Incident Atrial Fibrillation Increases Risk of Adverse Events in Patients with Decreased eGFR

Chronic kidney disease (CKD), a common medical problem with a high disease burden, is associated with increased risk for hospitalization, adverse cardiovascular outcomes, and death. The increased risks for adverse cardiovascular outcomes include an increase in the risk for atrial fibrillation (AF). Worldwide, AF affects approximately 33 million people and is the most common arrhythmia. The incidence of AF continues to increase, negatively affecting quality of life and creating increases in healthcare costs.

The development of cardioembolic stroke is common among patients with AF, and there are evidence-based guidelines in use for therapeutic interventions among patients with no CKD. AF is also associated with increased risk for other cardiovascular outcomes including congestive heart failure (CHF) and myocardial infarction (MI); AF is also associated with the development of CKD.

Earlier studies of AF have focused on patients with end-stage kidney disease (ESRD), a population with an increased incidence of AF and the associated risk for adverse outcomes. David Massicotte-Azarniouch, MD, and colleagues recently conducted a population-based retrospective cohort study in Ontario, Canada to examine the association between AF and cardiovascular outcomes (CHF and MI), ESRD, and all-cause in mortality patients with a decrease in estimated glomerular filtration rate (eGFR). The researchers also sought to determine whether different categories of eGFR would influence those associations and to test the hypothesis that AF would be associated with higher risk for adverse events in patients with low eGFR and the risk would vary by eGFR level. Results were reported in the American Journal of Kidney Diseases [2018;71(2):191-199].

Using linked databases to ascertain patient characteristics, the researchers identified patients with at least one outpatient eGFR obtained from April 1, 2006, to March 31, 2015, and then looked forward 12 months for first evidence of AF. Only patients with eGFRs <90 mL/min/1.73 m2 were included. The index date for each patient was the first eligible incident AF diagnosis. A random index date for patients who did not develop AF during the 12-month index period was assigned based on the distribution of index dates. Exclusion criteria included <18 years of age, any history of kidney transplantation, evidence of dialysis, or evidence of CHF or AF 5 years prior to the patient’s index date.

Estimated glomerular filtration rate was calculated using the CKD Epidemiology Collaboration creatinine equation and was categorized as 60 to <90, 30 to <60, and <30 mL/min/1.73 m2. Incident clinically identified AF was defined by a single International Classification of Diseases or billing code on first diagnosis during hospitalization, during a visit to the emergency department, or during an ambulatory office visit. Patients were followed up forward in time until receipt of dialysis, kidney transplantation, all-cause mortality, CHD, MI, or end of the study follow-up period (March 31, 2015).

Using propensity scores, eGFRs, and index dates, the final study cohort matched 93,414 individuals with CKD with AF to 93,414 individuals with CKD without AF. Prior to matching, patients with AF were older, had more comorbid conditions (diabetes, hypertension, stroke, and cardiovascular disease), and had greater use of healthcare (hospitalizations, visits to the emergency department, and physician visits) compared with participants without AF. Following matching, the two groups were balanced for all measured covariates with no detected statistically significant differences.

During total follow-up time of 795,131 person-years, the incidence of all study outcomes was higher among patients in the AF group; this was pronounced during the first 6 months following the AF diagnosis. The subdistribution hazard ratios (sHRs) within 6 months were 11.57 (95% confidence interval [CI], 10.26-13.05) for CHF; 4.76 (95% CI, 4.17-5.35) for MI; and 5.69 (95% CI, 3.97-8.16) for ESRD. The HR for mortality was 2.62 (95% CI, 2.50-2.76).

The risk for all outcomes attenuated 6 months following the AF diagnosis, but remained elevated: CHF, sHR, 2.64 (95% CI, 2.55-2.74); MI, sHR 1.24 (95% CI, 1.18-1.30); ESRD, sHR, 1.75 (95% CI, 1.57-1.95); and mortality, HR, 1.07 (95% CI, 1.04-1.10).

Analysis revealed that eGFR as an effect modifier on the association between AF, and CHF, MI, and mortality (P<.001, P=.04, P=.05, respectively) within the 6 months following the diagnosis of AF. Due to too few events at certain eGFRs, there was no analysis of ESRD for effect modification within the first 6 months. The risk for CHF, MI, and morality was highest among participants with eGFRs of 60 to <90 mL/min/1.73 m2 within the first 6 months after AF diagnosis: sHR for CHF, 14.03 (95% CI, 11.69-16.85) and for MI, 5.71 (95% CI, 4.74-6.86); HR for mortality, 2.68 (95% CI, 1.73-4.49). Six months after the AF diagnosis, eGFR was an effect modifier for CHF only (P<.001); the highest risk remained among those with eGFRs of 60 to <90 mL/min/1.73 m2.

The researchers cited some limitations to the study, including the observational design of the study that may have allowed residual confounding; the lack of data for medication use and treatments for AF; the use of a single outpatient serum creatinine value to define reduced eGFR with no inclusion of albuminuria; and exclusion of patients with a history of AF and CHF.

The researchers summarized by saying, “Patients with reduced eGFRs have a high incidence of AF and the development of AF is a risk factor for adverse events. A diagnosis of AF is associated with 11.6-, 4.8-, 5.7-, and 2.6-fold higher risks for CHF, MI, ESRD, and all-cause mortality, respectively, in the short term, and this attenuated but persisted in the long term compared with those without AF. Given the high burden of disease of patients with AF and reduced eGFRs, future areas of research should examine the mechanisms by which AF leads to adverse events and possible therapeutic means of prevention.”

Takeaway Points

  1. Researchers in Canada conducted a population-based retrospective cohort study to examine the association between atrial fibrillation (AF), estimated glomerular filtration rate (eGFR), and adverse events.
  2. Outcomes of interest were congestive heart failure, myocardial infarction, end-stage renal disease, and all-cause mortality.
  3. There was an association between incident AF and a high risk for adverse outcomes among patients with eGFRs <90 mL/min/1.73m2, particularly during the first 6 months after the AF diagnosis.