Digoxin has been used for centuries to treat heart failure and atrial fibrillation. Its continued widespread clinical use is primarily due to its proven ability to enhance myocardial contractility and slow heart rate. Therapeutic drug monitoring for digoxin began in the late 1960s with the introduction of radioimmunoassay. This was initially used mainly in cases of suspected poisoning or when patients failed to take their medication. In the 1970s, the therapeutic range for digoxin was established as 0.8–2.0 ng/mL. However, a 2003 retrospective study by Rathore et al. found that blood levels above 1.2 ng/mL were linked to higher mortality rates and little extra benefit. This led to a recommendation to lower the target range to 0.5–0.9 ng/mL. Nevertheless, many laboratory orders and textbooks still cite the old standard, and physicians are accustomed to using higher values. Combined with the atypical nature of poisoning symptoms, which are often overlooked, the risk persists. This study aims to evaluate the effectiveness of the newly proposed criteria by examining patient data from a recent period.

This study examined data on gender, age, digoxin concentrations and other information from 37,489 patients treated at Uppsala Hospital between 2004 and 2025. Until 2021, the Abbott method was used for digoxin testing; after this date, the Roche method was adopted.

Of all the patients, 17,771 were male and 19,718 were female. The median age was 77 years for males and 83 years for females. The median digoxin concentration was 0.9 nmol/L (interquartile range (IQR) 0.6–1.3) for males and 1.0 nmol/L (IQR 0.7–1.4) for females. Thirty percent of patients had digoxin concentrations greater than 1.2 nmol/L. As shown in Figure 1, the number of patients taking digoxin increased from 2004 to 2010, but then decreased until 2024. As shown in Figure 2, the median age of patients decreased slightly from 83 years in 2004 to 80 years in 2024.

Figure 1. Number of reported digoxin results per year.

Figure 2. Median patient age per year 2004-2024

Figure 3 shows that the median digoxin level and interquartile range decreased over time. The median digoxin level in 2004 was 1.1 nmol/L (interquartile range (IQR) 0.7–1.7 nmol/L), whereas in 2024 it was 0.8 nmol/L (IQR 0.6–1.2 nmol/L), representing a 27% decrease. Compared to 2004, the upper quartile decreased significantly more (29%) than the lower quartile (14%). Following the switch from Abbott to Roche testing methods in January 2021, digoxin levels increased by 15%. This methodological change likely contributed to the elevated results that year. Digoxin levels showed a positive correlation with age (Spearman correlation coefficient R= 0.063; p < 0.000001).

Figure 3. Median upper and lower quartiles for digoxin results per year

This study reviewed 20 years of digoxin blood concentration data in Sweden, revealing a consistent annual decline in median and upper quartile concentrations. This trend is consistent with the subsequent lowering of target concentrations to 0.5–0.9 ng/mL in European and American guidelines, as proposed by the 2003 Rathore study. This study found that concentrations exceeding 1.2 ng/mL (approximately 1.5 nmol/L) were associated with increased mortality. Nevertheless, approximately 30% of samples still exceeded 1.5 nmol/L, suggesting ongoing challenges in the implementation of these guidelines. Concentrations showed a negative correlation with year and a weak positive correlation with patient age, reflecting how declining renal function and polypharmacy predispose older patients to higher levels. The decreasing average patient age also contributed to the overall downward trend. Seasonal variations had a minimal impact on the results.

This study has several limitations, including a lack of clinical information such as indications, renal function, concomitant medications and blood sampling times. The authors suggest that the persistently high proportion of patients exceeding therapeutic ranges indicates a lack of awareness among physicians of the relevant guidelines, as well as delayed dose adjustments due to atypical manifestations of toxicity. This highlights the need for enhanced guidance and management.