Impact of kidney function on cardiovascular outcome in patients with device-detected atrial fibrillation in NOAH-AFNET 6
A Goette, M Lemoine, A Fierenz, A Brandes, A J Camm, G A Dan, J R De Groot, G Y H Lip, A Lubinski, B Merkely, L Mont, U Schotten, S Willems, A Zapf, P KirchhofAbstract
Background and aims
The impact of kidney function on cardiovascular outcome in patients with device-detected atrial fibrillation (DDAF) is unknown.
Methods
This prespecified analysis of NOAH-AFNET 6 trial (n = 2534 patients) compared anticoagulation (edoxaban) to no anticoagulation in patients with DDAF, depending on the stages of chronic kidney disease (CKD, from G1 to G4) at baseline; patients with G5 CKD were excluded from the trial. Exploratory analyses estimated the effect of kidney function (estimated glomerular filtration rate - eGFR) as a continuous parameter on both primary outcomes. The primary efficacy outcome was a composite of stroke, systemic embolism, or cardiovascular death. The safety outcome was major bleeding or death.
Results
Efficacy outcome events increased with increasing severity of CKD: G1: 9 events/265 patient-years (3.4%), G2: 72/2854 (2.5%), G3a: 59/1245 (4.7%), G3b: 28/540 (5.2%) and G4: 16/147 (10.9%). Of note, stroke rate was low across all CKD stages: G1: 2/265 (0.8%), G2: 25/2883 (0.9%), G3a: 16/1253 (1.3%), G3b: 5/542 (0.9%) and G4: 1/147 (0.7%). The incidence of the safety outcome was affected to a lesser extent by the degree of CKD: G1: 12/258 (4.6%), G2: 110/2870 (3.8%), G3a: 78/1232 (6.3%), G3b: 46/534 (8.6%) and G4: 17/146 (11.6%). Edoxaban therapy did not interact with estimated glomerular filtration (eGFR) or CKD groups. A plot of the smoothed risk for the efficacy outcome event depending on the for patients with and without anticoagulation using locally estimated scatterplot smoothing (LOESS) suggests that, in patients with advanced kidney disease, anticoagulation could be more effective (see Figure A, within CKD G3b/4 HREdoxaban vs Placebo 0.50; 0.26-0.94; p = 0.032). A LOESS plot of the smoothed risk for the safety outcome event rates suggested their increase with reduced kidney function with no difference between treatment groups (see Figure B).
Conclusions
Age and eGFR predict cardiovascular events, in particular cardiovascular death, in patients with DDAF and clinical stroke risk factors. However, the incidence of cardiovascular events in this population is low. Pending external validation, eGFR may help to refine cardiovascular risk estimation in patients with DDAF.