SAN FRANCISCO – Starting digoxin treatment in patients with newly diagnosed atrial fibrillation was linked with a doubled risk of death during the subsequent 8 months, on the basis of an observation of more than 23,000 patients in a California health system.
"This is a very big deal. We will tell Kaiser physicians not to prescribe digoxin to patients with atrial fibrillation," Dr. James V. Freeman said while presenting a poster at the annual meeting of the American College of Cardiology.
Although the mechanism is not known, he said he suspects it may involve arrhythmia generation and sudden death. The hospitalization rate among patients who began digoxin ran the same as in those not on the drug, even though mortality doubled.
While not widely used, digoxin received significant interest from the treating physicians in California during January 2006 through June 2009, with 13% of newly diagnosed atrial fibrillation patients receiving prescriptions. The analysis excluded patients who were already on digoxin at the time of their incident atrial fibrillation, so "these were not just physicians keeping patients on digoxin for years," Dr. Freeman said in an interview.
Existing recommendations from the ACC, last updated in 2012, list digoxin as one of eight drug choices for heart rate and rhythm control in atrial fibrillation patients. The college lists the eight drugs alphabetically and does not rate any drug as better or worse than the others.
Mitchel L. Zoler/IMNG Medical Media
Dr. James V. Freeman
The study presented by Dr. Freeman involved 23,272 adults aged 21 years or older with an index diagnosis of atrial fibrillation in the Kaiser Permanente Northern California and Southern California health plans during the 3.5-year period examined. Among these patients who had never used digoxin before and who were without heart failure, 2,997 filled at least one prescription for the drug.
During a median follow-up of 10 months, the mortality rate was 9.49/100 person-years of follow-up among those who received digoxin, versus 4.27/100 person-years in nonusers. After adjusting for a list of possible demographic and clinical confounders, the investigators found that patients who received digoxin had a 2.06 higher mortality rate, a statistically significant difference, reported Dr. Freeman, a cardiologist at Stanford (Calif.) University. Subgroup analyses showed similar increased mortality rates with digoxin use across both women and men and among patients younger than 75 years, 75-84, and 85 years or older.
The hospitalization rate was roughly 3 cases for every 100 person-years of follow-up in both groups.
Dr. Freeman conceded that as an observational study, the analysis was vulnerable to possible confounding factors, but in this case he thought that the risk of significant confounding was minimal. The Kaiser system collects voluminous data on comorbidities and other treatments that patients received, and the pattern showed a slight tilt toward sicker patients with more comorbidity among those who did not receive digoxin. For example, the nonusers had a 76% prevalence of hypertension, a 4% prevalence of prior coronary stenting, and a 4% rate of prior stroke, compared with prevalence rates of 74%, 3%, and 3%, respectively, among digoxin recipients. However, the digoxin recipients had slightly higher rates of cancer and lung disease, 7% compared with 6%, and 23% vs. 20%, respectively.
The patients not on digoxin also had slightly higher rates of medication with other potentially beneficial drugs. For example, treatment with an ACE inhibitor occurred in 31% of digoxin nonusers compared with 29% of the digoxin users; beta-blocker treatment occurred in 46% of digoxin nonusers, compared with 40% of the digoxin users.
Dr. Freeman said he had no relevant financial disclosures.
On Twitter @mitchelzoler