The n e w e ng l a n d j o u r na l of m e dic i n e

review article

Drug Therapy

Use of Diuretics in Patients

with Hypertension
Michael E. Ernst, Pharm.D., and Marvin Moser, M.D.

T
hiazide diuretics became available in the late 1950s and were From the Department of Pharmacy Prac-
the first effective oral antihypertensive agents with an acceptable side-effect tice and Science, College of Pharmacy,
and Department of Family Medicine,
profile.1,2 A half-century later, thiazides remain important medications for the Carver College of Medicine, University of
treatment of hypertension. These agents reduce blood pressure when administered Iowa, Iowa City (M.E.E.); and the Section
as monotherapy, enhance the efficacy of other antihypertensive agents, and reduce of Cardiovascular Medicine, Yale Univer-
sity School of Medicine, New Haven, CT
hypertension-related morbidity and mortality. This review focuses on thiazides, the (M.M.). Address reprint requests to Dr.
diuretics most often indicated for long-term therapy for hypertension; loop diuret- Ernst at the Department of Family Medi-
ics and potassium-sparing agents are briefly considered. cine, University of Iowa Hospitals and
Clinics, 200 Hawkins Dr., 01291-A PFP,
Iowa City, IA 52242, or at michael-ernst@
Cl inic a l Ph a r m ac ol o gy of Thi a zide s uiowa.edu.

This article (10.1056/NEJMra0907219) was

Chemistry and Mechanism of Action updated on November 3, 2010, at NEJM
Diuretic therapy for hypertension originated in 1937 with the discovery that sulfon- .org.
amides caused acidemia and mild diuresis by inhibiting carbonic anhydrase in the
N Engl J Med 2009;361:2153-64.
proximal tubule.3,4 Chlorothiazide, a benzothiadiazine derivative, was isolated dur- Copyright 2009 Massachusetts Medical Society.
ing a search for more potent inhibitors of carbonic anhydrase5; chlorothiazide was
found to be a more effective diuretic and also to unexpectedly increase the excretion
of chloride, rather than bicarbonate.6 This effect on excretion eventually led to iden-
tification of the upstream portion of the distal convoluted tubule as the major site
of action of the thiazides, where they interfere with sodium reabsorption by inhib-
iting the electroneutral sodiumchloride symporter (Fig. 1).7 Activity against car-
bonic anhydrase, although maintained by some thiazides, is considered irrelevant
to their mechanism of action, since sodium that is rejected proximally is reabsorbed
downstream in the renal tubule in the thick ascending limb. Despite structural varia-
tion among the different congeners, the term thiazide diuretic includes all diuretics
believed to have a primary action in the distal tubule.

Pharmacokinetics
The pharmacokinetic activities of thiazides are not uniform. All are absorbed oral-
ly and have volumes of distribution equal to or greater than body weight (Table
1).8-10 Thiazides are extensively bound to plasma proteins, which helps limit their
filtration by the glomeruli, in effect trapping the diuretic in the vascular space and
allowing for its delivery to secretory sites of the renal proximal tubular cells. Or-
ganic anion transporters may also assist by concentrating thiazides in the tubular
lumen.11
The onset of action occurs after approximately 2 to 3 hours for most thiazides,
with little natriuretic effect beyond 6 hours.12 Their metabolism, bioavailability,
and plasma half-lives are variable. The latter two pharmacokinetic features are the
most clinically relevant, as they determine the dose and frequency of administra-

n engl j med 361;22 nejm.org november 26, 2009 2153

The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

intake enhances absorption.8,9 Some thiazides un-

Proximal Distal convoluted
convoluted tubule tubule (5%)
dergo extensive metabolism, whereas others are
(6570%) excreted nearly intact in urine. A 50% reduction
Filtered blood out Na+ in hydrochlorothiazide absorption is observed in
(efferent arteriole) patients with heart failure. Relatively little else is
Glomerulus
known about the influence of disease on the phar-
Cl macokinetics of thiazides.8
Na+ Thiazides Most thiazides have a half-life of approximately
Collecting 8 to 12 hours, just permitting effective once-
duct daily dosing.9 Among thiazides, chlorthalidone
Unfiltered (12%) is uniquely long-acting, with an elimination half-
blood in
(afferent arteriole) life of 50 to 60 hours, owing to its large volume
NaHCO3 Na+ of distribution.13 Nearly 99% of chlorthalidone
Aldosterone
K+ is bound to erythrocyte carbonic anhydrase, and
Carbonic the drug has a stronger inhibitory effect than
anhydrase K+-sparing diuretics and other thiazides against several catalytically active
inhibitors mineralocorticoid-receptor
antagonists
mammalian isoforms.14,15 The substantial parti-
tioning of chlorthalidone into erythrocytes creates
Thick descending Na+/K+ a tissue reservoir that allows the constant seep-
Loop
limb 2Cl diuretics ing of chlorthalidone back into plasma, through
Cortex
which it exerts its therapeutic effect.14 This depot
Renal
Medulla effect of chlorthalidone may be advantageous in
artery Thick ascending patients who occasionally miss doses, and the
Kidney limb agent appears to retain measurable efficacy when
given less frequently than once daily.16,17

Cortex Urine
Renal
vein Loop of Henle
Medulla
(25%)
Figure 1. Sites of Diuretic Action in the Nephron.
The percentage of sodium reabsorbed in a given region is indicated in pa-
rentheses. K+ -sparing agents collectively refers to the epithelial sodium-
channel inhibitors (e.g., amiloride and triamterene) and mineralocorticoid-
receptor antagonists (e.g., spironolactone and eplerenone). Sodium is
reabsorbed in the distal tubule and collecting ducts through an aldoster-
one-sensitive sodium channel and by activation of an ATP-dependent so-
diumpotassium pump. Through both mechanisms, potassium is secreted
into the lumen to preserve electroneutrality. Sodium-channel inhibitors
preserve potassium by interfering with the sodiumpotassium pump,
whereas mineralocorticoid-receptor antagonists spare potassium through
their inhibitory effect on aldosterone. NaHCO3 denotes sodium bicarbonate.
tion. Chlorothiazide is relatively insoluble in lip-
ids, and large doses are required to achieve levels
sufficient to reach the site of action. Hydrochlo-
rothiazide has relatively greater bioavailability;
approximately 60 to 70% is absorbed, and food
Pharmacodynamics
Hemodynamic Effects
The hemodynamic effects of thiazides can be sep-
arated into short-term and long-term phases (Ta-
ble 2).18 Initial decreases in blood pressure are
attributed to the reductions in extracellular fluid
and plasma volumes, leading to depressed cardiac
preload and output.19 Dextran administration dur-
ing this short-term phase will restore plasma vol-
ume and blood pressure to pretreatment levels.
Counterregulatory activation of the sympathetic
nervous system and the reninangiotensinaldo-
sterone system induces a transient rise in periph-
eral vascular resistance, although this is not
usually sufficient to negate the blood-pressure
reduction.20 Combining a thiazide with an angio-
tensin-convertingenzyme (ACE) inhibitor or an
angiotensin IIreceptor blocker (ARB) can oppose
this transient rise in resistance and increase the
antihypertensive response.
The long-term antihypertensive response to
thiazides cannot be reliably predicted by the de-
gree of initial reduction in plasma volume, which
eventually returns to near-normal levels. Volume
expansion due to the use of dextran at this stage
no longer restores the blood pressure to pretreat-
COLOR FIGURE

Version 7 11/06/09
Author Ernst
Fig # 1
Title Diuretic Use in Hyper.
2154 n engl j med 361;22 nejm.org november
ME JL
26, 2009
DE JRI
The New England Journal of Medicine
Artist LAM
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only.PLEASE
AUTHOR No other
NOTE: uses without permission.
Figure has been redrawn and type has been reset
Copyright 2009 Massachusetts Medical Society. All rights reserved.
Please check carefully

Issue date 11/26/09

 drug ther apy

Table 1. Pharmacokinetic Characteristics of the Thiazide Diuretics Approved for Use in the United States.*

Relative Carbonic
Anhydrase Volume of Elimination
Diuretic Inhibition Oral Bioavailability Distribution Protein Binding Route of Elimination Half-Life
percent liters per kilogram percent hr
Thiazide-type
Chlorothiazide ++ 1530 1 70 100% Renal 1.52.5
Hydrochlorothiazide + 6070 2.5 40 95% Renal 910
Methychlothiazide Hepatically metabolized
Polythiazide + 25% Renal 26
Bendroflumethiazide 0 90 1.01.5 94 30% Renal 9
Thiazide-like
Chlorthalidone +++ 65 313 99 65% Renal 5060
Metolazone + 65 113 (total) 95 80% Renal 814
Indapamide ++ 93 25 (total) 75 Hepatically metabolized 14

* All the diuretics listed are available in generic form in the United States as monotherapy, except polythiazide (not currently available) and
bendroflumethiazide (available only in combination with nadolol). Dashes indicate an absence of data.
The terms thiazide-type and thiazide-like are used to group thiazides on the basis of the presence of a benzothiadiazine molecular structure.
Thiazide-like diuretics lack the benzothiadiazine structure but have a mechanism of action similar to that of thiazide-type diuretics, which
have the benzothiadiazine structure.
Plus signs indicate inhibition, with greater numbers of plus signs reflecting increased inhibition (lower inhibition constants); the zero indi-
cates an inhibition constant of 0.
The volumes of distribution of metolozone and indapamide are given for the total volume, in liters; data on liters per kilogram were not
available.

ment levels.21-23 A more likely explanation for the Tolerance to Diuretics

persistent antihypertensive effects of most thia
zides is an overall reduction in systemic resistance,
although the exact mechanisms are unclear.24 Evi-
dence suggests that chlorthalidone may not lower
systemic resistance, even after 8 to 12 months of
therapy, indicating that other mechanisms may
be responsible.17 It is not yet clear whether thi-
azides have direct vasodilatory properties or in-
duce a reverse autoregulation phenomenon; they
have also been proposed to cause structural mem-
brane changes or altered ion gradients.24 A sim-
pler possibility is that a low level of prolonged
diuresis produced by long-term thiazide admin-
istration may maintain a nominal state of volume
contraction, thereby promoting a downward shift
in vascular resistance.12
Thiazides have substantial residual effects after
discontinuation. Rapid volume expansion, weight
gain, and decreased renin levels occur after stop-
ping thiazides, but blood pressure rises slowly
and does not immediately reach pretreatment lev-
els.18,25 With adherence to lifestyle modifications
(e.g., weight loss, reduction in sodium and alco-
hol intake), nearly 70% of patients may not require
antihypertensive medications for up to 1 year after
long-term thiazide-based therapy is withdrawn.26
The use of diuretics elicits both short- and long-
term adaptations intended to protect intravascu-
lar volume. Short-term tolerance may result from
a period of post-dose antinatriuresis triggered by
the initial reduction in extracellular fluid volume,
corresponding to a decline in the drug level in
plasma and tubular fluid to below the diuretic
threshold.27 Activation of the reninangiotensin
aldosterone system and the sympathetic nervous
system, as well as suppression of the secretion of
atrial natriuretic peptide and renal prostaglandin,
also contribute to short-term tolerance.27 Post-dose
sodium retention is significantly influenced by di-
etary sodium intake. Sodium restriction promotes
an overall negative sodium balance and enhances
the therapeutic response to thiazides, whereas per-
sistently high dietary sodium offsets this effect.28
Long-term diuretic adaptation, or the braking
effect, refers to a gradual return of the sodium
chloride balance to an electroneutral level. Persis-
tent volume removal appears to trigger long-term
activation of the reninangiotensinaldosterone
system, increasing circulating angiotensin II levels,
which in turn promotes increased proximal so-
dium reabsorption and limits the overall delivery
of sodium to the distal site. Other volume-inde-

n engl j med 361;22 nejm.org november 26, 2009 2155

The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 2. Hemodynamic and Physiological Effects after Initiation and Cessation of Diuretic Therapy.

Short-Term Phase
Variable (first 24 wk) Long-Term Phase (mo) Post-Therapy Period
Cardiac output Decrease Increase (return to pretreatment No change
level)
Plasma volume Decrease Increase (near-return to pretreat- Increase (possibly exceeding pre-
ment level) treatment level)
Plasma renin activity Increase Increase Decrease (return to pretreatment
level)
Peripheral resistance Transient increase Gradual decrease Increase (gradual return to pre-
treatment level)
Blood pressure Decrease Decrease Gradual increase

pendent mechanisms may be involved, including
the up-regulation of sodium transporters down-
stream from the primary site of diuretic action
and structural hypertrophy of distal nephron
segments.29-31
Tolerance to diuretics can be overcome by ad-
ministering higher doses or combinations of di-
uretics. For example, synergistic diuresis occurs
when a thiazide is added to loop-diuretic mono-
therapy in patients with edema.32 Occult volume
expansion can be present in patients with resis-
tant hypertension, despite existing diuretic ther-
apy; increasing the dose of diuretics may improve
the blood pressure.33 High doses and combina-
tions of diuretics must be used carefully to avoid
renal injury and marked electrolyte disturbances.
Diur e t ic Ther a py
for H y per tensi v e Pat ien t s
Many physicians consider thiazides the diuretics
of choice for long-term therapy. On average, after
adjustment for reductions seen with the use of pla-
cebo, thiazides induce a reduction in the systolic
and diastolic blood pressures of 10 to 15 mm Hg
and 5 to 10 mm Hg, respectively. Hypertension
responding preferentially to thiazides is considered
to be low-renin or salt-sensitive hypertension. The
elderly, blacks, and patients with characteristics
associated with high cardiac output (e.g., obesity)
tend to have this type of hypertension. Thiazides
also correct the hypertension and electrolyte ab-
normalities associated with pseudohypoaldos-
teronism type 2 (Gordons syndrome), a rare
mendelian form of hypertension in which the so-
diumchloride symporter is excessively active. Thi-
azides potentiate other antihypertensive agents
when they are used in combination, often produc-
ing an additive decrease in blood pressure. The ad-
dition of a thiazide minimizes racial differences
usually observed in response to monotherapy with
inhibitors of the reninangiotensinaldosterone
system, and the use of such an inhibitor can lessen
the degree of the hypokalemia and the metabolic
perturbations that may be evoked by thiazides.
The dosing of thiazides has evolved in parallel
to our progressive understanding of their mech-
anism of action and doseresponse relationships.
Earlier use of high doses was based on the belief
that efficacy was directly linked to the amount of
renal sodium excretion and reduction in plasma
volume; the larger the dose, the greater the as-
sumed reduction in blood pressure. However, thi-
azides are now used in substantially smaller doses,
and the term low-dose thiazide has become syn-
onymous with hydrochlorothiazide at a dose of
12.5 to 25 mg per day (or the equivalent dose of
another thiazide). Approximately 50% of patients
will respond initially to these low doses. In the
Systolic Hypertension in the Elderly Program
(SHEP),34 chlorthalidone given at a dose of 12.5 mg
per day controlled blood pressure, for several
years, in more than 50% of patients. Increasing the
dose of hydrochlorothiazide from 12.5 to 25 mg
per day may result in a response in an additional
20% (approximately) of patients; at 50 mg per day,
80 to 90% of patients should have measurable de-
creases in blood pressure.35 Increased electrolyte
losses at the higher doses of diuretics may pre-
clude their routine use.
For many hypertensive patients, several agents
will be needed to achieve desired blood-pressure
targets. Combination regimens that include a thi-
azide can achieve therapeutic synergy while min-

2156 n engl j med 361;22 nejm.org november 26, 2009

The New England Journal of Medicine

Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
 drug ther apy

imizing adverse effects.36 The lack of appropriate per minute per 1.73 m2 of body-surface area) or
diuretic use is often identified as the primary drug- by volume overload (e.g., in congestive heart fail-
related cause of resistance to treatment.37 ure or the nephrotic syndrome); in patients with
such complications, loop diuretics provide consis-
Diuretics in Renal Impairment tent natriuresis and diuresis. Furosemide should
Thiazides are typically considered ineffective when be administered twice daily, whereas torsemide is
the glomerular filtration rate decreases below 30 a longer-acting alternative that may be adminis-
to 40 ml per minute per 1.73 m2 of body-surface tered once daily.
area, although direct evidence is lacking. Small
studies have shown that thiazides can elicit an an- Potassium-Sparing Agents and
tihypertensive response in patients with chronic Mineralocorticoid-Receptor Antagonists
kidney disease38,39; however, their use in patients Potassium-sparing agents induce only minimal
with severe renal impairment remains impracti- natriuresis and are relatively ineffective in lower-
cal, for two reasons: first, the reduced glomeru- ing blood pressure (Fig. 1). Their primary value is
lar filtration rate limits the overall filtered sodi- their ability to reduce the loss of potassium when
um load reaching the distal tubule; and second, they are used with thiazides. They also avert the
reabsorption in the distal tubule is only modestly urinary loss of magnesium, which is important,
effective as compared with that in the large-capac- since restoration of magnesium balance is neces-
ity, thick ascending limb.12 These features under- sary for optimal correction of diuretic-induced
score the rationale for substituting so-called high- hypokalemia. Triamterene is commonly admin-
ceiling diuretics that act more proximally, in the istered with hydrochlorothiazide, although other
loop of Henle, in hypertensive patients with renal fixed-dose combinations of thiazides and potas-
impairment. sium-sparing agents are available. Amiloride, an
Metolazone, a quinazoline derivative, is an ex- epithelial sodium-channel blocker, is reportedly
ception among thiazides because it retains its ef- more effective than spironolactone as therapy in
ficacy in patients who have renal insufficiency or blacks who have resistance to treatment.42
other diuretic-resistance states.40 Its effect is lim- The phenomenon of aldosterone escape, where-
ited by slow, erratic absorption; the more predict- by aldosterone activity is incompletely suppressed
able bioavailability of other thiazides makes them in hypertensive patients who are receiving inhibi-
better suited as long-term therapy of hypertension. tors of the reninangiotensinaldosterone system,
Metolazone should be reserved for use in combi- can lead to increased salt and water retention.
nation with loop diuretics in patients with volume Agents that block the effect of aldosterone are use-
overload whose fluid and electrolyte balance are ful for treating this retention. Spironolactone,
being closely monitored. It is administered daily a nonselective mineralocorticoid-receptor antago-
for a short period (3 to 5 days), with administra- nist, is well absorbed and has a long half-life (ap-
tion reduced to thrice weekly after this period or proximately 20 hours), attributable to its active
after euvolemia is achieved.27 metabolites.43 Spironolactone not only corrects
thiazide-induced potassium and magnesium loss-
Loop Diuretics es, but also, in low doses (12.5 to 50 mg per day),
Diuretics that act in the loop of Henle can lower provides additive hypotensive effects in patients
blood pressure but are less effective in the long who have resistance to treatment.44 Spironolactone
term than thiazides.41 Most loop diuretics have a remains effective when renal function is impaired,
short duration of action (approximately 6 hours), but patients must be monitored carefully for the
resulting in an initial diuresis that is followed development of hyperkalemia. Eplerenone, a newer
closely by a period of antinatriuresis lasting up to agent that is more selective for aldosterone than
18 hours per day when the drug is administered for androgen and progesterone receptors, is asso-
once daily. A net neutral sodium balance, or even ciated with less gynecomastia and breast tender-
a positive balance, can occur with the use of loop ness than is found with spironolactone. However,
diuretics. These agents are most appropriate for direct comparisons of the efficacies of eplerenone
the treatment of hypertension that is complicated and spironolactone in patients with treatment-
by a reduced glomerular filtration rate (<30 to 40 ml resistant hypertension are lacking.

n engl j med 361;22 nejm.org november 26, 2009 2157

The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
Diur e t ic s in T r i a l s w i th
C a r diova scul a r End P oin t s
The efficacy of thiazides in reducing the risk of
major cardiovascular events was first established
in 1967, in the landmark Veterans Affairs Coop-
erative Study.45 In the ensuing years, thiazide reg-
imens have proved highly effective in the preven-
tion of stroke, coronary heart disease, and heart
failure (see the Supplementary Appendix, available
with the full text of this article at NEJM.org).
Combined meta-analyses and systematic re-
views report that, as compared with placebo, thi-
azide-based therapy reduces relative rates of heart
failure (by 41 to 49%), stroke (by 29 to 38%), coro-
nary heart disease (by 14 to 21%), and death from
any cause (by 10 to 11%).46-48 These and other
analyses also show that the benefit from thia
zides is broadly similar to that from other antihy-
pertensive drugs, with results consistent across
age and sex strata.46-51 The largest randomized,
blinded study performed to date, the Antihyper-
tensive and Lipid-Lowering Treatment to Prevent
Heart Attack Trial (ALLHAT),52 compared initial
treatment consisting of chlorthalidone, at a dose
of 12.5 mg per day, with treatment consisting of
amlodipine, lisinopril, or doxazosin. The 42,418
high-risk study participants were 55 years of age
or older, and 35% were black. The doxazosin treat-
ment was stopped prematurely owing to a signifi-
cantly greater incidence of heart failure events than
with chlorthalidone. At the end of the study, no
significant differences were found between the
chlorthalidone group and any other treatment
group in the rate of the composite primary end
point of fatal coronary heart disease or nonfatal
myocardial infarction; however, chlorthalidone
was superior with respect to several predefined
secondary end points, including heart failure (vs.
amlodipine and lisinopril) and stroke (vs. lisino-
pril). The explanation for these findings may be
related to the improved blood-pressure control
maintained throughout the study in the patients
receiving chlorthalidone as compared with those
receiving another treatment. Analyses of data
stratified on the basis of race and metabolic status
consistently indicate that chlorthalidone, used as
initial therapy, was not surpassed by any of the
three comparison drugs with regard to antihyper-
tensive efficacy or event-rate reduction.53-57
Although chlorthalidone and hydrochlorothi-
azide have been commonly used in the United
2158 n engl j med 361;22 nejm.org november 26, 2009
The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
States, other thiazides, such as bendroflumethia
zide and indapamide, have also been studied. Re-
cently, the use of a sustained-release indapamide-
based regimen as initial therapy led to relative
reductions of 39%, 64%, and 21% in the rates of
fatal stroke, heart failure, and death, respectively,
in patients older than 80 years of age.58
Ther a peu t ic C on t rov er sie s
The concept of a class effect among thiazides has
been promulgated in the absence of comparative
studies. In the United States, this debate largely
centers on whether to use hydrochlorothiazide or
chlorthalidone.59 Chlorthalidone is frequently sug-
gested for use, since the rate of cardiovascular
events has been reduced in every study in which
it was used, including at the low doses in the
SHEP study and ALLHAT, whereas the effects of
other thiazide regimens have been less consis-
tent.60 Trials in which hydrochlorothiazide was
associated with favorable reductions in the rate
of cardiovascular events have typically used higher
doses (between 25 and 50 mg per day) than cur-
rently advocated. These dose-based differences are
underappreciated, and the perception that thia
zides are interchangeable is reinforced by the abun-
dance of fixed-dose combinations containing low-
dose hydrochlorothiazide. Two trials of low-dose
hydrochlorothiazide in combination with other
medications showed less effectiveness as com-
pared with the nonthiazide regimen.61,62
Hydrochlorothiazide and chlorthalidone are of-
ten incorrectly described as equipotent.10 How-
ever, early trials comparing the two drugs often
involved the administration of hydrochlorothia
zide twice daily to achieve reductions in blood
pressure similar to those effected by chlorthali
done.63,64 When evaluated on the basis of the
dose required to achieve equivalent reductions in
the systolic blood pressure, 24-hour ambulatory
blood-pressure monitoring confirmed that chlor
thalidone is approximately twice as potent as
hydrochlorothiazide.65 These findings may be at-
tributable to the ability, with chlorthalidone thera-
py, to maintain efficacy throughout the night and
reduce the gradual rise in blood pressure that may
occur during the interval between daily doses.
Whether hydrochlorothiazide and chlorthali-
done are interchangeable in reducing the risk of
cardiovascular events is questionable. An analy-
sis in the Multiple Risk Factor Intervention Trial
 drug ther apy
(MRFIT) suggested a lower mortality rate in the
group randomly assigned to undergo a special
intervention in clinics predominantly using
chlorthalidone than in those predominantly us-
ing hydrochlorothiazide, but the assignment to a
specific diuretic was not randomized, and the
study design precluded an ability to reliably sepa-
rate diuretic effects from those of the other con-
current interventions.66 Conversely, a meta-analy-
sis that did not include the MRFIT findings has
suggested no significant differences between the
two agents.67
Optimal Role of Diuretics Initial or Add-On
Therapy?
Despite a long, well-established record of efficacy,
the role of thiazides in hypertension therapy con-
tinues to provoke debate.68 Based on the benefits
observed in studies in which thiazides were used
as initial therapy in a stepped-care approach, with
agents added sequentially, guidelines for hyper-
tension therapy in the United States have advocated
thiazides as initial treatment since 1977, when
the first Joint National Committee report was is-
sued.69 In contrast, British guidelines recommend
diuretics more selectively, such as for use in the
elderly and blacks.70 European guidelines endorse
several antihypertensive agents, including diuret-
ics, as initial therapy.71 An update from the Joint
National Committee is anticipated in the near
future.
Slightly more than one third of the patients in
the United States who are eligible for thiazide ther-
apy actually receive it.72 The reasons for this low
rate of use may include the absence of corporate
promotion of diuretics and the heavy marketing of
newer medications.73 Pleiotropic effects in addition
to the blood-pressurelowering effect have been
hypothesized for ACE inhibitors and ARBs but
not for thiazides.74 In addition, there is discussion
about the importance of adverse electrolyte and
metabolic changes in association with thiazides
and whether these changes abrogate the overall
blood-pressurelowering benefits of the drugs.75,76
Cumulative evidence from trials suggests that
the magnitude of blood-pressure lowering is the
most important determinant in reducing the car-
diovascular risks associated with hypertension.49
As diuretics are eventually required in many hy-
pertensive patients to achieve blood-pressure goals,
debate about their role as initial or add-on thera-
py may be predominantly academic.
n engl j med 361;22 nejm.org november 26, 2009
The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
S a fe t y a nd A dv er se Effec t s
of Thi a zide s
Much of the criticism against thiazides is directed
toward their adverse-effect profile, but low doses
are usually well tolerated and have been shown to
improve quality-of-life measures.77 Most compli-
cations of thiazide therapy are related to the dose
and duration of use (Fig. 2).
Thiazides can reduce the excretion of calcium
and uric acid and thereby increase their plasma
levels, and the drugs increase potassium and
magnesium excretion, potentially leading to hy-
pokalemia and hypomagnesemia. On average,
measured plasma potassium levels decrease by
about 0.3 to 0.4 mmol per liter with typical dos-
ing of thiazide monotherapy,78 but coadministra-
tion of an ACE inhibitor or an ARB can reduce the
incidence of clinically relevant hypokalemia. In
ALLHAT, the average potassium level among pa-
tients receiving chlorthalidone decreased from
4.3 mmol per liter to 4.1 mmol per liter over a
4-year period.52 The plasma potassium level should
be measured at baseline, before the initiation of
thiazide monotherapy; a potassium-sparing agent
may be considered if the baseline level is less
than 3.8 mmol per liter.
Maintaining potassium homeostasis is essen-
tial, since epidemiologic evidence implicates hy-
pokalemia in the pathogenesis of thiazide-induced
dysglycemia.79 The importance of potassium bal-
ance is also emphasized by findings from the
SHEP study, in which the rate of coronary events
among patients with potassium levels below 3.5
mmol per liter was reduced to a lesser degree than
among patients with higher potassium levels.80
Hypokalemia can be managed with the use of a
potassium-sparing agent or supplemental potas-
sium chloride. Potassium-sparing agents are pre-
ferred because they correct the underlying cause.
Dietary salt restriction can also minimize thia
zide-induced potassium losses.28
New-onset diabetes has been reported in pa-
tients receiving thiazides; however, newly diag-
nosed diabetes occurs over time in many hyper-
tensive patients, regardless of which class of
antihypertensive agent is used. Data suggest that
receipt of thiazides, as compared with other anti-
hypertensive agents, over several years may lead to
an excess of 3 to 4% of new cases of diabetes.79
In ALLHAT, the rates of fatal and nonfatal myo-
cardial infarction were similar among patients re-
2159
 The n e w e ng l a n d j o u r na l of m e dic i n e

Increased proximal Na+ reabsorption Increased arterial resistance

Possible limitation of blood-pressure lowering

Decreased plasma volume Increased proximal Ca++ reabsorption

Decreased Ca++ clearance

Decreased cardiac output Orthostasis
Adaptations to diuretics

Diuretic
Diuresis Hypomagnesemia

Increased distal Ca++ reabsorption

Decreased uric acid clearance Hyperuricemia

Impaired renal H2O excretion Hyponatremia

Decreased renal blood flow Prerenal azotemia

Increased activation of RAAS Kaliuresis Hypokalemia

Impaired insulin resistance Dysglycemia

Figure 2. Potential Complications of Diuretics and Their Associated Mechanisms.

RAAS denotes the reninangiotensinaldosterone system.

ceiving chlorthalidone, despite the fact that newly
diagnosed diabetes was more frequent in the
chlorthalidone group (seen in 11.6% of patients)
than in the amlodipine group (9.8%) or the
lisinopril group (8.1%).54 To date, no analyses of
ALLHAT data have indicated that the develop-
ment of diabetes obviates the benefit of the
thiazide.53-57 Similar findings have been report-
ed in a large meta-analysis and in long-term
follow-up data of the SHEP cohort.81-83 Despite
these reassurances, newly diagnosed diabetes in
patients receiving thiazides remains a potential
concern, since the period of monitoring for long-
term diabetes-related adverse effects on cardiovas-
cular outcomes may have been too short.
Few clinically relevant drug interactions oc-
cur with the use of thiazides; most notably, non-
steroidal antiinflammatory drugs diminish the
therapeutic effect of thiazides by causing sodium
retention. Thiazides can increase serum lipid lev-
els, primarily total cholesterol and low-density li-
poprotein cholesterol levels, by approximately 5 to
7% in the first year of therapy.84 A summary of
common clinical problems encountered with thi-
azides, and potential solutions, are detailed in
Table 3. COLOR FIGURE

Version 4 10/29/09
Author Ernst
Fig # 2
Title Diuretic Use in Hyper.
ME JL
JRI
2160 n engl j med 361;22 nejm.org november 26, 2009
DE
Artist LAM
AUTHOR PLEASE NOTE:
The New England Journal of Medicine Figure has been redrawn and type has been reset
Please check carefully
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Issue date 11/26/09
Copyright 2009 Massachusetts Medical Society. All rights reserved.
 drug ther apy

Table 3. Common Clinical Problems Encountered with Thiazides, and Possible Solutions.*

Clinical Problem Possible Solution

Attack of acute gouty arthritis
Hypokalemia (serum potassium 3.5 mmol/liter)
Increase in serum creatinine from baseline level
No apparent response to hydrochlorothiazide at a dose
of 25 mg/day
Report of dizziness on standing
Discovery of asymptomatic hyponatremia
Thiazide therapy recommended for patient with docu-
mented history of allergy to a sulfa antibiotic
Report of muscle cramps
Impaired fasting glucose level or diabetes at baseline
Development of diabetes during thiazide therapy
Nocturia or incontinence
Baseline GFR <3040 ml/min/1.73 m2 of body-surface
area
Development of sun sensitivity
Obtain uric acid level, and discontinue thiazide if level is elevated. Recheck
level after resolution of the attack, and assess the need for prophylaxis.
Use another antihypertensive agent if uricosuric prophylaxis is not
tolerated or indicated.
Correct hypomagnesemia if present. Add potassium-sparing agent or supple-
mental potassium chloride. Advise salt restriction. If blood pressure is not
controlled, consider adding a RAAS inhibitor.
Assess hydration status and discontinue any concurrent and unnecessary
nephrotoxic drugs (e.g., NSAIDs). Recognize that a slight elevation in cre-
atinine may be the result of improved blood-pressure control in patients
with microvascular disease, in whom renal function is dependent on
blood pressure for adequate perfusion.
Assess lifestyle and advise salt restriction if needed. Consider increase in di-
uretic dose, switch to longer-acting thiazide such as chlorthalidone, or
addition of RAAS inhibitor.
Check for orthostasis. Reduce diuretic dose if necessary. Assess hydration
status, and insure diuretic is administered in the morning. Instruct the
patient to stand up slowly.
Assess concurrent medications (e.g., SSRIs) and determine risks and benefits
of continuing thiazide. Evaluate patient for excessive water intake.
Sulfa antibiotic allergy is not a contraindication to receiving a thiazide. The
risk for cross-sensitivity appears to be more dependent on an underlying
propensity for atopy than on any specific cross-reactivity among the
classes. If true allergy to thiazide is documented, ethacrynic acid
(a nonsulfa-containing loop diuretic) can be used.
Check serum potassium level and normalize if low. Consider another diuretic
if electrolyte levels are normal.
Institute appropriate management of cardiovascular risk factors. Thiazide use
is not precluded.
Institute appropriate management of diabetes and related cardiovascular risk
factors. The average excess increase in glucose attributed to thiazide use
is approximately 35 mg/dl (0.20.3 mmol/liter). Thiazides will probably
be necessary for achieving blood-pressure targets. The addition of a RAAS
inhibitor should be considered, especially if blood pressure is not con-
trolled.
Avoid thiazide dosing in the afternoon or evening. Limit evening fluid intake.
Consider discontinuing thiazide if symptoms are intolerable.
Substitute furosemide or torsemide. A practical formula for determining the
dose of furosemide, in milligrams to be given twice daily, is as follows:
(patient age + blood urea nitrogen level) 2. Torsemide can be given
once daily.
Encourage sunscreen use.

* GFR denotes glomerular filtration rate, NSAID nonsteroidal antiinflammatory drug, and SSRI selective serotonin-reuptake inhibitor.
Examples of reninangiotensinaldosterone system (RAAS) inhibitors are angiotensin-convertingenzyme inhibitors and angiotensin
IIreceptor blockers.

C onclusions pharmacologic discoveries have advanced the treat-

ment of any disease in such a profound and en-
Diuretics are a heterogeneous class of antihyper- during manner.
tensive medications that have long demonstrated Dr. Ernst reports receiving advisory fees and consulting fees
effectiveness in reducing blood pressure and the from Takeda Pharmaceuticals; and Dr. Moser, advisory fees
risk of cardiovascular events. With proper atten- from Takeda Pharmaceuticals and GlaxoSmithKline and lecture
fees from Kaiser and the Consortium for Southeastern Hyper-
tion to appropriate selection, dosing, and moni- tension Control. No other potential conflict of interest relevant
toring, diuretic-based regimens can greatly improve to this article was reported.
the ability to achieve blood-pressure goals. Few

n engl j med 361;22 nejm.org november 26, 2009 2161

The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l
References
1. Freis ED, Wanko A, Wilson IM, Par-
rish AE. Treatment of essential hyperten-
sion with chlorothiazide (Diuril): its use
alone and combined with other antihyper-
tensive agents. J Am Med Assoc 1958;166:
137-40.
2. Moser M, Macaulay AI. Chlorothia
zide as an adjunct in the treatment of es-
sential hypertension. Am J Cardiol 1959;3:
214-9.
3. Southworth H. Acidosis associated
with the administration of para-amino-
benzene-sulfonamide (Prontylin). Proc Soc
Exp Biol Med 1937;36:58-61.
4. Strauss MB, Southworth H. Urinary
changes due to sulfonamide administra-
tion. Bull Johns Hopkins Hosp 1938;63:
41-5.
5. Novello FC, Sprague JM. Benzothiadi-
azine dioxides as novel diuretics. J Am
Chem Soc 1957;79:2028-9.
6. Beyer KH Jr, Baer JE, Russo HF, Noll
R. Electrolyte excretion as influenced by
chlorothiazide. Science 1958;127:146-7.
7. Ellison DH, Velzquez H, Wright FS.
Thiazide-sensitive sodium chloride cotrans-
port in early distal tubule. Am J Physiol
1987;253:F546-F554.
8. Beermann B, Groschinsky-Grind M.
Clinical pharmacokinetics of diuretics.
Clin Pharmacokinet 1980;5:221-45.
9. Welling PG. Pharmacokinetics of the
thiazide diuretics. Biopharm Drug Dispos
1986;7:501-35.
10. Jackson EK. Diuretics. In: Brunton LL,
Lazo JS, Parker KL, eds. Goodman & Gil-
mans the pharmacological basis of thera-
peutics. 11th ed. New York: McGraw-Hill,
2006:737-70.
11. Hasannejad H, Takeda M, Taki K, et
al. Interactions of human organic anion
transporters with diuretics. J Pharmacol
Exp Ther 2004;308:1021-9.
12. Sica DA, Moser M. Diuretic therapy in
cardiovascular disease. In Black HR, Elliott
WJ, eds. Hypertension: a companion to
Braunwalds heart disease. Philadelphia:
W.B. Saunders, 2007.
13. Riess W, Dubach UC, Burckhardt D,
Theobald W, Vuillard P, Zimmerli M.
Pharmacokinetic studies with chlorthali-
done (Hygroton) in man. Eur J Clin Phar-
macol 1977;12:375-82.
14. Dieterle W, Wagner J, Faigle JW. Bind-
ing of chlorthalidone (Hygroton) to blood
components in man. Eur J Clin Pharmacol
1976;10:37-42.
15. Temperini C, Cecchi A, Scozzafava A,
Supuran CT. Carbonic anhydrase inhibi-
tors: sulfonamide diuretics revisited
old leads for new applications? Org Biomol
Chem 2008;6:2499-506.
16. Bengtsson C, Johnsson G, Sanner-
stedt R, Werk L. Effect of different doses
of chlorthalidone on blood pressure, serum
potassium, and serum urate. Br Med J
1975;1:197-9.
2162 n engl j med 361;22 nejm.org november 26, 2009
The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
of m e dic i n e
17. Lund-Johansenn P. Hemodynamic
changes in long-term diuretic therapy of
essential hypertension: a comparative study
of chlorthalidone, polythiazide and hydro-
chlorothiazide. Acta Med Scand 1970;187:
509-18.
18. Tarazi RC, Dustan HP, Frohlich ED.
Long-term thiazide therapy in essential
hypertension: evidence for persistent al-
terations in plasma volume and renin ac-
tivity. Circulation 1970;41:709-17.
19. Wilson IM, Freis ED. Relationship be-
tween plasma and extracellular fluid vol-
ume depletion and the antihypertensive
effect of chlorothiazide. Circulation 1959;
20:1028-36.
20. Roos JC, Boer P, Koomans HA, Gey-
skes GG, Dorhout Mees EJ. Haemodynam-
ic and hormonal changes during acute and
chronic diuretic treatment in essential hy-
pertension. Eur J Clin Pharmacol 1981;19:
107-12.
21. Gifford RW Jr, Mattox VR, Orvis AL,
Sones DA, Rosevear JW. Effect of thiazide
diuretics on plasma volume, body electro-
lytes, and excretion of aldosterone in hy-
pertension. Circulation 1961;24:1197-205.
22. de Carvalho JG, Dunn FG, Lohmller
G, Frohlich ED. Hemodynamic correlates
of prolonged thiazide therapy: compari-
son of responders and nonresponders.
Clin Pharmacol Ther 1977;22:875-80.
23. Shah S, Khatri I, Freis ED. Mechanism
of antihypertensive effect of thiazide di-
uretics. Am Heart J 1978;95:611-8.
24. Hughes AD. How do thiazide and
thiazide-like diuretics lower blood pres-
sure? J Renin Angiotensin Aldosterone
Syst 2004;5:155-60.
25. Finnerty FA Jr. Step-down treatment
of mild systemic hypertension. Am J Car-
diol 1984;53:1304-7.
26. Stamler R, Stamler J, Grimm R, et al.
Nutritional therapy for high blood pres-
sure: final report of a four-year random-
ized controlled trial the Hypertension
Control Program. JAMA 1987;257:1484-
91.
27. Ellison DH. Diuretic resistance: physi-
ology and therapeutics. Semin Nephrol
1999;19:581-97.
28. Ram CV, Garrett BN, Kaplan NM.
Moderate sodium restriction and various
diuretics in the treatment of hyperten-
sion. Arch Intern Med 1981;141:1015-9.
29. Almeshari K, Ahlstrom NG, Capraro
FE, Wilcox CS. A volume-independent
component to post-diuretic sodium reten-
tion in humans. J Am Soc Nephrol 1993;3:
1878-83.
30. Ellison DH, Velzquez H, Wright FS.
Adaptation of the distal convoluted tubule
of the rat: structural and functional ef-
fects of dietary sodium intake and chronic
diuretic infusion. J Clin Invest 1989;83:
113-26.
31. Kim GH. Long-term adaptation of re-
nal ion transporters to chronic diuretic
treatment. Am J Nephrol 2004;24:595-605.
32. Ellison DH. The physiologic basis of
diuretic synergism: its role in treating di-
uretic resistance. Ann Intern Med 1991;
114:886-94.
33. Taler SJ, Textor SC, Augustine JE. Re-
sistant hypertension: comparing hemody-
namic management to specialist care.
Hypertension 2002;39:982-8.
34. SHEP Cooperative Research Group.
Prevention of stroke by antihypertensive
drug treatment in older persons with iso-
lated systolic hypertension: final results
of the Systolic Hypertension in the Elderly
Program (SHEP). JAMA 1991;265:3255-64.
35. Materson BJ, Reda DJ, Cushman WC,
et al. Single-drug therapy for hyperten-
sion in men: a comparison of six antihy-
pertensive agents with placebo. N Engl J
Med 1993;328:914-21. [Erratum, N Engl J
Med 1994;330:1689.]
36. Chobanian AV, Bakris GL, Black HR,
et al. The Seventh Report of the Joint Na-
tional Committee on Prevention, Detection,
Evaluation, and Treatment of High Blood
Pressure: the JNC 7 report. JAMA 2003;289:
2560-72. [Erratum, JAMA 2003;290:197.]
37. Calhoun DA, Jones D, Textor S, et al.
Resistant hypertension: diagnosis, evalu-
ation, and treatment: a scientific statement
from the American Heart Association Pro-
fessional Education Committee of the
Council for High Blood Pressure Research.
Hypertension 2008;51:1403-19.
38. Knauf H, Mutschler E. Diuretic effec-
tiveness of hydrochlorothiazide and furo-
semide alone and in combination in chron-
ic renal failure. J Cardiovasc Pharmacol
1995;26:394-400.
39. Dussol B, Moussi-Frances J, Morange S,
Somma-Delpero C, Mundler O, Berland Y.
A randomized trial of furosemide vs hy-
drochlorothiazide in patients with chron-
ic renal failure and hypertension. Nephrol
Dial Transplant 2005;20:349-53.
40. Paton RR, Kane RE. Long-term di-
uretic therapy with metolazone of renal
failure and the nephrotic syndrome. J Clin
Pharmacol 1977;17:243-51.
41. Finnerty FA Jr, Maxwell MH, Lunn J,
Moser M. Long-term effects of furosemide
and hydrochlorothiazide in patients with
essential hypertension: a two-year com-
parison of efficacy and safety. Angiology
1977;28:125-33.
42. Saha C, Eckert GJ, Ambrosius WT, et
al. Improvement in blood pressure with
inhibition of the epithelial sodium chan-
nel in blacks with hypertension. Hyper-
tension 2005;46:481-7.
43. Brater DC. Diuretic therapy. N Engl J
Med 1998;339:387-95.
44. Nishizaka MK, Zaman MA, Calhoun
DA. Efficacy of low-dose spironolactone
in subjects with resistant hypertension.
Am J Hypertens 2003;16:925-30.

45. Veterans Cooperative Study Group on
Antihypertensive Agents. Effects of treat-
ment on morbidity in hypertension: re-
sults in patients with diastolic blood pres-
sures averaging 115 through 129 mm Hg.
JAMA 1967;202:1028-34.
46. Psaty BM, Lumley T, Furberg CD, et
al. Health outcomes associated with vari-
ous antihypertensive therapies used as first-
line agents: a network meta-analysis. JAMA
2003;289:2534-44.
47. Law MR, Morris JK, Wald NJ. Use of
blood pressure lowering drugs in the pre-
vention of cardiovascular disease: meta-
analysis of 147 randomised trials in the
context of expectations from prospective
epidemiological studies. BMJ 2009;338:
b1665.
48. Wright JM, Musini VM. First-line
drugs for hypertension. Cochrane Data-
base Syst Rev 2009;3:CD001841.
49. Turnbull F. Effects of different blood-
pressure-lowering regimens on major car-
diovascular events: results of prospectively-
designed overviews of randomised trials.
Lancet 2003;362:1527-35.
50. Blood Pressure Lowering Treatment
Trialists Collaboration. Do men and wom-
en respond differently to blood pressure-
lowering treatment? Results of prospec-
tively designed overviews of randomized
trials. Eur Heart J 2008;29:2669-80.
51. Idem. Effects of different regimens to
lower blood pressure on major cardiovas-
cular events in older and younger adults:
meta-analysis of randomised trials. BMJ
2008;336:1121-3.
52. ALLHAT Officers and Coordinators
for the ALLHAT Collaborative Research
Group. Major outcomes in high-risk hy-
pertensive patients randomized to an-
giotensin-converting enzyme inhibitor
or calcium channel blocker vs diuretic:
the Antihypertensive and Lipid-Lowering
Treatment to Prevent Heart Attack Trial
(ALLHAT). JAMA 2002;288:2981-97. [Er-
ratum, JAMA 2003;289:178, 2004;291:
2196.]
53. Whelton PK, Barzilay J, Cushman WC,
et al. Clinical outcomes in antihyperten-
sive treatment of type 2 diabetes, impaired
fasting glucose concentration, and nor-
moglycemia: Antihypertensive and Lipid-
Lowering Treatment to Prevent Heart At-
tack Trial (ALLHAT). Arch Intern Med
2005;165:1401-9.
54. Barzilay JI, Davis BR, Cutler JA, et al.
Fasting glucose levels and incident diabe-
tes mellitus in older nondiabetic adults
randomized to receive 3 different classes
of antihypertensive treatment: a report
from the Antihypertensive and Lipid-Low-
ering Treatment to Prevent Heart Attack
Trial (ALLHAT). Arch Intern Med 2006;
166:2191-201.
55. Wright JT Jr, Harris-Haywood S, Pres-
sel S, et al. Clinical outcomes by race in
hypertensive patients with and without
n engl j med 361;22 nejm.org november 26, 2009
The New England Journal of Medicine
Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.
drug ther apy
the metabolic syndrome: Antihyperten-
sive and Lipid-Lowering Treatment to Pre-
vent Heart Attack Trial (ALLHAT). Arch
Intern Med 2008;168:207-17.
56. Black HR, Davis B, Barzilay J, et al.
Metabolic and clinical outcomes in non-
diabetic individuals with the metabolic
syndrome assigned to chlorthalidone, am-
lodipine, or lisinopril as initial treatment
for hypertension: a report from the Anti-
hypertensive and Lipid-Lowering Treatment
to Prevent Heart Attack Trial (ALLHAT).
Diabetes Care 2008;31:353-60.
57. Wright JT Jr, Probstfield JL, Cushman
WC, et al. ALLHAT findings revisited in
the context of subsequent analyses, other
trials, and meta-analyses. Arch Intern Med
2009;169:832-42.
58. Beckett NS, Peters R, Fletcher AE, et
al. Treatment of hypertension in patients
80 years of age or older. N Engl J Med
2008;358:1887-98.
59. Carter BL, Ernst ME, Cohen JD. Hy-
drochlorothiazide versus chlorthalidone:
evidence supporting their interchange-
ability. Hypertension 2004;43:4-9.
60. Elliott WJ, Grimm RH Jr. Using di-
uretics in practice one opinion. J Clin
Hypertens (Greenwich) 2008;10:856-62.
61. Wing LMH, Reid CM, Ryan P, et al.
A comparison of outcomes with angio-
tensin-convertingenzyme inhibitors and
diuretics for hypertension in the elderly.
N Engl J Med 2003;348:583-92.
62. Jamerson K, Weber MA, Bakris GL, et
al. Benazepril plus amlodipine or hydro-
chlorothiazide for hypertension in high-
risk patients. N Engl J Med 2008;359:2417-
28.
63. Bowlus WE, Langford HG. A com-
parison of the antihypertensive effect of
chlorthalidone and hydrochlorothiazide.
Clin Pharmacol Ther 1964;5:708-11.
64. Finnerty FA Jr. A double-blind study
of chlorthalidone and hydrochlorothiaz-
ide in an outpatient population of mod-
erate hypertensives. Angiology 1976;27:
738-44.
65. Ernst ME, Carter BL, Goerdt CJ, et al.
Comparative antihypertensive effects of
hydrochlorothiazide and chlorthalidone
on ambulatory and office blood pressure.
Hypertension 2006;47:352-8.
66. The Multiple Risk Factor Intervention
Trial Research Group (MRFIT). Mortality
after 10 years for hypertensive participants
in the Multiple Risk Factor Intervention
Trial. Circulation 1990;82:1616-28.
67. Psaty BM, Lumley T, Furberg CD.
Meta-analysis of health outcomes of
chlort halidone-based vs nonchlorthali-
done-based low-dose diuretic therapies.
JAMA 2004;292:43-4.
68. Messerli FH, Bangalore S, Julius S.
Risk/benefit assessment of beta-blockers
and diuretics precludes their use for first-
line therapy in hypertension. Circulation
2008;117:2706-15.
69. Report of the Joint National Commit-
tee on Detection, Evaluation, and Treat-
ment of High Blood Pressure: a coopera-
tive study. JAMA 1977;237:255-61.
70. National Collaborating Centre for
Chronic Conditions. Hypertension: man-
agement in adults in primary care: phar-
macological update. London: Royal Col-
lege of Physicians, 2006.
71. Mancia G, De Backer G, Dominiczak
A, et al. 2007 ESH-ESC practice guidelines
for the management of arterial hyperten-
sion. J Hypertens 2007;25:1751-62. [Erra-
tum, J Hypertens 2007;25:2184.]
72. Muntner P, Krousel-Wood M, Hyre
AD, et al. Antihypertensive prescriptions
for newly treated patients before and after
the main Antihypertensive and Lipid-low-
ering Treatment to Prevent Heart Attack
Trial results and Seventh Report of the
Joint National Committee on Prevention,
Detection, Evaluation, and Treatment of
High Blood Pressure guidelines. Hyper-
tension 2009;53:617-23.
73. Moser M. Why are physicians not pre-
scribing diuretics more frequently in the
management of hypertension? JAMA 1998;
279:1813-6.
74. Zhou MS, Schulman IH, Jaimes EA,
Raij L. Thiazide diuretics, endothelial func-
tion, and vascular oxidative stress. J Hy-
pertens 2008;26:494-500.
75. McInnes GT, Yeo WW, Ramsay LE,
Moser M. Cardiotoxicity and diuretics:
much speculation, little substance. J Hy-
pertens 1992;10:317-35.
76. Freis ED. The efficacy and safety of
diuretics in treating hypertension. Ann
Intern Med 1995;122:223-6.
77. Grimm RH Jr, Grandits GA, Cutler JA,
et al. Relationships of quality-of-life mea-
sures to long-term lifestyle and drug
treatment in the Treatment of Mild Hy-
pertension Study. Arch Intern Med 1997;
157:638-48.
78. Cushman WC, Khatri I, Materson BJ,
et al. Treatment of hypertension in the
elderly. III. Response of isolated systolic
hypertension to various doses of hydro-
chlorothiazide: results of a Department of
Veterans Affairs cooperative study. Arch
Intern Med 1991;151:1954-60.
79. Carter BL, Einhorn PT, Brands M, et
al. Thiazide-induced dysglycemia: call for
research from a working group from the
National Heart, Lung, and Blood Insti-
tute. Hypertension 2008;52:30-6.
80. Franse LV, Pahor M, Di Bari M, Somes
GW, Cushman WC, Applegate WB. Hy-
pokalemia associated with diuretic use
and cardiovascular events in the Systolic
Hypertension in the Elderly Program. Hy-
pertension 2000;35:1025-30.
81. Turnbull F, Neal B, Algert C, et al. Ef-
fects of different blood pressure-lowering
regimens on major cardiovascular events
in individuals with and without diabetes
mellitus: results of prospectively designed
2163
 drug ther apy

overviews of randomized trials. Arch In-
tern Med 2005;165:1410-9.
82. Curb JD, Pressel SL, Cutler JA, et al.
Effect of diuretic-based antihypertensive
treatment on cardiovascular disease risk
in older diabetic patients with isolated
systolic hypertension. JAMA 1996;276:
1886-92. [Erratum, JAMA 1997;277:1356.]
83. Kostis JB, Wilson AC, Freudenberger
RS, Cosgrove NM, Pressel SL, Davis BR.
Long-term effect of diuretic-based thera-
py on fatal outcomes in subjects with
isolated systolic hypertension with and
without diabetes. Am J Cardiol 2005;95:
29-35.
84. Savage PJ, Pressel SL, Curb JD, et al.
Influence of long-term, low-dose, diuret-
ic-based, antihypertensive therapy on glu-
cose, lipid, uric acid, and potassium levels
in older men and women with isolated
systolic hypertension: the Systolic Hyper-
tension in the Elderly Program. Arch In-
tern Med 1998;158:741-51.
Copyright 2009 Massachusetts Medical Society.

Coyote Bluffs, Arizona Howie Garber, M.D.

2164 n engl j med 361;22 nejm.org november 26, 2009

The New England Journal of Medicine

Downloaded from nejm.org by LUIS VILLAR on July 28, 2017. For personal use only. No other uses without permission.
Copyright 2009 Massachusetts Medical Society. All rights reserved.