Background/Aim: There is an increasing prevalence of chronic heart failure (HF). It is well known that patients with HF and disturbances in the potassium level have an increased mortality risk. The aim of this study was to investigate the prognosis of a second plasma-potassium measurement after an episode with hyperkalaemia on short-term mortality in patients with chronic HF. Methods and results: From Danish national registers, 2,339 patients with chronic HF and hyperkalaemia (>4.6 mmol/L) at first potassium measurement within 14–365 days from concomitant treatment were identified. To be included, a second measurement was required within 6–30 days subsequent to the first measurement and the 60-day mortality was observed. Based on the second measurement, the patients were divided into five groups: <3.5 mmol/L (n = 257), 3.5–4.0 mmol/L (n = 709), 4.1–4.6 mmol/L (n = 1,204, reference), 4.7–5.0 mmol/L (n = 89) and >5.0 mmol/L (n = 80). To assess all-cause and cardiovascular mortality, we used the Cox regression model. The multivariable analysis showed that patients with potassium concentrations <3.5 mmol/L (hazard ratio (HR): 3.03; 95% CI: 2.49–3.70) and 3.5–4.0 mmol/L (HR: 1.81; 95% CI: 1.54–2.14) had a worse prognosis compared to the reference. We observed similar results when calculating the risk of cardiovascular mortality. A restricted cubic spline curve showed a U-shaped relationship between plasma-potassium and all-cause mortality. Conclusion: Patients with chronic HF and hyperkalaemia who became hypokalaemic after 6–30 days were associated with a higher 60-day all-cause and cardiovascular mortality compared to the reference. This also applied for patients with low normal potassium concentrations (3.5–4.0 mmol/L).