3307jum_online_Layout 1 6/20/14 9:00 AM Page 1265
ORIGINAL RESEARCH
Hair Morphology in Androgenetic Alopecia Sonographic and Electron Microscopic Studies Ximena Wortsman, MD, Robinson Guerrero, MD, Jacobo Wortsman, MD
Video online at www.jultrasoundmed.org
Objectives—To assess hair morphology in androgenetic alopecia on sonography and electron microscopy. Methods—A prospective study was performed in 33 patients with androgenetic alopecia and 10 unaffected control participants. In vivo sonography of the hair follicles of the scalp and in vitro sonography and electron microscopy of the hair shafts were performed according to a standardized protocol that included analysis of the right frontal and occipital regions. The upper frequency limit of the ultrasound probes ranged between 15 and 18 MHz. Results—Scalp hair follicles and hair shafts were recognizable on sonography in all cases. Hair follicles in alopecia cases had significantly lower depths (P < .05). The hair shafts in alopecia also had a different distribution of their laminar pattern on in vitro sonography, with a greater presence of mixed (trilaminar and bilaminar) and solely bilaminar tracts in comparison with the controls (mostly trilaminar). On electron microscopy, the alopecia hair tracts showed irregularities and commonly a “melted candle” appearance of the cuticle.
Received October 15, 2013, from the Departments of Radiology and Dermatology, Institute for Diagnostic Imaging and Research of the Skin and Soft Tissues, Clinica Servet, Faculty of Medicine, University of Chile, Santiago, Chile (X.W.); Department of Dermatology, Alopecia Clinic, Fundacion Medica San Cristobal, Santiago, Chile (R.G.); and Department of Medicine, Southern Illinois University School of Medicine, Springfield, Illinois USA (J.W.). Revision requested October 30, 2013. Revised manuscript accepted for publication November 18, 2013. We thank Nancy Olea, BSc (Electron Microscopy Unit, Faculty of Medicine, University of Chile), for support with the electron microscopic samples. Address correspondence to Ximena Wortsman, MD, Departments of Radiology and Dermatology, Institute for Diagnostic Imaging and Research of the Skin and Soft Tissues, Clinica Servet, Faculty of Medicine, University of Chile, Lo Fontecilla 201, of 734, Las Condes, Santiago, Chile. E-mail: [email protected]
doi:10.7863/ultra.33.7.1265
Conclusions—Sonography and electron microscopy uncover distinct abnormalities in the morphology of hair in androgenetic alopecia, which may potentially support the diagnosis and management of this common condition. Key Words—androgenetic alopecia; androgenetic alopecia sonography; androgenetic alopecia ultrasound; androgenic alopecia; hair sonography; hair ultrasound; scalp sonography; scalp ultrasound; superficial structures
B
esides the esthetic connotations, hair serves important roles in cutaneous functions such as dispersion of sweat gland products and regulation of body temperature.1 Hair loss, or alopecia, is therefore clinically important as it may be also associated with psychologic disturbances from disruptions of the selfimage of affected patients.2 Since hair loss may imply hair fragility or structural abnormalities, a detailed morphologic evaluation should be an important step for studying its pathologic characteristics and supporting management. In this regard, recent developments in the field of sonography have allowed details of the hair anatomy to be uncovered.3 On sonography, normal scalp hair follicles appear as oblique hypoechoic bands in the dermis, extending to variable depths according to the hair cycle phase; thus, active hair follicles (anagen phase) are located both in the superficial and deep dermis and can even reach the upper hypodermis, whereas inactive hair
©2014 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2014; 33:1265–1272 | 0278-4297 | www.aium.org
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1266
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
follicles (telogen phase) are restricted to the upper dermis. The external hair segments of the scalp, also called hair shafts or tracts, are reported as trilaminar hyperechoic sonographic structures with an outside cuticle-cortex complex and an inner medulla.3,4 The hair cuticle (outer layer of the hair tract) can be defined at high resolution with electron microscopy, an imaging method that has been used in this manner for the study of pili annulati, a genetic abnormality of the hair shaft.5 The most common type of hair loss is androgenetic alopecia, which develops mostly in genetically predisposed men in the presence of androgens (androgen dependent), although it can also affect women. It is characterized by progressive miniaturization of the hair follicles and generally appears as nonscarring scalp alopecia of the frontotemporal and vertex regions.6 The prevalence of androgenetic alopecia is highest in white men, reaching frequencies of up to 80% in men older than 70 years.6,7 Current methods for the evaluation of hair characteristics may present disadvantages that can limit their use. For example, the widely available hair count has a subjective component and can be tedious and time consuming. Hair pull tests that are based on the concept that easy-toextract hairs correspond to hair in the telogen phase (inactive) may be painful and variable from person to person. The standard trichogram requires the extraction of 60 to 80 hairs with a rubber-armed forceps, to allow a microscopic evaluation, a procedure that could be painful and unpleasant for people losing their hair. Moreover, the trichogram usually misses small hairs from early anagen; whereas the plucking procedure itself is known to modify the local hair cycle. Dermoscopy (gross magnification) provides an amplified image of the skin surface, hair shafts, and follicular ostia, but the underlying hair follicles are inaccessible. The shortcomings of confocal microscopy, an optical imaging method that provides high resolution, almost comparable to histologic analysis, and optical coherence tomography, an imaging method based on the signal and acquisition properties of light, include their limited tissue penetration (up to 0.3 mm for confocal microscopy and up to 2 mm for optical coherence tomography), with restricted or no access to the deep dermis and hypodermis.8,9 The reference standard technique in the evaluation of hair abnormalities is histologic analysis, which is an invasive modality based on the examination of a biopsy-obtained sample of scalp tissue, usually single and 4 mm in diameter, that, because of local variations in hair involvement, is not always adequate as being representative of the global process. In addition to being painful, scalp biopsy is difficult to repeat in the same area for evaluation during and after therapy.10,11
1266
Therefore, noninvasive imaging modalities seem to be needed to overcome some of these limitations and reveal anatomic details. Hence, the aim of this study was to assess the morphologic patterns of the hair using sonography and electron microscopy in patients with androgenetic alopecia.
Materials and Methods We prospectively studied 33 male patients referred by a dermatologist with the diagnosis of androgenetic alopecia (mean age, 32 years; range, 17–41 years) and 10 unaffected male control participants (mean age, 29 years; range, 19–49 years). The inclusion criterion was patients with androgenetic alopecia presenting disease severity of type II or greater (Hamilton-Norwood classification; Figure 1).12,13 Exclusion criteria were concomitant antialopecia treatment (oral or topical), coexistence of other cutaneous diseases (local or systemic), and a history of hair implants and applied products (hot irons, dyes, or chemical products). The study protocol followed the Helsinki principles of medical ethics and was approved by the Institutional Review Board; all participants signed a written informed consent form. Figure 1. Hamilton-Norwood classification of alopecia.
J Ultrasound Med 2014; 33:1265–1272
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1267
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
The ultrasound machines were LOGIQ E9 XD Clear (GE Healthcare, Milwaukee, WI) and HDI 5000 (Philips Healthcare, Bothell, WA) systems matched to compact linear probes tunable to frequencies of 8 to 18 and 7 to 15 MHz, respectively. Sonography was performed by scanning sequentially the right frontal and occipital regions of the scalp after careful manual separation of the hair shafts and application of a copious amount of gel to the surface of the skin. Examination sites were as follows: for the right frontal region, the area located 10 cm above the center of the right eyebrow; and for the right occipital region, the area located 4 cm behind the posteroinferior border of the right ear pinna. Sonographic parameters (millimeters) included dermal thickness, scalp hair follicle depth and transverse diameter (width), and hair shaft thickness. The sonograms followed the axes of the hair follicles and shafts for assessing their morphology. The measurements were performed at the maximum distance point (depth, width, or thickness following the axis of the hair). Immediately before the sonographic examination a total of 20 hair shafts approximately 4 cm long were trimmed from the same frontal and occipital regions (10 tracts each) for sonographic and electron microscopic studies. The hair shafts were trimmed at the level of the scalp surface. Thus, for in vitro sonographic studies, 5 shafts from each region were scanned while immersed in 80 mL of saline in a 100-mL disposable plastic cup. For the electron microscopic studies, hair shaft fragments were placed on aluminum mounts and covered with carbon adhesive tabs. The samples were covered with atomic particles of palladium/gold in a layer of 20 nm in thickness using a Polaron E-5000 sputter coater (Bio-Rad, Cambridge, England) and were observed on a DMS 940 scanning electron microscope (Zeiss, Oberkochen, Germany). Lateral and cross-sectional views of the hair specimens were then exposed to the electron beam. After surface analysis for assessing
the hair cuticle morphology, digital photomicrographs were taken at magnifications of ×500, ×1000, and ×2000. Data were analyzed with CaEst version 1.2 statistical software (Softonic, Barcelona, Spain) for assessing significance (at P < .05).
Results On in vivo sonography, hair follicles were recognizable, although in patients with androgenetic alopecia, both frontal and occipital follicles were shorter, extending to significantly lower depths than those of controls (frontal region, P = .0346; occipital region; P = .0047). In androgenetic alopecia, in addition to the shrinkage of the hair follicles, spaces without detectable hair follicles were observed in the dermis. However, neither the dermal thickness as a whole, nor the hair follicle transverse diameter was significantly different from that of controls (Table 1 and Figures 2–5). In vitro sonographic studies included a total of 430 hair shafts: 330 from the patients with androgenetic alopecia and 100 from the controls. Patients with alopecia and controls showed differences in the morphologic distribution of the laminar hair shaft pattern. A trilaminar hyperechoic pattern was predominantly seen in the controls. Mixed trilaminarbilaminar and solely bilaminar hyperechoic patterns were more frequently seen in the androgenetic alopecia cases. With regard to the specific zones, the hair shafts from controls consistently showed a trilaminar hyperechoic appearance in the frontal region; however, in the occipital region, the pattern was composed of two variants, where 80% (n = 8) of the controls showed the trilaminar hyperechoic appearance but 20% (n = 2) instead had a bilaminar hyperechoic appearance. The mean thicknesses of the frontal and occipital hair shafts in controls were 0.5 and 0.6 ± 0.1 (SD) mm, respectively.
Table 1. Sonographic Parameters of Scalp Hair From Patients With Androgenetic Alopecia and Controls Alopecia Parameter
Mean
SD
Age, y Frontal dermis thickness, mm Frontal hair follicle depth, mm Frontal hair follicle transverse diameter, mm Occipital dermis thickness, mm Occipital hair follicle depth, mm Occipital hair follicle transverse diameter, mm Frontal hair shaft thickness, mm Occipital hair shaft thickness, mm
32.0 1.7 1.4 0.6 1.9 2.1 0.6 0.5 0.5
5.8 0.4 0.4 0.1 0.3 0.5 0.3 0.1 0.1
Control Mean 29.0 1.5 1.7 0.6 2.1 2.6 0.6 0.5 0.6
SD
P
10.9 0.2 0.3 0 0.4 0.3 0.1 0.1 0.1
.2573 .1373 .0346a >.99 .0954 .0047a >.99 >.99 .0084a
aSignificant (P < .05).
J Ultrasound Med 2014; 33:1265–1272
1267
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1268
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
In patients with androgenetic alopecia, sonography of the hair shafts in the frontal region showed a trilaminar hyperechoic pattern in 45% (n = 15), a mixed pattern (trilaminar and bilaminar) in 45% (n = 15), and solely bilaminar hyperechoic hair shafts in 10% (n = 3). In the occipital region, 70% (n = 23) of patients with androgenetic alopecia showed a trilaminar hyperechoic appearance, 24% (n = 8) showed a mixed pattern (trilaminar and bilaminar), and 6% (n = 2) showed only bilaminar hyperechoic hair shafts (Figure 6 and Video 1). The mean sonographic transverse diameters of the frontal and occipital hair shafts in the alopecia cases were similar and corresponded to 0.5 ± 0.1 for each region. No significant differences were found when comparing the thickness of the hair shafts of the frontal regions in controls versus androgenetic alopecia cases (P > .99).
However, alopecia cases showed a significantly lower thickness in the occipital region compared to controls (P = .0084). Electron microscopy of hair shafts from the frontal and occipital regions of the controls showed a smooth, sharp, well-defined surface of the cuticle scales. In the patients with androgenetic alopecia, the surface of the cuticle on the hair shafts was irregular, with a poorly defined scale surface and contour, commonly giving a “melted candle” appearance (Figures 7 and 8).
Discussion To our knowledge, this study is the first to describe the sonographic appearance of hair follicles and hair shafts in patients with androgenetic alopecia: the target structures of
Figure 2. Normal hair follicles in the frontal region. A and B (hair follicles outlined) are grayscale sonograms that follow the axis of the hair follicles and show several oblique hypoechoic bands in the dermis; b indicates bony margin of the skull; d, dermis; h, hypodermis; and m, epicranius muscle. A
B
Figure 3. Hair follicles in a patient presenting with androgenetic alopecia (frontal region). A and B (hair follicles outlined) are grayscale sonograms that follow the axis of the hair follicles and show sparsity and shrinkage of the hair follicles in the dermis. Notice the spaces (asterisks) without detectable hair follicles. Abbreviations are as in Figure 2. A
1268
B
J Ultrasound Med 2014; 33:1265–1272
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1269
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
Figure 4. Normal hair follicles in the occipital region. A and B (hair follicles outlined) show several oblique hypoechoic bands located in the dermis and contacting the upper hypodermis. Abbreviations are as in Figure 2. A
B
Figure 5. Hair follicles in androgenetic alopecia (occipital region). A and B (hair follicles outlined) are gray scale sonograms following the axis of the hair follicles that show slight miniaturization of the hair follicles, with spaces without detectable hair follicles (asterisks). Abbreviations are as in Figure 2. A
B
Figure 6. Sonographic patterns of the hair shafts of the scalp in vitro. A, Trilaminar. B, Bilaminar. A B
J Ultrasound Med 2014; 33:1265–1272
1269
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1270
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
this condition. Thus, the abnormalities in the hair follicles of androgenetic alopecia are consistent with the purported pathophysiologic mechanisms of the condition, implying miniaturization (atrophy or lower depth) of the hair follicle with the predominant involvement of the frontal region. Hair follicle shrinkage with a progressive decrease in the proportion during the mature (anagen) stage was previously linked to a decreased proliferation rate of follicular keratinocytes.14 Whether the primary cause is vascular dysfunction or local deficits in follicular trophic factors is currently unknown. Histologic studies have reported the miniaturization of the hair follicles in androgenetic alopecia15,16; however, indications for scalp biopsies in androgenetic alopecia are carefully considered due to their invasiveness and limited transregional information, which may not disclose the extent of the involvement. Moreover, biopsies usually can include only 3- to 4-mm samples of scalp tissue, and they are rarely performed in patients with androgenetic alopecia.6 Hence, histologic examinations have shown that there is marked variability in the hair follicle diameter and absence of substantial deep or peribulbar inflammation. Accordingly, transverse-axis histologic counting of the hair follicles appears normal in the superficial dermis; however, this counting is decreased when performed in the lower dermis due to shrinkage of the hair follicles.16
The noninvasive sonographic observation of the morphology of the hair follicles that persist in the dermis of patients with androgenetic alopecia may be, by itself, important and could represent a prognostic factor for the treatment of this condition. Thus, the sonographic detection of hair follicles in patients with androgenetic alopecia could signal potential follicular growth and amenability to treatment, while avoiding the potential discomfort and scarring caused by serial scalp biopsies. Furthermore, sonography can locate and measure the length of the hair follicles in the upper and lower dermis, which perhaps may also support management. Since the follicular transverse diameter (width) is increased in inflammatory processes such as hidradenitis suppurativa,3,4,17 the observed lack of differences in that parameter between patients with androgenetic alopecia and controls is consistent with previous histologic reports and potentially might predict unresponsiveness to anti-inflammatory therapies. Also of interest is the maintained dermal sonographic thickness in androgenetic alopecia, which helps localize its abnormalities to the hair follicles and not the surrounding dermis. The sonographic trilaminar or bilaminar appearance of the hair shaft seems to be provided by the keratin, an arrangement that resembles the sonographic bilaminar structure of the nail plate, another keratinized organ.3,4,18
Figure 7. Electron micrographs of hair shafts (original magnification ×2000). A, Normal. B, Androgenetic alopecia. Notice the loss of the sharpness and irregularities in the scales of the cuticle of the alopecia hair shaft. A
1270
B
J Ultrasound Med 2014; 33:1265–1272
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1271
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
A novel sonographic finding is represented by the observations of both trilaminar and bilaminar appearances in normal hair shafts, although the trilaminar pattern predominated, especially in the frontal region. This sonographic heterogeneity is consistent with previous histologic reports of mixtures of mature and vellus (ie, fine, soft, and unmedullated) hair shafts in the normal scalp in a proportion of 7:1.15,16,19 Since vellus hair tracts do not show the typical hyperechoic medulla, they may have decreased resistance to torsional stress, resulting in weaker hair that is more susceptible to fragmentation. Interestingly, the abnormal hair tract distribution (bilaminar versus trilaminar) in androgenetic alopecia differs from previous reports of the absence of abnormalities in the hair tract on confocal microscopy,8 a discrepancy that could be related to the higher tissue penetration of ultrasound. Thus, the higher proportion of a weaker type of hair (vellus-bilaminar) could help explain the hair loss in androgenetic alopecia and may, perhaps, be a predictor of future hair loss in unaffected controls. The observations that there were no significant differences when comparing the thickness of the hair shafts of the frontal regions in controls versus androgenetic alopecia cases and that alopecia cases showed a significantly lower hair shaft thickness in the occipital region when compared to controls should be taken with caution, since the submillimeter range of measurements and the
relatively low number of cases could have affected these statistical values. However, we have provided a description of the approximate range of the sonographic thickness values of the hair shafts in controls and androgenetic alopecia cases, which has not been previously described in the literature. To the best of our knowledge the morphologic abnormalities of androgenetic alopecia hair have not been previously reported with electron microscopy. Our findings suggest that the cuticle of the hair is substantially affected, resulting in structural disorganization and further weakness of the keratinous complex. Since this work was an imaging study, we did not investigate the biochemical integrity of hair components, particularly the all-important coiled-coil binding between the α-helices and keratin molecules. We do not know whether the morphologic changes correspond to multiple residue changes or to a single abnormal interaction. Moreover, a controlled comparison of electron microscopic studies comparing shafts of phenotypic fine hair (with no androgenetic alopecia) must be performed to elucidate whether these features are unique to androgenetic alopecia and are also not present in fine normal hair as well. A limitation of our study was the relatively small number of patients. This factor might motivate further investigation in this field. Another limitation was the lack of correlation of the size of the hair shafts between imaging
Figure 8. Electron micrographs of hair shafts in androgenetic alopecia (original magnification ×2000). A and B show a melted candle appearance of the cuticle, with prominent disappearance of the scales in B. A
J Ultrasound Med 2014; 33:1265–1272
B
1271
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1272
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
modalities (sonography and electron microscopy) and histologic analysis. The latter was due to the difficulty of achieving an exactly similar place of measurement in very small structures (hair shafts) using a real-time imaging technique (sonography) and modalities that require strong fixation and dehydration processes and show highly zoomed magnifications (electron microscopy and histologic analysis). In addition, histologic analysis was not considered in our study due to the possibility of scarring sequelae, since the correlation required would imply large sample sizes and inclusion of the hypodermis to match the sonographic views. Nevertheless, taken together, the abnormalities in the cuticle (on electron microscopy) and hair shaft medulla (on sonography) and the sonographic distribution pattern suggestive of increased follicular shrinkage give clear signals of the full involvement of all hair parts in androgenetic alopecia. These observations fit with previous reports of DNA damage, abnormalities in cell proliferation, and increased apoptosis in androgenetic alopecia.9 Furthermore, the findings show that sonography, a widely available imaging technique, can actually depict a morphologic pattern in androgenetic alopecia, which may be amenable to its potential use in the management of this condition in daily practice. Last, noninvasive imaging studies of androgenetic alopecia seem to answer a need and could support both management and research in a growing number of pharmaceutical and cosmetic treatments of this condition. Furthermore, the sonographic depiction of the presence of remaining hair follicles in the dermis in androgenetic alopecia may suggest the amenability of these cases to medical treatment. In contrast, the lack of hair follicles on sonography in androgenetic alopecia may perhaps support the early provision of more aggressive forms of treatment such as hair transplants. An informed and anatomically based decision on the management of androgenetic alopecia may also help patients who strongly demand successful results, providing them with a better way to understand medical decisions. In conclusion, sonography and electron microscopy reveal abnormalities in the morphology of the hair in patients with androgenetic alopecia. These data, collected noninvasively, could potentially support the diagnosis and management of this very common condition.
1272
References 1. 2.
3. 4.
5. 6. 7. 8. 9.
10.
11. 12. 13. 14.
15. 16. 17. 18. 19.
Paus R, Cotsarelis G. The biology of hair follicles. N Engl J Med 1999; 341:491–497. Gupta MA, Gupta AK. Depression and suicidal ideation in dermatology patients with acne, alopecia areata, atopic dermatitis and psoriasis. Br J Dermatol 1998; 139:846–850. Wortsman X, Wortsman J, Matsuoka L, et al. Sonography in pathologies of scalp and hair. Br J Radiol 2012; 85:647–655. Wortsman X, Wortsman J. Sonography of the scalp and hair. In: Wortsman X, Jemec GBE (eds). Dermatologic Ultrasound With Clinical and Histologic Correlations. 1st ed. New York, NY: Springer; 2013:477– 503. Akoglu G, Emre S, Metin A, et al. Pili annulati with fragility: electron microscopic findings of a case. Int J Trichology 2012; 4:89–92. Hillmann K, Blume-Peytavi U. Diagnosis of hair disorders. Semin Cutan Med Surg 2009; 28:33–38. Rathnayake D, Sinclair R. Male androgenetic alopecia. Expert Opin Pharmacother 2010; 11:1295–1304. Wortsman X, Jemec G. Common inflammatory diseases of the skin: from the skin to the screen. Adv Psoriasis Inflammatory Skin Dis 2010; 2:9–15. Fujimoto JG, Pitris C, Boppart SA, Brezinski ME. Optical coherence tomography: an emerging technology for biomedical imaging and optical biopsy. Neoplasia 2000; 2:9–25. Blume-Peytavi U, Blumeyer A, Tosti A, et al; European Consensus Group. S1 guideline for diagnostic evaluation in androgenetic alopecia in men, women and adolescents. Br J Dermatol 2011; 164:5–15. Dhurat R, Saraogi P. Hair evaluation methods: merits and demerits. Int J Trichology 2009; 1:108–119. Hamilton JB. Patterned loss of hair in man: types and incidence. Ann NY Acad Sci 1951; 53:708–728. Norwood OT. Male pattern baldness: classification and incidence. South Med J 1975; 68:1359–1365. El-Domyati M, Attia S, Saleh F, Bassyouni MI, El-Fakahany H, AbdelWahab H. Proliferation, DNA repair and apoptosis in androgenetic alopecia. J Eur Acad Dermatol Venereol 2009; 23:7–12. Stefanato CM. Histopathology of alopecia: a clinicopathological approach to diagnosis. Histopathology 2010; 56:24–38. Childs JM, Sperling LC. Histopathology of scarring and nonscarring hair loss. Dermatol Clin 2013; 31:43–56. Wortsman X. Common applications of dermatologic sonography. J Ultrasound Med 2012; 31:97–111. Wortsman X, Jemec GB. Ultrasound imaging of nails. Dermatol Clin 2006; 24:323–328. Restrepo R. Microscopic anatomy of the hair follicle. Rev Assoc Colombian Dermatol 2010; 18:123–138.
J Ultrasound Med 2014; 33:1265–1272
ORIGINAL RESEARCH
Hair Morphology in Androgenetic Alopecia Sonographic and Electron Microscopic Studies Ximena Wortsman, MD, Robinson Guerrero, MD, Jacobo Wortsman, MD
Video online at www.jultrasoundmed.org
Objectives—To assess hair morphology in androgenetic alopecia on sonography and electron microscopy. Methods—A prospective study was performed in 33 patients with androgenetic alopecia and 10 unaffected control participants. In vivo sonography of the hair follicles of the scalp and in vitro sonography and electron microscopy of the hair shafts were performed according to a standardized protocol that included analysis of the right frontal and occipital regions. The upper frequency limit of the ultrasound probes ranged between 15 and 18 MHz. Results—Scalp hair follicles and hair shafts were recognizable on sonography in all cases. Hair follicles in alopecia cases had significantly lower depths (P < .05). The hair shafts in alopecia also had a different distribution of their laminar pattern on in vitro sonography, with a greater presence of mixed (trilaminar and bilaminar) and solely bilaminar tracts in comparison with the controls (mostly trilaminar). On electron microscopy, the alopecia hair tracts showed irregularities and commonly a “melted candle” appearance of the cuticle.
Received October 15, 2013, from the Departments of Radiology and Dermatology, Institute for Diagnostic Imaging and Research of the Skin and Soft Tissues, Clinica Servet, Faculty of Medicine, University of Chile, Santiago, Chile (X.W.); Department of Dermatology, Alopecia Clinic, Fundacion Medica San Cristobal, Santiago, Chile (R.G.); and Department of Medicine, Southern Illinois University School of Medicine, Springfield, Illinois USA (J.W.). Revision requested October 30, 2013. Revised manuscript accepted for publication November 18, 2013. We thank Nancy Olea, BSc (Electron Microscopy Unit, Faculty of Medicine, University of Chile), for support with the electron microscopic samples. Address correspondence to Ximena Wortsman, MD, Departments of Radiology and Dermatology, Institute for Diagnostic Imaging and Research of the Skin and Soft Tissues, Clinica Servet, Faculty of Medicine, University of Chile, Lo Fontecilla 201, of 734, Las Condes, Santiago, Chile. E-mail: [email protected]
doi:10.7863/ultra.33.7.1265
Conclusions—Sonography and electron microscopy uncover distinct abnormalities in the morphology of hair in androgenetic alopecia, which may potentially support the diagnosis and management of this common condition. Key Words—androgenetic alopecia; androgenetic alopecia sonography; androgenetic alopecia ultrasound; androgenic alopecia; hair sonography; hair ultrasound; scalp sonography; scalp ultrasound; superficial structures
B
esides the esthetic connotations, hair serves important roles in cutaneous functions such as dispersion of sweat gland products and regulation of body temperature.1 Hair loss, or alopecia, is therefore clinically important as it may be also associated with psychologic disturbances from disruptions of the selfimage of affected patients.2 Since hair loss may imply hair fragility or structural abnormalities, a detailed morphologic evaluation should be an important step for studying its pathologic characteristics and supporting management. In this regard, recent developments in the field of sonography have allowed details of the hair anatomy to be uncovered.3 On sonography, normal scalp hair follicles appear as oblique hypoechoic bands in the dermis, extending to variable depths according to the hair cycle phase; thus, active hair follicles (anagen phase) are located both in the superficial and deep dermis and can even reach the upper hypodermis, whereas inactive hair
©2014 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2014; 33:1265–1272 | 0278-4297 | www.aium.org
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1266
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
follicles (telogen phase) are restricted to the upper dermis. The external hair segments of the scalp, also called hair shafts or tracts, are reported as trilaminar hyperechoic sonographic structures with an outside cuticle-cortex complex and an inner medulla.3,4 The hair cuticle (outer layer of the hair tract) can be defined at high resolution with electron microscopy, an imaging method that has been used in this manner for the study of pili annulati, a genetic abnormality of the hair shaft.5 The most common type of hair loss is androgenetic alopecia, which develops mostly in genetically predisposed men in the presence of androgens (androgen dependent), although it can also affect women. It is characterized by progressive miniaturization of the hair follicles and generally appears as nonscarring scalp alopecia of the frontotemporal and vertex regions.6 The prevalence of androgenetic alopecia is highest in white men, reaching frequencies of up to 80% in men older than 70 years.6,7 Current methods for the evaluation of hair characteristics may present disadvantages that can limit their use. For example, the widely available hair count has a subjective component and can be tedious and time consuming. Hair pull tests that are based on the concept that easy-toextract hairs correspond to hair in the telogen phase (inactive) may be painful and variable from person to person. The standard trichogram requires the extraction of 60 to 80 hairs with a rubber-armed forceps, to allow a microscopic evaluation, a procedure that could be painful and unpleasant for people losing their hair. Moreover, the trichogram usually misses small hairs from early anagen; whereas the plucking procedure itself is known to modify the local hair cycle. Dermoscopy (gross magnification) provides an amplified image of the skin surface, hair shafts, and follicular ostia, but the underlying hair follicles are inaccessible. The shortcomings of confocal microscopy, an optical imaging method that provides high resolution, almost comparable to histologic analysis, and optical coherence tomography, an imaging method based on the signal and acquisition properties of light, include their limited tissue penetration (up to 0.3 mm for confocal microscopy and up to 2 mm for optical coherence tomography), with restricted or no access to the deep dermis and hypodermis.8,9 The reference standard technique in the evaluation of hair abnormalities is histologic analysis, which is an invasive modality based on the examination of a biopsy-obtained sample of scalp tissue, usually single and 4 mm in diameter, that, because of local variations in hair involvement, is not always adequate as being representative of the global process. In addition to being painful, scalp biopsy is difficult to repeat in the same area for evaluation during and after therapy.10,11
1266
Therefore, noninvasive imaging modalities seem to be needed to overcome some of these limitations and reveal anatomic details. Hence, the aim of this study was to assess the morphologic patterns of the hair using sonography and electron microscopy in patients with androgenetic alopecia.
Materials and Methods We prospectively studied 33 male patients referred by a dermatologist with the diagnosis of androgenetic alopecia (mean age, 32 years; range, 17–41 years) and 10 unaffected male control participants (mean age, 29 years; range, 19–49 years). The inclusion criterion was patients with androgenetic alopecia presenting disease severity of type II or greater (Hamilton-Norwood classification; Figure 1).12,13 Exclusion criteria were concomitant antialopecia treatment (oral or topical), coexistence of other cutaneous diseases (local or systemic), and a history of hair implants and applied products (hot irons, dyes, or chemical products). The study protocol followed the Helsinki principles of medical ethics and was approved by the Institutional Review Board; all participants signed a written informed consent form. Figure 1. Hamilton-Norwood classification of alopecia.
J Ultrasound Med 2014; 33:1265–1272
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1267
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
The ultrasound machines were LOGIQ E9 XD Clear (GE Healthcare, Milwaukee, WI) and HDI 5000 (Philips Healthcare, Bothell, WA) systems matched to compact linear probes tunable to frequencies of 8 to 18 and 7 to 15 MHz, respectively. Sonography was performed by scanning sequentially the right frontal and occipital regions of the scalp after careful manual separation of the hair shafts and application of a copious amount of gel to the surface of the skin. Examination sites were as follows: for the right frontal region, the area located 10 cm above the center of the right eyebrow; and for the right occipital region, the area located 4 cm behind the posteroinferior border of the right ear pinna. Sonographic parameters (millimeters) included dermal thickness, scalp hair follicle depth and transverse diameter (width), and hair shaft thickness. The sonograms followed the axes of the hair follicles and shafts for assessing their morphology. The measurements were performed at the maximum distance point (depth, width, or thickness following the axis of the hair). Immediately before the sonographic examination a total of 20 hair shafts approximately 4 cm long were trimmed from the same frontal and occipital regions (10 tracts each) for sonographic and electron microscopic studies. The hair shafts were trimmed at the level of the scalp surface. Thus, for in vitro sonographic studies, 5 shafts from each region were scanned while immersed in 80 mL of saline in a 100-mL disposable plastic cup. For the electron microscopic studies, hair shaft fragments were placed on aluminum mounts and covered with carbon adhesive tabs. The samples were covered with atomic particles of palladium/gold in a layer of 20 nm in thickness using a Polaron E-5000 sputter coater (Bio-Rad, Cambridge, England) and were observed on a DMS 940 scanning electron microscope (Zeiss, Oberkochen, Germany). Lateral and cross-sectional views of the hair specimens were then exposed to the electron beam. After surface analysis for assessing
the hair cuticle morphology, digital photomicrographs were taken at magnifications of ×500, ×1000, and ×2000. Data were analyzed with CaEst version 1.2 statistical software (Softonic, Barcelona, Spain) for assessing significance (at P < .05).
Results On in vivo sonography, hair follicles were recognizable, although in patients with androgenetic alopecia, both frontal and occipital follicles were shorter, extending to significantly lower depths than those of controls (frontal region, P = .0346; occipital region; P = .0047). In androgenetic alopecia, in addition to the shrinkage of the hair follicles, spaces without detectable hair follicles were observed in the dermis. However, neither the dermal thickness as a whole, nor the hair follicle transverse diameter was significantly different from that of controls (Table 1 and Figures 2–5). In vitro sonographic studies included a total of 430 hair shafts: 330 from the patients with androgenetic alopecia and 100 from the controls. Patients with alopecia and controls showed differences in the morphologic distribution of the laminar hair shaft pattern. A trilaminar hyperechoic pattern was predominantly seen in the controls. Mixed trilaminarbilaminar and solely bilaminar hyperechoic patterns were more frequently seen in the androgenetic alopecia cases. With regard to the specific zones, the hair shafts from controls consistently showed a trilaminar hyperechoic appearance in the frontal region; however, in the occipital region, the pattern was composed of two variants, where 80% (n = 8) of the controls showed the trilaminar hyperechoic appearance but 20% (n = 2) instead had a bilaminar hyperechoic appearance. The mean thicknesses of the frontal and occipital hair shafts in controls were 0.5 and 0.6 ± 0.1 (SD) mm, respectively.
Table 1. Sonographic Parameters of Scalp Hair From Patients With Androgenetic Alopecia and Controls Alopecia Parameter
Mean
SD
Age, y Frontal dermis thickness, mm Frontal hair follicle depth, mm Frontal hair follicle transverse diameter, mm Occipital dermis thickness, mm Occipital hair follicle depth, mm Occipital hair follicle transverse diameter, mm Frontal hair shaft thickness, mm Occipital hair shaft thickness, mm
32.0 1.7 1.4 0.6 1.9 2.1 0.6 0.5 0.5
5.8 0.4 0.4 0.1 0.3 0.5 0.3 0.1 0.1
Control Mean 29.0 1.5 1.7 0.6 2.1 2.6 0.6 0.5 0.6
SD
P
10.9 0.2 0.3 0 0.4 0.3 0.1 0.1 0.1
.2573 .1373 .0346a >.99 .0954 .0047a >.99 >.99 .0084a
aSignificant (P < .05).
J Ultrasound Med 2014; 33:1265–1272
1267
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1268
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
In patients with androgenetic alopecia, sonography of the hair shafts in the frontal region showed a trilaminar hyperechoic pattern in 45% (n = 15), a mixed pattern (trilaminar and bilaminar) in 45% (n = 15), and solely bilaminar hyperechoic hair shafts in 10% (n = 3). In the occipital region, 70% (n = 23) of patients with androgenetic alopecia showed a trilaminar hyperechoic appearance, 24% (n = 8) showed a mixed pattern (trilaminar and bilaminar), and 6% (n = 2) showed only bilaminar hyperechoic hair shafts (Figure 6 and Video 1). The mean sonographic transverse diameters of the frontal and occipital hair shafts in the alopecia cases were similar and corresponded to 0.5 ± 0.1 for each region. No significant differences were found when comparing the thickness of the hair shafts of the frontal regions in controls versus androgenetic alopecia cases (P > .99).
However, alopecia cases showed a significantly lower thickness in the occipital region compared to controls (P = .0084). Electron microscopy of hair shafts from the frontal and occipital regions of the controls showed a smooth, sharp, well-defined surface of the cuticle scales. In the patients with androgenetic alopecia, the surface of the cuticle on the hair shafts was irregular, with a poorly defined scale surface and contour, commonly giving a “melted candle” appearance (Figures 7 and 8).
Discussion To our knowledge, this study is the first to describe the sonographic appearance of hair follicles and hair shafts in patients with androgenetic alopecia: the target structures of
Figure 2. Normal hair follicles in the frontal region. A and B (hair follicles outlined) are grayscale sonograms that follow the axis of the hair follicles and show several oblique hypoechoic bands in the dermis; b indicates bony margin of the skull; d, dermis; h, hypodermis; and m, epicranius muscle. A
B
Figure 3. Hair follicles in a patient presenting with androgenetic alopecia (frontal region). A and B (hair follicles outlined) are grayscale sonograms that follow the axis of the hair follicles and show sparsity and shrinkage of the hair follicles in the dermis. Notice the spaces (asterisks) without detectable hair follicles. Abbreviations are as in Figure 2. A
1268
B
J Ultrasound Med 2014; 33:1265–1272
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1269
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
Figure 4. Normal hair follicles in the occipital region. A and B (hair follicles outlined) show several oblique hypoechoic bands located in the dermis and contacting the upper hypodermis. Abbreviations are as in Figure 2. A
B
Figure 5. Hair follicles in androgenetic alopecia (occipital region). A and B (hair follicles outlined) are gray scale sonograms following the axis of the hair follicles that show slight miniaturization of the hair follicles, with spaces without detectable hair follicles (asterisks). Abbreviations are as in Figure 2. A
B
Figure 6. Sonographic patterns of the hair shafts of the scalp in vitro. A, Trilaminar. B, Bilaminar. A B
J Ultrasound Med 2014; 33:1265–1272
1269
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1270
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
this condition. Thus, the abnormalities in the hair follicles of androgenetic alopecia are consistent with the purported pathophysiologic mechanisms of the condition, implying miniaturization (atrophy or lower depth) of the hair follicle with the predominant involvement of the frontal region. Hair follicle shrinkage with a progressive decrease in the proportion during the mature (anagen) stage was previously linked to a decreased proliferation rate of follicular keratinocytes.14 Whether the primary cause is vascular dysfunction or local deficits in follicular trophic factors is currently unknown. Histologic studies have reported the miniaturization of the hair follicles in androgenetic alopecia15,16; however, indications for scalp biopsies in androgenetic alopecia are carefully considered due to their invasiveness and limited transregional information, which may not disclose the extent of the involvement. Moreover, biopsies usually can include only 3- to 4-mm samples of scalp tissue, and they are rarely performed in patients with androgenetic alopecia.6 Hence, histologic examinations have shown that there is marked variability in the hair follicle diameter and absence of substantial deep or peribulbar inflammation. Accordingly, transverse-axis histologic counting of the hair follicles appears normal in the superficial dermis; however, this counting is decreased when performed in the lower dermis due to shrinkage of the hair follicles.16
The noninvasive sonographic observation of the morphology of the hair follicles that persist in the dermis of patients with androgenetic alopecia may be, by itself, important and could represent a prognostic factor for the treatment of this condition. Thus, the sonographic detection of hair follicles in patients with androgenetic alopecia could signal potential follicular growth and amenability to treatment, while avoiding the potential discomfort and scarring caused by serial scalp biopsies. Furthermore, sonography can locate and measure the length of the hair follicles in the upper and lower dermis, which perhaps may also support management. Since the follicular transverse diameter (width) is increased in inflammatory processes such as hidradenitis suppurativa,3,4,17 the observed lack of differences in that parameter between patients with androgenetic alopecia and controls is consistent with previous histologic reports and potentially might predict unresponsiveness to anti-inflammatory therapies. Also of interest is the maintained dermal sonographic thickness in androgenetic alopecia, which helps localize its abnormalities to the hair follicles and not the surrounding dermis. The sonographic trilaminar or bilaminar appearance of the hair shaft seems to be provided by the keratin, an arrangement that resembles the sonographic bilaminar structure of the nail plate, another keratinized organ.3,4,18
Figure 7. Electron micrographs of hair shafts (original magnification ×2000). A, Normal. B, Androgenetic alopecia. Notice the loss of the sharpness and irregularities in the scales of the cuticle of the alopecia hair shaft. A
1270
B
J Ultrasound Med 2014; 33:1265–1272
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1271
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
A novel sonographic finding is represented by the observations of both trilaminar and bilaminar appearances in normal hair shafts, although the trilaminar pattern predominated, especially in the frontal region. This sonographic heterogeneity is consistent with previous histologic reports of mixtures of mature and vellus (ie, fine, soft, and unmedullated) hair shafts in the normal scalp in a proportion of 7:1.15,16,19 Since vellus hair tracts do not show the typical hyperechoic medulla, they may have decreased resistance to torsional stress, resulting in weaker hair that is more susceptible to fragmentation. Interestingly, the abnormal hair tract distribution (bilaminar versus trilaminar) in androgenetic alopecia differs from previous reports of the absence of abnormalities in the hair tract on confocal microscopy,8 a discrepancy that could be related to the higher tissue penetration of ultrasound. Thus, the higher proportion of a weaker type of hair (vellus-bilaminar) could help explain the hair loss in androgenetic alopecia and may, perhaps, be a predictor of future hair loss in unaffected controls. The observations that there were no significant differences when comparing the thickness of the hair shafts of the frontal regions in controls versus androgenetic alopecia cases and that alopecia cases showed a significantly lower hair shaft thickness in the occipital region when compared to controls should be taken with caution, since the submillimeter range of measurements and the
relatively low number of cases could have affected these statistical values. However, we have provided a description of the approximate range of the sonographic thickness values of the hair shafts in controls and androgenetic alopecia cases, which has not been previously described in the literature. To the best of our knowledge the morphologic abnormalities of androgenetic alopecia hair have not been previously reported with electron microscopy. Our findings suggest that the cuticle of the hair is substantially affected, resulting in structural disorganization and further weakness of the keratinous complex. Since this work was an imaging study, we did not investigate the biochemical integrity of hair components, particularly the all-important coiled-coil binding between the α-helices and keratin molecules. We do not know whether the morphologic changes correspond to multiple residue changes or to a single abnormal interaction. Moreover, a controlled comparison of electron microscopic studies comparing shafts of phenotypic fine hair (with no androgenetic alopecia) must be performed to elucidate whether these features are unique to androgenetic alopecia and are also not present in fine normal hair as well. A limitation of our study was the relatively small number of patients. This factor might motivate further investigation in this field. Another limitation was the lack of correlation of the size of the hair shafts between imaging
Figure 8. Electron micrographs of hair shafts in androgenetic alopecia (original magnification ×2000). A and B show a melted candle appearance of the cuticle, with prominent disappearance of the scales in B. A
J Ultrasound Med 2014; 33:1265–1272
B
1271
3307jum_online_Layout 1 6/20/14 9:00 AM Page 1272
Wortsman et al—Sonography and Electron Microscopy of Androgenetic Alopecia
modalities (sonography and electron microscopy) and histologic analysis. The latter was due to the difficulty of achieving an exactly similar place of measurement in very small structures (hair shafts) using a real-time imaging technique (sonography) and modalities that require strong fixation and dehydration processes and show highly zoomed magnifications (electron microscopy and histologic analysis). In addition, histologic analysis was not considered in our study due to the possibility of scarring sequelae, since the correlation required would imply large sample sizes and inclusion of the hypodermis to match the sonographic views. Nevertheless, taken together, the abnormalities in the cuticle (on electron microscopy) and hair shaft medulla (on sonography) and the sonographic distribution pattern suggestive of increased follicular shrinkage give clear signals of the full involvement of all hair parts in androgenetic alopecia. These observations fit with previous reports of DNA damage, abnormalities in cell proliferation, and increased apoptosis in androgenetic alopecia.9 Furthermore, the findings show that sonography, a widely available imaging technique, can actually depict a morphologic pattern in androgenetic alopecia, which may be amenable to its potential use in the management of this condition in daily practice. Last, noninvasive imaging studies of androgenetic alopecia seem to answer a need and could support both management and research in a growing number of pharmaceutical and cosmetic treatments of this condition. Furthermore, the sonographic depiction of the presence of remaining hair follicles in the dermis in androgenetic alopecia may suggest the amenability of these cases to medical treatment. In contrast, the lack of hair follicles on sonography in androgenetic alopecia may perhaps support the early provision of more aggressive forms of treatment such as hair transplants. An informed and anatomically based decision on the management of androgenetic alopecia may also help patients who strongly demand successful results, providing them with a better way to understand medical decisions. In conclusion, sonography and electron microscopy reveal abnormalities in the morphology of the hair in patients with androgenetic alopecia. These data, collected noninvasively, could potentially support the diagnosis and management of this very common condition.
1272
References 1. 2.
3. 4.
5. 6. 7. 8. 9.
10.
11. 12. 13. 14.
15. 16. 17. 18. 19.
Paus R, Cotsarelis G. The biology of hair follicles. N Engl J Med 1999; 341:491–497. Gupta MA, Gupta AK. Depression and suicidal ideation in dermatology patients with acne, alopecia areata, atopic dermatitis and psoriasis. Br J Dermatol 1998; 139:846–850. Wortsman X, Wortsman J, Matsuoka L, et al. Sonography in pathologies of scalp and hair. Br J Radiol 2012; 85:647–655. Wortsman X, Wortsman J. Sonography of the scalp and hair. In: Wortsman X, Jemec GBE (eds). Dermatologic Ultrasound With Clinical and Histologic Correlations. 1st ed. New York, NY: Springer; 2013:477– 503. Akoglu G, Emre S, Metin A, et al. Pili annulati with fragility: electron microscopic findings of a case. Int J Trichology 2012; 4:89–92. Hillmann K, Blume-Peytavi U. Diagnosis of hair disorders. Semin Cutan Med Surg 2009; 28:33–38. Rathnayake D, Sinclair R. Male androgenetic alopecia. Expert Opin Pharmacother 2010; 11:1295–1304. Wortsman X, Jemec G. Common inflammatory diseases of the skin: from the skin to the screen. Adv Psoriasis Inflammatory Skin Dis 2010; 2:9–15. Fujimoto JG, Pitris C, Boppart SA, Brezinski ME. Optical coherence tomography: an emerging technology for biomedical imaging and optical biopsy. Neoplasia 2000; 2:9–25. Blume-Peytavi U, Blumeyer A, Tosti A, et al; European Consensus Group. S1 guideline for diagnostic evaluation in androgenetic alopecia in men, women and adolescents. Br J Dermatol 2011; 164:5–15. Dhurat R, Saraogi P. Hair evaluation methods: merits and demerits. Int J Trichology 2009; 1:108–119. Hamilton JB. Patterned loss of hair in man: types and incidence. Ann NY Acad Sci 1951; 53:708–728. Norwood OT. Male pattern baldness: classification and incidence. South Med J 1975; 68:1359–1365. El-Domyati M, Attia S, Saleh F, Bassyouni MI, El-Fakahany H, AbdelWahab H. Proliferation, DNA repair and apoptosis in androgenetic alopecia. J Eur Acad Dermatol Venereol 2009; 23:7–12. Stefanato CM. Histopathology of alopecia: a clinicopathological approach to diagnosis. Histopathology 2010; 56:24–38. Childs JM, Sperling LC. Histopathology of scarring and nonscarring hair loss. Dermatol Clin 2013; 31:43–56. Wortsman X. Common applications of dermatologic sonography. J Ultrasound Med 2012; 31:97–111. Wortsman X, Jemec GB. Ultrasound imaging of nails. Dermatol Clin 2006; 24:323–328. Restrepo R. Microscopic anatomy of the hair follicle. Rev Assoc Colombian Dermatol 2010; 18:123–138.
J Ultrasound Med 2014; 33:1265–1272
