the level of GFR and other adverse outcomes, such as cardiovascular disease.11 Hence, if ACEI/ARB users had on average even slightly higher levels of GFR, confounding may account for the rather modest observed effect size (approximately 6%). BlockadeofRAAS is known to lower hematocrit levels in CKDpatients12; hence, use of anESAmay be triggered amongACEI/ARBusers at higher GFR levels than ACEI/ARB nonusers. In addition,patientswithadvancedCKDwhoaremoreadherent to treatment regimens may be more likely to be prescribedACEIs/ARBs because nephrologistsmay trust them to come reliably for follow-up appointments and toundergo freRelated article page 347 Renin-Angiotensin System Blockade in CKD Original Investigation Research
Read full abstract