Jose M. Villalon-Gomez, MD, MPH
In patients with hypertension, are thiazides associated with increased risk for hyponatremia compared with nonthiazide treatment?
Retrospective cohort study using data from the Research Patient Data Registry and Social Security Death Index. Follow-up was ≤ 10 years.
2 academic hospitals and affiliated clinics in Massachusetts, USA.
2613 adult outpatients (mean age 62 y, 51% men) with hypertension who had a first prescription for an acceptable antihypertensive medication and continued to receive antihypertensive prescriptions throughout the follow-up period. Patients who had a previous diagnosis of hyponatremia, died ≤ 30 days after study enrollment, or took an unacceptable antihypertensive agent (e.g., clonidine, hydralazine) were excluded. 220 patients (thiazide-exposed cohort) were new users of thiazides (96% hydrochlorothiazide), and 2393 patients (non–thiazide-exposed cohort) had not received a prescription for a thiazide but received another acceptable antihypertensive drug (angiotensin-converting enzyme inhibitor, angiotensin II–receptor blocker, β-adrenergic blocker, or calcium-channel blocker) and no thiazide during follow-up.
Hyponatremia (serum sodium level ≤ 130 mmol/L). Secondary outcomes included hyponatremia-associated hospitalizations (hospitalizations with laboratory evidence of hyponatremia on the day of admission) and all-cause mortality.
Thiazide exposure was associated with increased risk for hyponatremia but not hyponatremia-related hospitalizations or all-cause mortality (Table). The estimated number needed to harm for hyponatremia over 5 years was 16 (95% CI 8 to 161).
In patients with hypertension, thiazides were associated with increased risk for hyponatremia compared with nonthiazide treatment, but not hospitalizations related to hyponatremia.
Associations between thiazides and hyponatrenmia or mortality in patients with hypertension*
*Adjusted for age, sex, ethnicity, congestive heart failure, chronic kidney disease, cancer, Charlson comorbidity index, and nonsteroidal antiinflammatory drugs.
†Hospitalizations with laboratory evidence of hyponatremia on the day of admission.
Villalon-Gomez JM. Thiazides increased hyponatremia, but not hospitalizations, in patients with hypertension. Ann Intern Med. ;156:JC4–11. doi: 10.7326/0003-4819-156-8-201204170-02011
Download citation file:
Published: Ann Intern Med. 2012;156(8):JC4-11.
Results provided by:
Copyright © 2019 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use