. ADVANCES IN
`Dern1atology®
`
`VOLUME 17
`
`Editor-in-Chief
`William D. James, MD
`Albert M. Kligman Professor of Dermatology, Direc tor of Resident
`Program and Faculty Clinic, University of Pennsylvania Schoo l of
`Medicine, Philadelphia
`Associate Editors
`Clay J. Cockerell, MD
`Director, Division of Dermatopathology, Clinical Professor of Pa th ology
`and Dermatology, University of Texas Southwes tern Medica l Cent er,
`Dallas
`Mary E. Maloney, MD
`Professor of Medicine (Dermatology), Cli ni cal Division Chi d and
`Director of Dermatological Surgery, University of Massachuse tt s Medica l
`School, Worcester
`Amy S. Paller, MD
`Professor of Pediatrics and Dermatology, Northwes tern University
`Medical School; Head, Division of Dermatology, Children's Memori al
`Hospital, Chicago
`Kim B. Yancey, MD
`Professor and Chair, Department of Dermatology, Medi cal College of
`Wisconsin, Milwaukee
`
`~T~ Mosby
`
`1
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`~T~ Mosby
`
`Publisher: Cynthia Baudendistel
`Developmental Editor: Victoria Cernich
`Manage1; Continuity Production: !delle L. Winer
`Senior ProdLiction Editor: Pat Costigan
`Project Supervism; Production: Joy Moore
`Composition Specialist, Production: Betty Dockins
`
`Copyright © 2001 by Mosby, Inc
`
`All rights reserved. No part of this publication may be reproduced,
`stored in a retrieval system, or transmitted, in any form or by any
`means, electronic, mechanical, photocopying, recording, or otherwise,
`without prior written permission from the publisher.
`
`Permission to photocopy or reproduce solely for internal or personal use
`is permitted for libraries or other users registered with the Copyright
`Clearance Center, provided that the base fee of $35.00 per chapter is
`paid directly to the Copyright Clearance Center, 21 Congress Street,
`Salem, MA 01970. This consent does not extend to other kinds of
`copying, such as copying for general distribution, for advertising or
`promotional purposes, for creating new collected works, or for resale.
`
`Printed in the United States of America
`Composition by Thomas Technology Solutions, Inc
`Printing/binding by the Maple-Vail Book Manufacturing Group
`
`Editorial Office:
`Mosby, Inc
`11830 Westline Industrial Drive
`StLouis, MO 63146
`Customer Service: hhspcs@harcourt.com
`
`International Standard Serial Number: 0882-0880
`International Standard Book Number: 0-323-01535-2
`
`2
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`CHAPTER2
`
`Pathogenesis of Rosacea
`
`Mark V. Dahl, MD
`Professor and Chairman, Department of Dermatology, Mayo Clini c
`Scottsdale, Scottsdale, Ariz
`
`EDITOR'S COMMENT
`Rosacea is a source of constant embarrassment and worry for our affected
`patients. Bright red cheeks, enlarged noses, and erythematous papules and
`pustules interfere with social and occupational interactions. Depression,
`anxiety, and inferences about alcoholism haunt sufferers of this fi ery di sea se.
`To compound their misery, very few hard data exist regarding the
`etiopathogenesis of rosacea . Dr Mark Dahl, who has a long interest in
`helping alleviate the effects of this illness, shares hi s insights in thi s superbly
`written article. Mark separates rosacea into several subsets of specifi c
`findings . Explanations behind the appearance of each of them are reviewed.
`In the concluding section, a unifying concept is proposed to bring all of
`these contributing factors into focus and explain the totality of the signs a nd
`symptoms of this chronic and very public disease. Expanding thi s knowledge
`base and using our discoveries to improve therapy are goals we look
`forward to being realized.
`
`William D. James, MD
`
`R osacea is a disease composed of many elements or subsets.
`
`Rosacea is more like a syndrome than a single disease. Al(cid:173)
`though all patients with rosacea develop some erythema of their
`skin, not all patients develop all other stigmata. Some patients have
`centrofacial flushing and edema. Others are more troubled by
`papules and pustules. Still others are .more troubled by burning and
`stinging sensations, or by rhinophyma. Subsets of rosacea are listed
`in Table 1.
`The cause of the rosacea complex is unknown, and unifying
`hypotheses are speculative. Whatever the cause, facial skin is
`
`Advances in Derm atology®, vol 17
`© 2001, Mosby, Inc
`
`29
`
`3
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`30
`
`M. V. Dahl
`
`TABLE 1.
`Subsets of Rosacea
`
`Erythema
`Flushing
`Edema
`Papules and pustules
`Telangiectasias
`Stinging and burning sensations
`Rosacea dermatitis
`Rosacea cellulitis
`Fibrosis/rhinophyma
`
`predisposed. The disease affects only the facial skin and cornea.
`The tip of the nose is affected first, followed sequentially by the
`cheeks, forehead, and chin.
`
`ERYTHEMA
`Many people have ruddy complexions. Not all of these people have
`rosacea. The ruddy erythema of rosacea tends to persist (erythema
`congestivum) and usually waxes and wanes markedly in response
`to changes of body temperature, ingestion of certain foods, evoked
`emotions, or other factors. Sometimes the erythema involves the
`whole face, but more often it involves just the central face.
`Erythema is often striped by telangiectatic blood vessels.
`A red complexion characterizes people of European extraction,
`particularly Celts. In addition to this obvious genetic element,
`erythema is more pronounced in some families than in others. The
`amount of blood in the facial skin and blood vessels is high. Total
`movement of red blood cells in flush areas of rosacea skin is 3 to 5
`times higher than in controls. 1
`· Finally, the erythema of rosacea seems aggravated by chronic
`sun exposure and photodamage. People without rosacea can de(cid:173)
`velop poikiloderma of Civatte (erythromelanosis coli), and this
`same propensity toward erythema and telangiectasia seems to
`augment the baseline erythema among patients with rosacea and
`lightly pigmented skin. Solar damage is not a prerequisite for the
`development of rosacea.2
`3 Patients with rosacea are not more
`•
`prone to sunburn. 4
`The blushlike redness that develops on rosacea faces during
`and after exposure to sunlight is more likely caused by heat from
`infrared radiation than by photons from ultraviolet radiation. The
`radiant heat-derived redness develops quickly during (rather than
`
`4
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`Pathogenesis of Rosacea
`
`31
`
`after) exposure. Exposing facial skin to other sources of rad iant
`heat, such as heat from a fireplace or stove, reproduces it.
`Immunofluorescence staining of rosacea skin often finds immu(cid:173)
`noglobulins, especially IgG and IgM, at the derma l-epid ermal
`junction of sun-damaged skin and facial skin of pati ents with
`rosacea. This can be an artifact produced by preserving s pecimen s
`in Michel's transport media, which seems prone to increase non(cid:173)
`specific staining of dermis, dermal-epidermal junction, and blood
`vessels , especially in sun-damaged skin (unpublished observa(cid:173)
`tions). In other patients, the appearance of immunoglobulin depos(cid:173)
`its at the dermal-epidermal junction (especially in derma l pap ill ae)
`is an artifact allegedly produced by compression of autofluorescing
`elastic tissue against the dermal-epidermal junction by edema and
`dilated papillary dermal blood vessels (Burnham 's "fibrillar
`pseudoband"). 5
`Perhaps in others, antibodies bind to sun-damaged connecti ve
`tissue. 6
`9 Circulating IgG and IgM antibodies to collagen , an ti (cid:173)
`-
`nuclear antibodies, and anti-Demodex antibodies have been found
`in patients' sera. In addition, eluates from autologous circu lating
`lymphocytes in peripheral blood of patients with rosacea have
`reacted with nuclei of "dermal cells," endothelial cells, and eccrin e
`duct cells. 7
`
`FLUSHING
`Both flush and erythema occur when blood vessels dilate and blood
`flow increases in superficial vessels of the skin. Whereas th e
`erythema is chronic and persistent and varies subtl y, the flu shing
`can wax and wane quickly. Flushing usually involves the central
`face and is associated with a warming of surface skin.
`Patients with mastocytosis and carcinoid syndrome can de(cid:173)
`velop all the stigmata of rosacea, including ocular rosacea, faci al
`1 ~ Flus hing
`telangiectasia, and connective tissue hypertrophy.10
`-
`can be stimulated by various environmental trigger fa ctors as li sted
`in Table 2.
`In hyperthermic, but not normothermic healthy human beings,
`cool venous blood from the face and sca lp enters the dural veno us
`sinus via vascular pathways or emissary veins. The cooler blood
`intermingling with the brain and meningeal blood vessels keeps th e
`intracranial temperature cool. Brinnel et al' :I found that venous
`blood flow from the skin to the brain was suppressed in patients
`with rosacea, thus inhibiting selective brain cooling in hyperther(cid:173)
`mic conditions. Apparently, the facial skin acts li ke an autom obil e
`radiator to lose heat by radiation , convection , and co nducti on. fn
`
`5
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`w
`N
`
`~
`~
`
`~ :::,..
`......
`
`TABLE 2.
`Environmental Trigger Factors
`
`Beverages
`
`Alcohol
`(especially
`red wine)
`Hot drinks
`
`Temperature
`Related
`
`Saunas
`Simple over(cid:173)
`heating
`Excessively
`\Varm envi(cid:173)
`ronments
`Hot baths
`Radiant heat
`
`Weather Sun
`(radiant heat)
`
`Emotional
`Influences
`
`Drugs
`
`Physical
`Exertion
`
`Skin Care
`Products
`
`Menopause
`
`Rage
`Anger
`Stress
`Embarrassment
`
`Hot, humid
`temperatures
`Strong winds
`Cold
`Spring
`season
`
`Alcoholic or
`acetone sub-
`stances
`Any substance
`that causes
`redness or
`stinging
`Some cosmetics
`
`Vasodilators
`Nicotinic acid
`Cholinergic
`agents
`Calcium chan-
`nel blockers
`Topical steroids
`Cyclosporin
`Opiates
`Tamoxifen
`Rifampin
`
`Exercise
`Medical
`conditions
`Menopause
`Caffeine
`withdrawal
`Straining
`Frequent
`flushing
`Chronic
`cough
`Stress and
`strain jobs
`
`Foods
`
`Cheese (not cot(cid:173)
`tage cheese)
`Vanilla
`Thermally hot
`foods
`Chocolate
`Soy sauce
`Spicy foods
`Liver
`Yogurt
`Vinegar
`Eggplant
`Avocados
`Spinach
`Broad -leaf beans
`Citrus fruits
`
`(Adapted from Dahl MV: Optimal strategies for the treatment of rosacea. Skin & Aging March 1999, pp 44-50, and a survey of patients by the National Rosa(cid:173)
`cea Society.)
`
`6
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`Path ogenesis of Rosacea
`
`33
`
`patients with rosacea, the cooled blood does not flow well from the
`central face. The veins of the central face (venae supratrochl eari s,
`venae palpebrales superioris et inferioris, venae n asales externae,
`venae faciei profunda, venae labialis superior, venae labiali s infe(cid:173)
`riores, venae submentalis, etc) drain through the cavernous sinu s to
`co-mingle with arteries in the dura to cool them and to cool the
`brain. If this were impaired, flushing and venous congestion would
`occur. The brain would keep sending vessel dilatory signals to the
`blood vessels of the central face in hopes of getting more cooling
`action. Alternatively, the congestion could simply be passive,
`based on a relative block of efferent blood flow.
`Here's how this study by Brinnel et aP 3 was done. The
`tympanic, esophageal, forehead skin surface, and hand skin surface
`temperatures were recorded in 4 patients with rosacea and 2
`controls. Each took a 1-hour hot bath (38°C-39°C) . During the last
`half ofthe bath, the facial skin was "vigorously fanned. " Blood flow
`was measured by ultrasonic Doppler flowmeters in the sensory
`emissary veins of the cranium. During hyperthermia and fa ce
`fanning, all control subjects had their tympanic temperatures
`decrease below their esophageal temperatures , indicating that the
`facial vessels had cooled the eardrum (a measure of intracranial
`temperature) relative to the central (core, esophageal) temperature.
`In contrast, the tympanic temperature always remained higher than
`the esophageal temperatures in patients with rosacea. As expected,
`patients with rosacea had higher forehead temperature than did
`controls. Sweating rates on the face were equal among pati ents and
`controls. After hot baths, Doppler flowmeters showed a rapid
`inward blood flow from skin to brain in control subj ects, but there
`was no change in patients with rosacea. The authors concluded that
`venous blood tlow from the skin to the brain is suppressed in
`rosacea, thus inhibiting brain cooling in hyperthermic conditions.
`Flushing, at least from ingestion of alcoholic beverages, may be
`mediated by catecholamines derived from the gastric mucosa.
`Alcohol would liberate catecholamines, which would elevate lev(cid:173)
`els of bradykinin in blood, leading to t1ushing. 1 4 As might be
`predicted from such a hypothesis , clonidine (an a 2-agonist) could
`inhibit such flushing, and indeed it does. However, alcohol(cid:173)
`induced tlushing might also be medicated by opioids, since it also
`can be partly blocked by the opiate antagonist naloxone. 15
`Although caffeine and coffee are alleged to provoke Hushing,
`ingestion of neither caffeine nor coffee at zzoc leads to Hushing
`reactions. However, ingestion of both coffee and water at 60oC
`causes flushing reactions. Apparently the heat of the beverage, not
`
`7
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`34
`
`M. V. Dahl
`
`the caffeine or coffee in it, produces the flush in coffee drinkers. 16
`Migraine headaches occur more commonly among patients with
`rosacea, which is not surprising because both are associated with
`abnormal vascular reactivity, and both are aggravated by stressY
`Substance P may be involved in neural flush. 18
`Although a link between rosacea and gastrointestinal illness
`seems plausible, linking mechanisms, if any, are speculative.
`Clearly, administration of hot beverages, spicy foods, and other
`foods can evoke flushing and rosacea. Flushing in these cases might
`be caused by direct effects of secondary mediators on all blood
`vessels or by overzealous attempts to cool the body or brain. Both
`gastrin and vasoactive intestinal peptide have vasodilatory actions,
`so these agents could elicit flushes from gastrointestinal illnesses
`and stimulations. This is speculative. Jejunal biopsy specimens
`taken from 34 patients with rosacea and compared with those from
`34 patients with other skin diseases showed no consistent abnor(cid:173)
`mality. Careful history taking from patients showed no difference
`in the prevalence of gastrointestinal illness or symptoms in the 2
`groups. 19 Pancreatic exocrine function is normal, although the
`output of pancreatic lipases is lower among patients with rosacea. 20
`In a study comparing organisms isolated from feces, patients with
`rosacea harbored a significantly higher frequency of Penicillium
`lanoso-coeruleum and Aspergillus niger. Patients with rosacea also
`harbored a larger amount of antibodies to these organisms in their
`blood. 21 The significance of these findings is unknown.
`Helicobacter pylori is a bacterium that frequently colonizes the
`stomach and is associated with gastric ulcer disease. A study of 60
`rosacea patients with papules, pustules, erythema, and flushing
`found a prevalence of H pylori of 88% compared with 65% in
`control patients. Patients with rosacea were more likely to have
`gastritis in their stomach's antrum. After treatment for H pylori
`with omeprazole, clarithromycin, and metronidazole, symptoms of
`rosacea disappeared in 51 patients, markedly decreased in 1
`patient, and remained unchanged in 1 other patient. Vasogastrin
`levels decreased from 48 pmol/L before treatment to 17 pmol/1
`after treatment. Levels of pentagastrin, plasma tumor necrosis
`factor-a, and interleukin-8 were also reduced slightly in sera after
`treatment. The authors concluded that rosacea was a disorder
`closely linked to gastritis, especially in the antrum mucosa. Both
`occurred frequently in patients colonized by H pylori expressing
`the cytotoxin-associated gene A (CagA). 22 Other researchers also
`have noted that rosacea improves when H pylori is eradicated by
`
`8
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`Pathogenesis of Rosaceo
`
`35
`
`treatment, but the improvement may be more related to the well(cid:173)
`known (and perhaps independent) response of rosacea to many of
`the antibiotics that are used to kill the associated H pylori.·D-z:;
`
`EDEMA
`After blood vessels dilate, plasma can exude from the d i Iated
`vessels. If the outflow is high enough, the outflow surpasses the
`ability of lymphatics to carry the excess fluid away, resulting in
`clinical swelling. Facial swelling is observed from time to time
`among patients with rosacea, although usually not in a physician 's
`office. Edema clearly can develop after vasodilation. The role oF
`edema in producing other sequelae of rosacea (such as inflamma(cid:173)
`tory papules) is more speculative. Wilkins 26 believes that intlam (cid:173)
`matory edema can appear in these circumstances, especially when
`facial cutaneous lymphatic vessels are damaged. Exposure to
`ultraviolet radiation may damage lymphatic vessels and thereby
`slow clearance of tissue fluids. Edema can produce low-grade
`cellulitis. Rosacea occurs in patients with Morbihan disease
`(lymph vessel dysplasia), lending support to the notion that edema
`is the fuse that ignites inflammation.
`Stigmata of rosacea are typically found in areas of th e face
`overlying relatively inactive musculature. In these areas, edema
`caused by flushing might tend to persist, since edema fluid and
`venous blood in vessels in this area would not be subjected to
`pumping effects related to contraction of adjacent muscles.2 7 Inter(cid:173)
`relationships among edema, elastin degeneration, neutrophil pro(cid:173)
`teases, and cutaneous cellulitis have been reviewed. 26 Further(cid:173)
`more, transudation of factor XIII can cause connective ti ssue
`hypertrophy and fibroplasia that might lead to fibrosis and phy(cid:173)
`mas.28
`
`PAPULES AND PUSTULES
`Papules and pustules of rosacea so much resemble acne vul garis
`that the condition has been called acne rosacea. However, come(cid:173)
`dones are absent. Papules and pustules are limited to th e face
`(especially the flush and erythematous areas), and no bacterium has
`been consistently isolated from pustules. Oil glands secrete seburn
`at normal rates. 29 Follicle and surface lipid compositions are
`probably normal, 3 0 although epidermally derived surface choles(cid:173)
`terol levels may be increased. 3 1 The amounts of free fatty acids in
`surface lipids may decrease and amounts of free squalene may
`increase after treatment with tetracycline , although studies are
`
`9
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`36
`
`M. V. Dahl
`
`controversial. Jo,:n Serum levels of testosterone and estradiol are
`normal, as are cellular levels of estrogen and androgen. 32
`Biopsy specimens from patients with rosacea show an inflam(cid:173)
`matory cell infiltrate in the upper and mid-dermis consisting
`mostly of lymphocytes and histiocytes intermingled with polymor(cid:173)
`phonuclear leukocytes, plasma cells, and giant cells. Only about
`25% of biopsy specimens show areas of acute folliculitis. In the
`remainder, the inflammation is entirely independent of the pilose(cid:173)
`baceous unit. a3 Surprisingly, much of the inflammation resides far
`away from the pilosebaceous apparatus . In fact, Marks et aP 3 found
`acute folliculitis in only 25% of 24 serially sectioned skin biopsy
`specimens. In 51%, the infiltrate was partially distributed around
`the hair follicles, 78% showed some follicular plugging, and 19%
`showed recognizable Demodex. Granulomatous rosacea also shows
`no particular predilection to follicles. 34
`Despite these findings, I believe that the papules and pustules
`do have a bacterial origin. The prompt improvement and even
`clearing of papules and pustules that routinely follows the use of
`numerous oral and topical antibiotics of completely different
`chemical structures and microbial spectrums would seem to sup(cid:173)
`port this view. I believe that normal bacteria on the skin surface or
`in follicles behave differently in the milieu of elevated skin surface
`temperature. Indeed, staphylococci isolated from pustules secrete
`different products when cultured at different temperatures (Dahl
`MV, Shlievert P: submitted for publication). The role, if any, of
`staphylococcal exotoxins acting as superantigens to elicit cell(cid:173)
`mediated inflammation is unknown. Furthermore, the role, if any,
`of toxins or other products elaborated at various temperatures by
`normal anaerobic constituents of the follicles (such as Propioni(cid:173)
`bacterium acnes) is also unknown.
`Demodex mites are frequently found in facial follicles of
`3 5
`patients with rosacea. 3 3
`3 6 These organisms constitute part of the
`'
`'
`normal flora of the face, but the prevalence of Demodex mites in
`facial skin of patients with rosacea is greatly increased. 37 Although
`many follicles of patients with rosacea harbor numerous Demodex
`organisms, there is no difference in the number of mites in infested
`follicles of patients with rosacea compared with controls. 36
`The role of Demodex mites in producing follicular papules and
`pustules in patients with rosacea is speculative. Forton39 found a
`positive correlation between the presence of Demodex mites and
`the development of inflammation (P < .001). Elimination of Demo(cid:173)
`dex mites with treatment sometimes leads to improvement. Infre(cid:173)
`quently, Demodex parts have been seen in the inflammatory milieu
`
`10
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`Pathogenesis of Rosacea
`
`37
`
`FIGURE 1.
`Demodex folliculitis. This man developed these widespread facial red
`papules and crusted papules only 3 months previously. Biopsy specimens
`showed Demodex Jolliculorum and folliculm inflammation. The papul es
`responded promptly to treatment with a single overnight appli cation or
`lindane and systemic tetracycline.
`
`around a ruptured follicle of a patient with rosacea, suggesting that
`the inflammatory response resulted from the extruded organi sm.
`Whereas rosacea is usually successfully treated with tetracycline,
`mite counts do not decrease during treatment.:17 Perhaps Demodex
`folliculitis is a separate disease from rosacea that is frequently
`associated with rosacea because of the high prevalence of mit es on
`facial skin of patients with rosacea (Fig 1).
`
`lELANGIECTASIA
`Telangiectases are dilated blood vessels. They are frequently found
`in the milieu of erythema and Hush and are usually particularly
`prevalent in the area near the nasolabial fold. With long-standing
`erythema and flush, vessels become more apparent on the nose and
`
`11
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`38
`
`M. V Dahl
`
`cheeks. Endothelial walls are often thicker than normal, and a
`perivascular inflammatory cell infiltrate around telangiectatic
`blood vessels is not rare in rosacea compared with other telangi(cid:173)
`ectases in otherwise normal patients. 40
`4 1 A role for sun damage as
`'
`a cofactor seems likely because telangiectases are often seen in
`chronically sun-damaged skin. Furthermore, more elastotic change
`was found in patients with rosacea than in controls, despite the fact
`42
`that the controls reported more sun exposure. 41
`'
`
`STINGING AND BURNING SENSATIONS
`Patients with rosacea frequently complain that their skin is "sen(cid:173)
`sitive"; moisturizers, cosmetics, sunscreens, and other topically
`applied products cause unpleasant burning and stinging sensa(cid:173)
`tions. The skin feels tight or dry. The reason for this is speculative.
`Substance P is released from nerves by various stimuli, and
`substance P is found in abundance around papillary dermal blood
`vessels of patients with rosacea. 42 Even serum levels of substance P
`may be increased among patients with rosacea. 4 3 I speculate that
`acetylcholine released from keratinocytes after activation by soaps
`and detergent-like emulsifiers might also cause stinging and burn(cid:173)
`ing sensations. 44
`45 In these patients, the keratinocyte acts as the
`'
`end organ, secreting acetylcholine to activate cutaneous neurons.
`Barrier function may be compromised by rosacea dermatitis.
`The disruption or abnormality of the stratum corneum could allow
`increased penetration of sensory irritants. Alternatively, nerves of
`patients with rosacea may react abnormally to sensory irritants.
`When 5% lactic acid (or water) was applied to the cheeks of
`patients with rosacea, 24 of 32 patients with rosacea reported
`stinging as opposed to only 6 control subjects (P < .001). 46 All 7
`patients with erythematotelangiectasia reported stinging. This sug(cid:173)
`gests a relationship of blood vessels and pain (erythrodysesthe(cid:173)
`sia). 47
`Whatever the cause, patients often stop washing their faces
`because they believe soap not only causes burning sensations, but
`also because they believe soap-and-water washings cause dryness.
`Ayres and Ayres48 noted this habit decades ago. They believed
`failure to wash facial skin leads to higher densities of Demodex
`mites, which in turn leads to stinging, burning, and dryness, which
`in turn leads to still less frequent washing. Treatment with sulfur
`(which eliminates Demodex) seemed helpful. They advised these
`patients to resume face-washing.
`
`12
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`Pathogenesis of Rosacea
`
`39
`
`ROSACEA DERMATITIS
`Many patients with rosacea experience roughness and scaling (they
`call it "dryness"), especially on the cheeks. Clinicians frequentl y
`diagnose seborrheic dermatitis, but more typical sites for sebor(cid:173)
`rheic dermatitis such as the scalp and eyebrows are usuall y spared .
`Other clinicians believe the dermatitis results from irritation , and
`take the frequent association of stinging and burning sensation s
`as evidence. Pityriasis folliculorum consists of follicul ar plugs,
`and fine, rough, white scales occurring on flu shed faces . II may
`be a dermatitis caused by Demodex folliculorum. 48 Sometimes
`large numbers of Demodex organisms can even produce scaling
`patches.49
`The dermatitis in rosacea seems unique enough to me to be
`considered as an eczematous process in its own right. Roughness
`and sometimes mild erythema of the skin on the cheeks are prese nt ,
`especially near the nasolabial folds. The dermatiti s responds to
`topical steroids, but it also variably responds to topi cal and
`systemic antibiotics and to topical and systemic antifungal agents
`such as ketoconazole. Here again, higher densities of Demodex or
`staphylococcal organisms on the skin surface may play a role. The
`higher skin surface temperature of chronically eryth emato us skin
`may support higher densities of bacteria and enhanced toxin
`production.
`
`ROSACEA CELLULITIS
`The inflammatory response affecting patients with rosacea spreads
`well beyond facial follicles and may involve large areas. Rosacea
`cellulitis likely has causes similar to causes of papules and pus(cid:173)
`tules, as evidenced by the following: (1) the disorder spreads from
`typical papules and pustules; (2) the disorder is almost always
`associated with the presence of active inflammatory papules and
`pustules elsewhere on the skin; and (3) the disorder improves
`rapidly after administration of antibiotics that typically bel p th e
`papules and pustules of rosacea.
`
`PHYMAS
`In some but not all patients with rosacea, fibro sis develops in
`affected areas, especially in the noses of men. Rhinophyma is
`frequently associated with fibrosis elsewhere, such as on th e chin
`or above the nasolabial fold on the cheeks. Factor XIII in edema
`fluid has been blamed. Indeed, men wearing heavy glasses may
`develop rhinophyma distal to the nosepads as a consequence of
`
`13
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`40
`
`M. V. Dahl
`
`lymphatic obstruction ("pseudorhinophyma"; Fig 2). Microscopic
`examinations of rhinophyma usually show marked dermal thick(cid:173)
`ening, absence of pilosebaceous units, sclerotic collagen bundles,
`large amounts of mucin, and dilated blood vessels. 5° The inflam(cid:173)
`matory infiltrate are inconspicuous, but there are interstitial
`spindle cells reactive with factor XIIIa. The picture resembles the
`histology of chronic lymphedema.
`Alternatively, long-standing chronic inflammation of rosacea
`papules, pustules, dermatitis, or cellulitis may generate fibrosis.
`Indeed, patients with the common form of rhinophyma show a very
`different histologic picture from that of severe rhinophyma dis(cid:173)
`cussed above . Specifically, skin biopsy sections of common rhino(cid:173)
`phyma show findings of typical rosacea, except for prominent
`sebaceous hyperplasia. 5 0
`The reasons men seem so much more likely to develop phymas
`than women are unknown, but the sex difference suggests hor(cid:173)
`monal causes. Men may be less likely to seek treatment for rosacea,
`but it is premature to conclude that early treatment can prevent
`rhinophyma.
`
`FIGURE 2.
`Pseud orhinophyma. This man has a bulbous nose, presumed secondary to
`lymphatic obstru ction of heavy spectacle frames. Lymphedema can pro(cid:173)
`du ce fibro sis by actions of factor XIII.
`
`14
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`Pathogenesis of I1.osacen
`
`41
`
`OCULAR ROSACEA
`Most patients with rhinophyma also have ocular rosacea. Ocular
`manifestations include blepharitis, meibomitis, chalazia, styes, and
`hyperemic conjunctivae. 51 The cornea may be involved with ero(cid:173)
`sions or peripheral vascularization. Most patients compla in of
`pain, burning, or the sensation of foreign bodies in their eyes. Bo th
`eyes are usually affected. The decade of peak prevalence is 51lo (-JO
`years, but the disorder may affect teenagers.
`Bacteria have been incriminated, based on the frequent re(cid:173)
`sponse of ocular rosacea to administration of systemic antibiotics.
`However, cultures are often sterile. Staphylococci can often be
`isolated but may be a secondary invader.52 Demodex mites fre(cid:173)
`quently are found attached to epilated eyelashes, but barely with
`increased frequency among patients with rosacea compared with
`controls. 53
`Clinically, ophthalmologists frequently observe inspissation of
`secretions of the meibomian gland orifices. 54
`·"" Meibomian lipid
`composition is similar to that of controls. 5
`G
`Hoang-Xuan et al57 suggested that conjunctival inflammation
`results from a type IV (cell mediated) immune reaction. They found
`a 3.5-fold increase in the CD4/CD8 ratio of lymphocytes in rosacea
`conjunctivae compared with those of normal sub jects. Most of Lhe
`cells are in the substantia propria, and sometimes they team up
`with CD14 antigen-presenting cells and phagocy tes to form granu(cid:173)
`lomas there.
`More than half of patients with ocular rosacea had dry eyes
`with less than 8 mm of strip wetting seen on Schirmer Las ting,
`compared with 25% of controls (P < .02)." 8 Dry eyes may result
`from ocular rosacea and an inability to maintain a normal prcocular
`tear film. Because so many patients with ocular rosacea have
`normal tests, poor tear films are unlikely the cause of rosacea and
`more likely an effect of it.
`
`UNIFYING HYPOTHESIS
`If rosacea is indeed one disease or syndrome, the redness of t he l'acc
`must somehow be related to th e papules, pustules, sensations,
`skin-surface changes, and fibrosis (Fig 3). Erythema may res ult from
`odd brain-cooling mechanisms, sun damage, or genetic !'actors .
`Rosacea skin is warmer and blood vessels are dilaled. Ed ema and
`telangiectases may result. Additional redness and edema are el ic(cid:173)
`ited by flushing. Edema may set off infl ammation and Fibrosis.
`Alternatively, the warm facial skin may change mi crob ial (bac t l~ -
`
`15
`
`Galderma Laboratories, Inc. Ex 2009
`Dr. Reddy's Labs v. Galderma Labs., Inc.
`IPR2015-01777
`
`

`
`42
`
`M. V. Dahl
`
`FIGURE 3.
`Unifying hypothesis for cause. Erythema and flush cause