The risk of thiazide-associated hyponatremia may be greatest in older women

1. Thiazide-associated hyponatremia is highly age and sex-dependent, with the greatest excess risk seen in older adults, especially women aged 80 years or older.

2. In younger adults, the excess absolute risk appears small, whereas in older women, the risk is large enough to justify considering alternative antihypertensives and closer sodium monitoring after initiation.

3. The risk of hyponatremia was greatest in the first days to weeks after starting a thiazide, but it remained elevated over time.

Evidence Rating Level: 2 (Good)

Study Rundown: This propensity score-matched cohort study used the Stockholm Sodium Cohort to compare the risk of hyponatremia after new thiazide initiation versus calcium channel blocker initiation in 159,080 adults. The primary outcome was profound hyponatremia, defined as serum sodium less than 125 mEq/L, with secondary thresholds of less than 130 mEq/L and less than 135 mEq/L. Over two years, profound hyponatremia occurred in 0.80% of thiazide users and 0.46% of calcium channel blocker users. Risk was strongly modified by age and sex. Among women aged 80 years or older, the two-year incidence of profound hyponatremia was 3.06% with thiazides, with a number needed to harm of 53, while the risk was much lower in younger women, with a number needed to harm of 790 for those younger than 65 years. The authors conclude that thiazide-associated hyponatremia is clinically important in older adults, especially women, and supports closer sodium monitoring or alternative antihypertensive therapy in this group.

Click to read the study in JAMA Network Open

Relevant Reading: A systematic review and meta-analysis of thiazide-induced hyponatraemia: time to reconsider electrolyte monitoring regimens after thiazide initiation?

In-Depth [cohort study]: 

Thiazide diuretics are a mainstay of antihypertensive therapy because they are effective, inexpensive, and generally well tolerated. Their use, however, is limited by the risk of hyponatremia, a clinically important adverse effect that may present with fatigue, confusion, nausea, headache, balance disturbance, or seizures. Previous studies have suggested that women and older adults are especially susceptible to thiazide-associated hyponatremia, but most have focused on relative rather than absolute risk. Absolute risk is more clinically useful because it helps prescribers judge whether the benefit of thiazide therapy outweighs the harm in specific patient groups. This study therefore sought to quantify the age- and sex-specific absolute risk of hyponatremia after thiazide initiation compared with calcium channel blocker initiation in a large, real-world population.

This was a population-based, propensity score–matched cohort study using the Stockholm Sodium Cohort, a linked regional database containing laboratory measurements, prescription data, diagnostic information, socioeconomic variables, and mortality records. Adults aged 18 years or older who newly initiated a thiazide or calcium channel blocker between 2006 and 2018 were eligible. New thiazide exposure required at least one year without prior thiazide dispensing. The primary outcome was profound hyponatremia, defined as a serum sodium concentration below 125 mEq/L. Secondary outcomes were sodium concentrations below 130 mEq/L and below 135 mEq/L. Sodium was corrected for glucose when available. Patients were matched 1:1 using propensity scores derived from age, sex, comorbidities, medications associated with hyponatremia, and socioeconomic characteristics. The investigators then calculated cumulative incidence, relative risk, absolute risk difference, and number needed to harm over follow-up periods extending to two years.

After matching, 79,540 patients initiating thiazides were compared with 79,540 initiating calcium channel blockers, with good balance in baseline characteristics between groups. Over two years, the cumulative incidence of profound hyponatremia was 0.80% among thiazide users compared with 0.46% among calcium channel blocker users. Although the overall excess risk was modest, the subgroup analyses showed marked heterogeneity. Risk rose substantially with age and was consistently higher in women. Among thiazide users, the two-year cumulative incidence of profound hyponatremia was 0.40% in those younger than 65 years, 1.05% in those aged 65 to 79 years, and 2.40% in those aged 80 years or older. The highest-risk group was women aged 80 years or older, in whom the cumulative incidence reached 3.06%. In contrast, the corresponding rate among women younger than 65 years was much lower.

The absolute risk estimates were the most clinically informative findings. In women aged 80 years or older, thiazide use produced a number needed to harm of 53 for profound hyponatremia, 28 for sodium below 130 mEq/L, and 16 for sodium below 135 mEq/L. This indicates a clinically meaningful excess risk in this subgroup. By contrast, among women younger than 65 years, the number needed to harm for profound hyponatremia was very high, indicating minimal excess absolute risk. The relative risk of hyponatremia was greatest early after treatment initiation, particularly during the first 14 days, although the excess risk remained elevated over longer follow-up.

The authors conclude that thiazide-associated hyponatremia is not a uniform adverse effect and should be interpreted through the lens of patient age and sex. In younger adults, especially those younger than 65 years, the excess risk appears negligible. In older adults, particularly women, the excess risk is substantial enough to influence treatment selection and follow-up planning. These findings support considering alternative antihypertensive therapy in high-risk patients and, when thiazides are used, counseling patients about symptoms of hyponatremia and considering regular serum sodium monitoring, especially during the early weeks and months after initiation. As an observational study, residual confounding cannot be excluded, but the large sample, active comparator design, and detailed matching strengthen the relevance of the findings for routine clinical practice.

Image: PD

©2026 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.