Diuretics and renin–angiotensin–aldosterone system inhibitors are central in the treatment of hypertension, but may cause serum potassium abnormalities. We examined mortality in relation to serum potassium in hypertensive patients. From Danish National Registries, we identified 44 799 hypertensive patients, aged 30 years or older, who had a serum potassium measurement within 90 days from diagnosis between 1995 and 2012. All-cause mortality was analysed according to seven predefined potassium levels: <3.5 (hypokalaemia), 3.5–3.7, 3.8–4.0, 4.1–4.4, 4.5–4.7, 4.8–5.0, and >5.0 mmol/L (hyperkalaemia). Outcome was 90-day mortality, estimated with multivariable Cox proportional hazard model, with the potassium interval of 4.1–4.4 mmol/L as reference. During 90-day follow-up, mortalities in the seven strata were 4.5, 2.7, 1.8, 1.5, 1.7, 2.7, and 3.6%, respectively. Adjusted risk for death was statistically significant for patients with hypokalaemia hazard ratio (HR): 2.80, 95% confidence interval (95% CI): 2.17–3.62, and hyperkalaemia (HR: 1.70, 95% CI: 1.36–2.13). Notably, normal potassium levels were also associated with increased mortality: K: 3.5–3.7 mmol/L (HR: 1.70, 95% CI: 1.36–2.13), K: 3.8–4.0 mmol/L (HR: 1.21, 95% CI: 1.00–1.47), and K: 4.8–5.0 mmol/L (HR: 1.48, 95% CI: 1.15–1.92). Thus, mortality in relation to the seven potassium ranges was U-shaped, with the lowest mortality in the interval of 4.1–4.4 mmol/L. Potassium levels outside the interval of 4.1–4.7 mmol/L were associated with increased mortality risk in patients with hypertension.