650427

research-article2016
CMSXXX10.1177/1203475416650427Journal of Cutaneous Medicine and SurgeryAsai et al

Review Article

Journal of Cutaneous Medicine and Surgery

Canadian Clinical Practice Guidelines 2016, Vol. 20(5) 432445

The Author(s) 2016
Reprints and permissions:
for Rosacea sagepub.com/journalsPermissions.nav
DOI: 10.1177/1203475416650427
jcms.sagepub.com

Yuka Asai1, Jerry Tan2, Akerke Baibergenova3, Benjamin Barankin4,

Chris L. Cochrane5, Shannon Humphrey6, Charles W. Lynde3,
Danielle Marcoux7, Yves Poulin8, Jason K. Rivers5,9, Mariusz Sapijaszko10,
R. Gary Sibbald3, John Toole11, Marcie Ulmer5,12, and Catherine Zip13

Abstract
Rosacea is a chronic facial inflammatory dermatosis characterized by background facial erythema and flushing and may
be accompanied by inflammatory papules and pustules, cutaneous fibrosis and hyperplasia known as phyma, and ocular
involvement. These features can have adverse impact on quality of life, and ocular involvement can lead to visual dysfunction.
The past decade has witnessed increased research into pathogenic pathways involved in rosacea and the introduction
of novel treatment innovations. The objective of these guidelines is to offer evidence-based recommendations to assist
Canadian health care providers in the diagnosis and management of rosacea. These guidelines were developed by an expert
panel of Canadian dermatologists taking into consideration the balance of desirable and undesirable outcomes, the quality
of supporting evidence, the values and preferences of patients, and the costs of treatment. The 2015 Cochrane review
Interventions in Rosacea was used as a source of clinical trial evidence on which to base the recommendations.

Keywords
inflammatory dermatoses, laser, dermatology, rosacea

Rosacea is a chronic facial inflammatory dermatosis esti- negative impact on health-related QoL of patients with
mated to affect up to 10% of Western populations.1,2 rosacea,13 and other studies have identified increased anxi-
Extrapolation to the 2015 Canadian population would esti- ety and depression in this group.14 Stigmatization can also
mate that rosacea affects 3.6 million Canadians,3 and a recent be an issue with facial redness and with rhinophyma. In
survey found that 46% of respondents with rosacea had been particular, the latter has an erroneous cultural association
living with symptoms for over 10 years.4 with alcoholism, with synonyms including rum nose and
Rosacea is characterized by clinical features of facial whiskey nose.
flushing and redness, which may be accompanied by inflam-
matory papules and pustules, fibrotic changes and skin thick- 1
Division of Dermatology, Queens University, Kingston, ON, Canada
ening known as phyma, and ocular involvement.5,6 Based on 2
University of Western Ontario, Windsor, ON, Canada
clinical presenting features, rosacea has been classified into 3
University of Toronto, Toronto, ON, Canada
4 subtypes, which may occur concurrently: (1) erythemato- 4
Toronto Dermatology Centre, Toronto, ON, Canada
5
telangiectatic rosacea (ETR) is characterized by flushing and Bearing Biomedical Consulting, Vancouver, BC, Canada
6
persistent centrofacial erythema; (2) papulopustular rosacea Department of Dermatology and Skin Science, University of British
Columbia, Vancouver, BC, Canada
(PPR) by the presence of inflammatory papules and pustules 7
CHU Sainte-Justine, University of Montreal, Montreal, QC, Canada
with ETR; (3) phymatous by marked skin thickening and 8
Laval University, Quebec City, QC, Canada
surface nodularities, most commonly affecting the nose; and 9
Pacific DermAesthetics, Vancouver, BC, Canada
10
(4) ocular by blepharitis and conjunctivitis, which often Division of Dermatology, Department of Medicine, University of
occur in conjunction with other cutaneous features and can Alberta, Edmonton, AB, Canada
11
University of Manitoba, Winnipeg, MB, Canada
lead to visual dysfunction in severe cases.7-9 Representative 12
Carruthers & Humphrey, Vancouver, BC, Canada
photographs of rosacea features are presented in Figure 1. 13
University of Calgary, Calgary, AB, Canada
The impact of rosacea on quality of life (QoL) may not
Corresponding Author:
be fully appreciated.10-12 Embarrassment and desire to hide Jerry Tan, University of Western Ontario, 2224 Walker Rd, Ste 300,
the skin are common among Canadians with rosacea.4 A Windsor, Ontario N8W5L7, Canada.
recent systematic review found that all studies reported a Email: jerrytan@bellnet.ca
Asai et al 433

Target Audience
This document is intended for Canadian health care provid-
ers, including pharmacists, nurse practitioners, family physi-
cians, dermatologists, and other clinicians involved with care
of patients with rosacea.

Pathogenesis
While the pathogenesis of rosacea is incompletely under-
stood, recent investigations suggest involvement of the
innate immune system and cutaneous neurovascular dysreg-
ulation (reviewed in Two etal15 and Steinhoff etal16).
The cutaneous innate immune system is a primordial non-
specific body defense mechanism comprising the intact skin
barrier and its structural (keratinocytes, sebocytes) and
immune (mast cells, neutrophils, natural killer cells, den-
dritic cells) cellular elements. Antimicrobial peptides
(AMPs), cathelicidins, and defensins are soluble defense fac-
tors, which are primarily secreted by keratinocytes in
response to external triggers such as injury, UV radiation,
barrier disruption, and diverse pathogens such as bacteria,
viruses, and fungi. These peptides have proinflammatory and
vasoactive properties.
Activation and involvement of the immune system have
been demonstrated in all cutaneous subtypes of rosacea,
including ETR, previously considered to be due solely to
vascular dilation.17 In rosacea, there is increased expression
Figure 1. Clinical subtypes of rosacea. Representative photos of both the propeptide and 37amino acid cathelicidin,
for (A) mild to moderate fixed-background centrofacial erythema LL-37, as well as the enzyme KLK 5 that catalyzes conver-
(erythematotelangiectatic rosacea), (B) moderate inflammatory
papules with fixed-background centrofacial erythema
sion of the propeptide to LL-37.18 The latter may be due to
(papulopustular rosacea), (C) prominent nodules and soft tissue increased TLR-2 expression from keratinocytes. TLRs, or
hypertrophy at nasal tip and alar regions with nasal deformation Toll-like receptors, are pathogen-associated sensor mole-
(severe late-stage rhinophyma), and (D) ocular rosacea with lipid cules, which act to detect and signal the presence of micro-
inspissation of Meibomian glands lining the lid margin. bial structures. Potential triggers for TLR-2 in rosacea
include the saprophytic mite Demodex folliculorum, mite-
related bacteria such as Bacillus oleronius, and
Previously, treatment options were limited, but recent Staphylococcus epidermidis.15 TLRs may also be activated
advances in the understanding of rosacea pathogenesis have by reactive oxygen species resulting from ultraviolet
led to development of new treatments. exposure.
Reactive oxygen species may also activate neurogenic
receptors, such as transient receptor potential channels
Objective
(TRPs), that are expressed on neural tissues, keratinocytes,
The objective of this rosacea clinical practice guideline and endothelial cells.19 TRPV1 can be activated by heat, eth-
(CPG) is to assist Canadian health care providers in the diag- anol, or spicy food, and TRPA1 can be activated by cold,
nosis and management of rosacea. formalin, or other chemicals. TRP activation induces release
of substance P and calcitonin generelated peptide, leading
to pain/edema and vasodilation, respectively.20 Increased
Scope
serine protease activity may upregulate TRP and activate
The scope of these guidelines is rosacea management in protease-activated receptors, which may lead to the decreased
adults based on medications available in Canada. Specifically barrier function observed in rosacea.21
excluded are the following: pediatric rosacea, pyoderma A genome-wide association study of more than 22000
faciale, granulomatous rosacea, and rosaceiform dermatiti- individuals of European ancestry, of whom just over 10%
des and diagnostic mimics such as democidosis, acne vul- had a prior diagnosis of rosacea, investigated the influence of
garis, and folliculitis. genetics in rosacea. In this cohort, significant associations
434 Journal of Cutaneous Medicine and Surgery 20(5)

were found with a single-nucleotide polymorphism and 3 2015. The panel included a chair/methodologist trained in
human leukocyte antigen (HLA) alleles, supporting a prob- dermatology and epidemiology with no conflict of interest in
able genetic component in the pathogenesis of rosacea.22 this therapeutic area (Y.A.). All panelists received no
remuneration.
Prior to the meeting, the panelists were surveyed for their
Severity Grading
most preferred treatments for each of the following clinical
Clinical trials for rosacea interventions required the develop- presentations of rosacea: erythema, papules/pustules, and
ment of severity grading scales for persistent facial erythema phymatous and ocular features. These were then presented in
and papules/pustules.23 The Clinician Erythema Assessment aggregate to the group for feedback and further deliberation.
(CEA) and Patient Self-Assessment of Erythema (PSA) scales,
based on the categories clear/almost clear, mild, moderate and Literature Search, Review, and Adaptation
severe, have been shown to be valid24 and reliable.24,25 Severity
of papules and pustules is assessed as part of global grading The information derived from the panel survey was reviewed
scales and is typically rated as clear/almost clear, mild, moder- in conjunction with a summary of findings tables from a
ate, and severe.23 There are presently no specific scales for Cochrane review entitled Interventions for Rosacea,28
severity grading of phymatous or ocular rosacea. which provided information on efficacy and quality of evi-
A standard grading system encompassing all 4 signs, dence of interventions for relevant outcomes.
accompanied with patient photos, has been proposed by The Cochrane review did not address flushing or mainte-
Wilkin etal.26 nance therapy. To address the former, a literature search was
performed up to October 25, 2015, on PubMed to find stud-
ies evaluating treatment of flushing associated with rosacea.
Differential Diagnosis Keywords used were clonidine, nadolol, propranolol,
The diagnosis of rosacea is a clinical one, comprising many carvedilol, rosacea, and flushing. Only relevant studies con-
possible clinical features. For each feature, the following dif- ducted in English were included in this study. For mainte-
ferential diagnoses should be considered: nance, previously identified studies were screened for
randomised controlled trials (RCTs) of rosacea that evalu-
Flushing: carcinoid syndrome, systemic mastocytosis, ated outcomes beyond 12 weeks, preferably to 1 year.
benign cutaneous flushing and perimenopause, med-
ullary carcinoma of the thyroid, and pancreatic and Formulation and Interpretation of
renal cell tumors Recommendations
Centrofacial erythema: photodamage, systemic lupus
A Delphi voting process, whereby 75% was predetermined
erythematosus, facial dermatitis, seborrheic dermati-
to be the threshold for consensus, was undertaken to estab-
tis, psoriasis, and keratosis rubra pilaris faceii
Papules/pustules: acne vulgaris (characterized by lish the strength of recommendations. Provided with the evi-
presence of comedones) and folliculitis dence from the Cochrane review, the panelists applied
Phymatous changes: nonmelanoma skin cancer, gran- predetermined methodology to develop recommenda-
ulomatous infiltration (which may be infectious in tions.29,30 For each recommendation, a direction, either
origin such as rhinoscleroma or noninfectious such as against or for, and a strength, either weak or strong, was
sarcoidosis), and B- and T-cell lymphomas given. The panel considered factors within 4 domains (see
Table 1) in determining the recommendation strength.30
Recommendations likely to apply to all or virtually all
Methods patients would be given a strong recommendation.
Recommendations appropriate for some but not all patients
Nomination of Expert Panel (ie, those in whom the net benefit is small or uncertain, the
Two authors (J.T. and C.L.) recruited an expert panel, via evidence is not of high quality, patient preferences are vari-
invitations to the medical advisory board of the Acne and able or unknown, or costs or resource use present a barrier)
Rosacea Society of Canada and the Canadian Dermatology would be determined to be weak. Thus, even in the presence
Association, to deliberate and vote on treatments. Criteria for of very high-quality evidence (see Table 2) for a given treat-
panelist selection included prior guidelines development ment, a weak recommendation may be appropriate if other
experience; publication history and/or national prominence in treatments, or no treatment, present viable alternatives.
rosacea research; working knowledge of Grading of For scenarios in which confidence in effect estimates was
Recommendations, Assessment, Development, and low, trade-offs were closely balanced, patient specific values
Evaluation (GRADE)27; commitment to completion of online and preferences were highly variable, and/or costs (resource
Delphi surveys and authorship of specific sections of the implications) were unknown, a no recommendation category
guidelines; and attendance at 1 meeting in Toronto in February was available.30
Asai et al 435

Table 1. Determinants of a Recommendations Strength.

Domains That Contribute to the

Strength of Recommendation Comment
Balance between desirable and The larger the difference between desirable and undesirable outcomes, the more likely a strong
undesirable outcomes recommendation is warranted. The smaller the net benefit, the more likely a weak recommendation
is warranted
Confidence in effect estimates The higher the quality of evidence, the more likely a strong recommendation is warranted
(see Table 2)
Confidence in values and The greater the variability in patient values and preferences for the intervention or the greater the
preferences and variability uncertainty in those values and preferences, the more likely a weak recommendation is warranted
Resource use The higher the costs of an intervention (the more resources consumed), the less likely a strong
recommendation is warranted

Adapted from Andrews J, Guyatt G, Oxman AD, etal. GRADE guidelines: 14. Going from evidence to recommendations: the significance and
presentation of recommendations. J Clin Epidemiol. 2013;66(7):719-725.

Table 2. Significance of the 4 Levels of Evidence.30

Quality Level Current Definition

High We are very confident that the true effect lies close to that of the estimate of the effect
Moderate We are moderately confident in the effect estimates: the true effect is likely to be close to the estimate of the effect, but
there is a possibility that it is substantially different
Low Our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect
Very low We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the
estimate of the effect

Estimates of treatment cost were made available to the Recommendations

panelists for consideration during final approval of the rec-
ommendations and algorithm. This section contains the expert panels consensus recom-
mendations for specific clinical features of rosacea and the
rationale and evidence, if available, underpinning the recom-
Specific questions addressed in these guidelines mendations. Further guidance in selecting appropriate ther-
1. What treatments are available for the redness of rosa- apy based on clinical features may be found in the clinical
cea (erythema)? decision-making algorithms in Figures 2 to 5.
2. What treatments are available for the inflammatory Few treatments effectively address multiple clinical fea-
papules and pustules of rosacea? tures, so treatment should be targeted to the symptoms con-
3. What treatments are available for the skin thickening sidered most burdensome by the patient. Combination
(phymatous features) of rosacea? treatment should be considered if multiple clinical features
4. What treatments are available for the ocular signs and are present. Gentle skin care, moisturizers, use of sun protec-
symptoms of rosacea? tion, and avoidance of triggers are recommended, based on
expert opinion, for all patients with rosacea.

Disclaimer Erythema
The recommendations within these guidelines serve as gen- Prior to initiating treatment for background erythema, health
eral advice based on current evidence, not as legal standards. care providers should discuss with patients the possibility
Clinical research evidence is derived from well-defined, that papules and pustules, if present, might become more
tightly controlled group data and may not be adequately spe- visually prominent upon reduction of background erythema.
cific for the circumstances of individual patients or general- Initiating treatment for both features concurrently should be
izable to populations outside of the original study groups. considered. A treatment algorithm can be found in Figure 2.
Additionally, as evidence-based guidelines focus only on We suggest the following for treatment of erythema of
treatments with high-quality evidence, some effective treat- rosacea:
ment with inadequate evidence may not be represented.
Accordingly, the most appropriate treatment for an individ- A1. Topical brimonidine. (Weak recommendation: high con-
ual patient derives from informed decision sharing with his fidence in effect estimate but perceived variability in patient
or her physician. values and preferences)
436 Journal of Cutaneous Medicine and Surgery 20(5)

Azelaic acid has shown limited efficacy in 5 studies for

treatment of erythema of rosacea, with decreases in erythema
of 44% to 48% compared with 28% to 38% for placebo.40-43
No improvement was seen for telangiectasia.
Adverse events, notably irritation, are common, which
may prompt health care providers to suggest other treatments
first. Azelaic acid is covered by some provincial formularies
and private plans.

A4. Vascular laser or intense pulsed light therapy. (Weak rec-

ommendation: very low confidence in effect estimate)
Vascular laser treatment (Nd:YAG or pulsed dye laser
[PDL]) or intense pulsed light (IPL) may ameliorate ery-
thema. We rated our confidence in the effect estimate as
very low because the efficacy of vascular laser or IPL for
erythema has not been evaluated by a placebo controlled
trial; however, its use was supported by panelists based on
clinical experience. Low-quality evidence suggests that
PDL is more effective than Nd:YAG and IPL.44,45
The efficacy of this intervention depends on training and
expertise of the treating physician. Swelling and redness may
persist for several weeks with low risk of scarring.
Figure 2. Treatment algorithm for erythema of rosacea. IPL, Improvement can be rapid and significant; however, multiple
intense pulsed light.
sessions may be required. These interventions may be costly
and access may be limited. This treatment should be offered
Topical brimonidine was demonstrated in 2 high-quality to all patients, acknowledging that the patients preferences
studies to be an effective treatment for erythema of rosacea. and values and treatment cost will influence their decision.
On day 29 of treatment, a 2-grade improvement on the CEA
scale was seen in 31.5% of patients in the brimonidine group A5. Oral doxycycline.(Weak recommendation; high confi-
but only 9.2% of those on vehicle.31 dence in effect estimate but variability in patient values and
Brimonidine tartrate was well tolerated during 4 weeks of preferences with concerns regarding cost, potential adverse
daily treatment, and reported adverse events were mild and events, and uncertainty concerning efficacy for background
transient. The rate of worsening of erythema after the 4-week erythema without papules/pustules)
treatment was approximately 4%.31 The panel noted that bri- There is high-quality evidence showing efficacy of doxy-
monidine is costlier than some other treatments, notably met- cycline for reduction of erythema of rosacea. However, this
ronidazole, and is not covered by provincial drug formularies was based on a study of patients with papulopustular rather
at this time. However, it is available on several private plans. than erythematotelangiectatic rosacea. The effect may have
Case reports of worsening or rebound erythema are an addi- been due to reduction in perilesional erythema associated
tional consideration.32,33 with reduction in papules/pustules, so efficacy for back-
ground erythema alone is uncertain. Doxycycline 40 mg
A2. Topical metronidazole. (Weak recommendation: moder- (modified release) and 100 mg appear equivalent, with the
ate confidence in effect estimate) 100-mg dose being associated with a significantly higher
Metronidazole has been shown in 6 studies to confer sta- rate of adverse events, primarily gastrointestinal in nature.46
tistically and clinically significant improvement in erythema Doxycycline, like all tetracycline group antibiotics, is
by patient and physician assessment and is more effective contraindicated during pregnancy and has been associated
than placebo.34-39 with side effects such as photosensitivity and esophagitis.47
The rate of adverse events for metronidazole is similar to The 40-mg dose is significantly more costly than the 100-mg
placebo. Reactions, which were generally mild, included dose and is not covered by provincial drug formularies. For
itching, skin irritation, and dry skin. Metronidazole is avail- more information, please see Antibiotic Resistance.
able on many provincial formularies as well as several pri-
vate plans and has an extensive history of safe use in patients A6. Skin care and camouflage. (Weak recommendation: very
with rosacea. low confidence in effect estimate)
Over-the-counter skin care and cosmetic products are
A3. Topical azelaic acid. (Weak recommendation: high confi- used commonly and may ameliorate or conceal mild ery-
dence in effect estimate) thema associated with rosacea. We rated our confidence in
Asai et al 437

B3. Topical metronidazole. (Weak recommendation: moder-

ate confidence in effect estimate)
There is moderate-quality evidence that metronidazole
leads to reduction in the number of lesions; however, these
data were skewed and inadequately reported.28 One study
comparing the 1% and 0.75% concentrations found no sig-
nificant difference in efficacy.49 For safety and other infor-
mation, please see Recommendation A2.

B4. Oral doxycycline.(Weak recommendation, high confi-

dence in effect estimate but variability in patient values and
preferences regarding cost and potential adverse events)
There is high-quality evidence showing efficacy of doxy-
cycline for treatment of papules and pustules of rosacea,28
and doses of 40 mg (modified release) and 100 mg appear
equivalent. For safety and other information, please see
Recommendation A5 and Antibiotic Resistance.

Figure 3. Treatment algorithm for papules and pustules of B5. Oral tetracycline.(Weak recommendation: moderate
rosacea. confidence in effect estimate but variability in patient values
and preferences regarding potential adverse events)
Oral tetracycline is effective for papules and pustules,
the effect estimate as very low because efficacy has not with 1 study reporting a mean difference of 14.64 fewer
been evaluated by an RCT. However, these products are lesions for tetracycline compared with placebo.50
safe and widely available. Properly selected skin care regi- Oral tetracycline is generally well tolerated but can cause
mens may be beneficial in helping to repair the skins bar- gastrointestinal disturbances,47,51 and all tetracycline group
rier function, dysfunction of which may contribute to antibiotics are contraindicated during pregnancy.52
rosacea pathogenesis.21 Tetracycline is widely available and less costly than doxycy-
cline and most topicals; however, clinicians should discuss
Papules and Pustules with patients whether the benefits outweigh the risks of gas-
trointestinal distress and potential for selection for antibiotic
We suggest the following for treatment of papules and pus- resistance. For more info, please see Antibiotic Resistance.
tules of rosacea. A treatment algorithm can be found in
Figure 3. B6. Oral isotretinoin.(Weak recommendation; high confi-
dence in effect estimate but variability in patient values and
B1. Topical ivermectin. (Weak recommendation: high confi- preferences regarding potential adverse events)
dence in effect estimate but variability in patient values and In 1 study, low-dose isotretinoin (0.3 mg/kg) was at least as
preferences due to cost) effective for the reduction of lesions as 100 mg doxycycline.53
In 2 large RCTs with a combined total of 1371 patients, There have been no placebo-controlled trials; however, panel-
ivermectin was found to confer statistically significant ists reported good results in their own practices, and isotreti-
improvement in patient- and physician-assessed global out- noin may be a good choice for those in whom tetracycline
comes and also significant and clinically important reduc- group antibiotics were not effective or are contraindicated.
tions in lesions compared with placebo.48 Isotretinoin is associated with potential adverse events
Ivermectin was associated with fewer dermatologic requiring careful monitoring during treatment and is abso-
adverse events than placebo, with patients reporting less dry lutely contraindicated during pregnancy due to high risk of
skin and itching. The panel noted that ivermectin is costlier teratogenicity.54,55 Low-dose and intermittent-dose regimens
than other medications for this indication and is currently not may reduce the frequency and severity of adverse events.56,57
covered by provincial drug formularies; however, it is listed Because isotretinoin treatment for rosacea is likely to be
on several private plans. more long term than for acne, its use is cautioned in females
of childbearing potential.
B2. Topical azelaic acid. (Weak recommendation: moderate
confidence in effect estimate)
Phyma
In 1 study, azelaic acid resulted in a greater reduction in
papules and pustules than placebo.42 For safety and other We suggest the following for treatment of phymatous features
information, please see Recommendation A3. of rosacea. A treatment algorithm can be found in Figure 4.
438 Journal of Cutaneous Medicine and Surgery 20(5)

Figure 4. Treatment algorithm for phymatous rosacea.

C1. Topical retinoids. (Weak recommendation: very low con- Figure 5. Treatment algorithm for ocular rosacea.
fidence in effect estimate)
Topical retinoids may help minimize progression of rosa- The efficacy of these interventions depends on training
cea-associated phyma. We rated our confidence in the effect and expertise of the treating physician. Treatment may be
estimate as very low because efficacy has not been evaluated costly if not covered by provincial health plans, and access
by RCTs. may be limited. Swelling and redness may persist for sev-
The panelists did not report strong anecdotal evidence for eral weeks or longer. These risks are balanced against the
efficacy; however, given the lack of noninvasive treatment potential for excellent outcomes. This option, if available,
options for phymatous features of rosacea, topical retinoids should be offered to all patients, acknowledging that the
represent a safe option for those with mild to moderate patients preferences and values and treatment cost will
involvement that is less costly than procedural treatments. influence their decision.

C2. Oral tetracycline or doxycycline. (Weak recommendation: C4. Oral isotretinoin. (Weak recommendation: very low con-
very low confidence in effect estimate) fidence in effect estimate but variability in patient values and
Oral tetracycline and doxycycline may also be useful for preferences regarding potential adverse events)
mild phymatous rosacea, particularly if there is an inflamma- Oral isotretinoin may be effective at reducing early phy-
tory component. Although there have been no RCTs for this matous features of rosacea. For phymatous features, we rated
indication, clinicians on the panel reported anecdotal benefit our confidence in the effect estimate as very low because the
for patients. For safety and other information, please see outcome has not been validated; however, panelists felt that
Recommendation B5 and Antibiotic Resistance. it may have some benefit in patients with early phymatous
changes. For safety and other information, please see
C3. Ablative laser surgery, using CO2 or Er:YAG modalities, or Recommendation B6.
surgery, including electrosurgery and cryosurgery. (Weak recom-
mendation: very low confidence in effect estimate and vari-
Ocular
ability in patient values and preferences)
Ablative laser resurfacing, using CO2 or Er:YAG modali- We suggest the following for treatment of ocular features of
ties, and surgery, including electrosurgery, may significantly rosacea. A treatment algorithm can be found in Figure 5.
improve phymatous features of rosacea. We rated our confi-
dence in the effect estimate as very low because the efficacy D1. Lid care and artificial tears.(Weak recommendation:
of these procedural treatments for phymatous features has very low confidence in effect estimate)
not been evaluated by RCTs; however, their use was sup- Over-the-counter ocular hygiene products and artificial
ported by strong panelist sentiment based on clinical tears are used commonly and may alleviate some of the dis-
experience. comfort and irritation associated with ocular rosacea. We
Asai et al 439

rated our confidence in the effect estimate as very low rosacea treatment modalities. An ocular expert can also mon-
because efficacy has not been evaluated by RCTs. These itor for disease progression and mitigate the risk of
products are safe, widely available, and less costly than complications.
cyclosporine drops; however, a hygiene regimen may be
time-consuming and inconvenient, which compromises
Treatments for Flushing Associated With Rosacea
compliance.
While studies of flushing in rosacea exist, they suffer from
D2. Oral doxycycline.(Weak recommendation; low confi- methodological limitations, and results are, in some cases,
dence in effect estimate) contradictory. Propranolol, at doses of 20 mg to 40 mg, was
Doxycyclines efficacy for ocular rosacea has not been found to be effective in 2 open-label studies.62,63 Studies of
evaluated in a placebo-controlled trial; however, 1 open- clonidine have demonstrated variable effects with some
label study of 40 mg once daily demonstrated effective- showing reduction in flushing64,65 and another showing no
ness,58 and another comparing doxycycline with tetracycline evidence of benefit.66 One reported worsening of papulo-
reported equivalent improvement of ocular rosacea symp- pustular features in some patients.64 The single study
toms at 6 months.59 For safety and other information, please reported for nadolol found no benefit.67 Carvedilol, a
see Recommendation A5 and Antibiotic Resistance. nonselective -adrenergic antagonist with 1-antagonist
selectivity, was shown effective in a case of refractory rosa-
D3. Oral tetracycline.(Weak recommendation: low confi- cea-associated flushing.68 No formal recommendation is
dence in effect estimate) made concerning these treatments and they should only be
Oral tetracycline is commonly prescribed for ocular rosa- considered for patients with rosacea whose predominant
cea, despite an absence of studies specifically studying its feature is flushing.
efficacy for ocular symptoms. Its mechanism of action for
ocular rosacea is likely via an anti-inflammatory effect, and
some panelists reported improvement of ocular symptoms
Maintenance Therapy of Rosacea
with their own patients at doses of 500 to 1000 mg/d. For Rosacea requires ongoing care, as it is a chronic condition.
safety and other information, please see Recommendation While evidence for maintenance therapy is lacking, if
B5 and Antibiotic Resistance. improvement is inadequate after 8 to 12 weeks of treatment
or if symptoms worsen, escalation of frequency or dose or
D4. Cyclosporine drops. (Weak recommendation: low confi- use of an alternate treatment is advised. Conversely, treat-
dence in effect estimate) ment can be tapered by reduction in frequency or dose once
Cyclosporine inhibits T-lymphocyte activation and has improvement has been achieved. Adequate improvement
been shown to reduce the number of activated lympho- should be judged based on patient satisfaction and reduction
cytes in the conjunctiva.60 One double-blind RCT of 37 in severity to mild or clear/almost clear. Once rosacea symp-
patients found that topical cyclosporine 0.05% reduced toms are under control, those treatments recommended as
ocular surface disease index (OSDI) (P = .022) and first line for mild rosacea can be considered for long-term
improved tear production (P = .002) compared with artifi- maintenance (Figure 6).
cial tears.61 Currently, only 4 studies have investigated long-term effi-
The study found cyclosporine was well tolerated, although cacy and maintenance therapy, specifically looking at topical
1 patient withdrew from the study, reporting stinging. As metronidazole,37 topical brimonidine tartrate,69 topical iver-
cyclosporine suppresses the immune system, it is contraindi- mectin,70,71 and azelaic acid.70 Some of these were not
cated during conjunctival or ocular infection, and assessment included in the Cochrane review as they were either open-
by an eye expert may be advisable prior to beginning treat- label studies or were recently published.69-71
ment. Cyclosporine 0.05% drops are more costly compared Topical brimonidine has been shown to be effective for
with oral tetracycline, which is commonly prescribed for maintenance therapy of erythema in a 12-month open-
ocular rosacea, and their use requires monitoring and discon- label observational study.69 Long-term use of brimonidine
tinuation in the presence of ocular infection; thus, other treat- was found to be safe with no evidence of tachyphylaxis.
ments may be preferable. Adverse events included worsening of erythema in 6.5%,
worsening of rosacea in 3.6%, and contact dermatitis in
D5. Referral to an ocular expert (ophthalmologist pre- 2.2%.
ferred). (Weak recommendation: no evidence, based on Maintenance therapy with metronidazole was evaluated
expert opinion) as an extension of a trial of combination topical metronida-
Ocular experts may be able to rule out other ocular pathol- zole and oral tetracycline.37 Those that achieved at least 70%
ogy that may be mistaken for rosacea. This is particularly reduction in inflammatory lesions were randomized to
important if cyclosporine treatment is being considered to receive either topical metronidazole or vehicle as mainte-
rule out infection or if symptoms prove refractory to other nance therapy for a 6-month blinded study. Relapse of
440 Journal of Cutaneous Medicine and Surgery 20(5)

prefer doxycycline or tetracycline.47 Furthermore, they noted

the risk of side effects, especially from long-term use, and
price as reasons for avoiding nontetracycline antibiotics.

Implementation
Implementation of these guidelines will be facilitated by
widespread dissemination to professional societies involved
in the care of patients with rosacea, including presentation at
meetings, publications, and online medical education
resources.
Figure 6. Treatment algorithm for maintenance therapy for
rosacea. IPL, intense pulsed light.
Consultation, Endorsement, and Testing
rosacea symptoms occurred in 23% and 42% in the metroni- Prior to publication, input was sought from the following
dazole and control groups, respectively. stakeholders: patients with rosacea from the Acne and Rosacea
Long-term ivermectin was evaluated in a 40-week open- Society of Canada, the Canadian Dermatology Association,
label extension to a 12-week phase 3 vehicle-controlled trial the Canadian Skin Patient Alliance, the Canadian Pharmacists
for papules and pustules.70 In the extension, those originally Association, and Canadian Family Physician Association. A
on vehicle were treated with azelaic acid 15% gel while those listing of comments and feedback from these groups was com-
on topical ivermectin continued for a total of 52 weeks dura- piled, and shortcomings considered important and consistently
tion.71 There was a lower incidence of related adverse events identified were addressed in the manuscript. Pilot testing of
with topical ivermectin compared with azelaic acid gel. After the CPG was conducted in clinical practice of some of the
52 weeks, the proportion of subjects achieving global scores authors from November 2015 to January 2016.
of clear or almost clear was 73% with topical ivermectin. After
40 weeks of azelaic acid 15% gel, the proportion was 57%. Applicability
The potential for remission after a 16-week course of
treatment with ivermectin 1% cream once daily or metroni- Treatment recommendations can be applied at time of initial
dazole cream 0.75% cream twice daily was conducted in visit and modifications thereafter based on clinical and
patients with moderate to severe papulopustular rosacea. For patient-reported outcomes on follow-up visits. Specific
those achieving clearance/almost clearance, initial treatment advice for applying recommendations, clinical follow-up,
with ivermectin 1% cream was shown to significantly extend and treatment modification is provided in Figures 2 to 6.
the duration of remission of papules/pustules compared with
those initially treated with metronidazole 0.75% cream.71 Resource Implications of Applying the
Recommendations
Antibiotic Resistance These guidelines provide evidence-based treatment recom-
All antibiotics present a potential risk of selection for antibi- mendations along with cost information. We anticipate that
otic resistance in the microflora of the skin and other sites72-75; this aggregate information will provide prescribers a means
however, in 2 recent studies, tetracycline resistance was not of rationalizing treatment for individual patients with vary-
detected in Propionibacterium acnes isolated from acne ing values and preferences.
patients.76,77 No increase in the number or severity of resis-
tant organisms in skin flora was observed following twice-
Monitoring or Auditing Criteria
daily use of 20 mg doxycycline.78 Thus, sub-antimicrobial
doses of doxycycline and tetracycline may mitigate the risk Monitoring and audit criteria include use of appropriate clin-
of bacterial resistance. ical and patient-reported outcomes for rosacea severity,
For oral antibiotics other than doxycycline and tetracy- effectiveness, satisfaction with therapy, and adverse effects
cline (minocycline, trimethoprim, azithromycin, erythromy- during initial and follow-up visits.
cin, and metronidazole), there is no high-quality evidence A potential set of auditing criteria may include some or all
supporting their use in patients with papules and pustules of of the following elements:
rosacea. Panelists suggested that they be considered only for
patients in whom tetracycline group antibiotics are contrain- I. Initial assessment
dicated such as during pregnancy, are ineffective, or are A. Evaluation of rosacea signs/symptoms
poorly tolerated. Minocycline has been associated with a B. Evaluation of rosacea signs/symptoms severity
number of rare but severe side effects, leading panelists to C. Evaluation of impact of rosacea
Asai et al 441

D. Appropriate selection of treatment(s) based on Gaps in Knowledge

rosacea signs/symptoms and severity (see rec-
ommendations) Systematic, evidence-based CPGs intrinsically suffer from a
E. Treatment counseling development and publication lag time, despite using the most
a. Medication administration, application, and recent systematic review on rosacea that may omit newer lit-
potential adverse events erature. Furthermore, the Cochrane review only includes
b. Appropriate follow-up for monitoring: RCTs, so any useful information that could be used to make
within 8 to 12 weeks a recommendation, particularly data gathered from open-
label studies, case reports, or case series, may not be avail-
II. Ongoing management able to inform the strength of recommendation. For example,
A. Evaluation of rosacea signs/symptoms compensatory vasodilation and rebound erythema with use
B. Evaluation of rosacea signs/symptoms severity of topical brimonidine have been reported in the literature
C. Evaluation of impact of rosacea but are not captured given its recent addition to the treatment
D. Evaluation of treatment satisfaction arsenal.32,33 Long-term data on severe but rare adverse events
E. Evaluation of patient perceived improvement (or tend to be reported in this manner in the literature, especially
not) for newer treatments.
F. Inquiry into possible adverse events This process of CPG development and the Cochrane
G. Treatment modification if inadequate effective- review exposed knowledge gaps for rosacea related to (1)
ness or adverse event development treatment efficacy and risks, particularly with procedural
therapies and regimens using a combination of treatments;
(2) patient values concerning their rosacea symptoms and the
Updating various treatment options; and (3) a lack of standardized
This document will be updated for validity every 5 years. assessment of rosacea clinical features.
Updates may be provided sooner than scheduled to include
significant new developments such as evidence on existing Treatment Efficacy and Risks
benefits and harms of interventions, development of new
treatments, or changes in available treatments. For fixed erythema of rosacea, only brimonidine tartrate has
high-quality evidence for its efficacy, and nearly half of
patients show no improvement or worsen with its use.31
Discussion Other treatment options, such as metronidazole, azelaic acid,
Rosacea is a common condition that has a wide range of pre- and doxycycline, are less well supported by evidence and
sentations and an increasing array of treatment options.79 had little support for their use among the expert panel. No
CPGs can help navigate these options, particularly when high-quality evidence exists for treatment of transient ery-
developed by a panel providing expert experiential input thema (flushing).
within a foundation of best clinical evidence. Similarly, evidence is lacking on the best method for
treatment of phyma. Case reports support the role of surgery
for improvement of phyma; however, an important knowl-
Other Guidelines edge gap remains concerning the efficacy of medical therapy,
Prior guidelines on treatment of rosacea are largely consensus such as isotretinoin or oral antibiotics, in reducing
based.5,6,79,80 Here we present a CPG for rosacea based on the progression.
recently published update of the Cochrane review Interventions Furthermore, individuals with rosacea rarely present with
for Rosacea.28 The Cochrane Collaboration is a nonprofit, only one clinical feature; most have multiple components to
nongovernmental organization that conducts systematic their presentation, such as erythema and papulopustular
reviews of RCTs. Basing the CPG on this robust and recent lesions or papulopustular lesions and ocular involvement.
systematic review decreases redundancy in literature search, Clinicians often use combination therapy as needed depend-
quality assessment, and meta-analyses, which in turn reduces ing on clinical presentation. However, few studies have
the cost of personnel and funding. The Cochrane study reports addressed combination therapy. Adjunctive skin care therapy
effect estimates as relative risks with confidence intervals, in combination with medical therapy is also often recom-
which may be most accurate and helpful for experts in epide- mended but has little supportive evidence.
miology but can be less familiar to clinicians. This Cochrane
review also provides a summary of findings for interventions,
in which the body of evidence is graded based on assessments
Patient Values
for design limitations, inconsistency in results, indirectness, GRADE methodology requires the consideration of patient
imprecision, and potential for publication bias, whereby quality values in making recommendations; however, several aspects
is rated as very low, low, moderate, or high.81,82 of the patient perspective are uncertain. It is clear that the
442 Journal of Cutaneous Medicine and Surgery 20(5)

burden of seeking treatment will vary among Canadians, as Whilst the past decade has witnessed considerable progress
some will have to travel long distances to see a dermatolo- in clarifying some of the underlying basic mechanisms of dis-
gist. This would be particularly burdensome for treatments ease and the advent of new treatment options, we identify fur-
requiring repeated visits, such as vascular laser for erythema. ther needs in clinical research, including specific outcome
It is unclear how Canadian patients will weigh the financial measures relevant to patients regarding QoL, and knowledge
costs and inconvenience of such barriers to access. gaps in long-term efficacy, combination therapy, and mainte-
Also, further studies of QoL and other adverse effects are nance. Nevertheless, this evidence-based rosacea CPG trans-
necessary to quantitate the impact of this disease, particularly lates the summary of findings from the most recent Cochrane
the less-studied subtypes, phymatous and ocular rosacea. review on this topic and imbues it with experience and exper-
tise of dermatological experts to guide Canadian health care
providers in caring for those with this condition.
Lack of Standardized Assessment
The lack of evidence for treatment of specific clinical fea- Acknowledgments
tures of rosacea may be related to the instruments used to
We acknowledge Angela Ross and Dr Craig Crippen for critical
measure outcomes. Many are global assessments incorporat-
review of the manuscript. We also thank Dr Allan R. Slomovic for
ing multiple clinical features, often based on the rosacea sub- providing a representative image for ocular rosacea (Figure 1D).
types ETR and PPR. Thus, less common subtypes such as
phymatous may be neglected despite their morbidity. Ocular Conflicts and Editorial Independence
rosacea is another subtype that is poorly studied; no clear
The development of these guidelines was funded by Galderma,
definition of ocular rosacea exists, and criteria for appropri-
Pierre Fabre, and Valeant. These guidelines were developed inde-
ate referral to ocular experts need to be determined to avoid pendently by the authors, and the contents and treatment recom-
unnecessary consultations. mendations represent their collective opinion based on best
A recent review of methods used to evaluate the severity evidence. The following steps were implemented to ensure the rec-
of rosacea in clinical trials found only 3 of 32 identified stud- ommendations were free from external influence by industry, third-
ies used standardized assessment methods.23 Measurements party payers, or governmental agencies.
of improvement in rosacea differ greatly depending on ther-
apy used due to the variety of clinical features of rosacea (eg, 1. Exclusion of the panel members involved in solicitation of
brimonidine for erythema but not papules/pustules and oral funding (JT, CL) from writing and voting on treatment sections
isotretinoin for papules/pustules but not erythema). Future 2. Nondisclosure of funding pharmaceutical company
studies need to address improvement of specific features identities until the final draft of the manuscript was
submission-ready
using validated tools.
3. Exclusion of funding pharmaceutical company input into
the conception, design, and development of the CPG proj-
Conclusion ect or in the writing of the final manuscript
4. Invitation of all pharmaceutical and cosmetic companies
In developing these recommendations, we weighed (1) the offering rosacea products to participate as funding spon-
balance of desirable and undesirable outcomes, (2) the qual- sors for unrestricted educational grants
ity of the supporting evidence, (3) the values and preferences 5. Funds obtained were used for travel, accommodation,
of patients, and (4) the costs of treatment. For some clinical meals, and the administrative support group. Honoraria for
featuresnamely, brimonidine for erythema and ivermectin authors and expert panel participation were not provided.
for papules and pustules, respectivelythe benefits of treat-
ments outweigh their harms, are supported by strong evi- Contributor Statements
dence, are expected to be acceptable to patients, but are more Yuka Asai served as chair and methodologic expert, contributed to
costly than treatments supported by lower quality evidence. the conception and design, contributed to the drafting of the article,
For other features, such as phyma, no treatments have been provided critical review, and gave final approval of the version to
demonstrated to be effective in RCTs, and treatments with be published. Jerry Tan and Charles W. Lynde contributed to the
the strongest clinician support, such as surgery, are invasive, conception and design, contributed to the drafting of the article,
are more costly than medical treatment, and may not be read- provided critical review, and gave final approval of the version to
ily available to all Canadians. be published. Chris L. Cochrane contributed to the design, contrib-
uted to the drafting of the article, and provided final approval of the
Indeed, accessibility to therapy is a major issue. Many
version to be published. Akerke Baibergenova, Benjamin Barankin,
individuals cannot afford the most strongly recommended Shannon Humphrey, Danielle Marcoux, Yves Poulin, Jason K.
medications or procedural interventions (laser, IPL therapy). Rivers, Mariusz Sapijaszko, R. Gary Sibbald, John Toole, Marcie
New, effective topical products with lower cost will help Ulmer, and Catherine Zip provided clinical expertise in developing
with the gap in treatment of this disease and, by superseding recommendations, voted on and provided critical review of consen-
systemic antibiotics, could reduce the selection for antibi- sus recommendations, provided critical review of the manuscript,
otic-resistant bacteria. and gave final approval of the version to be published.
Asai et al 443

Declaration of Conflicting Interests 9. Suzuki T. Meibomitis-related keratoconjunctivitis: implica-

tions and clinical significance of meibomian gland inflamma-
The author(s) declared the following potential conflicts of interest
tion. Cornea. 2012;31:S41-S44.
with respect to the research, authorship, and/or publication of this
10. Moustafa F, Lewallen RS, Feldman SR. The psychological
article: All authors disclose funding from Galderma, Pierre Fabre, and
impact of rosacea and the influence of current management
Valeant; however, these funders were kept confidential from all
options. J Am Acad Dermatol. 2014;71(5):973-980.
authors except Jerry Tan and Charles W. Lynde until submission of
11. Aksoy B, Altaykan-Hapa A, Egemen D, Karagoz F, Atakan N.
the manuscript for publication. Akerke Baibergenova has served on an
The impact of rosacea on quality of life: effects of demographic
advisory board for Galderma, Valeant, and Abbvie. Benjamin
and clinical characteristics and various treatment modalities.
Barankin has received honoraria for CME lectures from Galderma,
Br J Dermatol. 2010;163(4):719-725.
Valeant, and Cipher and has been an investigator for Galderma.
12. Scheinfeld NS. Rosacea. SKINmed. 2006;5(4):191-194.
Shannon Humphrey has served as a speaker, consultant, and/or inves-
13. Van Der Linden M, Van Rappard DC, Daams JG, Sprangers
tigator to Allergan, Galderma, GSK, Johnson & Johnson, Kythera,
MA, Spuls PI, De Korte J. Health-related quality of life in
Loreal, Procter & Gamble, Valeant, Revance, and Zeltiq. Charles W.
patients with cutaneous rosacea: a systematic review. Acta
Lynde has served as a clinical investigator, speaker, or consultant to
Dermatovenereol. 2015;95(4):395-400.
Cipher Pharma, Galderma, Johnson & Johnson, Stiefel, Valeant, and
14. Bhm D, Schwanitz P, Stock Gissendanner S, Schmid-

Bayer. Danielle Marcoux has received personal fees from Galderma
Ott G, Schulz W. Symptom severity and psychological
and Valeant. Jason K. Rivers has served on advisory boards for
sequelae in rosacea: results of a survey. Psychol Health Med.
Galderma, Allergan, and Valeant and been a speaker and investigator
2014;19(5):586-591.
for Allergan. R. Gary Sibbald has served as an investigator or advisor
15. Two AM, Wu W, Gallo RL, Hata TR. Rosacea, part I: intro-
for Leo, Galderma, Valeant, and Abbott/Abbvie. Mariusz Sapijaszko
duction, categorization, histology, pathogenesis, and risk fac-
has served as a consultant, speaker, or investigator for Roche, Valeant,
tors. J Am Acad Dermatol. 2015;72(5):749-758.
Galderma, Leo, and Allergan. Jerry Tan has been an advisor, consul-
16. Steinhoff M, Schauber J, Leyden JJ. New insights into rosa-
tant, speaker, and/or trialist for Cipher, Galderma, Pierre-Fabre, and
cea pathophysiology: a review of recent findings. J Am Acad
Valeant. John Toole has served as a clinical investigator for Cutanea
Dermatol. 2013;69(6):S15-S26.
and Galderma. Marcie Ulmer has been a clinical investigator, advisor/
17. Buhl T, Sulk M, Nowak P, etal. Molecular and morpho-
consultant, or speaker for Allergan, Cipher, Galderma, Johnson &
logical characterization of inflammatory infiltrate in rosacea
Johnson, LOreal, Procter & Gamble, and Valeant. Catherine C. Zip
reveals activation of Th1/Th17 pathways. J Invest Dermatol.
has participated on advisory boards for Valeant, Galderma, and Bayer
2015;135(9):2198-2208.
Healthcare. Yuka Asai, Chris L. Cochrane, and Yves Poulin have
18. Yamasaki K, Di Nardo A, Bardan A, etal. Increased serine
nothing further to disclose.
protease activity and cathelicidin promotes skin inflammation
in rosacea. Nat Med. 2007;13(8):975-980.
Funding 19. Graepel R, Fernandes ES, Aubdool AA, Andersson DA, Bevan
The author(s) received no financial support for the research, author- S, Brain SD. 4-oxo-2-nonenal (4-ONE): evidence of transient
ship, and/or publication of this article. receptor potential ankyrin 1-dependent and-independent noci-
ceptive and vasoactive responses in vivo. J Pharmacol Exp
Ther. 2011;337(1):117-124.
References 20. Sulk M, Seeliger S, Aubert J, etal. Distribution and expression
1. Abram K, Silm H, Oona M. Prevalence of rosacea in an of non-neuronal transient receptor potential (TRPV) ion chan-
Estonian working population using a standard classification. nels in rosacea. J Invest Dermatol. 2012;132(4):1253-1262.
Acta Dermatovenereol. 2010;90(3):269-273. 21. Dirschka T, Tronnier H, Flster-Holst R. Epithelial barrier
2. Berg M, Liden S. An epidemiological study of rosacea. Acta function and atopic diathesis in rosacea and perioral dermatitis.
Dermatovenereol. 1988;69(5):419-423. Br J Dermatol. 2004;150(6):1136-1141.
3. Quarterly Demographic Estimates. Vol 29. No 3. Ottawa: 22. Chang ALS, Raber I, Xu J, etal. Assessment of the genetic
Statistics Canada; 2015. Report No.: Contract No.: 91-002-X. basis of rosacea by genome-wide association study. J Invest
4. Rosacea Patient Journey Report. Ottawa: Canadian Skin Dermatol. 2015;135(6):1548-1555.
Patient Alliance; 2015. 23. Hopkinson D, Tuchayi SM, Alinia H, Feldman SR. Assessment
5. Reinholz M, Tietze JK, Kilian K, etal. RosaceaS1 guideline. of rosacea severity: a review of evaluation methods used in
J Dtsch Dermatol Ges. 2013;11(8):768-780. clinical trials. J Am Acad Dermatol. 2015;73(1):138-143.
6. Del Rosso JQ. Advances in understanding and managing rosa- 24. Tan J, Leoni M. Erythema of rosacea: validation of

cea: part 1: connecting the dots between pathophysiological patients self-assessment grading scale. J Drugs Dermatol.
mechanisms and common clinical features of rosacea with 2015;14(8):841-844.
emphasis on vascular changes and facial erythema. J Clin 25. Tan J, Liu H, Leyden JJ, Leoni MJ. Reliability of clinician
Aesthetic Dermatol. 2012;5(3):16. erythema assessment grading scale. J Am Acad Dermatol.
7. Wilkin J, Dahl M, Detmar M, etal. Standard classification 2014;71(4):760-763.
of rosacea: report of the National Rosacea Society Expert 26. Wilkin J, Dahl M, Detmar M, etal. Standard grading system
Committee on the Classification and Staging of Rosacea. J Am for rosacea: report of the National Rosacea Society Expert
Acad Dermatol. 2002;46(4):584-587. Committee on the classification and staging of rosacea. J Am
8. Al Arfaj K, Al Zamil W. Spontaneous corneal perfora- Acad Dermatol. 2004;50(6):907-912.
tion in ocular rosacea. Middle East Afr J Ophthalmol. 27. Brozek JL, Akl EA, Alonso-Coello P, etal. Grading quality of
2010;17(2):186. evidence and strength of recommendations in clinical practice
444 Journal of Cutaneous Medicine and Surgery 20(5)

guidelines. Part 1 of 3. An overview of the GRADE approach microsecond 1064-nm neodymium:yttrium-aluminum-garnet

and grading quality of evidence about interventions. Allergy. laser for treatment of diffuse facial erythema: a double-blind ran-
2009;64(5):669-677. domized controlled trial. J Am Acad Dermatol. 2013;69(3):438-
28. van Zuuren EJ, Fedorowicz Z, Carter B, van der Linden MM, 443.
Charland L. Interventions for rosacea. Cochrane Database Syst 45. Nymann P, Hedelund L, Haedersdal M. Long-pulsed dye laser
Rev. 2015;4:CD003262. vs. intense pulsed light for the treatment of facial telangiec-
29. Andrews J, Guyatt G, Oxman AD, etal. GRADE guidelines: tasias: a randomized controlled trial. J Eur Acad Dermatol
14. Going from evidence to recommendations: the signifi- Venereol. 2010;24(2):143-146.
cance and presentation of recommendations. J Clin Epidemiol. 46. Del Rosso JQ, Schlessinger J, Werschler P. Comparison of anti-
2013;66(7):719-725. inflammatory dose doxycycline versus doxycycline 100 mg in
30. Andrews JC, Schnemann HJ, Oxman AD, etal. GRADE the treatment of rosacea. J Drugs Dermatol. 2008;7(6):573-576.
guidelines: 15. Going from evidence to recommendation 47. Smith K, Leyden JJ. Safety of doxycycline and minocycline: a
determinants of a recommendations direction and strength. systematic review. Clin Ther. 2005;27(9):1329-1342.
J Clin Epidemiol. 2013;66(7):726-735. 48. Stein L, Kircik L, Fowler J, etal. Efficacy and safety of iver-
31. Fowler J Jr, Jackson M, Moore A, etal. Efficacy and safety of mectin 1% cream in treatment of papulopustular rosacea:
once-daily topical brimonidine tartrate gel 0.5% for the treat- results of two randomized, double-blind, vehicle-controlled
ment of moderate to severe facial erythema of rosacea: results pivotal studies. J Drugs Dermatol. 2014;13(3):316-323.
of two randomized, double-blind, and vehicle-controlled piv- 49. Dahl MV, Jarratt M, Kaplan D, Tuley MR, Baker MD. Once-
otal studies. J Drugs Dermatol. 2013;12(6):650-656. daily topical metronidazole cream formulations in the treatment
32. Gillihan R, Nguyen T, Fischer R, Rajpara A, Aires D. Erythema of the papules and pustules of rosacea. J Am Acad Dermatol.
in skin adjacent to area of long-term brimonidine treat- 2001;45(5):723-730.
ment for rosacea: a novel adverse reaction. JAMA Dermatol. 50. Marks R, Ellis J. Comparative effectiveness of tetracy-

2015;151(10):1136-1137. cline and ampicillin in rosacea: a controlled trial. Lancet.
33. Werner K, Kobayashi TT. Dermatitis medicamentosa: severe 1971;298(7733):1049-1052.
rebound erythema secondary to topical brimonidine in rosacea. 51. Dreno B, Bettoli V, Ochsendorf F, Layton A, Mobacken H,
Dermatol Online J. 2015;21(3). Degreef H. European recommendations on the use of oral anti-
34. Bitar A, Bourgouin J, Dor N, etal. A double-blind randomised biotics for acne. Eur J Dermatol. 2004;14(6):391-399.
study of metronidazole (Flagyl) 1% cream in the treatment of 52. Koda-Kimble MA, Alldredge BK, Corelli RL, Ernst ME. Koda-
acne rosacea. Drug Invest. 1990;2(4):242-248. Kimble and Youngs Applied Therapeutics: The Clinical Use of
35. Bleicher PA, Charles JH, Sober AJ. Topical metronidazole Drugs. Philadelphia, PA: Lippincott Williams & Wilkins; 2012.
therapy for rosacea. Arch Dermatol. 1987;123(5):609-614. 53. Gollnick H, Blume-Peytavi U, Szab EL, etal. Systemic

36. Breneman D, Stewart D, Hevia O, Hino P, Drake L. A double- isotretinoin in the treatment of rosaceadoxycycline-and pla-
blind, multicenter clinical trial comparing efficacy of once- cebo-controlled, randomized clinical study. J Dtsch Dermatol
daily metronidazole 1 percent cream to vehicle in patients with Ges. 2010;8(7):505-514.
rosacea. Cutis. 1998;61(1):44-47. 54. Cunliffe WJ, van de Kerkhof PC, Caputo R, etal. Roaccutane
37. Dahl MV, Katz HI, Krueger GG, etal. Topical metroni-
treatment guidelines: results of an international survey.
dazole maintains remissions of rosacea. Arch Dermatol. Dermatology. 1997;194(4):351-357.
1998;134(6):679-683. 55. Caffery B, Josephson J. Ocular side effects of isotretinoin ther-
38. Koak M, Yagli S, Vahapolu G, Ekiolu M. Permethrin apy. J Am Optom Assoc. 1988;59(3):221-224.
5% cream versus metronidazole 0.75% gel for the treatment 56. Agarwal US, Besarwal RK, Bhola K. Oral isotretinoin

of papulopustular rosacea. Dermatology. 2002;205(3):265- in different dose regimens for acne vulgaris: a random-
270. ized comparative trial. Indian J Dermatol Venereol Leprol.
39. Nielsen PG. Treatment of rosacea with 1% metronidazole cream: 2011;77(6):688-694.
a double-blind study. Br J Dermatol. 1983;108(3):327-332. 57. Lee JW, Yoo KH, Park KY, etal. Effectiveness of conven-
40. Bjerke R, Fyrand O, Graupe K. Double-blind compari-
tional, low-dose and intermittent oral isotretinoin in the treat-
son of azelaic acid 20% cream and its vehicle in treat- ment of acne: a randomized, controlled comparative study. Br
ment of papulo-pustular rosacea. Acta Derm Venereol. J Dermatol. 2011;164(6):1369-1375.
1999;79(6):456-459. 58. Sobolewska B, Doycheva D, Deuter C, Pfeffer I, Schaller M,
41. Carmichael A, Marks R, Graupe K, Zaumseil R. Topical
Zierhut M. Treatment of ocular rosacea with once-daily low-
azelaic acid in the treatment of rosacea. J Dermatol Treat. dose doxycycline. Cornea. 2014;33(3):257-260.
1993;4(suppl 1):S19-S22. 59. Frucht-Pery J, Sagi E, Hemo I, Ever-Hadani P. Efficacy of dox-
42. Draelos ZD, Elewski B, Staedtler G, Havlickova B. Azelaic ycycline and tetracycline in ocular rosacea. Am J Ophthalmol.
acid foam 15% in the treatment of papulopustular rosacea: a 1993;116(1):88-92.
randomized, double-blind, vehicle-controlled study. Cutis. 60. Allergan. Restasis [package insert]. Irvine, CA: Allergan;

2013;92(6):306-317. 2004.
43. Thiboutot D, Thieroff-Ekerdt R, Graupe K. Efficacy and safety 61. Schechter BA, Katz RS, Friedman LS. Efficacy of topical
of azelaic acid (15%) gel as a new treatment for papulopustular cyclosporine for the treatment of ocular rosacea. Adv Ther.
rosacea: results from two vehicle-controlled, randomized phase 2009;26(6):651-659.
III studies. J Am Acad Dermatol. 2003;48(6):836-845. 62. Park JM, Mun JH, Song M, etal. Propranolol, doxycycline and

44. Alam M, Voravutinon N, Warycha M, etal. Comparative combination therapy for the treatment of rosacea. J Dermatol.
effectiveness of nonpurpuragenic 595-nm pulsed dye laser and 2015;42(1):64-69.
Asai et al 445

63. Craige H, Cohen JB. Symptomatic treatment of idiopathic and disorders: focus on antibiotic resistance. Cutis. 2007;79(6)
rosacea-associated cutaneous flushing with propranolol. J Am (suppl):9-25.
Acad Dermatol. 2005;53(5):881-884. 74. Oprica C, Nord CE; ESCMIDStudy Group onAntimicrobial
64. Cunliffe W, Dodman B, Binner JG. Clonidine and facial flush- Resistancein Anaerobic Bacteria. European surveillance study
ing in rosacea. BMJ. 1977;1(6053):105. on the antibiotic susceptibility of Propionibacterium acnes.
65. Guarrera M, Parodi A, Cipriani C, Divano C, Rebora A.
Clin Microbiol Infect. 2005;11(3):204-213.
Flushing in rosacea: a possible mechanism. Arch Dermatol 75. Leyden JJ, McGinley KJ, Cavalieri S, Webster GF, Mills
Res. 1982;272(3-4):311-316. OH, Kligman AM. Propionibacterium acnes resistance
66. Wilkin JK. Effect of subdepressor clonidine on flushing reactions to antibiotics in acne patients. J Am Acad Dermatol.
in rosacea: change in malar thermal circulation index during pro- 1983;8(1):41-45.
voked flushing reactions. Arch Dermatol. 1983;119(3):211-214. 76. Giannopoulos L, Papaparaskevas J, Refene E, Daikos G,

67. Wilkin JK. Effect of nadolol on flushing reactions in rosacea. Stavrianeas N, Tsakris A. MLST typing of antimicrobial-resis-
J Am Acad Dermatol. 1989;20(2):202-205. tant Propionibacterium acnes isolates from patients with mod-
68. Hsu C-C, Lee JY-Y. Carvedilol for the treatment of refrac- erate to severe acne vulgaris. Anaerobe. 2015;31:50-54.
tory facial flushing and persistent erythema of rosacea. Arch 77. Schafer F, Fich F, Lam M, Grate C, Wozniak A, Garcia P.
Dermatol. 2011;147(11):1258-1260. Antimicrobial susceptibility and genetic characteristics of
69. Moore A, Kempers S, Murakawa G, etal. Long-term safety and Propionibacterium acnes isolated from patients with acne. Int
efficacy of once-daily topical brimonidine tartrate gel 0.5% for the J Dermatol. 2013;52(4):418-425.
treatment of moderate to severe facial erythema of rosacea: results 78. Skidmore R, Kovach R, Walker C, etal. Effects of subanti-
of a 1-year open-label study. J Drugs Dermatol. 2014;13(1):56-61. microbial-dose doxycycline in the treatment of moderate acne.
70. Stein GL, Kircik L, Fowler J, etal. Long-term safety of ivermec- Arch Dermatol. 2003;139(4):459-464.
tin 1% cream vs azelaic acid 15% gel in treating inflammatory 79. Goldgar C, Keahey DJ, Houchins J. Treatment options for acne
lesions of rosacea: results of two 40-week controlled, investiga- rosacea. Am Fam Physician. 2009;80(5):461-468.
tor-blinded trials. J Drugs Dermatol. 2014;13(11):1380-1386. 80. Del Rosso JQ. Advances in understanding and managing

71. Taieb A, Khemis A, Ruzicka T, etal. Maintenance of remis- rosacea: part 2: the central role, evaluation, and medical man-
sion following successful treatment of papulopustular rosa- agement of diffuse and persistent facial erythema of rosacea.
cea with ivermectin 1% cream vs. metronidazole 0.75% J Clin Aesthetic Dermatol. 2012;5(3):26.
cream: 36-week extension of the ATTRACT randomized 81. Higgins JPT, Green S (editors). Cochrane Handbook for
study. J Eur Acad Dermatol Venereol. 2016;30(5):829-836. Systematic Reviews of Interventions Version 5.1.0 [updated
72. Levy RM, Huang EY, Roling D, Leyden JJ, Margolis DJ. March 2011]. The Cochrane Collaboration; 2011. Available
Effect of antibiotics on the oropharyngeal flora in patients with from www.cochrane-handbook.org.
acne. Arch Dermatol. 2003;139(4):467-471. 82. Balshem H, Helfand M, Schunemann HJ, etal. GRADE guide-
73. Leyden JJ, Del Rosso JQ, Webster GF. Clinical considerations lines: 3. Rating the quality of evidence. J Clin Epidemiol.
in the treatment of acne vulgaris and other inflammatory skin 2011;64(4):401-406.

Scarpetta. Sam Hanna, MD