Abstract Background and Aims The renin-angiotensin-aldosterone inhibitor (RASI) is one of the important medications for chronic kidney disease (CKD) with potential decrease of glomerular filtration rate (GFR) after initiation. There are few studies about the effect of GFR change by the RASI on variations of GFR and albuminuria. Method This observational cohort study was the extension of an open-label, case-controlled randomized clinical trial (NCT01552954) including adult hypertensive CKD patients with estimated GFR ≥30 ml/min/1.73 m2 and random urine albumin-to-creatinine ratio (UACR) ≥ 30 mg/g creatinine. In the trial phase, the angiotensin receptor blocker (ARB: Olmesartan 40 mg/day) was prescribed to participants after cessation of RASIs for 8 weeks of screening phase. (0-week). Participants had the second visit at 8 weeks after initiation of ARB (8-week) with randomization into the intensive education for low salt diet and the conventional education. Third visit was conducted at 16 weeks after 0-week (16-week). After 27 (19∼38) months later, we recruited participants for the cohort observation and finished at 38 (29∼48) months after 0-week visit. We grouped patients by tertiles of percent change of initial GFR during the first 8 weeks (from 0-week to 8-week) (pcGFRi) and compared the change of UACR from 24-hour urine and estimated GFR during trial and observational period. Results We analyzed data from 174 people finished 38-month observation among 235 participants enrolled for the trial phase. Initial GFR and 24-h urine UACR were 67 ± 24 ml/min/1.73 m2 and 829 ± 829 mg/g cr, respectively. There were −16.0 (−41.5∼−8.1) %, −2.8 (−7.3∼1.6) %, and 8.3 (1.8∼28.4) % pcGFRi in the first tertile group (group 1), the second tertile group (group 2), and the third tertile group (group 3), respectively. There were no differences in age, gender, GFR, 24-hour UACR, and 24-hour sodium excretion among the tertile groups at 0-week visit. The pcGFRi and the tertile group showed positive correlation with GFR change during 16 weeks of the trial, however, were not related to the percent change of 24-hour UACR during the trial phase. They were also positively correlated to the percent change of GFR during whole study period of 38 months (factor B = 8.180 and p-value< 0.001 for the group and factor B = 0.801 and p-value< 0.001 for pcGFRi), although, they did not show relationships with the percent change of 24-hour UACR during whole study period and UACR at last visit. The GFR slopes per year during 38 (29∼48) months were −2.9 (−4.0∼ −1.8), −1.3 (−2.4∼-0.3), and 0.1 (−1.0∼1.2) ml/min/1.73 m2/year in group 1, group 2, and group 3, respectively (p = 0.001 by the ANCOVA test). The relative risks to estimate GFR decrease ≥ 30% during whole study period were 0.237 (0.07∼0.798) (p = 0.020) and 0.035 (0.004∼0.302) (p = 0.035) in group 2 and group 3 compared to group 1, respectively. The AUC to estimate GFR decline ≥ 30% by the pcGFRi was 0.787 (0.685∼0.889, p< 0.001). The cut-off value 7.25% of pcGFRi estimates GFR decline ≥ 30% with 71.4% sensitivity and 70.9% specificity. Conclusion The decline of GFR by initiation of RASI predicts long-term renal event of GFR decline ≥ 30% in users of RASI, strongly. We need to compare the renal outcomes between RASI users with initial decrease of GFR and non-users to make guideline adjusting RASI usage according to initial change of GFR with RASI prescription.
Read full abstract