Effects of antihypertensives, lipid-modifying drugs, glycaemic control drugs and sodium bicarbonate on the progression of stages 3 and 4 chronic kidney disease in adults: a systematic review and meta-analysis

Abstract

Objective: To evaluate the effects of drug interventions that may modify the progression of chronic kidney disease (CKD) in adults with CKD stages 3 and 4.

Design: Systematic review and meta-analysis.

Methods: Searching MEDLINE, EMBASE, Database of Abstracts of Reviews of Effects, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, International Clinical Trials Registry Platform, Health Technology Assessment, Science Citation Index, Social Sciences Citation Index, Conference Proceedings Citation Index and Clinical Trials Register, from March 1999 to July 2018, we identified randomised controlled trials (RCTs) of drugs for hypertension, lipid modification, glycaemic control and sodium bicarbonate, compared with placebo, no drug or a drug from another class, in ≥40 adults with CKD stages 3 and/or 4, with at least 2 years of follow-up and reporting renal function (primary outcome), proteinuria, adverse events, maintenance dialysis, transplantation, cardiovascular events, cardiovascular mortality or all-cause mortality. Two reviewers independently screened citations and extracted data. For continuous outcomes, we used the ratio of means (ROM) at the end of the trial in random-effects meta-analyses. We assessed methodological quality with the Cochrane Risk of Bias Tool and confidence in the evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.

Results: We included 35 RCTs and over 51 000 patients. Data were limited, and heterogeneity varied. Final renal function (estimated glomerular filtration rate) was 6% higher in those taking glycaemic control drugs (ROM 1.06, 95% CI 1.02 to 1.10, I²=0%, low GRADE confidence) and 4% higher in those taking lipid-modifying drugs (ROM 1.04, 95% CI 1.00 to 1.08, I²=88%, very low GRADE confidence). For RCTs of antihypertensive drugs, there were no significant differences in renal function. Treatment with lipid-modifying drugs led to a 36% reduction in cardiovascular disease and 26% reduction in all-cause mortality.

Conclusions: Glycaemic control and lipid-modifying drugs may slow the progression of CKD, but we found no pooled evidence of benefit nor harm from antihypertensive drugs. However, given the data limitations, further research is needed to confirm these findings.

Prospero registration number: CRD42015017501.

Keywords: Cardiovascular events; Chronic renal failure; Drug treatment; Meta-analysis; Renal function.

Figure 1

PRISMA diagram. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses. CKD, chronic kidney disease; MDRD, Modification of Diet in Renal Disease.

Figure 2

ROM of estimated glomerular filtration rate at the end of the trials for antihypertensives versus comparator (boxes) and pooled estimates across studies (diamonds) calculated by the random-effects DerSimonian and Laird method, split by CKD stage. CHARMa and CHARMb, CKD stages 3 and 4, respectively, with estimated glomerular filtration rate measured by the MDRD equation. IDNTa, intervention is irbesartan; IDNTb, intervention is amlodipine. CKD, chronic kidney disease; MDRD, Modification of Diet in Renal Disease; ROM, ratio of means.

Figure 3

ROM of estimated glomerular filtration rate at the end of the trials for lipid-modifying drugs versus comparator (boxes) and pooled estimates across studies (diamonds) calculated by the random-effects DerSimonian and Laird method, split by CKD stage. ASCOT-LLAa, atorvastatin and amlodipine versus placebo and amlodipine; ASCOT-LLAb, atorvastatin and atenolol versus placebo and atenolol; GISSI-HFa, CKD stage 3; GISSI-HFb, CKD stage 4; GREACEa, atorvastatin versus usual care for the group with metabolic syndrome; GREACEb, atorvastatin versus usual care for the group without metabolic syndrome; SHARPa, CKD stages 1–3; SHARPb, CKD stage 4. CKD, chronic kidney disease; ROM, ratio of means; SHARP, Study of Heart and Renal Protection.

Figure 4

ROM of estimated glomerular filtration rate at the end of the trials for glycaemic control drugs versus comparator (boxes) and pooled estimates across studies (diamonds) calculated by the random-effects DerSimonian and Laird method, split by CKD stage. Bode 2015L, intervention is low-dose canagliflozin (100 mg per day); Bode 2015H, intervention is high-dose canagliflozin (300 mg per day); EMPA-REG H, intervention is high-dose empagliflozin (25 mg daily); EMPA-REG L, intervention is low-dose empagliflozin (10 mg daily); Kohan 2014L, intervention is low-dose dapagliflozin (5 mg per day); Kohan 2014H, intervention is high-dose dapagliflozin (10 mg per day); CANVASa, CKD stage 3a; CANVASb, CKD stage 3b; TECOSa, CKD stage 3a; TECOSb, CKD stage 3b. CKD, chronic kidney disease; ROM, ratio of means.

Figure 5

Effect of treatment on estimated glomerular filtration rate by drug class: CKD stages 3 and 4 only. ROM >1 favoured the intervention group (higher values are better). ‘Studies’ refers to the study of different populations. Of the 35 included studies, 15 reported renal function of patients with CKD stage 3 and/or stage 4. Of these, 9 reported renal function of a single population and 6 reported results for two CKD subgroups or dose-dependent subgroups. CKD, chronic kidney disease; ROM, ratio of means.

Figure 6

Relative risk (RR) of cardiovascular disease during trials of lipid-modifying drugs versus comparator (boxes) and pooled estimates across trials (diamonds) calculated by the random-effects DerSimonian and Laird method, split by CKD stage. GREACEa, atorvastatin versus usual care for the group with metabolic syndrome; GREACEb, atorvastatin versus usual care for the group without metabolic syndrome. CKD, chronic kidney disease.

Figure 7

Relative risk (RR) of all-cause mortality during trials of lipid-modifying drugs versus comparator (boxes) and pooled estimates across trials (diamonds) calculated by the random-effects DerSimonian and Laird method, split by CKD stage. GREACEa, atorvastatin versus usual care for the group with metabolic syndrome; GREACEb, atorvastatin versus usual care for the group without metabolic syndrome. CKD, chronic kidney disease.