Abstract
INTRODUCTION
There are concerns that hydrochlorothiazide may increase the risk of incident nonmelanoma (cutaneous squamous cell carcinoma [cSCC], basal cell carcinoma [BCC]) and melanoma skin cancer, with regulatory agencies and societies calling for additional studies.
METHODS
We conducted a propensity score-matched population-based cohort study using the United Kingdom Clinical Practice Research Datalink. A total of 20,513 new users of hydrochlorothiazide were propensity score matched, in a 1:1 ratio, to new users of other thiazide diuretics between January 1, 1988 and March 31, 2018, with follow-up until March 31, 2019. Cox proportional hazards models were used to estimate hazard ratios (HRs) with 95% confidence intervals (CIs) for cSCC, BCC, and melanoma, comparing use of hydrochlorothiazide with use of other thiazide diuretics overall, by cumulative duration of use, and cumulative dose.
RESULTS
After an 8.6-year median follow-up, hydrochlorothiazide was associated with an increased risk of cSCC (HR 1.50, 95% CI 1.06-2.11). HRs increased with cumulative duration of use, with evidence of an association after 5-10 years (HR 2.10, 95% CI 1.20-3.67) and highest after > 10 years (HR 3.70, 95% CI 1.77-7.73). Similarly, HRs increased with cumulative dose, with higher estimates for ≥ 100,000 mg (HR 4.96, 95% CI 2.51-9.81). In contrast, hydrochlorothiazide was not associated with an increased risk of BCC (HR 1.01, 95% CI 0.91-1.13) or melanoma (HR 0.82, 95% CI 0.63-1.08), with no evidence of duration- or dose-response relationships.
CONCLUSIONS
Use of hydrochlorothiazide was associated with an increased risk of cSCC and with evidence of a duration- and dose-response relationship. In contrast, no association was observed for BCC or melanoma.
Collapse