An observational descriptive survey of rosacea in the Chinese population: clinical features based on the affected locations

Introduction

Rosacea is a common chronic inflammatory cutaneous disorder, predominantly presenting on the faces of adults, which is characterized by a tendency of frequent facial flushing, central facial erythema, papulopustules, telangiectasias, ocular manifestations, and phymatous changes primarily on the nose. The exact cause of rosacea remains unclear. It may contain hereditary components, and has been hypothesized to be associated with disorders of the innate immune system, dysfunction of facial vascular regulation, neurogenic inflammation, and elevated levels of Demodex mites, among others (Gibson, 2004; Yamasaki & Gallo, 2009; Abram et al., 2010; Steinhoff et al., 2011; Van Zuuren et al., 2015; Margalit et al., 2016). Morbidity of rosacea varies greatly among different ethnic populations, with a higher prevalence amongst fair-skinned individuals of northern European or Celtic ancestry (Spoendlin et al., 2012; Tuzun et al., 2014). There has yet to be any published data regarding the incidence of rosacea in the Chinese population.

The current classification system for rosacea describes four distinct clinical subtypes: erythematotelangiectatic, papulopustular, phymatous, and ocular (Wilkin et al., 2002). Erythematotelangiectatic rosacea is characterized by flushing and persistent central facial erythema with or without telangiectasia. Papulopustular rosacea is associated with persistent central facial erythema with transient, central facial papules or pustules or both. Phymatous rosacea is characterized by skin thickening, irregular surface nodularities, and enlargement, which can affect the nose, chin, forehead, ears, and eyelids. The most commonly affected area is the nose, which is also called rhinophyma. There are three grades of phymatous changes: (1) mild, manifested as puffiness and mildly patulous follicles with no clinically apparent hypertrophy of connective tissue or sebaceous glands and no change in contour; (2) moderate, manifests as moderate swelling and moderately dilated patulous follicles with clinically mild hypertrophic change in nasal contour but no nodular components; (3) severe, manifests as marked swelling and large dilated follicles with distortion of contour with a nodular component. Ocular rosacea is a subtype that displays a series of non-specific ocular symptoms. For the first three subtypes, the present method of assessing the severity of this disease classifies the progression of rosacea into four general stages (Wilkin et al., 2004): Stage 1, which is characterized by frequent blushing; Stage 2, which is characterized by transient erythema of the central areas of the face, and obvious, but slight, telangiectasias; Stage 3, which includes more severe facial erythema, increased telangiectasias, and papule and pustule formation; and Stage 4, which is the most severe, and is also known as rhinophyma (Zuber, 2000). Based on this classification system, it could be inferred that the stages of rosacea might evolve from one to another and rhinophyma seemed to be the “end-stage” (Wilkin, 1994; Jansen & Plewig, 1997). A number of studies have also confirmed the possibility of progression between subtypes in western countries (Crawford, Pelle & James, 2004; Powell, 2005; Tan & Berg, 2013; Tan et al., 2013), but this theory is still being questioned (Crawford, Pelle & James, 2004). Moreover, during the progression process, patients can display a number of subtypes simultaneously, which makes the classification of rosacea vague and indistinct. Nevertheless, no other classification standards can define the clinical parameters more scientifically and reasonably. During our clinic work, we noticed some interesting phenomenon in the Chinese rosacea patients. For example, the clinical features varied among patients with different affected areas. Patients whose lesions first appeared on the nose could easily develop rhinophyma with thickened and nodular skin. While patients whose initial lesions occurred outside the nose area rarely progressed to clinically apparent hypertrophic changes of rhinophyma. In addition, based on our preliminary observation, we supposed that the involved location might be the possible sign of the natural development of rosacea. However, few studies have evaluated the differences in clinical features and disease outcomes among rosacea patients on the basis of involved locations so far.

In our present descriptive survey, we recruited 586 rosacea patients from south of China, aiming to evaluate and compare the clinical features of rosacea at different sites of the face, assess quantitative details regarding the rosacea-associated symptoms, signs and indicators of epidermal barrier function, and analyze the potential for progression among different affected areas in the Chinese population, hoping to provide some evidences for future investigations on a better classification scheme.

Patients and Methods

Methods

As shown in Fig. 1, the face can be divided into five main parts: the forehead, eyes, cheeks, nose, and perioral area. In this study, the pattern of skin involvement was classified into four groups: (1) full-face group (rosacea lesion occupied no less than two parts of the face); (2) cheek group (rosacea lesions limited to the cheeks); (3) nasal group (rosacea lesions limited to the nose); (4) perioral group (rosacea lesions limited to the perioral area).

Demographic and clinical data, including patient age, sex, disease duration, and self-reported symptoms, including burning, dryness, itching, stinging, skin tension, swelling, ants line sense, and pain (graded from 0 to 10, with 0 representing no symptoms and 10 representing most severe symptoms) were recorded by the patients. Moreover, three dermatologists graded the rosacea-associated symptoms based on the National Rosacea Society’s grading system independently (0–3).

Results

Discussion

Although the National Rosacea Society Expert Committee (NRSEC) diagnosis and classification system has been incorporated in many basic, clinical, and epidemiological investigations (Bae et al., 2009; Abram et al., 2010; Abram, Silm & Oona, 2010; Aksoy et al., 2010; Khaled et al., 2010; Lazaridou et al., 2010; Tan et al., 2013), shortcomings of the NRSEC recommendations do exist, which does not make it universally accepted (Wilkin et al., 2002; Crawford, Pelle & James, 2004; Van Zuuren et al., 2015). The present descriptive study design evaluated the differences in clinical features among rosacea patients with lesions at four different sites for the first time.

A dysfunction of the epidermal barrier was observed in rosacea patients in the previous studies (Dirschka, Tronnier & Folster-Holst, 2004; Lee, Jeong & Ahn, 2006; Ramos-e Silva & Jacques, 2012); we detected the barrier-associated physiological indicators that showed lower water content, higher TEWL, and hemoglobin in our rosacea patients compared with the healthy control. An interesting result of our study was the epidermal barrier function of the cheeks and chin in the nasal group was close to normal, which might explain, at least partly, why the nasal group had the least amount of burning, dryness, itching and overall symptoms. On the other hand, the more severe self-report symptoms observed in the full face or cheeks or perioral groups might be attributed to the impaired epidermal barrier. These distinctive features might be associated with the distinct physiological characteristics of local facial skin. In previous studies, the symptoms of burning/stinging were compared between patients with erythematotelangiectatic rosacea or papulopustular rosacea. However, no consistent conclusions were reached (Lonne-Rahm, Fischer & Berg, 1999; Crawford, Pelle & James, 2004). We inferred that the inconsistent results of these studies might have been attributed to the differences of the affected areas of the patients included. These findings need to be further confirmed, but if it is the different sites involved that contribute to the epidermal barrier functions and the clinical symptoms, the obtained conclusion may be beneficial for optimizing therapy schemes. For example, for patients who have full-face, cheek or perioral involvement, restoring the normal epidermal barrier function would be particularly important (Bikowski, 2001).

In this study, transitions between the four groups were addressed by surveying the sequence of the involved locations in rosacea patients by retrospective analysis. We found that (1) there were no transitions between the perioral groups and the other groups; (2) there were only a few patients in the full-face group whose lesions first occurred on the nose with the appearance of erythematotelangiectasia or papulopustule, and most of them keep the erythematotelangiectasia or papulopustule appearance on the nose; (3) Mild rhinophyma was occasionally seen in the full-face group, especially for those first lesions occurring in the nose, but moderate to severe rhinophyma, which means distortion of contour, only occurred in the nasal group, and transitions between the pre-existing rhinophyma in the nasal group with other groups were not observed; and (4) the cheek group could progress to a full-face. Considering the fact that the self-reported, dermatologist-evaluated grading of symptoms and physiological indicators of epidermal barrier function between the cheek and full-face groups did not have much difference, and the full-face group could have their lesions start at the cheek in majority of the patients, moreover, all the patients have lesions on the cheek when they were included in, we have speculated the cheek group might be a mild preceding form of the full-face group. However, the findings mentioned above require further investigation in a larger population with some following-up data.

Combined with the differences in the severity of symptoms, physiological indicators of epidermal barrier function, and disease outcomes between the nasal group and the other groups mentioned above, the following was observed: rosacea that occurred in the cheek or in the perioral area or in the full face were characterized by obvious self-conscious symptoms and abnormal epidermal barrier function, along with a relatively low risk of rhinophyma formation; thus, treatment should be focused on restoring the normal epidermal barrier function. On the contrary, rosacea with locations restricted to the nose meant milder symptoms and relatively normal epidermal barrier function but a potential progression to rhinophyma. Early medical treatments then become indispensable to avoid irreversible disfiguring of the nose.

Consistent with previous studies, the majority of patients included in our study were women, but more than half of the nasal group were men, further confirming that men are at greater risk of having an affected nasal area (Jansen & Plewig, 1997; Kyriakis et al., 2005; Abram et al., 2010; Spoendlin et al., 2012; Tuzun et al., 2014). It might be associated with the different hormone levels or the facial skin physiology between women and men, while no related studies have been presented so far. As for the ocular subtypes of rosacea, because patients were recruited mainly in dermatologic clinics, patients with single ocular disorders were not included in our study, but 46 (7.9%) presented with ocular manifestations concomitantly, the prevalence of which is within the range of recent investigations (6–72%) (Spoendlin et al., 2012). Interestingly, all the patients with ocular symptoms were in the full-face group. It has been hypothesized that ocular rosacea tends to occur in patients with more extensive lesions on the face. Thus, careful ocular examination is recommended.

Conclusions

In summary, herein, we revealed that rosacea lesions on the cheek or full face, nose and perioral of the face are associated with distinct features, including the proportion of male vs. female patients, self-reported symptoms, doctor-evaluated severity of symptoms and signs, and physiological indicators of epidermal barrier function. We also found that the localization of each group was relatively stable and barely transformed into other types. Our data may be beneficial for modifying the existing classification and stage definitions, optimizing clinical treatments, and facilitating mechanism researches of rosacea in the future.

Supplemental Information