In androgenetic alopecia (AGA), hypoechogenicity of the involved hair follicles has been reported, a finding whose pathophysiology remains unclear and which could be attributed to inflammation.8 In a prospective study of 33 patients with AGA and 10 healthy controls,9 shallower follicles in the dermis and a replacement of the normal predominantly trilaminar pattern of hair shafts by a bilaminar one and areas of absence of hair follicles were found in AGA.

In alopecia areata (AA), signs such as empty hair follicles and perifollicular hyperechogenicity in the subcutaneous tissue8 have been identified, which could facilitate the direct evaluation of the severity of the disease and thus help design therapeutic strategies. In turn, the findings recently reported by Mikiel et al.10 are interesting, who characterized the ultrasound images of 25 patients with AA and its phases, 10 with AGA, and 12 with seborrheic dermatitis, correlating the findings with trichoscopy (Table 1). In active AA, clear and widened water drop-like follicles were described; the inactive phase was characterized by a smaller number of follicles without well-demarcated borders; and in the regrowth phase, follicles of different widths, elongated and with widened distal parts, were found. Both the width and echogenicity of the follicles in AA, regardless of the stage of the disease, were significantly greater. In cases of AGA, the width of the follicles was the smallest of all the groups studied.

The first publications that used HFUS in the evaluation of scarring alopecias showed that dermal inflammation appears hypoechoic and subcutaneous inflammation, hyperechoic, and, with Doppler, the affected areas are hypovascular,2,4,9,11 However, increased vascularization in the scarred areas using microvascular imaging modalities and even greater thickness and stiffness with shear wave elastography12 have also been reported.

Recently, HFUS was evaluated as a diagnostic method in 8 patients with lichen planopilaris (LPP), 11 with discoid lupus erythematosus (DLE), 14 with frontal fibrosing alopecia (FFA), and 11 healthy volunteers.13 The active form of the disease in all groups was characterized by follicular structures with morphology and parameters different from those observed in the control group; the inactive forms were characterized by a significant reduction in the number of follicles (Table 2). The echogenicity of the skin and follicular structures was significantly higher than in healthy volunteers, which is likely due to the underlying fibrosis process. Of note, the absence of follicular openings in trichoscopy often did not correlate with the total lack of follicular structures in ultrasound images. All these findings indicate that HFUS allows a more precise evaluation of when the scarring process is complete.

For its part, UHFUS in FFA has reported findings such as perifollicular hypoechogenicity in the mid-dermis, distal ambiguity of hair follicles, and undulating hyperechoic bands in the subcutaneous cellular tissue (Fig. 2). The first 2 were also observed in LPP and corresponded in histopathology to perifollicular fibrosis and inflammation in the promontory and to inflamed or destroyed follicles in the isthmus or infundibulum, respectively. In turn, the hyperechoic bands in the subcutaneous tissue corresponded to fibrous septa distributed linearly in the hypodermis.8 Another ultrasound study in 99 women with FFA and 40 controls showed greater vascular diameter and greater flow in the hairline area vs the alopecic band area and the control group, which can be explained by both active inflammation in the frontal line and fibrosis in the alopecic band. In addition, the diameter of the vessels was significantly greater in the healthy scalp area of patients with FFA vs the control group, which could be explained by subclinical inflammation and could be a predictor of future progression.14

Figure 2.

Ultrasound of frontal fibrosing alopecia. (A) Loss of hair follicles (*) and subcutaneous adipose tissue. Loss of definition of the dermohypodermal border (oblique arrow), increased subcutaneous echogenicity (o), and prominent subcutaneous lamellar structures (horizontal arrow). (B) Color Doppler of the same region demonstrates increased dermal vascularization.

(0.12MB).

Dissecting cellulitis presents ultrasound findings similar to hidradenitis suppurativa and includes dilation of hair follicles, pseudocysts (<1cm), hypoechoic collections (>1cm), tunnels that run through the dermis and subcutaneous tissue and may present variable degrees of fibrosis and, with color Doppler, variable degrees of dermohypodermal hypervascularization in the periphery of the lesions.15

Rudnicka et al.3 published a small case series in which RCM was used to characterize some hair diseases. The structure of the hair shaft presented the same appearance in healthy controls, AA, and AGA; however, heterogeneity in the thickness of the shafts was observed in the latter.

RCM criteria for the diagnosis of non-scarring alopecias such as delimited dermal papillae, miniaturized hair shafts, and follicular keratinization have been reported (Fig. 3). These findings were published in a 2016 retrospective study,18 in which several descriptive parameters of scarring (28 with LPP and 9 with DLE) and non-scarring alopecias (30 with AGA and 19 with AA) were established, and statistically significant characteristics for the 2 types of alopecia were identified (Table 3). Although no significant distinctive features were identified for AGA vs AA, inflammatory cells were found in the upper epidermis and around adnexal structures in several patients with AA, which could be explained by its autoimmune origin.

Figure 3.

(A) Erythematous-desquamative plaque on the vertex. (B) Dermoscopy: diffuse erythema, follicular plugs, and fine arborizing vessels. (C) Epidermal RCM: dendritic-like cells with peri- and interfollicular distribution (

). (D) RCM at the DEJ: follicular plug (*), inflammatory infiltrate composed of round-shaped cells (

), and dilated vessels (o).

(1.77MB).

In tinea capitis, homogeneous rounded hyperreflective structures of 5–10μm have been described in the proximal area of the hair shaft, corresponding to conidia that, in vivo, were observed compact, while ex vivo, they were more dispersed.19,20

For its part, distinctive findings have also been found in chemotherapy-induced alopecia. In a prospective observational study of 16 women with breast cancer, confocal and trichoscopic characteristics specific to the evolutionary phases of chemotherapy-induced alopecia were described.21

With RCM, inflammatory cells in the dermis, epidermis, and around adnexal structures have been described, which is consistent with the inflammatory or autoimmune pathogenesis of scarring alopecias. In turn, 3 highly specific characteristics have been identified: inflammatory cells in the epidermis, absence of miniaturization, and absence of follicular keratinization.18 In fact, one of the main criteria that allows discrimination between scarring and non-scarring alopecias is active inflammation in the adnexal epithelium or dermoepidermal junction, with the corresponding darkening of the dermal papillae and absence of their normal border. Another important criterion is dermal sclerosis, characterized by increased and thickened fibers arranged radially around the follicle. Although these findings are typical of scarring alopecias, they have been reported in some cases of chronic non-scarring alopecia, such as in the late phase of AGA, which is attributed to severe sun damage and chronic inflammation.18,22

Regarding LPP and FFA, there is a specific study from 202023 in which 10 patients with LPP and 2 with FFA were evaluated, and inflammatory cells in the epidermis next to the adnexal infundibulum were described in almost half of the cases, a finding consistent with the histopathology of early stages.18,23 In long-standing lesions, dilated blood vessels, an increase in poorly defined thick dermal fibers, and extensive perifollicular fibrosis were observed (Fig. 4).

Figure 4.

(A) Scarring alopecic plaque in the interparietal area. (B) Trichoscopy: erythema, perifollicular hyperkeratosis, and follicular tufting. (C) RCM: infundibular hyperkeratosis with whitish material filling the infundibulum of the hair follicle and surrounding the hair shaft. (D) Dermis: increased number of poorly defined and thick dermal fibers.

(1.09MB).

In LPP and DLE, characteristics that have shown correlation with histopathology have also been described. There is a predominance of inflammatory infiltrate in the epidermis and diffuse interface changes in LPP, focal changes in DLE with dermoepidermal junction involvement, and changes located exclusively in the basal layer of the adnexal epithelium in LPP with prominent follicular involvement. These findings were described in a study of 7 patients with a previous histological diagnosis of LPP and DLE.24

Particularly for cases of DLE, the diagnostic process can be difficult, so the use of RCM is promising. An observational study of 12 patients with DLE found that interface dermatitis and infundibular dilation were the characteristics that exhibited the highest level of agreement with histopathology.25 In turn, confocal analysis associated with 3 trichoscopic findings managed to improve the agreement with histopathology.

Morphea and folliculitis decalvans have also been reported. In a patient diagnosed with coup de saber morphea, RCM showed marked hyperreflective areas with severe atrophy of eccrine glands, disappearance of sebaceous glands, and a few small follicles,26 which contributed to the diagnosis of early (with a predominance of lymphocytic infiltrate) and late stages (with more prominent sclerosis). Regarding folliculitis decalvans, RCM in 14 patients identified highly reflective tubular structures, bright round or oval structures surrounded by a regular border, and interfollicular hyperkeratosis, corresponding in trichoscopy to tufted hairs, pustules, and scales or crusts, respectively.27,28

To determine hair morphology and vitality with OCT, 2 parameters are used that allow comparisons and analyses in studies: cross-section (CS) and form factor (FF), the latter determined by the ratio of the maximum and minimum diameters of the hair. The rate of variation in CS in healthy volunteers is reportedly<3%, while the corresponding FF varies by ≤4%30 and has been observed to worsen in patients undergoing chemotherapy31 and with AA.32

The examination of 20 hairs from the frontal and occipital areas, ex vivo, in women with breast cancer on tamoxifen (17) vs chemotherapy (17) showed that the mean diameter after chemotherapy was 15% lower and the CS was significantly smaller than in the hairs obtained prior to therapy; although the FF did not vary for the frontal area, it did in the occipital area. No changes in CS or FF were observed in patients on tamoxifen.

García Bartels et al.32 published morphological changes detected with OCT in 9 patients with AA. The CS in the alopecic patches was significantly smaller vs unaffected areas; however, the FF did not show significant differences. These results indicate that structural abnormalities of the shafts are found only in active AA lesions. Another publication monitored the progression of 2 patients with patchy AA and 1 with universal AA treated with 3 sessions of platelet-rich plasma33. They found an increase in the length and number of follicles in patchy AA and no significant changes in the patient with universal AA. A different study on AA found that 22 out of 30 scans of the dermoepidermal junction had more follicles than on the surface,4 indicating the presence of growing hairs in AA that are not visible on the surface.

A recent publication proposes an atlas of tomographic characteristics obtained from the evaluation of 6 healthy subjects, 12 with scarring alopecia, and 14 with non-scarring alopecia.34 They highlight the findings of increased epidermal thickness and decreased follicle count in the scarring alopecia group vs the non-scarring alopecia and control groups. Although few patients are analyzed, this study provides a valuable set of images potentially useful for the monitoring and management of alopecias, including scarring alopecias (Table 4).

In FFA, OCT has been used to show that epidermal thickness is greater at the hairline and smaller in the scarring alopecic band.35 These findings are explained by the presence of inflammatory infiltrate and edema in areas of activity and atrophy in the scarred areas, respectively. In addition, as the inflamed areas progress to scarred areas, the vascular flow of the superficial plexus progressively decreases, and the deeper flow increases, which would occur due to obliteration of superficial vessels in the scarring stage with redistribution of flow toward the deeper ones.