Hydrochlorothiazide: Discovery, Development, Safety and Pharmacovigilance
Hydrochlorothiazide is a thiazide diuretic used in the treatment of hypertension and oedema. It is inexpensive, familiar and frequently present in fixed-dose antihypertensive combinations. Its long clinical history should not obscure the need for careful pharmacovigilance: disturbances of sodium and potassium balance, metabolic effects, idiosyncratic ocular reactions and an association between cumulative exposure and non-melanoma skin cancer create distinct surveillance problems. Some risks emerge soon after starting treatment; others depend on years of exposure and are best evaluated using pharmacoepidemiology rather than isolated reports.
- Hydrochlorothiazide: Discovery, Development, Safety and Pharmacovigilance
- The discovery and early development of thiazide diuretics
- Pharmacology and clinical rationale
- Clinical evidence and comparative-effectiveness boundaries
- From long-established use to modern safety questions
- Electrolyte, renal and metabolic adverse reactions
- Acute myopia and secondary angle-closure glaucoma
- Photosensitivity and non-melanoma skin cancer
- Interactions, special populations and medication errors
- Distinguishing the major safety time courses
- Operational pharmacovigilance framework
- Practical checklist
- Key takeaways
- References
- Regulatory Note
The discovery and early development of thiazide diuretics
The first modern thiazide diuretics emerged from mid-twentieth-century medicinal chemistry research on sulfonamide-derived compounds and renal electrolyte transport. Chlorothiazide was an early member of the class. Hydrochlorothiazide, its 3,4-dihydro derivative, was developed as a more potent orally active agent. Hydrochlorothiazide entered clinical use in 1959; the US National Institutes of Health's LiverTox history dates its US approval to 1959, not 1958 [9]. The commercial story is not attributable to one uncontested inventor or manufacturer. A US Federal Trade Commission historical review documents competing patent applications by Ciba and Merck, later royalty-free cross-licensing, and 1959 US marketing by Ciba (Esidrix), Merck (Hydroduril) and Abbott (Oretic) [10]. Contemporary 1959 clinical reports already compared hydrochlorothiazide with chlorothiazide, confirming that the compound's therapeutic profile was being evaluated during its introduction [13]. Its subsequent adoption occurred alongside recognition that lowering blood pressure could prevent cardiovascular complications, although those outcome benefits must be attributed to the relevant antihypertensive trial regimens and populations rather than assumed from an individual drug's market longevity.
The compound is generally marketed as a generic active substance, both alone and in combinations with other antihypertensives. This creates a product-identification challenge: a combination tablet may expose the patient to hydrochlorothiazide even when the reporter remembers only the brand or the other active ingredient. A lifecycle article about the hydrochlorothiazide INN must therefore distinguish its own pharmacology and safety from the effects of angiotensin receptor blockers, ACE inhibitors, beta blockers and other combination partners.
Pharmacology and clinical rationale
Site of action
Hydrochlorothiazide inhibits sodium–chloride reabsorption in the distal convoluted tubule, increasing urinary sodium and chloride excretion and promoting water loss. Potassium loss can accompany enhanced downstream sodium delivery. The effect on blood pressure cannot be reduced to urine volume alone: chronic antihypertensive responses involve additional haemodynamic adaptation, and the precise contribution of different mechanisms is not completely resolved in product labelling [2].
The March 2026 US tablet label describes onset of diuresis within approximately two hours, peak effect around four hours and a duration of six to twelve hours [2]. These pharmacodynamic timings are useful for evaluating early dizziness, dehydration or dosing errors, but do not determine the latency of all adverse reactions. Electrolyte abnormalities can develop later, particularly during intercurrent illness, changes in fluid intake or co-medication.
Indications and product boundaries
US hydrochlorothiazide labels describe hypertension as monotherapy or as an adjunct to other antihypertensives, and oedema associated with conditions including congestive heart failure, hepatic cirrhosis and certain renal disorders [2,3]. These are product-label indications; individual combination products have their own authorised populations, contraindications and dose instructions. A jurisdiction's prescribing guideline is not identical to a specific product's marketing authorisation.
The balance between hydrochlorothiazide and longer-acting thiazide-like agents is a clinical guideline question rather than evidence that the substances are interchangeable. Trials involving chlorthalidone or indapamide must not be relabelled as direct hydrochlorothiazide outcome trials. Similarly, combination-trial benefits should not be allocated entirely to the hydrochlorothiazide constituent.
Clinical evidence and comparative-effectiveness boundaries
The evidentiary base for hydrochlorothiazide is unusually heterogeneous. Early clinical work assessed diuresis and blood-pressure reduction, often before modern large outcome trials became routine. Later hypertension studies evaluated different thiazides, combination regimens, or the consequences of switching between established treatments. A pharmacovigilance reference should explain what those trials tested instead of implying that every thiazide cardiovascular outcome is direct evidence for hydrochlorothiazide.
ACCOMPLISH: an active-comparator combination trial
In ACCOMPLISH, 11,506 patients with hypertension at high cardiovascular risk were randomised to benazepril plus amlodipine or benazepril plus hydrochlorothiazide. The primary cardiovascular composite occurred in 9.6% and 11.8%, respectively, favouring the benazepril–amlodipine combination (hazard ratio 0.80; 95% CI 0.72–0.90) [11]. The finding is clinically important, but it does not establish that hydrochlorothiazide monotherapy lacks benefit, nor that amlodipine is universally preferable in every patient. Both study arms received benazepril; the tested question concerned two combination strategies in a defined high-risk population. The relatively low hydrochlorothiazide doses and differences in pharmacokinetics among diuretics also limit extrapolation to every thiazide regimen.
Diuretic Comparison Project: no superiority demonstrated for switching
The pragmatic Diuretic Comparison Project randomised 13,523 predominantly older US Veterans Affairs patients already receiving hydrochlorothiazide to continue it or switch to chlorthalidone. Over a median 2.4 years, the primary cardiovascular composite occurred in 10.0% of the hydrochlorothiazide group and 10.4% of the chlorthalidone group (hazard ratio for chlorthalidone versus hydrochlorothiazide 1.04; 95% CI 0.94–1.16; P=0.45) [12]. Hypokalaemia was more frequent after switching to chlorthalidone (6.0% versus 4.4%). This study did not demonstrate chlorthalidone superiority; absence of a significant difference is not, by itself, proof of equivalence in every clinical subgroup. The study's older, predominantly male, previously hydrochlorothiazide-treated population matters when considering generalisability.
How to interpret indirect evidence
ALLHAT's diuretic arm used chlorthalidone, not hydrochlorothiazide. Its outcome findings should therefore be identified as thiazide-like-diuretic evidence rather than mislabelled as a hydrochlorothiazide-specific trial. Equally, blood-pressure reduction is an intermediate effect, whereas myocardial infarction, stroke, heart failure and mortality are patient-important outcomes. Differences in population, comparator, dose and duration explain why apparently inconsistent findings across ACCOMPLISH and the Diuretic Comparison Project need not represent a contradiction. For a patient already controlled on a combination product, clinical decisions must account for the entire regimen, tolerability and comorbidities, not only the hydrochlorothiazide component.
From long-established use to modern safety questions
Early safety experience concentrated on diuresis, hypotension, renal function and electrolyte disturbances. As cumulative exposure expanded, rare idiosyncratic events and long-latency outcomes became more visible. In 2018 the European Medicines Agency's Pharmacovigilance Risk Assessment Committee recommended product-information changes concerning non-melanoma skin cancer after epidemiological evidence of a cumulative-dose association [4]. The US FDA approved corresponding labelling changes on 20 August 2020 following an analysis in its Sentinel surveillance system [15]. These developments demonstrate that mature generic products can undergo important safety reappraisal decades after introduction.
The IARC Working Group announced its Group 1 carcinogenic-hazard evaluation in November 2024 and published the full Volume 137 monograph in February 2026 [1,7,14]. It found sufficient evidence in humans for squamous cell carcinoma of the skin and cancer of the lip; evidence for basal cell carcinoma and several other skin tumours was more limited [14]. This classification addresses strength of evidence for carcinogenic hazard; it does not quantify the absolute risk for a particular patient, mean every exposure causes cancer, or by itself direct universal treatment discontinuation. The dose, duration, skin phenotype, ultraviolet exposure and underlying cardiovascular indication remain relevant to clinical benefit–risk decisions.
Electrolyte, renal and metabolic adverse reactions
Hyponatraemia, hypokalaemia and volume depletion
Hydrochlorothiazide can cause hyponatraemia, hypokalaemia and hypochloraemic alkalosis. The risk is affected by renal function, age, diet, fluid intake, intercurrent vomiting or diarrhoea and other medicines. Symptoms may be nonspecific—weakness, fatigue, confusion, cramps, dizziness or falls—and serious cases may involve seizures or arrhythmias. A pharmacovigilance report should record baseline and serial sodium, potassium, creatinine, clinical volume status, dose changes and co-medications. The severity of a low laboratory value and the seriousness of its clinical consequences are related but distinct concepts.
Thiazide-associated hyponatraemia is especially important in older patients and those taking other medicines that influence water balance. For a suspected event, timing relative to initiation or dose increase, fluid intake, infection, other diuretics and medications such as antidepressants may help establish competing explanations. A report of a low sodium concentration without its date, laboratory range, symptoms or follow-up cannot support a robust assessment of causality or preventability.
Hypokalaemia can increase susceptibility to cardiac arrhythmia and enhance digitalis-related toxicity. Conversely, potassium-sparing co-therapy can change the risk profile and must be recorded rather than presumed protective in every patient. The US product information advises periodic serum electrolyte measurement at appropriate intervals, not one universal monitoring schedule for all indications and risk groups [2,3].
Kidney function and other metabolic effects
Volume depletion may cause prerenal kidney dysfunction, while severe underlying renal disease can alter the benefit–risk profile of a thiazide. Hyperuricaemia may precipitate gout; impaired glucose tolerance and other metabolic changes may be clinically relevant. In hepatic impairment, even modest fluid or electrolyte changes can worsen encephalopathy risk. These labelled precautions require individualised clinical evaluation; they do not establish that every laboratory change during therapy was caused by hydrochlorothiazide.
A useful case narrative should include the clinical indication, renal and hepatic baseline status, other diuretics, blood-pressure changes, laboratory trajectory, concurrent illness and response to intervention. Medication reconciliation must identify combination products, which may otherwise lead to duplicate thiazide exposure.
Acute myopia and secondary angle-closure glaucoma
Hydrochlorothiazide is a sulfonamide-derived drug associated with an idiosyncratic syndrome of acute transient myopia and secondary angle-closure glaucoma. US labelling describes sudden visual deterioration or ocular pain, often within hours to weeks of starting treatment. Untreated acute angle closure can threaten permanent vision loss [3]. This is not the same mechanism as the gradual development of chronic primary open-angle glaucoma.
Case assessment should capture visual symptoms, laterality, intraocular pressure, ocular examination findings, start and stop dates, other medications and treatment outcome. Product labelling directs rapid discontinuation of hydrochlorothiazide and urgent appropriate ophthalmic management. Pharmacovigilance teams should not turn this into a generic instruction for patients to change treatment without clinician involvement; it is a time-sensitive clinical safety concern requiring immediate assessment.
Photosensitivity and non-melanoma skin cancer
What the epidemiology established
Hydrochlorothiazide has photosensitising properties. Danish registry-based case–control research found a cumulative-dose association with basal cell carcinoma (BCC) and squamous cell carcinoma (SCC), stronger for SCC. At cumulative exposure of at least 50,000 mg, the reported adjusted odds ratios were 1.29 (95% CI 1.23–1.35) for BCC and 3.98 (95% CI 3.68–4.31) for SCC [5]. These are odds ratios in a particular observational study, not the probability that an individual patient will develop cancer. The authors noted lack of information on sun exposure as a limitation.
The 2018 EMA PRAC recommendation described the association, suggested photosensitisation as a possible mechanism and recommended updating product information [4]. UK MHRA advice similarly emphasised sun protection, checking for suspicious lesions and reconsidering hydrochlorothiazide use in patients with prior non-melanoma skin cancer [6]. These are risk-minimisation recommendations, not proof that a specific skin tumour was caused by the medicine.
Absolute risk and generalisability
The US Sentinel analysis incorporated into later US labels estimated an additional SCC case of approximately one per 16,000 patients per year in the overall studied population and approximately one per 6,700 per year among white patients with cumulative exposure of at least 50,000 mg [3,8]. The latter estimate is not a risk for every patient receiving 50,000 mg: it is an absolute excess rate in the studied subgroup, based on observational comparisons. Absolute estimates depend on the source population and cannot be applied without qualification to patients with different ancestry, skin type, UV exposure or baseline risk. Cumulative dose should be calculated from actual exposure when feasible, not inferred from the date of first prescription alone.
IARC's Group 1 assessment, announced in 2024 and published in full in 2026 [1,14], is a hazard identification conclusion. The Working Group judged evidence sufficient for cutaneous squamous cell carcinoma and lip cancer, while evidence for basal cell carcinoma was limited [14]. Clinical management still requires assessment of cardiovascular benefit, alternative treatments, skin cancer history and patient preferences. Neither an epidemiological association nor a hazard classification automatically justifies stopping antihypertensive treatment without a replacement plan.
Safety reporting and follow-up
For a suspected skin-cancer case, obtain pathology and anatomical site, BCC/SCC subtype, date of diagnosis, prior lesions, dermatological history, cumulative hydrochlorothiazide dose, ultraviolet exposure where known, immunosuppression and other relevant risk factors. The latency may span years, so conventional short-term dechallenge reasoning is generally uninformative. Case-series counts without exposed-person denominators cannot estimate incidence or quantify an exposure–response relationship.
Interactions, special populations and medication errors
Clinically important interaction patterns
Lithium is a particularly important co-medication because thiazide-induced changes in sodium handling can reduce lithium clearance and raise toxicity risk. Product information advises that lithium generally should not be given with diuretics [3]. Concurrent digoxin may become more hazardous in the presence of hypokalaemia. Corticosteroids can worsen potassium loss, while other antihypertensives may intensify hypotension. Some nonsteroidal anti-inflammatory drugs may blunt diuretic or antihypertensive effects and complicate renal function. The magnitude of any interaction depends on the patient, dose and co-exposures; a case should not be classified as an interaction solely because two products were prescribed together.
Fixed-dose combinations create two distinct errors: failing to recognise that hydrochlorothiazide is present, and attributing all reported effects to the combination's better-known active ingredient. Source verification should use the exact pack, strength, product identifier and dispensing history. A suspected overdose or duplicate therapy report should include every hydrochlorothiazide-containing prescription.
Pregnancy, lactation and older patients
Thiazides cross the placenta and appear in breast milk. Product-specific labelling describes fetal or neonatal concerns and advises individualised decisions concerning nursing [3]. Exposure during pregnancy should be assessed by trimester, indication, maternal disease, dose, co-treatment and outcome; it should not be presumed either harmless or causative of an adverse pregnancy outcome.
Older patients may be more susceptible to hypotension, dehydration, electrolyte disturbances and falls. These effects are particularly difficult to assess when several antihypertensives, sedatives or medicines affecting sodium balance are taken together. Risk management should emphasise the actual clinical context rather than an age-only rule.
Distinguishing the major safety time courses
The following matrix is a case-assessment aid, not a diagnostic rule or a substitute for local prescribing information. Events may occur outside the usual time frame, and the absence of a characteristic latency does not exclude a suspected reaction.
| Safety question | Typical evidentiary time horizon | Most informative follow-up | Common interpretation error |
|---|---|---|---|
| Diuresis, hypotension or electrolyte change | Hours to weeks after initiation, dose change or intercurrent illness; sometimes later | Dosing history, serial sodium/potassium, renal function, volume status, co-medications | Assuming the first abnormal laboratory result is the date the problem began |
| Acute myopia or secondary angle closure | Often hours to weeks after starting therapy in product labelling | Ophthalmic examination, intraocular pressure, exact exposure dates, clinical course | Mistaking the acute idiosyncratic syndrome for chronic glaucoma |
| Suspected drug interaction | Related to introduction or withdrawal of a second medicine; variable | Complete combination-product and co-medication history, levels where relevant | Assigning causality merely because two drugs were prescribed |
| Non-melanoma skin cancer | Long-term, cumulative exposure; latency may be years | Histopathology, cumulative dose, skin phenotype, UV exposure and previous cancer | Treating an odds ratio as the individual's probability or relying on short-term dechallenge |
Operational pharmacovigilance framework
Individual case processing
The case record should distinguish reporter observations, confirmed laboratory or diagnostic findings, the treating clinician's assessment and the marketing authorisation holder's causality evaluation. For an electrolyte event, capture onset, values, symptoms, medical interventions, dechallenge and alternative causes. For acute ocular reactions, obtain specialist findings and exact latency. For skin cancer, pathology, cumulative exposure and background risks are central.
An illustrative case of severe hyponatraemia shortly after a new prescription would prompt investigation of fluid intake, concomitant medicines, renal function and sodium trajectory. An illustrative SCC case after many years of treatment requires a different evidentiary approach: cumulative exposure and pathology matter more than whether the lesion resolved immediately after stopping the drug. Both are hypothetical teaching examples, not real regulatory findings.
Aggregate review and signal evaluation
A mature safety programme should not rely solely on spontaneous-report counts. Electrolyte events may be frequent but incompletely reported; rare ocular events may be distinctive but require careful diagnostic confirmation; long-latency cancer risks are better examined with linked exposure and outcome datasets. Useful evidence includes spontaneous reports, published epidemiology, regulatory signal assessments, clinical studies and real-world data with transparent denominators and confounder handling.
An aggregate safety review should document the precise safety question, competing explanations, source quality, observed patterns, known limitations, any label or risk-minimisation action, and follow-up evidence. Reporting obligations are governed by the applicable jurisdiction and authorisation; suggested operational practices are not automatically legal mandates.
Inspection readiness and governance
Inspection-ready documentation should allow an independent reviewer to trace a signal from detection through validation, assessment, decision, implementation and effectiveness review. For hydrochlorothiazide, a meaningful governance example is the translation of the 2018 skin-cancer signal into product information and patient advice, while continuing to evaluate subsequent evidence. A local team should be able to show which version of product information was used and when any risk communication or training was implemented.
Practical checklist
- Confirm the hydrochlorothiazide INN, formulation, strength and any combination ingredients.
- Document indication, actual exposure, start/stop dates, dose changes and cumulative dose when relevant.
- Review electrolytes, renal function, volume status and relevant symptoms for suspected metabolic or renal events.
- For visual complaints, document onset and ophthalmic findings and ensure urgent clinical evaluation.
- For suspected skin cancer, seek pathology, exposure history and relevant background risk factors.
- Assess lithium, digoxin, other diuretics, anti-inflammatory medicines and antihypertensives as potential contributors.
- Separate labelled risks, epidemiological associations, individual causality assessments and preventive advice.
- Record the local product-information version and the rationale for benefit–risk decisions.
Key takeaways
Hydrochlorothiazide's long history combines established antihypertensive and diuretic uses with immediate and long-latency safety challenges. Electrolyte depletion, renal effects and drug interactions call for patient-specific clinical monitoring. Acute myopia and secondary angle-closure glaucoma are time-sensitive labelled reactions. Observational evidence of a cumulative-dose association with non-melanoma skin cancer led to European and US label changes, and IARC announced a Group 1 carcinogenic-hazard evaluation in 2024, publishing the full monograph in 2026. Hazard classification, relative risk, absolute risk and individual causality are different concepts; good pharmacovigilance preserves those distinctions.
References
- International Agency for Research on Cancer. IARC Monographs Volume 137: Hydrochlorothiazide, Voriconazole, and Tacrolimus (2026). https://www.ncbi.nlm.nih.gov/books/NBK622116/
- National Library of Medicine. DailyMed: Hydrochlorothiazide tablets, updated 5 March 2026. https://www.dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=2ec371a5-18ba-4ce2-9725-7f99bb73454a
- National Library of Medicine. DailyMed: Hydrochlorothiazide tablets, US labelling including ocular warning, interactions and skin-cancer precautions. https://www.dailymed.nlm.nih.gov/dailymed/fda/fdaDrugXsl.cfm?setid=9cd5d16e-3db7-4b25-847d-6de91543463b&type=display
- European Medicines Agency. PRAC recommendations on signals adopted 3–6 September 2018: hydrochlorothiazide—skin cancer (EPITT 19138). https://www.ema.europa.eu/en/documents/prac-recommendation/prac-recommendations-signals-adopted-3-6-september-2018-prac-meeting_en.pdf
- Pedersen SA et al. Hydrochlorothiazide use and risk of nonmelanoma skin cancer: a nationwide case-control study from Denmark. J Am Acad Dermatol. 2018. https://pubmed.ncbi.nlm.nih.gov/29217346/
- UK Medicines and Healthcare products Regulatory Agency. Hydrochlorothiazide: risk of non-melanoma skin cancer, particularly in long-term use. 14 November 2018. https://www.gov.uk/drug-safety-update/hydrochlorothiazide-risk-of-non-melanoma-skin-cancer-particularly-in-long-term-use
- International Agency for Research on Cancer. Volume 137: Hydrochlorothiazide, voriconazole, and tacrolimus. https://monographs.iarc.who.int/news-events/volume-137-hydrochlorothiazide-voriconazole-and-tacrolimus-2/
- National Library of Medicine. DailyMed: Hydrochlorothiazide tablet, US labelling updated 16 September 2026, with Sentinel absolute-risk estimates. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=5ba2325f-5fc1-f6f0-e063-6294a90aefc6
- National Institute of Diabetes and Digestive and Kidney Diseases. LiverTox: Diuretics, updated 13 October 2021; historical US approval dates. https://www.ncbi.nlm.nih.gov/books/NBK548808/
- US Federal Trade Commission. Bond RS, Lean DF. Sales, Promotion, and Product Differentiation in Two Prescription Drug Markets. February 1977; historical hydrochlorothiazide patent and marketing account. https://search.ftc.gov/sites/default/files/documents/reports/sales-promotion-and-product-differentiation-two-prescription-drug-markets/197702salespromo.pdf
- Jamerson K et al. Benazepril plus amlodipine or hydrochlorothiazide for hypertension in high-risk patients (ACCOMPLISH). N Engl J Med. 2008;359:2417–2428. https://doi.org/10.1056/NEJMoa0806182
- Ishani A et al. Chlorthalidone vs. hydrochlorothiazide for hypertension—cardiovascular events (Diuretic Comparison Project). N Engl J Med. 2022;387:2401–2410. https://pubmed.ncbi.nlm.nih.gov/36516076/
- Fleming PR et al. Hydrochlorothiazide: a comparison with chlorothiazide. Lancet. 1959. https://pubmed.ncbi.nlm.nih.gov/13666019/
- International Agency for Research on Cancer. Volume 137: Hydrochlorothiazide, voriconazole, and tacrolimus—evaluation announcement. 29 November 2024. https://monographs.iarc.who.int/news-events/volume-137-hydrochlorothiazide-voriconazole-and-tacrolimus/
- US Food and Drug Administration, Sentinel Initiative. FDA labeling change: hydrochlorothiazide and non-melanoma skin cancer. 20 August 2020. https://www.sentinelinitiative.org/news-events/fda-safety-communications-labeling-changes/fda-labeling-change-hydrochlorothiazide-and
Regulatory Note
This article is educational and does not replace current locally authorised prescribing information, medical judgement or jurisdiction-specific pharmacovigilance requirements. Label language and combination-product indications vary by product and region. The IARC carcinogenic-hazard classification and epidemiological estimates must not be mistaken for individual causality determinations or universal instructions to discontinue treatment.