Primary Scarring Alopecia: Clinical-Pathological Review of 72 Cases and Review of the Literature

here are currently various PSA classifications, but the most accepted is the one proposed by the North American Hair Research Society (NAHRS) [8], whose last revision was in 2003. This classification divides PSAs into 2 large groups based on the dominant type of inflammatory cell infiltrate. This concept had already been suggested by other authors earlier, but it was improved by this work group, who also added the lymphocytic and neutrophil-associated subgroups, as well as mixed (such as acne keloidalis) and nonspecific idiopathic scarring alopecias with inconclusive clinical and histological findings, most of which probably correspond to the final stage of other groups.

In the authors’ experience, PSAs can represent a true clinical challenge when clinical-pathological correlation is difficult to assess, which more commonly occurs during late and advanced stages. This makes histopathological diagnosis more difficult because the main classification criterion (the type of inflammatory infiltrate) might be impossible to assess. In these cases, additional studies such as direct immunofluorescence (DIF) can be useful. Reference sources are limited with respect to their true epidemiology, and randomized, double-blind clinical trials regarding treatment are currently lacking; consequently, therapy is mostly based on case series and the clinicians’ personal experience.

Materials and Methods

In order to obtain new clinical and pathological knowledge about PSAs, we decided to study the epidemiologic, demographic, clinical, and histological characteristics (classified in agreement with the NAHRS) of PSAs cases diagnosed over a 7-year period, at the Department of Dermatology of Hospital Clinic, Barcelona, Spain. We also evaluated the therapies used in each case, their clinical evolution, and the correlation of these results with the predominant type of inflammatory infiltrate on histopathology.

The first step included reviewing all the diagnoses of scalp biopsies, and searching the histopathological database of our patients, using the names of the PSA clinical entities according to the NAHRS classification, (such as LPP dissecting folliculitis, frontal fibrosing alopecia [FFA], etc.) between 2006 and 2012. We included only cases of PSA subtypes as a diagnosis, and for which there was access to the clinical history and histological findings. Collected data included age at diagnosis, sex, ethnic group, clinical pattern, predominant histological infiltrate, DIF result (if available), final clinical diagnosis, time of onset before diagnosis, treatments used, and clinical evolution. The final clinical diagnosis was based on clinical-pathological correlation. Exclusion criteria were diagnoses of secondary scarring alopecia (e.g., tumors, infections, traumas, sarcoidosis, etc.), scarring alopecia outside of the scalp, and lack of access to the clinical history.

Results

A total of 1,085 cases were found during the search for general scalp biopsies and 72 biopsies with the diagnosis of PSA. The most frequent type of predominant inflammatory infiltrate was lymphocytic (31 cases), followed by absent or nonspecific inflammatory infiltrate (30 cases), neutrophil-associated (8 cases), and other/mixed type (2 granulomatous, 1 plasmatic). The ratio of neutrophilic scarring alopecia (NSA) to lymphocytic scarring alopecia (LSA) was 1:4 (Fig. 1, 2).

The ethnic groups were distributed as follows: 93% European-Caucasian, 5% Latin-American, 1% oriental and 1% African-American. Of the 72 cases, 21 (29%) were males and 51 (71%) were females, with a mean age of 51 ± 6 years. Both the LSA and NSA types tended to affect women with a mean age of 60 and 47 years, respectively. The mean onset time prior to diagnosis was 1.71 years, with the time being longer for LSAs than NSAs (1.95 vs. 1.11 years, respectively).

The most frequent clinical pattern was the follicular pattern (20 cases), followed by 18 cases with a single large patch, 13 cases with multiple patches, 7 cases with marginal pattern, 6 cases with a “footprints in the snow” pattern, 5 cases with FD pattern and 3 cases with acne keloidalis pattern (Fig. 3, 4).

According to the time of onset, 8 cases had an acute onset and 31 a subacute onset; 33 presented a chronic subclinical form. The most frequent type was acute-subacute onset in the NSAs and subacute-subclinical in the LSAs. In cases in which the infiltrate was slight or absent, with a predominance of scar tissue, the type was mainly subclinical(Fig. 5).

The most frequent form of evolution was the stable form presented in 44 cases, followed by the slowly progressive form in 18 cases. In a lower proportion, there were 8 cases of acute outbreaks and 3 cases of subacute outbreaks. By correlating with the type of inflammatory infiltrate, the most frequent form of progression in LSA was the stable, while in NSA it was the acute outbreak type (Fig. 6). The correlation between the type of inflammatory infiltrate and the clinical pattern was heterogeneous and not very specific, more commonly in the lymphocytic forms (Table 2).

A total of 48 DIF studies were performed, mostly because of clinical suspicion of LE or LPP, including the FFA type. Only 18 were positive, 10 suggestive of LE, and 8 suggestive of LPP. In 10 of these positive cases, histopathology was inconclusive.

The most frequent final diagnoses were LPP and its variant FFA (24 and 7 cases, respectively). These were followed by LE (13 cases), 20 cases of nonspecific cicatricial alopecia (central centrifugal cicatricial alopecia, PB), 5 cases of FD, and 3 cases of acne keloidalis (Fig. 7).

The treatments used for these entities were multiple and heterogeneous, depending on whether the patient had clinical activity (Fig. 8). Treatment response was evaluated subjectively according to overall physician and patient assessments with a follow-up of at least 1 year in all the cases. Patients with LSA were offered a large variety of treatments, with the most used being potent topical or intralesional corticosteroids in combination with 5% minoxidil solution, systemic corticosteroids, and antimalarials (hydroxychloroquine). Topical corticosteroids were the first line of treatment in almost all cases. However, they were frequently combined with systemic therapies. Intralesional corticosteroids (triamcinolone acetonide 10-30 mg/session) were applied every 1-2 months in severe cases. For nonresponders or rapidly progressing cases, oral corticosteroids (prednisone or deflazacort) were administered to suppress the acute inflammatory response, and then they were gradually withdrawn. As for the use of antimalarials (specifically hydroxychloroquine), they were used in our center as the first line of therapy in cases with a subclinical onset and/or slowly progressive course, at dosages of 200-400 mg/day. The majority of patients were prescribed 5% minoxidil (together with topical or intralesional corticosteroids), more commonly during periods of remission.

For NSA cases, most common treatments included antibiotics (systemic and topical), oral isotretinoin, and corticosteroids (systemic and topical) in combination with 5% minoxidil. As a first-line treatment, a wide spectrum of systemic antibiotics with both anti-inflammatory and antimicrobial properties, such as doxycycline, minocycline, erythromycin and azithromycin, at variable dosages were used. The mean time of use of systemic antibiotics was 6 months (range, 2-12 months); however, since there was a high recurrence rate in the majority of cases when treatment was stopped, new therapeutic cycles were required. As another treatment option, isotretinoin was used in 3 cases at dosages commonly used in acne, with significant response, but with recurrences after drug suspension.

Discussion

The epidemiology of PSAs in the general population is unknown. There are 2 retrospective studies that have been carried out in clinical centers specializing in hair research, in which the estimated prevalence was 3.2% in one center and 7.3% in the other [2,3]. In our study, of the total scalp biopsies, PSAs corresponded to 6.6% of cases, so our results fell within this range. If we had considered all clinical consultations for disorders that affect the scalp, this number would probably be lower. A considerable number of PSA cases are not biopsied because they are diagnosed based on the clinical picture alone. The greater prevalence of these entities in females agrees with our results. A possible reason for this is that women are more likely to consult for hair diseases due to the aesthetic implications they involve. The ratio of LSA to NSA of 4:1 is what we constantly found in the series reviewed [3].

There are few data in the literature about the origin of PSAs. In the majority of PSA cases, histopathology reveals inflammation affecting the upper section of the hair follicle. This would explain why the process is irreversible, as this is the spot where stem cells are located, at the level of the bulge (where the hair erector muscle is inserted), below the infundibulum. The location of this inflammatory response is sometimes the result of antigenic stimulation of the Langerhans cells that are mostly found at the infundibulum and in lower numbers in the isthmus. Examples of these possible antigenic stimuli would be ultraviolet light in the case of LE, certain drugs in the case of LPP and Staphylococcus aureus in the case of FD. New knowledge about PSA origin proposes that a loss in the immune protection of the hair bulb stem cells might be involved [5,9], as well as a dysfunction in stem cell self-perpetuation, increased autoimmunity by proinflammatory cytokines, and genetic/environmental predisposition [10,11,12]. Recent data also suggest an association between lipid metabolism alteration and PSA development, in which dysfunction of the sebaceous glands could play an important role in its pathogenesis. Independent of the initial event, obliteration or permanent functional alteration of the crucial elements for follicular reconstitution causes permanent baldness [13,14,15,16,17].

In our study, we found 72 biopsied cases of PSA in a 7-year period, which is lower than the number found in similar studies in even shorter periods. An explanation for this could lie partially in the fact that many diagnoses are made clinically in agreement with patient history and clinical findings; this is the case of cutaneous LE, in which biopsy is often omitted for diagnosis in the context of a patient diagnosed with known systemic and/or cutaneous lupus and compatible clinical findings. The same is true with FFA, which can be diagnosed without need for a biopsy because of its typical clinical picture. Another explanation is that, since our Department is a hair research center of reference, patients commonly arrive from other services, with previously performed biopsies, and they are consequently not included in the database.

With respect to final clinical diagnoses, the most frequent was LPP and its frontal fibrosing variant (43%), followed by nonspecific scarring alopecia (28%), LE (18%), FD (7%), and acne keloidalis (4%). It should be emphasized that it is impossible to perform a precise diagnosis in almost a third of the patients, which represents a true diagnostic and therapeutic challenge for both dermatologists and patients [2,18]. These numbers are similar to those in recent studies, which highlight the importance to perform biopsies at the initial stages to achieve a more accurate management and prognosis. In the study by Whiting [2], nonspecific alopecias were the most frequent forms (40.6%), followed by LPP, (12.6%) and FD (11.2%). In the series described by Tan et al. [3], discoid LE was the most common diagnosis, with 33.9%, followed by pseudopelade (nonspecific alopecia), with 24.1%, and, in the third place, LPP (22.3%). These discrepancies in the various series are probably due to differences in patient demographics and the pathological clinical criteria for diagnosis of pseudopelade/nonspecific alopecia. It is important to mention that diagnosis of PSAs is not a merely academic exercise, and it should be done in a prompt and accurate manner, given that early treatment of the inflammatory component can prevent the course of the disease, and avoid progression to fibrosis (which is irreversible).

Taking a correct scalp biopsy is crucial for diagnosing PSAs. Consequently, it should be performed at the initial presentation of the disorder, or shortly afterwards, taking a sample from the active (inflamed) borders of the affected area; this increases the possibility of finding more inflammatory infiltrate than fibrosis, therefore establishing a more accurate diagnosis and initiating proper and early treatment to achieve a better response. However, histopathology might not be diagnostic if the biopsy is taken from the incorrect site. This is especially relevant in PSAs since the disorder can be focal and the activity might be hard to see with the naked eye [10,19,20,21]. For these reasons, we would like to emphasize the importance of dermoscopy in studying scalp pathology, in addition to allowing the physician to examine the morphology of PSAs macro- and microscopically. It also makes possible to identify subtle clinical changes, confirm the naked-eye diagnoses, monitor the treatment and guide the scalp biopsy for optimum results. This last usefulness was confirmed by Miteva and Tosti [22] in a study with 80 patients with PSA in which the biopsies were selected based on the dermoscopy findings, reaching a definitive diagnosis in 95% of the cases; it is a quick, accurate method, even for identifying individually affected follicles in focal PSA or in early stages.

In normal clinical practice, we observe 2 distinct clinical patterns of presentation for scarring alopecia. The first, called “footprints in the snow,” is characterized by multiple irregular patches; the second, called “great patch,” corresponds to a large central patch of scarring alopecia surrounded by various smaller similar patches in the form of cicatricial satellitosis scarring. If we analyze other forms of clinical presentations of scarring alopecia, we can see that there are characteristic patterns, which we could call specific patterns: marginal (associated with FFA and traction alopecia), follicular (associated with LPP, FD, alopecia parvimaculata, acne necrotica), tufted pattern (folliculitis in tufts, acne keloidalis nuchae type), abscessing inflammatory pattern (erosive pustular scalp dermatosis, perifolliculitis capitis abscedens et suffodiens), and diffuse pattern (acute LPP, “red scalp” syndrome) [18].

In our series, the most frequent clinical patterns in LSAs were follicular and multiple patches, which coincides with the pattern described as characteristic of LPP. This was followed by the great central patch pattern, which fundamentally corresponded to LE, normally initiating as a single patch that gradually expanded. In the case of nonspecific alopecias, the 2 main patterns were the follicular and large single patch; this could be explained by the likelihood of final or inactive stages of some types of LSAs. It should be emphasized that clinical cases with the “footprints in the snow” pattern that are described as characteristic of the classic PB were finally classified as LPP or LE using other complementary study techniques (e.g., DIF); this would support the belief that this entity is probably not an entity in itself, but rather a final process of another. In the NSAs, the most frequent clinical pattern was FD, characterized by pustules in the follicular areas and sometimes associated with follicles in tufts; however, there were also 3 cases that presented as a large central patch. From these results, we can conclude that the clinical pattern can allow us to infer the type of inflammatory infiltrate of PSAs, but that it does not contribute to making an etiological diagnosis.

Conclusions

The PSAs are trichological emergencies. To reach an accurate diagnosis, proper clinical-pathological correlation is required. The great variability in the clinical presentation of PSAs makes their diagnosis and treatment a true challenge for the dermatologist. The entity LSA is the most frequent histological type, with LPP and its variant FFA being the most diagnosed disorders. Aggressive therapy is often needed to avoid progression of the disorder. The therapeutic alternatives available will sometimes depend on the type of histological infiltrate; both topical and systemic treatments are valid.

Scarring alopecias can be hard to diagnose. Biopsy should be performed during the initial stages of the disorder in order to make a more accurate diagnosis. Clinical presentation can sometimes be the decisive factor in making a diagnosis. A great number of cases end up being classified as nonspecific scarring alopecia and are currently called pseudopelade.

Characteristic clinical and pathological findings sometimes allow us to make a precise diagnosis. However, diagnostic certainty is often hard to achieve and reflects the limits of current knowledge of these disorders.

The fact that an important number of our patients were finally classified as having nonspecific scarring alopecias strengthens the importance of early referral of these patients for dermatological assessment to receive proper treatment.

Finally, appropriate patient counselling, support measures from other health professionals as clinical psychologist and beauticians and camouflage measures as reconstructive hair fibers and prostheses should be provided in order to benefit the quality of life of those patients. Patients with PSAs need the care of dermatologists specializing in hair disorders to help them in all their clinical and personal concerns.