THIAZIDE DIURETICS











Overview
  • The effect of thiazide diuretics on blood pressure have been studied in a number of trials
  • Results from a large randomized controlled trial and a Cochrane meta-analysis are detailed below
HCTZ vs Others for Hypertension in Male Veterans, NEJM (1993) [PubMed abstract]
  • The Veterans Affairs Cooperative study enrolled 1292 men with hypertension
Main inclusion criteria
  • Male veteran
  • DBP 95 - 109 mmHg off medications
Baseline characteristics
  • Average age 59 years
  • Average BP 152/99 mmHg
  • Black race - 48%
  • Current smoker - 32%
Randomized treatment groups
  • Group 1 (188 patients) - Hydrochlorothiazide 12.5 - 50 mg once daily
  • Group 2 (176 patients) - Atenolol 25 - 100 mg once daily
  • Group 3 (188 patients) - Captopril 25 - 100 mg/day given in 2 divided doses
  • Group 4 (177 patients) - Clonidine 0.2 - 0.6 mg/day given in 2 divided doses
  • Group 5 (182 patients) - Diltiazem SR 120 - 360 mg/day given in 2 divided doses
  • Group 2 (186 patients) - Prazosin 4 - 20 mg/day given in 2 divided doses
  • Group 2 (186 patients) - Placebo
  • There was a washout period of 4 - 8 weeks before randomization
  • Patients were titrated over a period of 4 - 8 weeks to a DBP < 90 mmHg or until they reached the maximum drug dose
Primary outcome: Attainment of blood pressure goal during titration (DBP < 90 mmHg) and DBP of < 95 mmHg at one year
Results

Average BP reduction at the end of the titration phase (SBP/DBP mmHg)
HCTZ Atenolol Captopril Clonidine Diltiazem Prazosin Placebo
14 / 10 11 / 12 9 / 10 16 / 12 13 / 14 12 / 11 3 / 5
  • Primary outcome: Diltiazem - 59%, Atenolol - 51%, Clonidine - 50%, HCTZ - 46%, Captopril - 42%, Prazosin - 42%, Placebo - 25%
  • All medications were significantly better than placebo for blood pressure reduction
  • The incidence of side effects with HCTZ was similar to placebo

Findings: Among men, race and age have an important effect on the response to single-drug therapy for hypertension. In addition to cost and quality of life, these factors should be considered in the initial choice of drug.
Blood pressure-lowering efficacy of monotherapy with thiazide diuretics for primary hypertension, Cochrane meta-analysis (2014) [PubMed abstract]
  • A Cochrane meta-analysis looked at the effects of thiazide diuretics on blood pressure in monotherapy trials

Average blood pressure reduction with HCTZ compared to placebo (SBP/DBP)
HCTZ 6.25 mg/day HCTZ 12.5 mg/day HCTZ 25 mg/day HCTZ 50 mg/day
4 / 2 6 / 3 8 / 3 11 / 5
  • Data based on 33 trials with an average baseline blood pressure of 155/100 mmHg

  • Average blood pressure reduction with chlorthalidone compared to placebo (SBP/DBP):
    • Chlorthalidone 12.5 - 75 mg/day: 12 / 4
    • Data based on 7 trials with an average baseline blood pressure of 163/88 mmHg
    • In the trials that were evaluated, chlorthalidone did not show a dose-dependent effect on blood pressure
Professional recommendations
  • See hypertension guidelines for a review of recommended therapies and treatment goals from various professional organizations
StraightHealthcare analysis:
  • Thiazide diuretics are cheap, widely used, and have a long history of improving outcomes in hypertension
  • For patients with hypertension and no compelling indication for another medication class, thiazide diuretics are a first-line choice


Overview
  • In the ALLHAT trial, the effect of chlorthalidone on cardiovascular outcomes was compared to lisinopril and amlodipine in patients with hypertension and at least one risk factor for heart disease.
  • Another arm of the trial used doxazosin, but it was stopped early. The results of that arm are detailed here - ALLHAT doxazosin.
ALLHAT study - Chlorthalidone vs Lisinopril vs Amlodipine for CVD, JAMA (2002) [PubMed abstract]
  • The ALLHAT study enrolled 33,357 patients with hypertension and at least one risk factor for heart disease
Main inclusion criteria
  • Age > 55 years
  • SBP ≥ 140 and/or DBP ≥ 90 or treated hypertension
  • One of the following: previous MI or stroke, left ventricular hypertrophy (LVH), type 2 diabetes, smoker, low HDL (< 35 mg/dl)
Main exclusion criteria
  • History of hospitalized or treated symptomatic heart failure
  • EF < 35%
Baseline characteristics
  • Average age 67 years
  • Race: White - 47% | Black - 32% | Hispanic - 16%
  • Women - 47%
  • Average BP - 146/84
  • Receiving treatment for hypertension - 90%
  • Qualifying risk factor: CVD - 52% | Diabetes - 36% | Smoker - 22% | LVH - 21% | Low HDL - 12%
Randomized treatment groups
  • Group 1 (15,255 patients) - Chlorthalidone 12.5 - 25 mg a day
  • Group 2 (9048 patients) - Amlodipine 2.5 - 10 mg/day
  • Group 3 (9054 patients) - Lisinopril 10 - 40 mg/day
  • Treatment was titrated to a BP goal of < 140/90
  • If BP goal was not met taking the maximum tolerated dosage of the initial medication, open-label Step 2 agent (atenolol, 25-100 mg/d, reserpine, 0.05-0.2 mg/d, or clonidine, 0.1-0.3 mg twice per day), or an open-label Step 3 agent (hydralazine, 25-100 mg twice per day) could be added
  • There was another arm of the study that used doxazosin. That arm was stopped early due to an increased risk of major CVD events. See ALLHAT doxazosin for more.
Primary outcome: Composite of fatal coronary heart disease or nonfatal myocardial infarction
Results

Duration: Average of 4.9 years
Outcome Chlorthalidone Amlodipine Lisinopril Comparisons
Primary outcome (6-year rate) 11.5% 11.3% 11.4% 1 vs 2 p=0.65 | 1 vs 3 p=0.81
Overall mortality (6-year rate) 17.3% 16.8% 17.2% 1 vs 2 p=0.20 | 1 vs 3 p=0.90
Stroke (6-year rate) 5.6% 5.4% 6.3% 1 vs 2 p=0.28 | 1 vs 3 p=0.02
Heart failure (6-year rate) 7.7% 10% 8.7% 1 vs 2 p<0.001 | 1 vs 3 p<0.001
Achieved BP goal (< 140/90) at 1 year 57.8% 55.2% 50.6% 1 vs 2 p<0.001 | 1 vs 3 p<0.001
Potassium < 3.5 mEq/L at 2 years 12.7% 2.6% 1.5% 1 vs 2 p<0.001 | 1 vs 3 p<0.001
New-onset diabetes at 4 years 11.6% 9.8% 8.1% 1 vs 2 p=0.04 | 1 vs 3 p<0.001

Findings: Thiazide-type diuretics are superior in preventing 1 or more major forms of CVD and are less expensive. They should be preferred for first-step antihypertensive therapy.
Summary
  • In the ALLHAT trial, chlorthalidone was equivalent to amlodipine and lisinopril for the primary outcome. Chlorthalidone was superior to lisinopril for stroke prevention and superior to both amlodipine and lisinopril for preventing heart failure. Hypokalemia and new-onset diabetes both occurred more frequently with chlorthalidone.
  • Thiazide diuretics are cheap and effective medications that make good first-line agents in most patients with hypertension
  • See hypertension guidelines for more








Overview
  • Osteoporosis is a condition marked by bone demineralization that leads to an increased risk of fracture
  • Thiazide diuretics cause the kidneys to retain calcium which may be beneficial in preventing osteoporosis
  • A small randomized controlled trial that evaluated the effects of HCTZ on osteoporosis outcomes is detailed below. A cohort study that looked at the risk of fracture in the ALLHAT study is also reviewed.
HCTZ vs Placebo for Osteoporosis Prevention, Annals of Internal Medicine (2000) [PubMed abstract]
  • A study in the Annals of Internal Medicine enrolled healthy men and women aged 60 - 79 years
Main inclusion criteria
  • Age 60 - 79 years
  • Normotensive
  • Bone mineral density (BMD) within 2 standard deviations of normal for their age (Z-score ± 2)
Main exclusion criteria
  • Taking bisphosphonates or hormone replacement therapy
  • Serious heart disease
  • Proteinuria
  • Elevated serum creatinine
  • Hyponatremia
  • Hypokalemia
  • Gout
  • Use of any diuretics or BP medications
  • Risk factors for osteoporosis (e.g. corticosteroids, malabsorptive conditions, etc.)
Baseline characteristics
  • Average age 68 years
  • Average BMD women (total hip) - 0.768 g/cm²
  • Average BMD men (total hip) ∼ 0.945 g/cm²
  • Average calcium intake (women) - 1100 - 1400 mg/day
  • Average calcium intake (men) - 770 - 944 mg/day
Randomized treatment groups
  • Group 1 (105 patients) Placebo once daily
  • Group 2 (108 patients) HCTZ 12.5 mg once daily
  • Group 3 (107 patients) HCTZ 25 mg once daily
  • All subjects were given advice to consume 1000 - 1500 mg of calcium a day
  • BMD measurements were taken every 6 months
Primary outcome: Percent change in BMD at the total hip over 3 years
Results

Duration: 3 years
Outcome Placebo HCTZ 12.5 mg HCTZ 25 mg Comparisons
Primary outcome (% change in women) -0.81% -0.06% +0.62% 1 vs 2, diff 0.75, 95%CI [-0.43 to 1.92] | 1 vs 3, diff 1.43, 95%CI [0.25 to 2.60]
Primary outcome (% change in men) +0.54% +1.36% +0.58% 1 vs 2, diff 0.83, 95%CI [-0.60 to 2.25] | 1 vs 3, diff 0.04, 95%CI [-1.42 to 1.51]
Still taking assigned medication at 3 years (men) 81.6% 89.7% 60.5% N/A
  • Groups 2 and 3 had more hypokalemia requiring potassium supplements than Group 1 [27]

Findings: In healthy older adults, low-dose hydrochlorothiazide preserves bone mineral density at the hip and spine. The modest effects observed over 3 years, if accumulated over 10 to 20 years, may explain the one-third reduction in risk for hip fracture associated with thiazide in many epidemiologic studies.
Association of chlorthalidone, amlodipine, and lisinopril with hip and pelvic fractures in the ALLHAT Trial, JAMA Internal Medicine (2017) [PubMed abstract]
  • Design: Cohort study (N=33,357) using data from ALLHAT study participants
  • Treatment: Chlorthalidone 12.5 - 25 mg a day vs Amlodipine 2.5 - 10 mg/day vs Lisinopril 10 - 40 mg/day
  • Primary outcome: Hip and pelvic fracture hospitalizations
  • Results:
    • Active trial (average 4.9 years): Risk of fracture was significantly lower in participants randomized to receive chlorthalidone vs lisinopril (HR 0.75, 95%CI [0.58 - 0.98], p=0.04) but not significantly different compared with those randomized to receive amlodipine (HR 0.82, 95%CI [0.63 - 1.08] p=0.17)
    • Active trial + post-trial (average 7.8 years): The cumulative incidence of fractures was not significantly lower in participants randomized to receive chlorthalidone vs lisinopril or amlodipine (HR 0.87, 95%CI [0.74 - 1.03], p=0.10) and vs each medication separately
  • Findings: These findings from a large randomized clinical trial provide evidence of a beneficial effect of thiazide-type diuretic therapy in reducing hip and pelvic fracture risk compared with treatment with other antihypertensive medications.
Summary
  • Thiazide diuretics may have a modest effect in preventing osteoporosis and fractures
  • Larger and longer trials are needed to draw firm conclusions





Low potassium (hypokalemia)
Overview
  • All thiazide diuretics can cause significant potassium loss
  • Potassium loss tends to increase with higher doses
Chlorthalidone
  • Two large randomized trials involving chlorthalidone reported average potassium changes among participants
  • In the ALLHAT trial, 15,255 patients who were randomized to chlorthalidone (12.5 - 25 mg a day) had an average decrease in potassium levels of 0.2 mEq/L after 4 years of therapy [97]
  • In the SHEP trial, 2365 patients who were randomized to chlorthalidone (12.5 - 25 mg a day) had an average decrease in potassium levels of 0.4 mEq/L after 1 year of therapy [87]
Hydrochlorothiazide
  • Compelling trials that measured potassium changes with HCTZ are scant
  • In a small trial, 268 patients who were receiving 25 - 50 mg of HCTZ a day had an average decrease in potassium levels of 0.47 mEq/L after 6 months of therapy (20% of patients were taking 25 mg HCTZ a day, 80% on 50 mg) [98]
Summary
  • Thiazides cause potassium loss typically in the range of 0.2 - 0.4 mEq/L
  • It is important to monitor potassium levels periodically when taking a thiazide
  • Combining a thiazide with an ACE inhibitor or a potassium-sparing diuretic can attenuate this effect
  • When given in comparable doses, chlorthalidone and HCTZ appear to cause the same amount of potassium loss

Increased urination

Low sodium (hyponatremia)
Overview
  • Thiazide diuretics work by blocking sodium reabsorption in the kidneys, and therefore, they may lead to significant sodium loss
STUDY
Risk of thiazide-induced hyponatremia in patients with hypertension, Am J Med (2011) [PubMed abstract]
  • A cohort study in the American Journal of Medicine looked at the risk of low sodium (defined as sodium ≤ 130 mEq/L) in patients taking thiazides compared to patients not taking thiazides
  • The study utilized a medical database to ascertain information
    • The following results were seen when patients taking thiazides were compared to nonusers:
      • Patients taking thiazides had a 61% greater risk for hyponatremia than nonusers
      • Incidence rates for hyponatremia were 14 cases per 100 people per year for thiazide-users compared to 8.7 cases per 100 people per year for nonusers
      • The risk for hospitalization associated with hyponatremia was not significantly different between the two groups
      • The risk for hyponatremia started early in therapy (within 90 days), and the median time to hyponatremia in thiazide-users was 1.75 years
      • The risk continued to increase over a 10-year period [99]
Summary
  • Thiazides can cause low sodium levels in a significant number of patients
  • It is important to check sodium levels periodically in patients taking thiazides
  • Risk factors for thiazide-induced hyponatremia may include advanced age, female sex, and certain medications (NSAIDS, SSRI antidepressants) [24, 99]

Elevated blood sugar / diabetes risk
Overview
  • Thiazide diuretics have been shown to cause blood sugar levels to rise, and subsequently, they can increase the risk of developing diabetes
SHEP trial
  • In the SHEP trial [PMID 2046107], patients receiving chlorthalidone had a significantly greater increase in fasting blood sugar after three years than patients receiving placebo
  • Chlorthalidone group fasting blood sugar increased by 9.2 mg/dl
  • Placebo group fasting blood sugar increased by 5.6 mg/dl
  • New cases of diabetes were not significantly different between the groups [32]
ALLHAT study
  • In the ALLHAT study, patients receiving chlorthalidone had a significantly greater increase in fasting blood sugar after two years than patients receiving amlodipine
  • Chlorthalidone group fasting blood sugar increased by 8.5 mg/dl
  • Amlodipine group fasting blood sugar increased by 5.5 mg/dl
  • The odds of developing diabetes was significantly higher in the chlorthalidone group when compared to the amlodipine and lisinopril groups [100]
Summary
  • Thiazide diuretics have a slight negative effect on blood sugar control
  • There is some evidence that this effect may be worsened by low potassium levels that can also be caused by thiazides. Treating low potassium levels may help attenuate the risk of increased blood sugar. [94]
  • The proven benefits of thiazides in hypertension outcomes will outweigh their negative effect on blood sugars in the majority of patients

Uric acid (gout risk)
Overview
  • Gout is an inflammatory joint disease caused by monosodium urate crystal formation in joints. Monosodium crystal formation is precipitated by high uric acid concentrations in the blood.
  • Thiazide diuretics cause the kidneys to retain uric acid which can worsen or precipitate gout
SHEP trial
  • In the SHEP trial [PMID 2046107], patients receiving chlorthalidone had a significantly greater increase in uric acid levels after three years than patients receiving placebo
  • Chlorthalidone group uric acid level increased by 0.90 mg/dl
  • Placebo group uric acid level increased by 0.30 mg/dl [32]
Summary
  • Thiazide diuretics raise uric acid levels and therefore increase the risk of gout (recurrent and new-onset)
  • Individual risk will depend on a patient's baseline uric acid level and the presence of other risk factors (ex. hypertension, heart failure, obesity, diet, etc.)
  • In general, patients who have well-controlled gout with normal uric acid levels will likely not be affected by thiazide therapy
  • Patients with uncontrolled gout should avoid thiazides

Low magnesium (hypomagnesemia)

High calcium levels (hypercalcemia)
Overview
  • Thiazide diuretics cause the kidneys to retain calcium. This may lead to elevated blood calcium levels
  • The overall significance and magnitude of this effect is not well-defined
  • We found one study that measured the incidence of thiazide-associated hypercalcemia using a medical database from a community in Minnesota. The incidence of thiazide-associated calcium level elevations was 7.7 cases per 100,000 thiazide-exposed patients per year [33]
Summary
  • Elevated calcium levels associated with thiazide therapy appear to be rare
  • It may take up to 3 months for thiazide-associated calcium elevations to return to normal once the thiazide is stopped [33]

Lipid parameters (cholesterol)
Overview
  • The effect of thiazides on lipid parameters has been inconsistent in trials with some studies showing no effect and others showing a negative effect [34]
  • The most consistent effect observed is a slight increase in triglyceride and LDL levels
  • In the SHEP trial [PMID 2046107], patients taking chlorthalidone for 3 years had a significant increase in triglycerides from baseline of 24.5 mg/dl (compared to a placebo increase of 8.0 mg/dl) [32]
Summary
  • It is unclear if thiazides negatively affect lipid parameters
  • Since thiazides promote fluid loss, part of the observed effect may be secondary to hemoconcentration (increased concentration of blood products secondary to fluid loss) [34]
  • The clinical significance of this is most likely negligible

Photosensitivity

Acute myopia and angle-closure glaucoma (HCTZ)




Kidney disease

Liver disease
Cirrhosis
Mild to moderate liver disease
  • Dosage adjustments are not typically needed

Sulfa allergy
Overview
  • Loop and thiazide diuretics contain a sulfonamide group in their structure
  • The sulfonamide group is different from the one that is found in sulfa-based antibiotics
  • Patients with a history of allergy to sulfa-based antibiotics may be at a slightly increased risk of an allergic reaction to thiazide and loop diuretics
  • A study that looked at the potential cross-reactivity is summarized below
STUDY
Cross-reactivity between sulfa-based antibiotics and sulfonamide nonantibiotics, NEJM (2003) [PubMed abstract]
  • A cohort study in the NEJM looked at the risk of an allergic reaction to nonantibiotic sulfonamides in patients who had a previous reaction to a sulfa-based antibiotic
  • The study found the following:
    • Patients with a history of sulfa-based antibiotic allergy who subsequently took nonantibiotic sulfonamides (including thiazide and loop diuretics) had a 10% risk of having a reaction to the nonantibiotic sulfonamide
    • In patients without a history of allergic reaction to a sulfa-based antibiotic, 1.6% had a reaction to a nonantibiotic sulfonamide
    • In addition, patients with a history of sulfa-based antibiotic allergy had a 14% chance of having a reaction to a penicillin antibiotic. This finding led researchers to conclude that a history of allergic reactions in general may be more predictive of a reaction than reactions to any specific medication [84]
Summary
  • Loop and thiazide diuretics may be prescribed to patients with a history of sulfa-based antibiotic allergy
  • Patients should be aware that a cross-sensitivity reaction may occur with a risk of around 10%
  • Patients with a history of severe reactions to sulfa-based antibiotics should avoid nonantibiotic sulfonamides if they can. If not, then the initial dosing should be done under medical supervision.

Gout

Prolonged QT syndrome

Diabetes

Lupus skin reactions

Hyperparathyroidism


Drug interactions
All thiazide diuretics
  • Bile acid sequestrants (Questran®, Welchol®, Colestid®) - Bile Acid Sequestrants can interfere with the absorption of thiazide and loop diuretics. Diuretics should be taken 1 hour before or 4 hours after bile acid sequestrants.
  • Lithium - Thiazides may reduce the clearance of lithium. Thiazides should be avoided with lithium if possible. Lithium levels should be monitored closely in patients taking thiazides.
  • NSAIDS (Advil®, ibuprofen, naprosyn, etc.) - NSAIDS can block the therapeutic effect of all diuretics. Patients should monitor for decreased effectiveness of diuretics when taking NSAIDS for extended periods.
HCTZ