Nội dung

51 5 ROSACEA – EPIDEMIOLOGY AND PATHOPHYSIOLOGY INTRODUCTION E P I D E M I O LO GY R Although diagnosed in patients of most ethnicities and races (51, 52), rosacea is most prevalent in fair-skinned individuals, especially of Northern and Eastern European descent, and is estimated to occur in 2.1–10% in this population (Bamford et al. 2006; Berg & Liden 1989). Unfortunately, large epidemiological studies have been hampered by the above-mentioned lack of precise and uniform clinical criteria that define this disease. Only a handful of studies have carefully examined the prevalence of rosacea by gender and age. In a frequentlycited study of Swedish office employees, rosacea was found to be nearly three times more common in women than in men (Berg & Liden 1989). However, because of the selected study population, elderly patients were OSACEA is a common cutaneous disorder that may present with a variety of clinical manifestations, including ocular involvement. It is, however, precisely because of such variability in presentation that a set of specific diagnostic criteria has long been elusive. Such pervasive confusion complicates not only clinical diagnosis and eventual choice of treatment modalities, but also research studies and investigations into the pathophysiology of this disease. A relatively recent consensus by a panel of experts established a new classification system based on relatively specific clinical features (Wilkin et al. 2002). Though not without its shortcomings, such a system represents an extremely important advance in rosacea. 51 51 Rosacea in an Asian patient. 52 52 Rosacea in an Hispanic patient. 52 under-represented. Other studies have noted an overall equal prevalence in both genders, with a tendency toward earlier presentation in females compared to males (Kyriakis et al. 2005). Gender predisposition also depends on the individual rosacea subtype, with rhynophyma occurring predominantly in male patients (Kyriakis et al. 2005). Overall, rosacea is most frequently diagnosed in patients between the ages of 30 and 50 years; however, presentation in the seventh, eighth, and even in the ninth decade in not unusual (Kyriakis et al. 2005). Childhood rosacea cases, though rare, have been documented in the literature (Chamaillard et al. 2008; Drolet & Paller 1992; Erzurum et al. 1993). DEFINITION OF ROSACEA No specific laboratory tests are available for rosacea; thus, a system of signs and symptoms must be utilized to define this disease. As per the expert committee consensus, rosacea may be diagnosed when one or more of the primary features are present, most commonly on the convex surfaces of the central face. The primary features include flushing (or transient erythema), persistent erythema, papules and pustules, and telangiectasias (Wilkin et al. 2002). Additional secondary features may include burning or stinging, rough and scaly appearance likely as a result of local irritation, edema, elevated red plaques, peripheral localization, ocular manifestations, and phymatous changes. Other authors have, however, suggested that these criteria may not be specific enough. They have thus proposed that persistent centrofacial erythema lasting at least 3 months with a tendency toward periocular sparing is most characteristic of rosacea (Crawford et al. 2004). Awareness of the potential rosacea mimickers is important. These include erythema and telangiectasias frequently noted in lupus erythematosus, dermatomyositis, and other connective tissue diseases, flushing associated with the carcinoid syndrome and mastocytosis, and plethora seen in polycythemia vera. Finally, if suspected, allergic contact dermatitis and photosensitivity can be excluded with the help of patch testing and phototesting, respectively. ROSACEA SUBTYPES Once diagnosed, each case of rosacea should be further classified as one of four recognized subtypes (Table 6). Erythematotelangiectatic subtype Papulopustular subtype Phymatous subtype Ocular subtype Granulomatous variant* *currently not recognized as a separate subtype Table 6 Rosacea classification This is an essential part of the diagnosis, as it has a direct impact on the choice of treatment modalities and the prognosis. The subtype is determined based on the predominant features present in a given patient. According to the expert committee, rosacea may be subdivided into erythematotelangiectatic (ET), papulopustular (PP), phymatous, and ocular subtypes, with granulomatous rosacea considered a special variant of the disease (Wilkin et al. 2002). On the other hand, several conditions previously considered variants of rosacea have now been reclassified as separate diagnosticentities. These include rosacea fulminans, also known as pyoderma faciale, steroid-induced acneiform eruption, and perioral dermatitis. It should, however, be noted that some authors consider rosacea to be a much more polymorphic disease with many more subtypes than those recognized by the expert panel (Kligman 2006). Still, the following discussion will focus on the latter, more widely-accepted classification system. Erythematotelangiectatic subtype Patients who belong to this subtype typically present with persistent centrofacial erythema and an extensive history of prolonged flushing in response to various stimuli (53, 54). Although not required for the diagnosis of this subtype, facial telangiectasias may also be present in the affected areas (55). Flushing may affect not only the central portions of the face, but also the ears, neck, and chest (Marks & Jones 1969). Unlike physiologic flushing, or blushing, prolonged facial vasodilation (lasting 10 minutes or longer and often accompanied by burning or stinging) is typically observed in such patients. It is important to note, R O S A C E A – E P I D E M I O LO G Y A N D PAT H O P H Y S I O LO G Y 53 55 54 53 Erythematotelangiectatic subtype of rosacea. 54 Erythematotelangiectatic subtype of rosacea resembling the stigmata of alcoholism. 55 Extensive telangiectasias in erythematotelangiectatic rosacea. however, that flushing associated with rosacea is never accompanied by sweating or light-headedness; in such cases, systemic causes of flushing should be sought. As well, perimenopausal flushing should not automatically evoke the diagnosis of rosacea, unless other symptoms and signs are present in a given patient. The stimuli of flushing, also known as triggers, may vary among patients and most commonly include hot showers, the extremes of ambient temperatures, hot liquids, spicy foods, alcohol, exercise, and emotional stress (Greaves & Burova 1997; Higgins & du Vivier 1999; Wilkin 1981). In addition, various foods, such as citrus fruits and tomatoes, have been described as occasional triggers, and detailed food diaries may be helpful in some patients. Patients with ET rosacea tend to exhibit poor tolerability of topically-applied products, often including those meant to ameliorate the condition. Itching, burning, and stinging following topical application are common complaints; over time, roughness and scaling may develop, likely as a consequence of low-grade irritation (Dahl 2001; LonneRahm et al. 1999). Although patch testing may at times be useful in these patients, most cases of contact dermatitis associated with ET rosacea appear to be irritant, rather than allergic, in nature (Jappe et al. 2005). 53 54 Papulopustular subtype This subtype of rosacea most resembles acne vulgaris, but lacks comedones. Patients present with persistent central facial erythema and transient papules and pustules, typically sparing the periocular regions (56). Edema may at times be present, but solid facial edema is rare (Harvey et al. 1998; Scerri & Saihan 1995). Flushing may occur, but is usually less common and less pronounced than that seen in patients with ET rosacea. Burning and stinging, as well as sensitivity to topical products, may be reported, but are also less frequent in PP rosacea as compared to the ET subtype (LonneRahm et al. 1999). Additionally, telangiectasias may be difficult to discern, as they are often obscured by the background of erythema. Progression to the phymatous subtype may occur in severe cases, but is most often limited to the male patients. The reasons for such a gender difference, however, are not fully understood. Phymatous subtype Phymatous rosacea is defined by thickened skin and irregular surface nodularities (Wilkin et al. 2002). Patulous follicles, as well as persistent erythema, papules and pustules, and telangiectasias, are also frequently seen in the areas of involvement. Although 56 56 Papulopustular subtype of rosacea. most common on the nose, where it is known as rhinophyma (57), this type of rosacea may also occur on the chin, forehead, ears, and eyelids. Despite a common misperception, most cases of rhinophyma are not associated with alcohol consumption (Curnier & Choudhary 2004). Four variants of rhinophyma, glandular, fibrous, fibroangiomatous, and actinic, have been recognized based on clinical and histological differences and a variety of grading scales have been devised (Aloi et al. 2000; Freeman 1970; Jansen & Plewig 1998). In severe cases, secondary nasal airway obstruction may occur; however, bony and cartilaginous structures are typically not affected (Rohrich et al. 2002). Ocular subtype Ocular rosacea should be considered in patients with such symptoms as burning, stinging, and itching of the eyes, foreign body sensation, light sensitivity, and blurred vision. Clinically, blepharitis and conjunctivitis are the most common presentations of ocular rosacea. Additional findings may include watery or dry eyes, interpalpebral conjunctival hyperemia, conjunctival telangiectasias, irregularity of the lid margin, eyelid and periocular erythema and edema, meibomian gland 57 57 Rhinophyma in an African-American patient. R O S A C E A – E P I D E M I O LO G Y A N D PAT H O P H Y S I O LO G Y dysfunction, and recurrent chalazia (Akpek et al. 1997; Chen & Crosby 1997; Lemp et al. 1984) (58). Although infrequent, keratitis, episcleritis, corneal perforations, and iritis may also occur and are potentially serious complications that may lead to blindness or require enucleation (Akpek et al. 1997; Browning & Proia 1986). The true incidence of ocular rosacea is difficult to ascertain secondary to conflicting reports in ophthalmologic and dermatologic literature, with estimates ranging from less than 5% to as high as 58% of all rosacea patients (Kligman 2006; Starr & Macdonald 1969). Ocular signs and symptoms may precede skin involvement in up to 20% of patients; however, the diagnosis of ocular rosacea without cutaneous findings is difficult, as most manifestations are nonspecific (Browning & Proia 1986). Granulomatous variant Classified by the expert panel as a special variant of rosacea, granulomatous rosacea often lacks many of the characteristic findings of the classic disease, including persistent erythema, flushing, and telangiectasias. It is also likely that lupus miliaris disseminatus faciei and granulomatous rosacea represent the same disorder, although this view is controversial (van de Scheur et al. 2003). Clinically, individual firm 1–5 mm brown-red to yellow papules and nodules appear on relatively normal, noninflamed skin. Involvement is not limited to the convexities of the face, with the eyelids, cheeks, and the upper lip being the most commonly-affected locations. Without treatment, lesions eventually resolve with scarring. Histologically, epithelioid granulomas with or without caseation necrosis have been observed; however, there is no relationship to Mycobacterium tuberculosis infection (Helm et al. 1991). Some authors believe that because of the significant clinical and histological differences from the other subtypes of rosacea, the granulomatous variant may, in fact, represent a distinct diagnostic entity (Crawford et al. 2004). PAT H O P H Y S I O LO GY O F R O S A C E A The study into the pathophysiology of rosacea has long been hampered by the lack of specific diagnostic criteria. In addition, many studies fail to specify the breakdown of the various subtypes, which may 58 58 Ocular rosacea. potentially have varied pathogenic mechanisms. Nonetheless, several fundamental findings have recently been made, and our understanding of the pathophysiological factors underlying the development of rosacea will likely improve significantly in the near future. Numerous mechanisms have been proposed over the years, including vascular abnormalities, inflammation and dermal matrix degradation, climactic exposures, pilosebaceous unit abnormalities, and various microbial organisms, and will now be examined at length. Vascular abnormalities Since flushing is often exaggerated in rosacea patients, inherent vascular abnormalities have been proposed as a causative factor in the pathogenesis of this disorder (Wilkin 1994). In a small study, a normal physiological response to hyperthermia of shunting blood away from facial circulation in order to increase blood flow to the brain was absent in rosacea patients (Brinnel et al. 1989). Rosacea patients have also been shown to flush more easily in response to various thermal stimuli. In the case of oral exposure to heat, such as that seen with ingestion of hot liquids, a countercurrent heat exchange between the internal jugular vein and the common carotid artery may be produced, thus triggering an anterior hypothalamic thermoregulatory reflex, resulting in cutaneous vasodilation (Wilkin 1981). 55 56 Why is flushing localized to the face? Both vasodilation in general and flushing in particular are controlled by neural stimuli and humoral factors. In fact, it has been shown that the proportional vasodilatory response to both neurally- and humorallymediated triggers is the same in cutaneous vasculature of the face and of the forearm (Wilkin 1988). However, the baseline cutaneous blood flow has been shown to be higher and the blood vessels larger, more numerous, and closer to the surface on the face as compared to other parts of the body (Tur et al. 1983; Wilkin 1988). Of interest, since both the blood flow and pain perception are regulated by C nerve fibers, low heat pain threshold has been found in the affected areas in patients with PP rosacea (GuzmanSanchez et al. 2007). More recently, the role of angiogenesis and vascular factors has been investigated. An increased expression of vascular endothelial growth factor (VEGF) and vascular endothelial marker CD31 has been demonstrated in the affected skin of rosacea patients (Gomaa et al. 2007). VEGF plays a dual role by inducing angiogenesis and by increasing vascular permeability with subsequent leakage of various proinflammatory factors, which may further contribute to the pathogenesis of the disease. In addition, tetracycline and similar agents work, at least in part, by inhibiting angiogenesis, further suggesting the role of neovascularization in rosacea (Dan et al. 2008; Fife et al. 2000; Gilbertson-Beadling et al. 1995). Of note, a high expression of D2-40, a marker of lymphatic vessels, in the affected skin has been demonstrated in both early and long-standing disorder, suggesting lymphangiogenesis as an early pathological process in rosacea (Gomaa et al. 2007). Inflammation and dermal matrix degradation Abnormalities of dermal connective tissue as seen in rosacea patients may be caused by the preceding vascular derangements (Neumann & Frithz 1998). Thus, inherent or acquired vasculopathy and the increased expression of VEGF may lead to leaky blood vessels and dermal accumulation of cytokines and other inflammatory mediators with subsequent dermal matrix deterioration. On the other hand, some researchers suggest a primary role for inflammation and connective tissue damage in the pathogenesis of vascular changes associated with the disease (Bevins & Liu 2007; Millikan 2004; Yamasaki et al. 2007). This is supported, in part, by the finding that ectatic blood vessels in rosacea are still able to dilate and contract in response to vasoactive agents (Borrie 1955a, b). Instead, solar exposure, as will be discussed in the next section, may cause deterioration of collagen and elastic fibers, resulting in poor structural support for the cutaneous vasculature (Fisher et al. 1999). The weakened or leaky blood vessel walls may lead to the extravasation of proinflammatory mediators and neutrophil chemotaxis. Activated neutrophils release reactive oxygen species (ROS) and various matrix metalloproteinases (MMPs), which further contribute to dermal matrix degradation and perpetuate the inflammatory response (Akamatsu et al. 1990; Jones 2004). Moreover, a decrease in the capacity of the antioxidant defense system, including superoxide dismutase, has been demonstrated in severe rosacea (Oztas et al. 2003). In addition, a study by Yazici et al. (2006) showed a significant correlation between rosacea and specific genetic polymorphisms in the glutathione S-transferase genes, also responsible for cellular defense against ROS damage. The newest findings involving the action of cathelicidin in the pathophysiology of rosacea gives further credence to the primary role of the immune system in rosacea (Yamasaki et al. 2007). These important findings will be discussed in a later section. Climactic exposures The notion that climactic exposures, most notably solar radiation, may lead to the development of rosacea has been advocated by many investigators (Wilkin 1994). This is supported by the observation that convex, sun-exposed surfaces are typically involved, sparing the sun-protected periorbital and submental areas. Prolonged ultraviolet (UV) radiation leads to the degradation of the elastic fiber network and collagen fibers in the dermis, resulting in the accumulation of solar elastotic material. As previously discussed, this leads to a weakened support structure for cutaneous vasculature. In addition, an upregulation of VEGF and subsequent angiogenesis has been demonstrated following irradiation of skin with UV-B light (Yano et al. 2005). R O S A C E A – E P I D E M I O LO G Y A N D PAT H O P H Y S I O LO G Y On the other hand, if excessive sun exposure were the primary etiological factor for rosacea, significant actinic damage prior to the development of the disease, as evidenced by a high incidence of actinic keratoses, would be expected. However, a very large study documented an increase in actinic keratoses only in female rosacea patients, but not in male patients (Engel et al. 1988). Additionally, despite a common misperception, rosacea patients do not show increased photosensitivity compared to the normal population. In fact, minimum erythema dose of either UV-A or UV-B radiation in rosacea patients is not decreased (Lee & Koo 2005). Thus, flares in response to sun exposure may actually be a reaction to heat rather than the light itself (Kligman 2006). Pilosebaceous unit abnormalities Despite certain similarities to acne vulgaris, it is not entirely clear whether the inflammatory lesions of rosacea are follicle-based. One study showed that only 20% of papules had follicular origin, while most histological studies of ET and PP rosacea have documented a low rate of periadnexal inflammation (Marks & Harcourt-Webster 1969; Ramelet & Perroulaz 1988). On the other hand, the glandular type of rhinophyma has been shown to be folliculocentric (Aloi et al. 2000). As well, Demodex folliculorum, a folliclebased mite, has been investigated on multiple occasions for its possible etiological function in rosacea, as will be described below. Thus, additional, more rigorous histological studies may be necessary to determine the role of the pilosebaceous unit in the development of this disease. Microbial organisms Three microbial organisms have been proposed as potentially pathogenic in rosacea: Demodex folliculorum, Bacillus oleronius, and Helicobacter pylori. Demodex mite is a common inhabitant of the human skin. In fact, a prevalence of nearly 100% has been demonstrated in healthy adult subjects using the modern, more sensitive identification techniques (Crosti et al. 1983). Mite density in tissue samples tends to increase with age, paralleling a similar trend in rosacea incidence (Andrews 1982). As its full name implies, Demodex usually resides in the follicles, most commonly on the nose, forehead, and cheeks (Bonnar et al. 1993). It has been suggested that an extrafollicular localization in the dermis may be pathogenic, as it then leads to a pronounced inflammatory reaction (Ecker & Winkelmann 1979; Hoekzema et al. 1995). Numerous studies have attempted to compare mite density in rosacea versus healthy patients. In two studies that employed highly sensitive techniques, the density of Demodex was found to be significantly higher in PP rosacea patients as compared to age-matched controls, whereas no statistical difference was demonstrated for patients with the ET subtype (Erbagci & Ozgoztasi 1998; Forton & Seys 1993). It is unclear, however, whether this difference in mite population is pathogenic or, instead, reflective of the presence of abnormal antimicrobial peptides, as will be discussed in the next section (Bevins & Liu 2007). Of interest, the Demodex density does not seem to decrease when standard oral antibiotics are used for the treatment of rosacea (Bonnar et al. 1993). In addition, though some investigators have noted perifollicular inflammatory infiltrates in the presence of the Demodex mite (Forton 1986), others have noted a lack of such correlation (Marks & Harcourt-Webster 1969; Ramelet & Perroulaz 1988). These discrepancies may, however, be secondary to the difficulty in detecting mites on standard histological sections. More recently, a potential role of a bacterial agent found inside the Demodex mites, Bacillus oleronius, has been investigated. When isolated, this bacterium was able to stimulate an immune response and caused peripheral mononuclear cell proliferation in 73% of patients with PP rosacea as compared to 29% of the control population (Lacey et al. 2007). Further studies are necessary; however, if these findings are confirmed, D. folliculorum may turn out to be essential as a vector of a pathogenic agent. Multiple studies have concentrated on the potential role of Helicobacter pylori in the etiology of rosacea; however, currently available data do not support such a role. Although extremely common in the general population, H. pylori rarely causes symptoms. Nonetheless, most cases of peptic ulcer disease and gastritis have now been linked to this organism, and some correlations between these gastrointestinal conditions and rosacea, such as seasonal variability, have been proposed. 57 58 A high prevalence of H. pylori in rosacea patients has been noted in several studies (Rebora et al. 1995; Szlachcic et al. 1999); most others have refuted such findings when the prevalence is compared to a control population (Jones et al. 1998; Sharma et al. 1998; Utas et al. 1999). Likewise, eradication of the bacterium did or did not improve the symptoms and signs of rosacea, depending on the study (Bamford et al. 1999; Gedik et al. 2005; Herr & You 2000; Utas et al. 1999). It should, however, be noted that the medications typically used to eradicate H. pylori–in particular, metronidazole–are known for their beneficial effect in rosacea, and the effectiveness of therapy does not, therefore, establish a causal association. In one study, elevated plasma levels of tumor necrosis factor (TNF)alpha and interleukin (IL)-8 in response to H. pylori were demonstrated in patients with symptoms of gastritis. Following treatment, most patients with concurrent rosacea experienced a significant improvement in their cutaneous condition, while their plasma cytokine levels normalized (Szlachcic et al. 1999). However, significantly elevated gastrin levels were also noted prior to therapy and may have been responsible for variations in skin temperature and vasomotor instability. In summary, without additional rigorous, well-controlled prospective studies a role for H. pylori in the pathogenesis of rosacea is doubtful. Newest findings The latest findings in the pathophysiology of rosacea seem to link many of the above-mentioned etiological factors; nonetheless, certain questions remain unanswered at this time. In a recent study, an overexpression and abnormal processing of cathelicidin have been demonstrated (Yamasaki et al. 2007). Also known as anti-microbial peptides for their action against Gram-positive and Gram-negative bacteria and some viruses, cathelicidins are part of the innate immune system with important links to adaptive immunity (Di Nardo et al. 2008; Howell et al. 2004; Nizet et al. 2001; Rosenberger et al. 2004; Yang et al. 2000). In the skin, cathelicidin is first secreted as a proprotein, known as 18-kDa cationic antimicrobial protein (CAP18), which is then cleaved by a serine protease, known as stratum corneum tryptic enzyme (SCTE) or kallikrein 5, to the active peptides (Yamasaki et al. 2006). Facial skin affected by rosacea demonstrated a highly-elevated expression of SCTE in all layers of the epidermis compared to normal facial skin, where the expression was also limited to the superficial layers. This was accompanied by a significantly higher expression of a biologically-active cathelicidin fragment, LL-37, and by the expression of several other fragments not encountered in normal skin. Furthermore, injection of these molecules into healthy mice rapidly induced clinical findings of erythema and vascular dilatation, as well as cutaneous inflammation, in a dose-dependent manner. Additionally, injection of SCTE into mice also resulted in similar changes. Finally, protease activity was also shown to be higher in facial skin as compared to other parts of the body, corresponding to the typical localization of rosacea (Yamasaki et al. 2007). Elevated levels of LL-37 lead to an increase in IL-8, a neutrophil chemoattractive cytokine (Yamasaki et al. 2007; Yang et al. 2000). As previously described, the influx of neutrophils initiates an inflammatory cascade and tissue degradation through the release of ROS and MMPs. Additionally, LL-37 is a strong angiogenic agent, thus further contributing to the observed rosacea phenotype (Koczulla et al. 2003). Nonetheless, several questions persist. First, a complete characterization of the additional proteases and protease inhibitors involved in the homeostasis of LL-37 is critical. Second, although the above findings represent a major breakthrough in the pathophysiology of rosacea, the initial insult or defect that eventuates in the overexpression of SCTE and cathelicidin LL-37 still needs to be identified. Finally, future research studies may attempt to develop specific mechanism-based treatments afforded by these new findings. 59 6 ROSACEA – CURRENT MEDICAL THERAPEUTICS INTRODUCTION A S with acne vulgaris, multiple topical and oral agents have been tried over the years for the treatment of rosacea. In fact, a large portion of the medications introduced in Chapter 2 of this book have been successfully utilized in rosacea (Table 7). These are especially important in the treatment of the acne–rosacea overlap, where clinical components of both diseases coexist in the same patient. On the other hand, additional therapeutic agents that may improve one disorder may not be useful in or even aggravate the other disease (Tables 8, 9). Rather than repeat the information already contained in a prior chapter, this chapter will focus mainly on the medications found to be of exclusive value in the treatment of rosacea and will only briefly touch on the previously-covered, but otherwise useful, rosacea agents. For the latter group of medications, the reader is invited to revisit the appropriate sections of Chapter 2. In addition, wherever available, current information on the proposed mechanism of action of the therapeutic agents will also be presented. Although efficacious in the treatment of papulopustular (PP) rosacea, both oral and topical agents tend to have less of an impact on the erythema of erythematotelangiectatic (ET) rosacea, and even so on telangiectasias. On the other hand, vascular-specific lasers may be especially useful in such presentations and will be covered in Chapter 7. Agent Mode Clindamycin Topical Retinoids Topical Azelaic acid Topical Sulfur Topical Sodium sulfacetamide Topical Tetracyclines Oral Azithromycin Oral Isotretinoin Oral Table 7 Agents generally appropriate for the treatment of both rosacea and acne vulgaris Agent Mode Metronidazole Topical and oral Tacrolimus Topical Pimecrolimus Topical Table 8 Agents generally appropriate for the treatment of rosacea, but not acne vulgaris G E N E R A L C O N S I D E R AT I O N S Before the forthcoming discussion on topical and oral therapeutics in rosacea, some important general considerations will now be addressed. First, patient exposure to rosacea triggers, as presented in the previous chapter, must be minimized. Thus, patients should be educated on the avoidance of their specific flushing stimuli. Additionally, the National Rosacea Agent Mode Benzoyl peroxide Topical Salicylic acid Topical Trimethoprim–sulfamethoxazole Oral Table 9 Agents generally appropriate for the treatment of acne vulgaris, but not rosacea 60 Society, which can be found on the internet at http://www.rosacea.org, is an excellent educational resource for the patients. General skin care should be addressed early on in the treatment of the disease. As mentioned in the previous chapter, poor tolerability of topical products is commonly encountered in rosacea, especially the ET subtype. The resultant irritant dermatitis typically presents as roughness and scaling, sometimes accompanied by itching, burning, or stinging (Dahl 2001; Lonne-Rahm et al. 1999). Thus, the selection of nonirritating cleansers, moisturizers, and make-up is essential, as harsh daily skin care regimens may negatively affect skin barrier function (Del Rosso 2005; Draelos 2004, 2006a; Laquieze et al. 2007). Some patients may also benefit from the use of green-tinted moisturizers and other green-colored cosmetics, as these tend to camouflage excessive facial redness. A tancolored foundation can then be applied to match the patient’s desired skin tone (Draelos 2008). Finally, photoprotection is advocated by many practitioners; however, the exact role of ultraviolet radiation in the pathogenesis of rosacea is still debated (Engel et al. 1988; Kligman 2006; Lee & Koo 2005, Wilkin 1994). When utilized, sun blocks containing zinc oxide or titanium dioxide tend to be well-tolerated by rosacea patients. TOPICAL AGENTS As with acne vulgaris, topical agents may be used alone or in combination with oral agents for maximum effect, especially during acute flares of the disease. In addition, topical therapy is generally required for long-term maintenance of remission (Dahl et al. 1998; Nielsen 1983). As mentioned above, rosacea patients may experience significant skin irritability, occasionally necessitating a discontinuation of the very same medications typically prescribed to improve the condition. This distinctive feature of the disease should be considered whenever a flare is observed with a new topical agent, especially if accompanied by itching, burning, or stinging. Antibiotics Metronidazole is one of the most commonly used topical agents in the treatment of rosacea. Although infrequently used in this condition, the oral form is also available for the more severe or recalcitrant cases. Topical metronidazole is available in different countries in a gel, cream, and lotion formulations, with concentrations ranging from 0.75 to 1%. Formulations may be used daily to twice daily (Yoo et al. 2006). A combination of topical metronidazole cream and sunscreen is also available outside the US. Oral metronidazole is available in 200 mg, 250 mg, 400 mg, and 500 mg tablets, as well as 750 mg extended-release tablets. The original study by Pye & Burton (1976) utilized a 200 mg dose taken twice daily, while later studies used a total of 500 mg per day (Aizawa et al. 1992). Metronidazole is a synthetic nitroimidazole antibiotic. It is active against a variety of Gram-positive and Gram-negative, as well as some anerobic, bacteria and certain protozoans, likely through the disruption of microbial DNA (Lamp et al.1999). However, its role in the treatment of rosacea appears to involve a different mechanism of action, as bacteria are unlikely to be involved in the pathophysiology of the condition. Thus, it has been demonstrated that metronidazole possesses significant anti-inflammatory properties in the skin. Specifically, the agent was found to modulate neutrophil function by suppressing neutrophil-generated reactive oxygen species (ROS) in a dose-related manner (Akamatsu et al. 1990; Miyachi et al. 1986). More recently, inherent ROS scavenging and inactivating properties of metronidazole were also demonstrated in a skin lipid model (Narayanan et al. 2007). Systemic absorption following cutaneous application appears to be very low (Elewski 2007). On the other hand, oral bioavailability of metronidazole is very high at over 90%. It is widely distributed following oral administration, including into breast milk and across the placenta. Studies on such distribution following cutaneous application to intact skin are lacking. Adverse effects following topical application to skin are few and typically include symptoms of localized irritant dermatitis. Rare cases of allergic contact dermatitis (sometimes to the base rather than to metronidazole itself) have also been documented (Choudry et al. 2002; Madsen et al. 2007). On the other hand, adverse effects associated with oral administration of metronidazole are fairly numerous and potentially serious, but are more frequent at higher doses and with long-term therapy (Martinez & Caumes 2001). These may include seizures, peripheral neuropathy, nausea,