advanced CKD: ACEi/ARB beneficial

 

A recent systematic review and individual participant-level meta-analysis of clinical trials found that prescribing ACE-inhibitors or ARBs to patients with advanced chronic kidney disease (stage 4 or 5) was beneficial: see ckd ACE ARB good for advanced CKD AnnIntMed2024 in dropbox or doi:10.7326/M23-3236)

 

Details:

-- 18 completed randomized controlled trials were found in a literature search from 1946 to 2024

-- 1739 participants were included, mean age 55, 52% female, 15% Black

-- median eGFR by creatinine was 23 (IQR 18-27), 32% had eGFR below 20; median albumin-creatinine ratio 1215mg/g (range 527-2463 mg/g) but 84% had “severe albuminuria” (which they did not define, >1000mg???)

-- mean systolic BP 155 mmHg

-- the issue of determining the overall demographics/comorbidities/etc is difficult given the huge differences in the individual studies, including year study done (varies from 1992 to 2007), different meds used (perindopril-indapamide, enalapril, valsartan, benazepril, captopril, olmesartan, ramipril, losartan, irbesartan), different numbers of participants followed different numbers of years, different baseline eGFR, or if patients with diabetes were included

    -- that being said,  they noted that in general the studies were pretty balanced, especially the larger studies that contributed more to the meta-analysis, though “some imbalances were noted by sex, race, and baseline systolic BP”

-- covariates assessed in the analysis: age, sex, race, BMI, eGFR, albuminuria, diabetes status, baseline systolic blood pressure, and history of coronary artery disease

-- primary outcome: onset of KFRT (kidney failure  with replacement therapy: ie, dialysis or getting kidney transplant)

-- seconday outcomes: death before the onset of KFRT, and a composite of KFRT or death

 

-- median followup 34 months

 

Results:

-- Kidney failure and mortality events:

    -- 624 patient (36%) developed KFRT

    -- 133 (8%) died

-- progression to KFRT: ACEi or ARB treatment initiation was associated with 34% lower risk in adjusted analysis, aHR 0.66 (0.55-0.79) [see graphs below: note that the curves splay apart after about 12 months and seem to continue to diverge]

    -- number-needed-to-treat (NNT): 7.5 people (5.1-14.6) at 4 years of followup

    -- no difference if comparator group was placebo or other antihypertensive meds (which in the individual studies were metoprolol or amlodipine, nisoldipine, atenolol or acebutolol, or nifedipine)

    -- each of the individual studies in their Forest plot found benefit to ACEi/ARBs, either as a statistically significant benefit or statistical trend to benefit (ie, no evidence of results suggesting worse effect by these meds)

 

 

 

 

-- all-cause death: ACEi or ARB treatment initiation was not associated with a statistically significant difference, HR 0.86 (0.58-1.28); there was no evidence of an interaction between randomization to ACEi/ARB initiation and baseline age, albuminuria, eGFR, or diabetes

-- risk of KFRT: no statistically significant interactions between randomized assignment to ACEi or ARB treatment initiation and baseline age, eGFR, diabetes, or albuminuria

    -- though for diabetes, there was only a strong trend to benefit (HR 0.78 (0.61-1.01)) vs no diabetes (HR 0.56 (0.43-0.72)), which was statistically significant

-- the risk of bias in the included studies was low in 10, high in 3, and unclear in 5 (lack of clarity surrounding allocation concealment), though there was no evidence of heterogeneity across the included trials)

 

Commentary:

-- the American Heart Association does comment in their recommendations on hypertension therapy that ACEi or ARB should be used as first-line therapy for patients with hypertension and  CKD stage III or higher; however, the actual use of these meds is low in those with CKD stage 4 or 5, perhaps related to the fact that there are limited studies with high-quality evidence to support the use of these drugs (the suggestion to use ACEi and ARB in advanced CKD, as per the AHA, was largely based on extrapolation from studies of patients with less severe CKD)

    -- in fact, patients with CKD in general are typically excluded from cardiovascular trials: https://gmodestmedblogs.blogspot.com/2024/06/cardiovasc-trials-patients-with-ckd.html

-- some of the concerns about ACEi/ARBs relate to the potential for acute kidney injury or hyperkalemia (the latter is most pronounced in those with diabetes). Some experts do believe that it might be beneficial to withdraw ACEi/ARBs in the setting of hyperkalemia, preparation for dialysis, or concerns about polypharmacy.

 

-- This current study was done to assess outcomes compiled in this 18-study systematic review, finding that the risk of progression to KFRT was reduced by 34% in those with advanced CKD, though no apparent effect on the risk of death.

    -- Notably, there was no difference in benefit in subgroup analysis by age, eGFR, albuminuria, or history of diabetes for the outcomes of KFRT or death, though the effect in those with diabetes was less pronounced and did not quite meet statistical significance.

-- this study was also consistent with the STOP-ACEi trial, which found that stopping an ACEi  or ARB in those with advanced kidney disease was not beneficial (Renin–Angiotensin System Inhibition in Advanced Chronic Kidney Disease | New England Journal of Medicine (nejm.org), though this study was about stopping  ACEi/ARB and not about initiating them in those with advanced CKD as in the above meta-analysis

 

-- The meta-analysis included patients with very significant proteinuria. ACEi/ARBs are particularly helpful in patients with high levels of proteinuria. Prior studies in nondiabetic patients have found that lowering the level of proteinuria to the 500-1000 mg/d (as noted in the quite old REIN studies with ramipril) is associated with decreases in renal deterioration. In patients with diabetes, the benefit seems to be at lower levels of proteinuria. And the antihypertensives best shown to decrease renal deterioration are ACEi/ARBs. The clinical evidence for spironolactone is a bit more mixed, though a 2018 article did find that long-term use was associated with decreasing CKD progression: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6262621/

  

-- as mentioned in several prior blogs, I am concerned about the accuracy of their assessment of CKD by using creatinine as the determinant, for a couple of reasons:

    -- cystatin-based eGFR has a better association with measured GFR, vs creatinine (eg see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3251200/ or https://www.nejm.org/doi/full/10.1056/NEJMoa1214234 ); KDIGO (Kidney Disease: Improving Global Outcomes) in their 2024 guidelines  recommended using cystatin as the marker for eGFR, especially when there is likely to be a concern about creatinine-based eGFR (eg low muscle mass, not infrequent in elderly and those with CKD), and that the eGFR equation incorporating both cystatin and creatinine seems to have the most accurate concordance with measured GFR (https://kdigo.org/wp-content/uploads/2024/03/KDIGO-2024-CKD-Guideline.pdf )

    -- there was a recent impressive Stockholm study that found that patients with lower eGFR as determined by cystatin vs creatinine had a much higher rate of adverse outcomes for both renal (KFRT and AKI) and nonrenal (atherosclerotic cardiovasc disease/heart failure/all-cause death) events, with creatinine-based eGFR not having much predictive value: https://gmodestmedblogs.blogspot.com/2023/12/cystatin-c-better-predictor-of-bad.html

 

Limitations:

-- there was no information on the incidence of hyperkalemia or acute kidney disease (largely because several trials did not report these events, but those that did found low rates of them)

-- also the included articles did not address those who actually took the meds or for how long. this was basically an "intention to start ACEi or ARB treatment"

    -- did this distort the results for patients with diabetes?  they are more likely to be hyperkalemic (type 4 RTA: hyporeninemic hypoaldosteronism). was there a selection bias created by their being included in a study on  ACEi/ARB for that reason? perhaps those with diabetes who were enrolled in the trials were more likely to stop these meds because of hyperkalemia, so including them in the analysis might distort the results because they stopped the meds soon after starting them and actually were more similar to patients in the placebo group?

-- several of the studies in this systematic review were very old. it is likely that practice patterns at that time, use of other meds, other comorbidities, other issues that affect kidney function (exposures, smoking, diet, exercise) were quite different from the current situation. and combining older and newer is fraught

-- the outcome of replacement therapy in those with severe CKD may not be a "hard endpoint" for all patients, since patients with end-stage renal disease may have elected not to have that therapy

-- their conclusion that albuminuria was not a significant risk factor for KFRT is problematic, since very few patients had only mild or moderate albuminuria. Would the results above apply to patients with severe CKD but without proteinuria??

-- we have no information about the blood pressure achieved during the studies. We do know that continued hypertension is perhaps the most important factor in further renal decline. It is likely that the placebo group continued with higher blood pressures than the group on ACEi/ARBs. We have no information on this group compared to other antihypertensives. It would also be useful to have information about the outcomes of these studies stratified by the year the study was done. The target of ideal achieved blood pressure has decreased over the years to a much lower goal now. So the year of the study might help clarify that

-- these studies predate the newer studies with GLP-1 agonists, which are renal protective for both patients with and without diabetes.  GLP-1 agonists have been shown to help in patients with stage 5 or end-stage kidney disease on dialysis (see dm GLP1 dec mortal or ckd progression in adv CKD JAMA2022 in dropbox, or doi:10.1001/jamanetworkopen.2022.1169). Another recent study found that the more aggressive approach of following patients with CKD with routine BNP determinations and treating them more aggressively if the BNP is elevated, seemed to decrease the likelihood of both renal deterioration as well as heart failure admissions (https://gmodestmedblogs.blogspot.com/2023/12/routine-bnp-assessment-helpful-for.html ). so, these older studies do need to be assessed in the context of newer therapies

    -- given the current thinking that there is a strong association between renal and cardiac outcomes, through the CKM model (cardiovascular-kidney-metabolic, which reflects the commonly found combination of CKD, heart failure and diabetes), it would have been useful in this study to assess the endpoints of heart failure and diabetes (though this information was likely not available in the included studies)

-- SGLT-2 inhibitors are also renoprotective in patients with and without diabetes, though the studies do  not include people with advanced CKD (see https://www.kidneymedicinejournal.org/article/S2590-0595(23)00012-2/fulltext , which notes that the study including patients with the lowest eGFR was the EMPA-KIDNEY study that had a cutoff eGFR of 20. Of note, a substantial portion of patients did have a significant worsening of eGFR on starting the SGLT-2's which equilibrated after a few months and still had renoprotection)

 

So, a few issues:

-- the likely concern that many of us have felt about ACEi/ARBs in patients with severe CKD is that they can be associated with acute kidney injury and hyperkalemia, and given the fragility of the kidneys in those with advanced CKD, there is concern that these meds could make things worse

-- the dominant conclusion here is that ACEi/ARB therapy is in general beneficial to patients with advanced CKD. This study assessed initiation of these meds in such patients. A study assessing stopping these meds in patients who develop advanced CKD found a trend to harm

-- so, it seems reasonable to initiate ACEi/ARBs in those with severe CKD. But there are, to me, a few unanswered questions:

    -- there can be a decrease in eGFR on starting these meds. at what level of CKD decrease do we stop the ACEi/ARB? it is clear that at least at lower CKD levels, there can be an increase in creatinine associated with these meds of up to 30%, though there still is renal protection (as noted in the SGLT-2 study cited above). Should we just continue with ACEi/ARBs for 2-3 months or so to see if the renal function stabilized? at what point should we stop these meds?

    -- are ACEi/ARBs so beneficial that those who develop hyperkalemia should continue them anyway, with therapy for the hyperkalemia (increased loop diuretics/use of lokelma or other agents)??

    -- what about the use of GLP-1 agonists? are they as renoprotective as ACEi/ARBs but without the same potential for hyperkalemia/acute kidney injury (the next blog suggests that GLP-1s and SGLT-2's also seem to decrease potassium levels)? is there further value to adding these agents, both for renal protection and less hyperkalemia?? (especially GLP-1s, since SGLT-2s should be avoided in those with severe CKD)

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:

 

 go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would like to see the articles in this blog, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list