BJD. Summary. British Journal of Dermatology DERMATOPATHOLOGY

Transcription

1 DERMATOPATHOLOGY BJD British Journal of Dermatology Actinic keratosis in the en-face and slice imaging mode of high-definition optical coherence tomography and comparison with histology T. Maier, 1 M. Braun-Falco, 1 R.P. Laubender, 2 T. Ruzicka 1 and C. Berking 1 1 Department of Dermatology and Allergology, Ludwig-Maximilian University of Munich, Frauenlobstr. 9 11, D Munich, Germany 2 Department of Biometry and Epidemiology, Institute of Medical Informatics, Ludwig-Maximilian University of Munich, Campus Grosshadern, Munich, Germany Summary Correspondence Tanja Maier. tanja.maier@med.uni-muenchen.de Accepted for publication 31 July 2012 Funding sources This work was supported by the Curd-Bohnewand- Fonds of the University of Munich (to T.M.), by the Matthias Lackas Foundation and the Dr Helmut Legerlotz Foundation (to C.B.). Conflicts of interest The HD-OCT Skintell Ò device used in this study was provided by Agfa HealthCare GmbH. Dr Maier Served as lectures for Agfa Healthcare GmbH. Dedication This work is dedicated to Prof. Hans-Christian Korting who always gave essential support and has recently passed away. DOI /j x Background Optical coherence tomography (OCT) allows real-time, in vivo examination of nonmelanoma skin cancer. An innovative high-definition (HD)-OCT with a horizontal (en-face) and vertical (slice) imaging mode offers additional information in the diagnosis of actinic keratosis (AK) and may potentially replace invasive diagnostic biopsies. Objectives To define the characteristic morphological features of AK by using HD- OCT in the two imaging modes compared with histopathology as gold standard. Methods In total, 20 AKs were examined by HD-OCT in the en-face and slice imaging modes and characteristic features were described and evaluated in comparison with the histopathological findings. Furthermore, the HD-OCT images of a subgroup of AKs were compared with those of the clinically normal adjacent skin. Results The preoperative in vivo diagnostics showed the following features in the en-face imaging mode of HD-OCT: disruption of stratum corneum, architectural disarray, cellular nuclear polymorphism in the stratum granulosum stratum spinosum, and bright irregular bundles in the superficial dermis. In the vertical slice imaging mode the following characteristics were found: irregular entrance signal, destruction of layering, white streaks and dots, and grey areas. In contrast, the clinically healthy adjacent skin showed mainly a regular epidermal honeycomb pattern in the en-face mode and distinct layering of the skin in the slice mode. Conclusions HD-OCT with both the en-face and slice imaging modes offers additional information in the diagnosis of AK compared with conventional OCT and might enhance the possibility of the noninvasive diagnosis of AK prior to treatment procedures and possibly in the monitoring of noninvasive treatment strategies. The incidence of nonmelanoma skin cancer including actinic keratosis (AK) is rising worldwide. As AKs are precursor lesions of potentially metastasizing squamous cell carcinoma (SCC) the increasing prevalence of AK is alarming. It is assumed that about 10% (6 16%) of patients with AK will develop over decades an SCC. 1,2 Epidemiological data show a high prevalence of AK in fair skin types and in countries with high exposure to ultraviolet radiation. The highest incidence of AK has been documented in Australia with 55% in men and 37% in women. 3 According to studies in the U.K. the prevalence of AK was 15% in men and 6% in women, and in the population over the age of 70 years it increased to 34% in males and 18% in females. 4 In the U.S.A. the incidence rate of AK is between 11% and 26%. 2 Additionally, the growing subgroup of immunosuppressed patients worldwide is known to develop a high rate of frequently aggressive AK, 5 who will need regular monitoring and treatment. AKs are mainly diagnosed clinically although histological confirmation is recommended to exclude an invasive SCC. As noninvasive treatments of AK such as topical immune response modifiers, topical chemotherapy, chemical peeling or photodynamic therapy are becoming more and more popular, the invasive biopsy for the diagnosis and treatment follow-up is often unfavourable. To date there is no commonly accepted classification of AK which accounts for the grade of invasiveness of the lesion. According to the Guidelines for Dermatology in Europe AK can be divided clinically into different subtypes (keratotic, atrophic, cornu cutaneum, verrucous, pigmented, lichenoid) 1

2 2 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. but without a distinct parameter for the prediction of malignant transformation. 6 In 2006 a review article tried to identify clinical features for the recognition of aggressive AK. 7 As risk factors for malignancy the following parameters were proposed: induration inflammation, diameter > 1 cm, rapid enlargement, bleeding, erythema and ulceration. In certain cases, the use of dermoscopy assists in the identification of pigmented AK. The strawberry sign has been described as a characteristic pattern of AK, meaning a red pseudonetwork of vessels around prominent hair follicles with a white halo in dermoscopy. 8,9 As AKs are the most common premalignant lesions worldwide a reliable diagnostic evaluation would be preferable as is established in pigmented lesions (ABCD rule). In this context the new imaging techniques such as reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) offer a noninvasive option for diagnostics and therapeutic monitoring. RCM has already been studied for the evaluation of melanoma 10,11 and nonmelanoma skin cancer such as AK, 8 and also OCT has been performed in different types of epithelial skin cancer including AK Nevertheless, the diagnosis of AK with RCM is time-consuming especially when screening multiple lesions. With respect to conventional OCT the visualization of AK is still insufficient because of the relatively low lateral and axial resolution of the available OCT devices. In this study, an innovative high-definition (HD)-OCT device with an improved resolution of about 3 lm in all dimensions and the possibility of vertical (slice-mode) and horizontal (en-face mode) imaging similar to the horizontal RCM scanning mode was used to evaluate noninvasively the clinical diagnosis of AK in correlation with the histological diagnosis. Subjects and methods Participants We investigated 20 lesions clinically suspicious for AK prior to the surgical intervention. The study was performed at the Department of Dermatology, Ludwig-Maximilian University of Munich after obtaining written informed consent from each patient. The study population consisted of nine men and four women with one or more lesions and an age range of years (Table 1). In a subgroup of five lesions (lesions 10, 16 19) the clinically healthy adjacent skin was additionally imaged by HD-OCT. After obtaining OCT images of the suspected lesions a punch or shave biopsy was performed preferentially from the centre of the lesions followed by the histopathological examination. There were no biopsies taken from the adjacent normal skin. The study was approved by the local ethics committee of the Ludwig-Maximilian University of Munich and followed the principles expressed in the Declaration of Helsinki. High-definition optical coherence tomography A commercially available full-field HD-OCT system (Skintell Ò ; Agfa HealthCare, Mortsel, Belgium) was used for OCT Table 1 Clinical patient data and histopathological characteristics of the 20 actinic keratoses (AKs) included in the high-definition optical coherence tomography study Lesion Patient age (years) sex Site Histopathological type of AK Parakeratosis Destruction of epidermal structure Pleomorphic keratinocytes 1 78 M Forehead Bowenoid M Eyebrow right Atrophic M Cheek right Unclassified M Temple left Unclassified M Cheek left Atrophic M Cheek left Hypertrophic F Cheek right Unclassified F Infraorbital right Initial ) + + ) 9 74 M Forehead Acantholytic F Eyebrow right Atrophic M Head left Atrophic M Head Unclassified F Cheek left Atrophic M Head right Unclassified ) M Occipital right Unclassified ) F Cheek right Atrophic F Temple left Initial ) ) + ) M Nose Hypertrophic M Cheek left Unclassified M Forehead left Unclassified , present; ), absent. Inflammatory infiltrate

3 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. 3 imaging. It works on the principle of a time domain OCT system including dynamic focus tracking which is a synchronized motion of the imaging lens system and the reference optical system. In the Skintell Ò (Agfa HealthCare) system a halogen lamp with a Gaussian filter and an ultrahigh bandwidth centred at 1300 nm (infrared radiation) serves as the light source which allows a high-depth resolution and no lateral cross-talk (speckling). Two different modes are available in this system: real-time b-scan and en-face which allows a fast capture of a three-dimensional tomogram. According to the producer s instructions, Skintell Ò allows a resolution of 3 lm in three dimensions. The penetration depth in skin is about 750 lm and the field of view in the en-face mode is 1Æ8 1Æ5 mm. As coupling medium an optical gel (Skintell Ò optical gel; Agfa HealthCare) is used and the probe is applied directly on to the skin. The clinically suspicious lesions were systematically evaluated by HD-OCT in the slice and en-face mode and images were recorded. The images were evaluated starting from the stratum corneum (SC) through the epidermis and upper dermis as far as it was displayed in the images. The HD-OCT probe was preferentially placed in the centre of the lesion, where the consecutive biopsy was taken later on. The imprint of the HD-OCT probe was visible right after the imaging and was marked with a pen for the consecutive biopsy. In a subgroup of patients adjacent clinically normal skin was additionally imaged by HD-OCT for comparison with the lesional skin. The lesions were examined by an experienced investigator (T.M.) prior to excision. Characteristic features were evaluated in relation to the established histological criteria of AK (Table 1). Histological evaluation Biopsies were taken of all 20 lesions clinically suspicious for AK and tissue sections were prepared for conventional histology. As there was no exact labelling of the HD-OCT position possible within the lesion the evaluation and consecutive biopsy were always done in the centre of each lesion. The histological evaluation was performed by a board-certified dermatopathologist (M.B.-F.). Results En-face mode of high-definition optical coherence tomography in actinic keratosis The evaluation of the documented images of the AKs revealed certain features which presented repeatedly in the examined lesions (see Figs 1 4). For the new en-face (horizontal) imaging mode in HD-OCT, we defined the following four criteria: disruption of SC, architectural disarray, cellular nuclear polymorphism in the stratum granulosum stratum spinosum (SG SS), and bright irregular bundles in the superficial dermis (Table 2). Disruption of SC was observed as irregular ruptures in the otherwise bright SC. Architectural disarray in the epidermis was defined as alterations in the usually regular honeycomb pattern with broadened intercellular demarcations and increased brightness. Cellular nuclear polymorphism was characterized by focal changes in shape and size of the keratinocytes and nuclei (bright polygonal formations with central dark oval structure consistent with the nucleus). In the superficial dermis bright irregular bundles were detectable when an increase of brightness and clusters of the otherwise regular network occurred. In the examined AKs we found disruption of the SC in 18 of 20 (90%) lesions, including three (15%) lesions with only moderate signs of disruption. Architectural disarray and cellular nuclear polymorphism were present in 20 of 20 (100%) examined AKs, while these changes were only moderate in six (30%) of these lesions. Bright irregular bundles were detected in 16 of 20 (80%) AKs with seven (35%) lesions showing only a moderate grade of these changes. Slice mode of high-definition optical coherence tomography in actinic keratosis In the classical vertical OCT imaging mode (slice mode), AKs showed a destruction of the typical layering of the epidermis and dermis as described earlier elsewhere. 17 In reference to the criteria of Jørgensen et al. 17 we chose the following characteristics for AK in the slice mode of HD-OCT: destruction of layering, white streaks and dots, and grey areas (Table 3). Additionally, we defined the feature of an irregular entrance signal meaning alterations in refractivity and thickness of the topmost band appearing in the HD-OCT image (Figs 2, 3). This irregular entrance signal was found in all of the 20 (100%) examined AKs. A destruction of layering was also present in all of the 20 (100%) lesions. White streaks and dots were located in 19 of 20 (95%) AKs. Lesion 16 did not show streaks and dots and was histologically defined as atrophic AK. Grey areas were observed in 18 of 20 AKs (90%). High-definition optical coherence tomography in normal skin The clinically healthy adjacent skin showed the following HD- OCT features: SC without interruption, regular epidermal honeycomb pattern in the SG SS, reticulated meshwork in the upper dermis, oval dark structures with bright border (hair follicles) in the en-face mode, and a regular bright entrance signal of the SC, and distinct layering of the epidermis dermis in the slice mode. In the en-face mode of HD-OCT the clinically normal skin at the level of the SG SS showed a regular pattern of small dark cells with bright cell borders without significant changes in shape and size (Fig. 1). These features were present in all of the five examined samples of clinically healthy skin. The transition of SG SS can be recognized by a slight decrease in size of the keratinocytes. The conversion of the dermoepidermal junction into the superficial dermis is not

4 4 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. (a) (e) (b) (f) (c) (d) (g) (h) Fig 1. (a d) High-definition optical coherence tomography (HD-OCT) of normal skin of the forehead in the slice (upper part) and en-face (lower part) mode. In the slice mode the regular layering of the epidermis and dermis is displayed, disrupted centrally by a hair follicle (HF). (a) Bright stratum corneum (SC) of normal skin is interrupted by dark skin folds and parts of regular stratum granulosum (SG) (arrow) in the en-face mode. (b) SG of normal skin presents as regular round dark structures surrounded by a bright web consistent with keratinocytes (arrow). (c) In the stratum spinosum (SS) the cells appear smaller and form the typical honeycomb pattern (arrow) in the en-face mode. (d) In the en-face mode the transition of stratum basale (SB) dermoepidermal junction (DEJ) to the upper dermis (UD) is depicted as bright bundles with oval-shaped dark structures. (e h) HD-OCT of an atrophic actinic keratosis (lesion 10) displays in the slice mode (upper part) the destruction of the epidermal dermal layering consistent with parakeratosis and the destruction of the epidermal structure. (f, g) The regular web and honeycomb pattern of the SG SS as displayed in the normal adjacent skin (a d) is disrupted and larger dark round structures surrounded by irregular bright borders are found coherent with pleomorphic keratinocytes. (h) The regular composition of the DEJ is disordered.

5 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. 5 Table 2 Evaluation criteria of actinic keratoses in the en-face imaging mode of high-definition optical coherence tomography Lesion Disruption of SC Architectural disarray in SG SS Cellular nuclear polymorphism in SG SS ) + ) + ) 6 ) + ) + ) + 7 ) + ) + ) ) ) ) ) ) + ) + ) + ) 16 + ) + ) + ) + ) ) ) ) 20 + ) + ) + ) + ) Bright irregular bundles in SD SC, stratum corneum; SG SS, stratum granulosum stratum spinosum; SD, superficial dermis; +, present; ), absent; + ), moderately present. Table 3 Evaluation criteria of the actinic keratoses in the slice imaging mode of high-definition optical coherence tomography Lesion Irregular entrance signal Destruction of layering White streaks White dots Grey areas ) ) + ) + ) + ) 6 + ) + ) + ) + ) ) + ) ) + ) ) + ) ) ) + ) + ) ) + ) ) ) ) ) + ) + ) + ) ) +, present; ), absent; + ), moderately present. sharp because of the flattened rete ridges of the facial skin. In other regions of the body with more prominent dermal papillae the papillary dermis was more clearly depicted in HD-OCT (data not shown). The superficial dermis is characterized by a regular network of bright bundles most probably consistent with the reticular collagen bundles. Histological evaluation and comparison with highdefinition optical coherence tomography The histological evaluation (Table 1) revealed eight unclassified, six atrophic, two hypertrophic, two initial, one acantholytic and one bowenoid AK. Typical histological features of AK

6 6 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. (a) (d) (b) (e) (c) Fig 2. (a) High-definition optical coherence tomography (HD-OCT) of actinic keratosis (AK) lesion 9 with irregular corneal layer displayed as inhomogeneous bright entrance signal (arrow) and destruction of the epidermal layering in the slice mode comparable with the histological parakeratosis (arrow, c). (b) In the en-face mode of HD- OCT at the basal layer acantholysis is depicted as dark fields within the brighter layers (white circle, micrometer bar 200 lm) consistent with the histopathological (haematoxylin and eosin) image of acantholysis (c, white circle, micrometer bar 50 lm). In contrast to the AK the adjacent clinically normal skin at the same level shows a intact layering of the epidermis and dermis in the slice mode (d) and a regular web pattern of the stratum granulosum (SG) in the en-face mode (e). were documented. Parakeratosis was found in (90%), destruction of the epidermal structure in (95%), pleomorphic keratinocytes in (100%) and inflammatory infiltrate in (80%) histologically examined AKs. The histological changes of the SC such as hyper- and parakeratosis and the pink and blue sign correlated with structural changes in HD-OCT, which we defined as irregular entrance signal and disruption of the SC. The presence of nuclei in the corneocytes as a sign of parakeratosis could not be detected distinctly in HD-OCT. The histologically described destruction of the epidermal structure correlated with the destruction of layering in the slice mode and the architectural disarray in the en-face mode of HD-OCT. The presence of pleomorphic keratinocytes correlated with cellular nuclear polymorphism in the en-face HD-OCT. The observed bright bundles in the superficial dermis presumably correlated with solar elastosis. The histologically described inflammatory infiltrate could not be definitely seen using HD-OCT, although clusters of small bright roundish structures were sometimes found in the upper layers. We correlated selected HD-OCT features with the corresponding histological findings as presented in Table 4. The following pairs were compared: parakeratosis in histology and disruption of SC in the en-face mode, destruction of epidermal structure in histology and architectural disarray in SG SS in the en-face mode, pleomorphic keratinocytes and cellular nuclear polymorphism in the en-face mode, parakeratosis and irregular entrance signal in the slice mode, destruction of epidermal structure and destruction of layering in the slice mode. The sensitivity, specificity and P-value were evaluated using McNemar s v 2 test with continuity correction. Due to the small sample size (n = 20) and the finding that the criteria are all positive in the histology (20 of 20) a specificity of 0% occurs with a very broad 95% confidence interval (0 84%) that covers the whole range of the domain for a proportion. The highest sensitivity was found in the correlating pair parakeratosis in histology and disruption of SC in the en-face mode (88%), and second highest sensitivity was 80% in the correlated pair pleomorphic keratinocytes in histology and cellular nuclear polymorphism in the en-face mode. Comparing the histological characteristics of a single lesion with the findings in the en-face HD-OCT images, individual histological features such as focal acantholysis and dyskeratosis could be identified. As depicted in Figure 3 we found in the SG single dark round structures with a bright irregular centre using the en-face mode of HD-OCT. These were larger than the regularly composed surrounding keratinocytes and thus possibly consistent with dyskeratotic cells as described by conventional histology. The histologically commonly found acantholysis was most probably presented by a disruption of the regular honeycomb pattern with blurred cell borders of keratinocytes and widened dark spaces between the single cells (Fig. 4). Altogether, the HD-OCT images correlated well with the features found by conventional histology. Discussion Because of the rising incidence of AK worldwide, the effective screening, treatment and monitoring of AK is becoming increasingly important. Therefore the purpose of this preliminary study was to test the improved HD-OCT device in a

7 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. 7 (a) (b) (c) Fig 3. (a) Actinic keratosis lesion 7 presents in the slice mode of high-definition optical coherence tomography (HD-OCT) with an irregular epidermal band alternating in width and brightness (white arrow), thus consistent with the histological picture of alternating ortho- and hyperkeratosis ( pink and blue sign) (black long arrow, c). (b) In the en-face mode the regular honeycomb pattern of the stratum granulosum stratum spinosum (SG SS) shows alterations with single dark round structures with bright centres (white arrows, micrometer bar 200 lm) possibly related to dyskeratosis of single cells as depicted in the histology (c, black arrows, micrometer bar 20 lm). (c) The cellular polymorphism and spongiosis of the keratinocytes as depicted in the histology (dark circle) could be correlated to variations in shape, size and brightness of cellular structures in the en-face mode of HD-OCT (b, white circle). series of AK to define and evaluate characteristic morphological features in comparison with the histological examination as the gold standard. The development of HD-OCT allows a higher resolution of morphological skin structures compared with the conventional OCT devices. It enables the visualization of skin changes not only in the typical vertical (slice) imaging mode but also in the horizontal (en-face) mode and has recently been shown to be a valuable tool in the diagnostic imaging of basal cell carcinoma (BCC). 18 The main findings of our study were: first, morphological alterations of AK could be visualized in the en-face mode of HD-OCT in correlation with histological findings. Second, lesional skin of AK showed different morphological features in the HD-OCT than nonlesional adjacent skin. Third, the en-face mode of HD-OCT allowed the description of cellular alterations present in AK including pleomorphism of keratinocytes, dyskeratosis and acantholysis. In earlier studies using conventional OCT, AKs were defined by the destruction of the epidermal dermal layering and the presence of white streaks, dots and grey areas corresponding to hyperkeratosis. 17 Furthermore, thinning of epidermis and disruption of the entrance signal were described in the OCT imaging of AK. Barton et al. 19 found a dark band in the SC characteristic for AK with a sensitivity of 79% and a specificity of 100% in conventional OCT. However, in an observerblinded study by Mogensen et al. 12 the sensitivity of recognizing AK and BCC in regular OCT was only 46%. They found that the OCT diagnosis was less reliable than the clinical diagnosis concerning BCC and AK, but the pathological lesions could be distinguished from healthy skin. These findings are comparable with our results of alterations in AK present in the slice mode of HD-OCT such as irregular entrance signal, destruction of layering, white streaks and dots, and grey areas. We found that there were distinct features present in the enface mode of HD-OCT imaging of AK. Herein, we were able to visualize cellular and nuclear polymorphism, which have not been described in conventional OCT so far. We found the en-face mode of HD-OCT superior to the slice mode in the imaging of individual histological alterations such as cellular polymorphism, dyskeratosis and acantholysis. In this context, our findings of morphological changes of AK in the en-face mode of HD-OCT could be better compared with features of AK described in RCM images than in conventional OCT. RCM features of AK were first described by Aghassi et al. 20 They found enlarged, pleomorphic nuclei varying in size, shape and orientation contrasting with small, uniform nuclei from normal skin. Later on, Ulrich et al. 8 refined evaluation parameters such as superficial disruption, architectural disarray, and cellular polymorphism at the granular and spinous layer. Additionally, RCM analysis was found helpful in the noninvasive monitoring of treatment response. 21,22 In RCM, keratinocytes are described as roundish structures with dark nuclei and bright cytoplasm forming a regular web, which is known as a honeycomb pattern. 23 This description is comparable with the image of SG SS in the en-face mode of HD-OCT, although the differences are not as prominent as in RCM, where the granular structure of the cells in the granular layer can be clearly visualized. Nevertheless, in consistency with RCM features for AK, we recognized alterations of keratinocytes in AK in the en-face mode of HD-OCT with enlargement of nuclei and varying size and shape of keratinocytes.

8 8 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. Table 4 High-definition optical coherence tomography (HD-OCT) criteria in the en-face and slice mode compared with the matching criteria in routine histology as the gold standard with individual sensitivity and P-value. Specificity was 0 in all cases because of the small sample size Histology Parakeratosis Destruction of epidermal structure Pleomorphic keratinocytes Sensitivity (%) P-value HD-OCT criteria Disruption of SC (en-face) Architectural disarray in SG SS (en-face) Æ136 Cellular nuclear polymorphism in SG SS (en-face) Æ133 Irregular entrance signal (slice) Æ683 Destruction of layering (slice) Æ13 SC, stratum corneum; SG SS, stratum granulosum stratum spinosum. (a) (b) (c) Fig 4. (a) High-definition optical coherence tomography imaging of actinic keratosis lesion 4 shows an irregular entrance band with bright and dark areas and focally broadened epidermal band in the slice mode consistent with hyperparakeratosis and acanthosis in histology. (b) In the en-face mode at the layer of the stratum granulosum stratum spinosum (SG SS) the regular web structure shows focally broadened bright interspaces and blurred dark cell borders (white arrows, micrometer bar 200 lm) possibly consistent with spongiotic disaggregation as displayed in the histology (c, dark arrows, micrometer bar 50 lm). Even though an exact mapping of the investigated area was not possible in the HD-OCT, we observed several large dark roundish bodies with a bright irregular centre in the lesions where histologically dyskeratotic cells were described. In the literature, the description of dyskeratotic cells in RCM is somehow inconsistent. In the latest RCM atlas large round bright cells with a dark nucleus are described suggestive of dyskeratotic keratinocytes in SCC, 24 whereas Terhorst et al. 25 found large cells with a bright centre and a dark halo corresponding possibly to apoptotic cells in wound healing. In our opinion, dyskeratotic and or apoptotic cells with a condensed nucleus are more likely to be displayed in HD-OCT and in RCM as dark cells with a bright centre. We presume that atypical, pleomorphic cells as found in neoplasia could presumably present as large bright cells with large darker centres. Another feature characteristic of AK in histology is acantholysis, which could be referred to as dark blurry spaces in HD- OCT in our study. This is consistent with the latest atlas of RCM describing microvesicle formation in spongiotic dermatitis as dark hollow spaces. In RCM inflammatory cells are depicted as scattered small bright spots. 24 In the images of enface mode of HD-OCT we found inconsistently clusters of small bright roundish structures in the upper layers possibly correlating with inflammatory cells. In comparison with conventional OCT devices, HD-OCT has the advantage of the additional en-face mode and a high resolution of about 3 lm in all dimensions which allows supplementary information on cellular changes of the investigated lesions. On the other hand, conventional OCT has a higher penetration depth of about 1 2 mm compared with approximately 750 lm using HD-OCT. In contrast, RCM offers the highest resolution of about 1Æ5 lm compared with conventional and HD-OCT and allows the visualization of the cellular and subcellular morphology. However, RCM imaging is limited to horizontal sections with a penetration depth of about 250 lm and the examination time is about 15 min depending on the investigator s experience. In comparison, HD-OCT imaging takes only about 5 7 min. Another advantage of RCM is the possibility of a topographical mapping of the investigated lesion with an integrated dermoscopic picture, which is to date not available with the HD-OCT device.

9 Actinic keratosis in high-definition optical coherence tomography, T. Maier et al. 9 Another promising skin imaging technique with high potential is multiphoton tomography (MPT), which is based on a titanium : sapphire laser system and provides horizontal optical sections up to 200 lm tissue depth with an axial and transverse resolution of < 2 lm and < 0Æ5 lm, respectively. It has shown promising results in the imaging of various dermatoses especially in combination with OCT or RCM. 26,27 Nevertheless, MPT currently has a relatively slow imaging process with a narrow field of view of about 0Æ3 mm 3, which makes it at this time point inconvenient as an imaging tool for daily clinical practice. Altogether, HD-OCT with the integrated slice and en-face mode offers an additional noninvasive investigation tool in the diagnostics of AK by combining certain advantages of both conventional OCT and RCM. For the determination of sensitivity and specificity levels of HD-OCT in the diagnosis of AK additional, observer-blinded studies are required. What s already known about this topic? Conventional optical coherence tomography (OCT) has been studied in the diagnosis of actinic keratosis (AK), without the discrimination of reliable criteria because of an insufficient resolution. What does this study add? Using the new en-face imaging mode of an innovative highdefinition OCT morphological features of AK are now described in correlation with the histological findings. References 1 Glogau RG. The risk of progression to invasive disease. J Am Acad Dermatol 2000; 42: Salasche SJ. Epidemiology of actinic keratoses and squamous cell carcinoma. J Am Acad Dermatol 2000; 42: Frost C, Williams G, Green A. High incidence and regression rates of solar keratoses in a Queensland community. J Invest Dermatol 2000; 115: Memon AA, Tomenson JA, Bothwell J et al. Prevalence of solar damage and actinic keratosis in a Merseyside population. Br J Dermatol 2000; 142: Stockfleth E, Ulrich C, Meyer T et al. Epithelial malignancies in organ transplant patients: clinical presentation and new methods of treatment. Recent Results Cancer Res 2002; 160: Sterry W, Griffiths C, Korting HC, eds. EDF Guidelines for Dermatology in Europe, Vol. 1. Berlin: ABW Wissenschaftsverlag Berlin, Quaedvlieg PJ, Tirsi E, Thissen MR et al. Actinic keratosis: how to differentiate the good from the bad ones? Eur J Dermatol 2006; 16: Ulrich M, Maltusch A, Rius-Diaz F et al. Clinical applicability of in vivo reflectance confocal microscopy for the diagnosis of actinic keratoses. Dermatol Surg 2008; 34: Zalaudek I, Giacomel J, Argenziano G et al. Dermoscopy of facial nonpigmented actinic keratosis. Br J Dermatol 2006; 155: Pellacani G, Cesinaro AM, Seidenari S. In vivo assessment of melanocytic nests in nevi and melanomas by reflectance confocal microscopy. Mod Pathol 2005; 18: Maier T, Sattler EC, Braun-Falco M et al. Reflectance confocal microscopy in the diagnosis of partially and completely amelanotic melanoma: report on seven cases. J Eur Acad Dermatol Venereol doi: j x. [Epub ahead of print]. 12 Mogensen M, Jørgensen TM, Nurnberg BM et al. Assessment of optical coherence tomography imaging in the diagnosis of nonmelanoma skin cancer and benign lesions versus normal skin: observer-blinded evaluation by dermatologists and pathologists. Dermatol Surg 2009; 35: Gambichler T, Orlikov A, Vasa R et al. In vivo optical coherence tomography of basal cell carcinoma. J Dermatol Sci 2007; 45: Welzel J. Optical coherence tomography in dermatology: a review. Skin Res Technol 2001; 7: Korde VR, Bonnema GT, Xu W et al. Using optical coherence tomography to evaluate skin sun damage and precancer. Lasers Surg Med 2007; 39: Gambichler T, Jaedicke V, Terras S. Optical coherence tomography in dermatology: technical and clinical aspects. Arch Dermatol Res 2011; 303: Jørgensen TM, Tycho A, Mogensen M et al. Machine-learning classification of non-melanoma skin cancers from image features obtained by optical coherence tomography. Skin Res Technol 2008; 14: Maier T, Braun-Falco M, Hinz T et al. Morphology of basal cell carcinoma in high definition optical coherence tomography: en-face and slice imaging mode, and comparison with histology. J Eur Acad Dermatol Venereol doi: j x. [Epub ahead of print]. 19 Barton JK, Gossage KW, Xu W et al. Investigating sun-damaged skin and actinic keratosis with optical coherence tomography: a pilot study. Technol Cancer Res Treat 2003; 2: Aghassi D, Anderson RR, Gonzalez S. Confocal laser microscopic imaging of actinic keratoses in vivo: a preliminary report. J Am Acad Dermatol 2000; 43: Ulrich M, Krueger-Corcoran D, Roewert-Huber J et al. Reflectance confocal microscopy for noninvasive monitoring of therapy and detection of subclinical actinic keratoses. Dermatology 2010; 220: Ulrich C, Johannsen A, Roewert-Huber J et al. Results of a randomized, placebo-controlled safety and efficacy study of topical diclofenac 3% gel in organ transplant patients with multiple actinic keratoses. Eur J Dermatol 2010; 20: González S, Gill M, Halpern AC, eds. Reflectance Confocal Microscopy of Cutaneous Tumors: An Atlas with Clinical, Dermoscopic and Histological Correlations. London: Informa Healthcare, Hofmann-Wellenhof R, Pellacani G, Malvehy J, Soyer HP, eds. Reflectance Confocal Microscopy for Skin Diseases, Vol. 1. Berlin: Springer- Verlag, Terhorst D, Maltusch A, Stockfleth E et al. Reflectance confocal microscopy for the evaluation of acute epidermal wound healing. Wound Repair Regen 2011; 19: Konig K, Speicher M, Buckle R et al. Clinical optical coherence tomography combined with multiphoton tomography of patients with skin diseases. J Biophotonics 2009; 2: Alex A, Weingast J, Weinigel M et al. Three-dimensional multiphoton optical coherence tomography for diagnostic applications in dermatology. J Biophotonics doi: jbio [Epub ahead of print].