A comparison of topical azelaic acid 20% cream

and topical metronidazole 0.75% cream in the

treatment of patients with papulopustular rosacea
Stuart Maddin, MD, FRCPC Vancouver, British Columbia, Canada
Background: Although it is important for physicians to have sufficient clinical data on
which to base treatment decisions, little comparative data exist regarding newer treatment
modalities for rosacea.
Objective: The goal of the study was to compare the efficacy and safety of topical azelaic
acid 20% cream and topical metronidazole 0.75% cream in the treatment of patients with
papulopustular rosacea. Parameters of patient satisfaction to treatment were also assessed.
Methods: Forty patients with the clinical manifestation of symmetric facial rosacea were
investigated in this single-center, double-blind, randomized, contralateral split-face comparison clinical trial.
Results: After 15 weeks of treatment, both azelaic acid and metronidazole induced significant, albeit equal reductions in the number of inflammatory lesions (pustules and
papules). A significantly higher physician rating of global improvement was achieved with
azelaic acid. Changes in the rosacea signs and symptoms of dryness, burning, telangiectasia, and itching were equal between treatments. A reduction in erythema tended toward significance with azelaic acid at week 15. A trace amount of stinging on application was noted
with azelaic acid; however, such discomfort did not appear to concern patients because
their overall impression of azelaic acid was superior to that of metronidazole.
Conclusion: Azelaic acid 20% cream provides an effective and safe alternative to metronidazole 0.75% cream with the added benefit of increased patient satisfaction.
(J Am Acad Dermatol 1999;40:961-5.)

Rosacea is a chronic, hyperemic, dermatologic

disorder of unknown origin characterized by persistent erythema and telangiectasia. Because of the
additional signs of edema, papules and pustules,
rosacea often resembles acne clinically, but its
pathogenesis does not involve hyperseborrhea,
abnormal follicular keratinization, or follicular
bacteria.1
Rosacea often proves difficult to treat. Current
therapy has focused on topically administered
broad-spectrum antibiotics such as erythromycin
and clindamycin.1,2 Although topical antibiotics
have demonstrated significant usefulness, longFrom the Division of Dermatology, University of British Columbia.
Supported by a grant provided by Allergan, Inc.
Reprint requests: Stuart Maddin, MD, FRCPC, University of British
Columbia, Division of Dermatology, Faculty of Medicine, Skin
Care Centre, 835 W 10th Ave, Vancouver, British Columbia, V5Z
4E8, Canada.
Copyright 1999 by the American Academy of Dermatology, Inc.
0190-9622/99/$8.00 + 0 16/1/97316

term systemic antibiotic therapy is not tolerated

well because of the side effects (eg, gastrointestinal irritation, candidal vaginitis). Topical corticosteroids have been effective in combating the
inflammation associated with rosacea, but longterm use is not recommended because of their skin
atrophogenic potential. Isotretinoin is effective for
severe cases of rosacea with a nodulocystic component,1,3 and lasers4,5 can be used effectively to
treat telangiectasia and occasionally control background erythema. However, because rosacea is a
chronic condition, the ideal treatment should be
safe and convenient for use over long periods.
Azelaic acid is a naturally occurring, saturated
dicarboxylic acid possessing antibacterial,
comedolytic, and anti-inflammatory activities.6
Applied topically as a 20% cream, it has been
shown to be effective in the treatment of comedonal acne and inflammatory acne as well as various cutaneous hyperpigmentary disorders characterized by hyperactive or abnormal melanocyte
961

962 Maddin
function.1 An inhibition of the production of reactive proinflammatory oxygen species (hydroxy
and super oxyradicals) from neutrophils possibly
accounts for azelaic acids anti-inflammatory
effect.7 This anti-inflammatory property may
account for the therapeutic efficacy observed in a
recent study where azelaic acid 20% cream was
more effective than its vehicle cream in reducing
the number of inflammatory lesions and degree of
erythema associated with rosacea.8
Although azelaic acid appears to be an effective
option for the treatment of rosacea, comparative
studies with other effective therapies are absent
from the literature. Studies with metronidazole, a
popular broad-spectrum antibiotic, have shown that
both the oral and topical (0.75% cream) formulations are effective in treating the erythema, pustules, and papules characteristic of rosacea.3,9 This
study was undertaken to compare the efficacy and
safety of topical azelaic acid 20% cream with that
of topical metronidazole 0.75% cream in the treatment of patients with papulopustular rosacea.
PATIENTS AND METHODS
This study was a single-center, randomized, doubleblind, contralateral, split-face comparison of topical
azelaic acid 20% cream (Allergan, Inc.) and topical
metronidazole 0.75% cream (Galderma Canada, Inc.) in
patients with papulopustular rosacea. Study medications were randomly assigned to either side of the face
(ie, azelaic acid cream on one side of the face and
metronidazole cream on the other side) and were
applied twice daily for 15 weeks. Study visits occurred
at baseline (week 0) and at weeks 3, 6, 9, and 15.
Patients were instructed as to the proper procedure for
applying study medication with emphasis on preventing
crossover contamination. Before commencing patient
activity, ethical approval was obtained from the
Integrated Research Inc. Ethics Review Committee.

Patients
Patients were recruited from the Division of
Dermatology Skin Care Centre at the University of
British Columbia. Patients of at least 21 years of age
and of any race or gender were eligible for study participation provided they presented with rosacea characterized by persistent symmetrical erythema affecting
the cheeks and had at least 10 inflammatory papules or
pustules. Patients were required to meet the following
washout periods for medications that could influence
the course of rosacea: topical metronidazole, 21 days;
topical corticosteroids or antibiotics, 14 days; systemic
corticosteroids, antibiotics, or any investigational med-

Journal of the American Academy of Dermatology

June 1999

ications, 28 days; isotretinoin or tretinoin, 1 year. All

patients were capable of understanding and signing an
informed consent form.
Patients were excluded from study participation if
they had nonsymmetric distribution of inflammatory
lesions (ie, a variance of more than 8 inflammatory
lesions between each side of the face), significant concomitant dermatologic disorders, a presence of other
conditions that could affect study results, allergy to any
component of the study medications, or a history of
noncompliance with medical treatment. Female patients
who were pregnant, breast-feeding, or of child-bearing
potential and not practicing a reliable method of birth
control were not enrolled.

Assessments
Assessments were made throughout the study by
both the investigator and the patients.
Investigator assessments. The investigator performed manual counts of inflammatory lesions (papules
and pustules) on each side of the face at baseline (week
0), and at 3, 6, 9, and 15 weeks. Disease severity and
signs and symptoms (erythema, telangiectasia, dryness,
itching, stinging, burning) were assessed on each side
of the face by means of a 4-point scale. Global improvement (ie, clinical responsiveness relative to baseline)
was assessed by means of a 6-point scale. Adverse
events and concomitant medications used since the last
visit were also recorded. Throughout the study, the
same observer made all assessments for any given
patient.
Patient assessments. Patients assessed disease
severity and signs and symptoms at baseline (week 0)
and at 9 and 15 weeks. Medication acceptability and
patient overall impression were assessed at weeks 9
and 15.

Statistical tests
On the basis of a 2-tailed of .05, 80% power, a standard deviation of 12 lesions, a correlation of 0.07, and a
drop-out rate of 20%, 40 patients were required to detect
a clinically significant difference of 5 inflammatory
lesions between treatment groups. Statistical analyses
were performed on all patients randomized into the study
(intent-to-treat). Appropriate statistical comparisons were
made by means of chi-square tests, paired t tests, and
analyses of variance (ANOVA). Primary comparisons
were based on the change in the response variable from
baseline to week 15 with additional analyses comparing
responses at each of the assessment visits (ie, at weeks 3,
6, 9, and 15). Descriptive statistics are reported as mean
SEM (standard error of the mean). All data from
patients who dropped out were retained and analyzed
according to the intent-to-treat analyses. Data from the
last available visit of these patients were carried forward.

Journal of the American Academy of Dermatology

Volume 40, Number 6, Part 1

Maddin 963

Table I. Demographics

No.
Mean age (y) (range)
Rosacea duration (y)
<1
1-2
3-5
6-10
> 10

Male

Female

11
52.2 (26-75)

29
49.6 (34-69)

0.0%
0.0%
36.4%
18.2%
45.4%

0.0%
6.9%
24.1%
20.7%
48.3%

RESULTS

A total of 40 white patients were enrolled into

the study, of which 37 completed 15 weeks of
treatment. A summary of patient demographics is
provided in Table I. One patient discontinued
because of an adverse event not related to treatment (cardiac arrest) and 2 patients discontinued
the study for personal reasons (relocated; insufficient time for study participation). The most common impact that rosacea had on patient lifestyle
was the need to use make-up to mask signs of
rosacea (28 patients, 70%). Although more than
half of the study patients (22 patients, 55%)
thought that rosacea had no impact on their overall
lifestyle, more than one fourth (10 patients, 26%)
avoided social functions because of rosacea.

Fig 1. Reductions in inflammatory lesions during treatment. Mean count ( SEM) at each study visit is indicated for azelaic acid (o) and metronidazole (x). No
significant differences were observed between treatment groups.

Efficacy assessments
Inflammatory lesions. Both treatments
demonstrated a significant reduction in the number
of inflammatory lesions after 15 weeks of treatment (P < .0001); however, the difference between
treatments was not significant. The mean number
SEM of inflammatory lesions decreased from
11.3 0.88 at baseline to 2.43 0.55 at week 15
for the azelaic acid treatment and from 11.40
1.03 to 3.49 0.71 for the metronidazole treatment
(P = .43; Fig 1) This represents a 78.5% and
69.4% reduction in inflammatory lesions for the
azelaic acid and metronidazole treatments, respectively. When pustules and papules were considered
separately, similar reductions were detected across
the 15 weeks of treatment for both treatments (P
.24; Table II).
Physician rating of individual signs and
symptoms. Assessments of erythema, telangiectasia, dryness, itching, stinging, and burning were

Fig 2. Change in physician rating of global improvement during treatment. Mean rating ( SEM) for azelaic acid (o) and metronidazole (x) is illustrated.
Significant differences between treatment groups were
observed at week 9 and week 15 (*P = .002, **P = .05;
ANOVA).

made on a scale of 1 to 4 (with half-point increments), where 1 = none, 2 = mild or trace, 3 =

moderate, and 4 = severe or marked.
Both treatments achieved a significant improvement in dryness and burning over time (P < .0001);
however, the treatments did not differ significantly
in the change in these severity scores from baseline to week 15 (P .59). Erythema improved over
time for both treatments (P = .01), and the difference between treatments tended towards significance in favor of azelaic acid at week 15. The
mean erythema score SEM decreased from 3.21

Journal of the American Academy of Dermatology

June 1999

964 Maddin

Table II. Effect of treatment on pustules and papules (mean SEM)

Azelaic acid

Metronidazole

Time (wk)

No.

Papules

Pustules

Papules

Pustules

0
3
6
9
15

40
39
38
37
37

10.33 0.79
8.21 0.77
6.61 0.70
4.41 0.64
2.30 0.53

0.98 0.36
0.51 0.24
0.21 0.11
0.24 0.11
0.14 0.11

10.73 0.92
9.00 0.88
7.05 0.86
4.97 0.90
3.24 0.60

0.68 0.29
0.95 0.39
0.47 0.23
0.41 0.20
0.24 0.17

Table III. Medication acceptability at week 15

Medication acceptability

Nondrying
Nongreasy
Cosmetic appearance
very acceptable
Greasiness appearance
very acceptable

Azelaic acid

Metronidazole

74%
61%
59%

71%
53%
43%

44%

38%

0.06 at baseline to 2.38 0.11 at week 15 with

azelaic acid and from 3.12 0.06 to 2.61 0.12
with metronidazole representing reductions of
25.5% and 18.7%, respectively (P = .07). Neither
treatment demonstrated a significant effect over
time on telangiectasia or itching. A trace amount
of transient stinging upon application was noted
with azelaic acid treatment. The mean stinging
severity score SEM increased from 1.30 0.11
at baseline to 1.45 0.10 at week 15 for azelaic
acid and decreased from 1.30 0.11 to 1.03
0.03 with metronidazole, representing an increase
of 11.5% and a decrease of 20.8%, respectively
(P = .02).
By means of a 5-point scale where 0 = none, 1
= trace, 2 = mild, 3 = moderate, and 4 = severe,
patient ratings of redness confirmed the physician
ratings of erythema. Patient assessment of redness
(mean SEM) was reduced from 2.61 0.14 at
baseline to 1.58 0.15 at week 15 with azelaic
acid and from 2.74 0.11 to 2.00 0.15 with
metronidazole, representing 39.5% and 27.0%
reductions, respectively (P = .05).
Physician rating of global improvement.
Global improvement was rated as 1 (100% clearance of disease signs and symptoms), 2 (75% to
99% clearance of disease signs and symptoms), 3
(50% to 74% clearance of disease signs and symp-

toms), 4 (< 50% clearance), 5 (no detectable

improvement from baseline), and 6 (exacerbation).
Global improvement ratings improved significantly
over time (P < .0001) for both treatments. However,
as illustrated in Fig 2, azelaic acid demonstrated
significantly more improvement than metronidazole by week 9 (P = .002), and by week 15 this
difference was borderline significant (P = .05).
Medication acceptability. Patients evaluated
the acceptability of the study medications on either
side of their face with respect to comfort, drying,
cosmetic appearance, and greasiness. Satisfaction
was rated on a scale of 0 to 4 with lower scores
indicating higher satisfaction. Although patients
were consistently more satisfied with azelaic acid,
differences between treatments were not significant (Table III).
Patient overall impression. Patients rated
their overall impression of study medication by
means of the same 5-point scale as above. At
week 15, azelaic acid and metronidazole were
rated as 1.87 0.12 and 2.33 0.15, respectively, indicating patient preference for azelaic acid
(P = .02). Additionally, at week 15 significantly
more patients indicated they would use azelaic
acid again versus metronidazole (92% versus
66%, P = .005), and when asked to compare the
study medication to other rosacea medication
used in the past, 57% and 21% of the patients
found azelaic acid and metronidazole to be more
effective than the previous medication, respectively (P = .01).
Compliance. Compliance to study medication
was very high with only 4 patients having reported
missing a maximum of 4 applications.
Safety assessments
No patients discontinued the study because of
adverse events. Other than the trace stinging noted

Journal of the American Academy of Dermatology

Volume 40, Number 6, Part 1

upon application of azelaic acid, no treatmentrelated adverse events were noted in this study.
DISCUSSION

This is the first published comparative study of

azelaic acid cream and metronidazole cream in
patients with rosacea. Results demonstrate that
these medications are equally effective in reducing
the number of inflammatory lesions and associated signs and symptoms of rosacea. However, the
physician rating of global improvement was significantly more favorable for azelaic acid than
metronidazole.
Provided that study patients present with symmetric disease, split-face comparative studies offer
the important advantage of each patient serving as
his/her own control. The primary disadvantage of
such a design is crossover contamination of study
medication. Although guarantees cannot be given
that it did not occur, emphasis was placed on
avoiding crossover contamination when patients
were instructed on proper medication application.
Although both medications were equally beneficial in their effect on erythema, dryness, and
burning, and lacked effect on telangiectasia and
itching, azelaic acid use was associated with trace
stinging upon application. Discomfort caused by
the stinging did not appear to concern patients
because their overall impression of azelaic acid
was superior to that of metronidazole and significantly more patients indicated that they would prefer to use azelaic acid again in the future.
When rosacea flares or becomes recalcitrant to
current therapy, it is important for the physician to
have safe and effective treatment alternatives.
Although some of the known risks of available
treatments for rosacea (ie, isotretinoin, cortico-

Maddin 965
steroids, oral agents) could be weighed against the
established safety of azelaic acid, published comparative efficacy and safety data have not been
available.10 The present study indicates that azelaic acid 20% cream provides a very effective and
safe alternative to metronidazole 0.75% cream
with the added benefit of increased patient satisfaction.
REFERENCES
1. Plewig G. Rosacea. In: Fitzpatrick TB, Eisen AZ, Wolff
K, Freedberg IM, Austen KF, editors. Dermatology in
general medicine, vol 1. New York: McGraw-Hill; 1993.
p. 727-35.
2. Nanda VS. Common dermatoses. Am J Obstet Gynecol
1995;173:488-95.
3. Bleicher PA, Charles JH, Sober AJ. Topical metronidazole therapy for rosacea. Arch Dermatol 1987;123:60914.
4. Lowe NJ, Behr KL, Fitzpatrick R, Goldman M, RuizEsparza J. Flash lamp pumped dye laser for rosaceaassociated telangiectasia and erythema. J Dermatol Surg
Oncol 1991;17:522-5.
5. Goodman GJ, Bekhor PS, Richards SW. Update on
lasers in dermatology. Med J Aust 1996;164:681-6.
6. Fitton A, Goa KL. Azelaic acid: a review of its pharmacological properties and therapeutic efficacy in acne and
hyperpigmentary skin disorders. Drugs 1991;41:780-98.
7. Akamatsu H, Miyaichi Y, Komura J. Effect of azelaic
acid on neutrophil function: a possible cause for its efficacy in treating pathogenetically unrelated diseases.
Arch Dermatol Res 1991;23:162-6.
8. Carmichael AJ, Marks R, Graupe KA, Zaumseil RP.
Topical azelaic acid in the treatment of rosacea. J
Dermatol Treat 1993;4:S19-S22.
9. Aronson IK, Rumsfield JA, West DP, Alexander J, Fisher
JH, Paloucek FP. Evaluation of topical metronidazole
cream in acne rosacea. Drug Intell Clin Pharm 1987;21:
346-51.
10. Graupe K, Cunliffe WJ, Gollnick HPM, Zaumseil RP.
Efficacy and safety of topical azelaic acid (20 percent
cream): an overview of results from European clinical
trials and experimental reports. Cutis 1996;57:20-35.

AVAILABILITY OF JOURNAL BACK ISSUES

As a service to our subscribers, copies of back issues of the Journal of the American Academy of Dermatology
for the preceding 5 years are maintained and are available for purchase from Mosby until inventory is depleted at a cost of $17.00 per issue. The following quantity discounts are available: 25% off on quantities of 12
to 23, and one third off on quantities of 24 or more. Please write to Mosby, Inc., Periodical Subscription
Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318, or call (800)453-4351 or (314)453-4351
for information on availability of particular issues. If unavailable from the publisher, photocopies of complete
issues may be purchased from UMI, 300 N. Zeeb Rd., Ann Arbor, MI 48106, (313)761-4700.