Potential of Contrast Agents to Enhance in Vivo Confocal Microscopy and Optical Coherence Tomography in Dermatology: a Review

Received: 12 December 2018 Revised: 2 March 2019 Accepted: 7 March 2019 DOI: 10.1002/jbio.201800462

REVIEW ARTICLE

Potential of contrast agents to enhance in vivo confocal microscopy and optical coherence tomography in dermatology: A review

Hans C. Ring1* | Niels M. Israelsen2 | Ole Bang2 | Merete Haedersdal1 | Mette Mogensen1

1Department of Dermatology, Bispebjerg Hospital, Nielsine Nielsens Vej 9, 2400 København NV, Distinction between Faculty of Health and Medical Sciences, normal skin and University of Copenhagen, Copenhagen, Denmark pathology can be a 2 DTU Fotonik, Department of Photonics diagnostic challenge. Engineering, Technical University of Denmark, Kongens Lyngby, Denmark This systematic *Correspondence review summarizes Hans C. Ring, Department of Dermatology, how various contrast Bispebjerg Hospital, University of Copenhagen, agents, either topically delivered or injected into the skin, affect distinction between Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark. skin disease and normal skin when imaged by optical coherence tomography (OCT) Email: [email protected] and confocal microscopy (CM). A systematic review of in vivo OCT and CM studies Funding information using exogenous contrast agents on healthy human skin or skin disease was per- Innovationsfonden, Grant/Award Number: formed. In total, nine CM studies and one OCT study were eligible. Four contrast 4107-00011A agents aluminum chloride (AlCl) n = 2, indocyanine green (ICG) n = 3, sodium fluorescein n = 3 and acetic acid n = 1 applied to CM in variety of skin diseases. ICG, acetic acid and AlCl showed promise to increase contrast of tumor nests in keratinocyte carcinomas. Fluorescein and ICG enhanced contrast of keratinocytes and adnexal structures. In OCT of healthy skin gold nanoshells, increased contrast of natural skin openings. Contrast agents may improve delineation and diagnosis of skin cancers; ICG, acetic acid and AlCl have potential in CM and gold nanoshells facilitate visualization of adnexal skin structures in OCT. However, as utility of bed- side optical imaging increases, further studies with robust methodological quality are necessary to implement contrast agents into routine dermatological practice.

KEYWORDS Au, contrast agent, contrast enhancer, fluorescence confocal microscopy, gold microparticles, medical skin imaging, optical coherence tomography, optical imaging, reflectance confocal microscopy

1 | INTRODUCTION diagnosis various skin diseases and to monitor treatment efficacy over time [1, 3–8]. Optical coherence tomography (OCT) and confocal micros- Distinction of normal skin architecture from skin pathol- copy (CM) are noninvasive medical imaging technologies ogy can be a diagnostic challenge in dermatological imag- capable of visualizing skin microarchitecture at the bedside. ing. Therefore, exogenous contrast agents that either These technologies are rapidly gaining clinical acceptance in highlight pathological changes in skin or enhance normal the field of dermatology and other medical specialties, for skin features are in demand. Medical imaging technologies example, ophthalmology [1, 2]. An array of clinical studies are increasingly applied in skin examinations, but contrast demonstrates a promising potential of CM and OCT to agents have yet to gain clinical acceptance. Our intention is

J. Biophotonics. 2019;12:e201800462. www.biophotonics-journal.org © 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1of9 https://doi.org/10.1002/jbio.201800462 2of9 RING ET AL. to provide an overview of current exogenous contrast agents relevant OCT and CM studies applying contrast agents available for CM and OCT imaging of skin and skin disease in vivo to healthy human skin or skin disease. Assess- in vivo by performing a systematic review. ment of study quality was undertaken using the Oxford Contrast agents, in medical terms, a contrast agent is Centre for Evidence-based Medicine levels of evidence used to enhance contrast of anatomical structures, patholo- criteria. gies or fluids within the body for medical imaging purposes. The term contrast means the degree to which light and dark areas of an image differ in brightness. Exogenous contrast 2 | MATERIAL/METHODS agents are chemical or physical substances used to increase contrast and enhance the visibility of tissues or anatomical 2.1 | Search strategy structures. Contrast agents can be delivered to skin topically, A systematic search of Embase and Medline was performed subcutaneously or intravenously. Although utilization of (January 01, 1980 to February 2, 2019) with the following contrast agents for optical imaging is an evolving field in two search queries: “contrast agent AND skin AND CM” dermatology, the method has not yet gained substantial clini- and “contrast agent AND skin AND OCT and “contrast cal impact. Indeed, application of contrast agents prior to agent AND skin AND fluorescence”. Furthermore, the refer- optical imaging, such as gold nanoparticles or fluorescent ence lists from the included articles were manually searched dyes such as indocyanine green (ICG) appear to have poten- for additional relevant studies. Two authors H.C.R. and tial for generating strong contrast signals in skin, thus poten- M.M. performed data extraction and quality assessment tially enhancing diagnostic accuracy of optical imaging independently. The overall search strategy is available in devices at the bedside [9–11]. Supporting Information Figure S1. OCT is a laser-based imaging technology that provides fast image acquisition, enabling real time, high-resolution, 2.2 | Inclusion criteria noninvasive, cross-sectional and en face representation and analysis of skin structures [3, 12, 13]. The imaging The target intervention was the use of a contrast agent in depthinskinis1to2mm,andresolution3to10μm OCT or CM imaging, and no restriction was placed on the depending on the OCT system and properties of the skin. type of OCT system or CM system. The target conditions OCT images are derived from measuring the signal of a were skin disease and normal skin. All human studies reference laser beam mixed with light scattered from were included with no restrictions on age, sex, ethnicity within the skin and is based on interferometry. OCT does or type of study. Case reports and case series were not detect fluorescence. Most commercially available sys- included. Only papers in English were included. More- tems have handheld probes with sizes such as modern over, only studies investigating gold nanoparticles with diameters above 80 nm were included. Particle size ultrasound systems and use an infrared laser centered at appears to constitute an important parameter defining the approximately 1300 nm. biological action of gold nanoparticles. This minimum Reflectance and fluorescence CM is a laser-based imag- particle size was selected based on the findings that ing technology, offering real-time microscopy to a depth of nanoparticles larger than 80 nm do not penetrate through 200 to 300 μm and resolution 0.5 to 1.25 μm. The laser light epidermis into dermis and cannot cause harm by passing focuses on a small skin area of interest and is reflected and the blood–brain barrier [14–16]. Exclusion criteria were scattered due to variations in refractive index of skin struc- other reviews, in vitro, ex vivo or phantom model studies. tures. The light travels back through the objective lens and Studies applying intravenous contrast agents were not reaches the selective aperture (pinhole) that prevents the pas- included, as we searched for contrast agents applied or sage of out-of-focus light. The confocality of the microscope injected directly to the skin. arises mainly from the diameter of the selective pinhole, which determines the resolution. The commercial CM sys- tem has a 0.5 × 0.5 mm field of view and image blocks can 2.3 | Methodological quality be captured to provide larger 8 × 8 mm image mosaics. Assessment of the quality of eligible studies was undertaken Whereas, contrast in reflectance CM (RCM) images is pro- using the Oxford Centre for Evidence-based Medicine vided by, for example, keratin and melanin that serve as (OCEBM) Levels of Evidence Working Group criteria (The endogenous chromophores, in fluorescence CM (FCM) con- Oxford 2011 Levels of Evidence). To rate the scientific qual- trast is derived from exogenous fluorescent dyes applied to ity of each study, all studies were graded according to or injected into the skin. Commercial systems usually consist OCEBM criteria using the question: “Is this diagnostic or of a flexible probe that can be applied to the skin area of monitoring test accurate?” and scored on a level from 1 to interest. 5. Oxford grades of recommendations ranges from A-D We systematically reviewed the literature using [17], which was also designated to each eligible study. See Embase and Medline databases to provide an overview of Tables S1. RING ET AL. 3of9

3 | RESULTS contrast to normal skin was increased compared with RCM images. The initial search yielded a total of 90 references, of which Fluorescein sodium: The normal skin architecture and 9 CM studies with 179 patients and 1 OCT study with psoriatic skin lesions have been investigated using fluores- 11 patients were eligible. All eligible studies were published cein sodium. In 22 patients, Swindel LD et al [20] demon- between 2003 and 2018. strated enhanced superficial skin architecture and clear The nine CM studies investigated four contrast agents: transition in keratinocyte size, shape and morphology aluminum chloride (AlCl) n = 2, ICG n = 3, sodium fluo- throughout epidermis. Dermal capillaries and blood flow rescein n = 3 and acetic acid n = 1, applied to a variety were clearly depicted. Moreover, in a comparative study of of skin diseases. A single OCT study was eligible, using 12 psoriasis patients and eight healthy controls, a combina- topically applied gold nanoshells (GNS) in 11 healthy par- tion of topical application and subsequent subepidermal ticipants. Tables 1 and 2 demonstrate the overall findings injections of fluorescein sodium characteristic features of when applying a contrast agent to CM and OCT imaging keratinocytes in lesional as well as nonlesional psoriatic in vivo. skin different from healthy skin was demonstrated. The horizontal width of dermal papillae of psoriatic skin was 3.1 | Grading of methodology in studies increased by approximately 50% when compared with healthy controls [21]. Overall, a good visualization of cell Studies included were rated according to the OCEBM nuclei and parakeratosis was demonstrated in psoriatic epi- Levels of Evidence Working Group criteria. All studies were dermis, and in FCM images of skin with fluorescein the – found to be level 4C. Level four is case-series or case contrast to normal skin was increased compared with RCM control studies, or poor-quality prognostic cohort studies, C images. designates the Oxford grades of recommendation and means A study investigated the application of sodium fluores- results based on level 4 studies or extrapolations from level cein lotion on the forearm of 15 healthy volunteers. During a 2 or 3 studies. See Table S2. 24-hour interval, in vivo CM was performed in a multi-wave mode. The fluorescence was imaged 30 minutes, 4 hours 3.2 | CM studies and 24 hours after application of the lotion. Interestingly, in A variety of skin diseases were studied by CM: basal cell all participants, CLSM images of stratum corneum and the carcinomas, BCC (n = 4), squamous cell carcinoma, SCC upper epidermis had detectable fluorescence after (n = 1), actinic keratosis (n = 1), psoriasis (n = 2), dermal 30 minutes and 4 hours. In 2/3 of the healthy volunteers, the naevi (1), contact dermatitis (1), pachyonychia congenital fluorescence was still detectable at the skin surface after (n = 1) and seborrheic keratosis (n = 1). 24 hours [22]. Aluminum chloride: The study by Flores et al [18] dem- In summary, BCC morphology was explored using ICG, onstrated that RCM imaging can facilitate Mohs surgery acetic acid or AlCl. All four studies demonstrated character- using AlCl. In 25 patients with keratinocyte carcinomas, istic BCC tumor features well known from histology, for AlCl was topically applied post-surgery to wounds. It example, tumor nests or bright nuclei of tumor cells. Two CM studies investigated AlCl applied to Mohs surgery, and offered a repeatable and consistently enhanced contrast of found a potential increase in efficacy of the surgical and his- tumor morphology at a cellular level. In 17 of 25 wounds, tological procedure when using AlCl as contrast agent. In reproducible and hyperreflective signals of nuclear BCC five CM studies, the normal skin architecture was depicted and SCC morphology were observed. Moreover, adnexal using either sodium fluorescein or ICG, and the latter structures such as hair follicles, sebaceous glands and enhanced the intricate structure of the collagen network in eccrine glands were easily differentiated in RCM. AlCl dermis. Fluorescein enhanced cellular details of appears to optimize contrast of keratinocyte carcinomas to keratinocytes. normal skin. Indocyanine green: Application methods of ICG dif- fered between studies. In the study by Jonak et al [19] a sta- 3.3 | OCT studies ble fluorescence signal was demonstrated in FCM images Gold nanoshells: a single OCT study investigated topical after injection of ICG into irritant dermatitis and basal cell application of 0.15 μm GNS to healthy skin of 11 partici- carcinoma. Moreover, the study demonstrated fluorescence pants. GNS deposits in natural skin openings were visualized of ICG for more than 48 hours. ICG is applicable for dis- using both a commercial and ultrahigh-resolution OCT tinction between normal and healthy skin since the intricate (UHR-OCT) system. The study found pixel intensity collagen structure of dermis, and hence BCC islands were around hair follicles significantly higher in UHR-OCT easier to detect. All studies demonstrated enhancement of images and showed that GNS worked well for enhancing adnexal structures such as hair follicles, sebaceous and contrast in the natural opening of the skin, that is, hair fol- eccrine glands. In FCM images of skin with ICG the licles and sweat gland ducts, the latter not otherwise 4of9 RING ET AL.

TABLE 1 Exogenous contrast agents evaluated in vivo by reflectance and/or fluorescence confocal microscopy

Brief description of study design and First authors/Oxford levelsa aim of study Application of contrast agent CM key features Contrast agent/ skin disease investigated Acetic acid/BCCs Sierra H et al [36] 2016 In 72 patients, normal skin and BCC Skin specimens were immersed in 5% Uptake of acetic acid provided Level 4, C tumor specimens were obtained to acetic acid for 30 s enhanced morphologic detail in characterize the depth of carbon RCM. Hair follicles, sebaceous dioxide laser-ablation and to evaluate glands, connective ducts and uptake of acetic acid. At 1-3 passes eccrine ducts were more clearly of 5.5 J/cm2, 6.5 and 7.5 J/cm2, and visualized in RCM. Presence of 1-2 passes of 8.5 J/cm2 residual tumor or complete clearance in adjacent skin was also demonstrated in RCM Contrast agent/ skin disease investigated Aluminum chloride/BCC and SCCs Flores ES et al [18]2015 Post-surgical procedure, AlCl was Wounds were topically swabbed with In 17 of 25 wounds, reproducible and Level 4, C applied and RCM imaging of ALCl (35%). Four swabs were hyperreflective signals of nuclear wounds in 25 patients with applied, each for 15 s, for total time BCC morphology were observed. keratinocyte carcinomas. Eight of 1 min In general, AlCl enhanced the patient specimens were shave contrast and detectability of excision wounds and 17 patients residual BCC and SCC tumors— were stage 1 Mohs. Immediately and tumor borders. Moreover, after surgery AlCl was applied on the adnexal structures such as hair wound follicles, sebaceous glands and eccrine glands were also easily differentiated in RCM Contrast agent skin disease investigated Aluminum chloride/BCC Tannous Z et al [24]2003 RCM scans from five patients with AlCl (20%) was topically applied In both in vivo and ex vivo, AlCl Level 4, C clinically visible and biopsy-proven provided a good immediate contrast BCC. The first Mohs layer was then between BCC cells and the excised, and fresh-frozen sections surrounding tissue. The tumor cells were correlated with the CM with AlCl exhibited intensely bright findings. AlCl was applied on the nuclei with an excellent contrast as Mohs defect followed by RCM on compared with the low-contrast one site from each lesion. dark nuclei without AlCl application. Contrast agent/skin disease investigated Fluorescein sodium/psoriasis Suihko C et al [21] 2012 In 12 patients with psoriasis, lesional Fluorescein sodium (0.1% w/v FCM images depicted characteristic Level 4, C and nonlesional skin were concentration in 0.9 g/L saline features of epidermis and investigated and compared with a solution): Combination of topical keratinocytes in lesional as well in control group of eight healthy application and subsequent nonlesional skin vs. healthy skin, participants using CM and sodium subepidermal injections 30% increased width of fluorescein keratinocytes was demonstrated compared with healthy skin controls. Number of keratinocytes per viewing field was reduced. Detailed visualization of cell nuclei and parakeratosis was demonstrated. Horizontal width of dermal papillae of psoriatic skin was increased by approximately 50% when compared to controls Contrast agent/object investigated Fluorescein sodium/normal skin Swindel LD et al [20] 2003 In 22 volunteers FCM was applied to Fluorescein sodium (20 uL of 0.2% FCM images showed a good Level 4, C visualize microarchitecture of normal wt/vpl) intradermally injected immediate visualization of the epidermis in vivo. Fluorescein superficial skin architecture. A sodium was then administered on the transition in keratinocyte size, volar forearm. The area of interest shape and morphology with was imaged continuously from 1 to progressive imaging intro deeper 15 min post injection. epidermal layers was demonstrated. Microscopic dermal capillaries with blood flow was clearly depicted Contrast agent/object investigated Fluorescein sodium/normal skin E. Sattler et al 2012 [22] In 15 healthy participants, sodium Fluorescein sodium lotion (4% Overall, sodium fluorescence was Level 4,C fluorescein lotion applied to forearm concentration). Topically applied. found a safe and efficient during a 24-h interval. CM was fluorophore and was indeed able to performed in a multi-wave mode. visualize kinetics of topically The fluorescence was imaged applied protective lotion. CM 30 min, 4 and 24 h after application images of stratum corneum and of the lotion. upper epidermis demonstrated detectable fluorescence after 30 min and 4 h. In 2/3 of the RING ET AL. 5of9

TABLE 1 (Continued)

Brief description of study design and First authors/Oxford levelsa aim of study Application of contrast agent CM key features healthy volunteers, the fluorescence was still detectable at the skin surface after 24 h Contrast agent/object investigated ICG/normal skin Skvara H et al [23] 2011 ICG was studied in healthy forearm ICG was intra-dermally injected (0.5%) In the stratum granulosum and Level 4, C skin of 10 volunteers and evaluated spinosum FM depicted a clearer by RCM and FCM outline of cell borders. Dermal papillae appeared as bright round structures. Capillary vessels appeared hyporeflective. Eccrine sweat ducts were depicted in both FCM and RCM. At depth 80 μm collagen fibers were only clearly visible in FCM. The distribution of ICG remained stable in all participants for at least 8 h Contrast agent/skin disease investigated ICG/BCC, actinic keratosis, seborrheic keratosis and psoriasis Skvara H et al [23]b 2011 ICG was applied on 10 patients with ICG was intra-dermally injected (0,5%) BCC: Both techniques depicted tumor Level 4, C either BCC, actinic keratosis, nests, characterized by polarization seborrheic keratosis and psoriatic of epithelial cells at the periphery of lesions and subsequently scanned by the islands; however, FCM detected RCM and FCM tumor nests with more precision. Actinic keratosis: Visualization of architectural disarray as well as of atypical keratinocytes varying in size. Seborrheic keratosis: In FCM and RCM, a clear distinction of backscattering from melanin-containing cells was demonstrated. Psoriasis: The capillaries within in dermal papillae were more visible in FCM than in RCM Contrast agent/skin disease investigated ICG/dermal nevus, irritant contact dermatitis and skin necrosis Jonak C et al [19] 2011 In nine healthy volunteers, in vivo ICG (0,5%) and sodium fluorescein Overall, the emission of fluorescein Level 4, C kinetics of ICG and fluorescein was (0,08%): 20 μL of the dye solution appeared higher than that of ICG compared using FCM. The signals was injected intradermally immediately post injection; were evaluated up to 48 h after however, it declined quickly and injection. Furthermore, in three was only faintly detectable after patients with dermal nevus, irritant 2 h, whereas ICG fluorescence was contact dermatitis or necrosis the still very evident. Moreover, the method was also investigated contrast of ICG in the epidermal and dermal layers was also superior over fluorescein. Finally, ICG was able to provide high resolution on subcellular information from various skin diseases. For example in contact dermatitis spongiotic areas were clearly visible Contrast agent/skin disease investigated ICG/ normal skin and PC Ra H et al [37] 2010 In a volunteer in vivo fluorescence Application of a formulated ICGcream In the healthy volunteer, the dye was Level 4, C (ICG) imaging of the skin (2 mg/mL). observed to penetrate to 80 μmin microanatomy was evaluated using a most regions. Below the SC, the handheld dual-axis confocal cream appeared to penetrate deeper microscope. After application of ICG layers of the normal epidermis at two distinct sites (calf of the leg unevenly. In the PC patient, the and plantar region). Same procedure fluorescence signal was imaged was performed with a PC patient down to a depth of 50 μm, and individual keratinocytes were observed. The signal was barely detectable below depth 70 μm.

Abbreviations: BCC, basal cell carcinoma, SCC, squamous cell carcinoma, CM, confocal microscopy, RCM, reflectance confocal microscopy, FCM, fluorescence con- focal microscopy, AlCl, aluminum chloride, ICG, indocyanine green, PC, pachyonychia congenita. aOCEBM Levels of Evidence Working Group. The Oxford 2011 Levels of Evidence. Oxford Centre for Evidence-Based Medicine. http://www.cebm.net/index.aspx? o=5653. bSubstudy from Skvara H et al 2011. 6of9 RING ET AL.

TABLE 2 Exogenous contrast agents evaluated in vivo by optical coherence tomography

Brief description of study design and First author/Oxford levelsa aim of study Application of contrast agent OCT key features Contrast agent/object investigated GNS/normal skin Mogensen M et al [11] 2018 Applicability of topical GNS as OCT GNS with silica core 150 nm in GNS enhanced contrast of natural skin Level 4, C contrast agents in 11 healthy diameter. GNS was massaged into openings such as hair follicles and controls. The OCT features were the skin sweat glands, the latter otherwise not assessed using both a Commercial detected in axillary skin. The different (Vivosight Dx) and an OCT systems show different GNS ultrahigh-resolution OCT system intensities

Abbreviations: OCT, optical coherence tomography; GNS, gold nanoshells. aOCEBM Levels of Evidence Working Group. The Oxford 2011 Levels of Evidence. Oxford Centre for Evidence-Based Medicine. http://www.cebm.net/index.aspx? o=5653. detected in axillary skin using OCT without GNS as con- the potential properties of gold nanoprisms as a contrast trast agent. agent in OCT imaging [27]. Because gold and silver nanoparticles may also be used as vehicles for delivery of, for example, psoriasis therapy 4 | DISCUSSION [28] and chemotherapy [29], we suggest that future treat- ment trials would benefit from using OCT and CM imaging This review, mainly based on smaller studies, demonstrate to track nanoparticles that have contrast agent properties that various chemical and physical contrast agents, either in vivo. topically delivered or injected into the skin, may increase In Figure 1 we illustrate a clinical case where ICG was the distinction between skin disease and normal skin, or topically applied to CO2 laser channels in normal skin of accentuate normal skin architecture when imaged by OCT the lower arm. ICG is a nontoxic fluorescent dye approved and CM. Regarding diagnostic accuracy, data and study for in vivo use and applied in medical diagnostics [30] setups did not allow for assessment of sensitivity and fluorescence is detected in FCM. In Figure 2 we illustrate specificity. All studies were graded with low numbers on how fluorescein enhances epidermal structures over time the assessment scale, thus demonstrating the relatively after topical application of the dye into CO2 laser channels low level of methodological robustness among eligible and miniature skin holes made by a dermaroller device on studies. The result does however suggest a promising normal thigh skin and imaged by FCM. In Figure 3, it is potential for contrast agents applied to OCT and CM demonstrated how deposition of 150 nm GNS in CO2 laser imaging, in particular ICG for FCM imaging of various channels are visualized by OCT in human thigh skin skin diseases, further ICG and AlCl have potential for in vivo. facilitating delineation of skin cancers in OCT and CM This review have illustrated potential applicable con- images [23, 24]. trast agents for in vivo imaging in humans, however, an Recently, Mogensen et al [11] suggested a potential for important aspect in this regard is the safety and potential GNS in diagnosing skin diseases confined to adnexal toxicity of the agents. Although ICG is now approved for in vivo use, the application of the agent may cause pain structures. Hence, GNS may potentially allow novel and necrosis when administered intracutaneously. Impor- insight into skin diseases where adnexal structures play a tantly, the clinical administration of ICG is not approved role in the pathogenesis e.g. acne vulgaris or hidradenitis for intracutaneous use and should therefore be administered suppurativa. In addition to dermatological studies, oph- with caution [31]. thalmological studies have demonstrated excellent optical Overall, fluorescein, AlCl and methylene blue are properties and diagnostic potential of gold nanoparticles considered safe to use in vivo topically. However, some using RCM [25]. Gold nano- and microparticles are thus limitations may be associated with application of con- applicable to both OCT and RCM imaging. Moreover, an trast agents. The use of proper and sufficient adminis- unprecedented in vivo rabbit study showed for the first tration contrast agents in OCT and CM may be rather time that blood vessels can be differentiated from lym- time-consuming. Moreover, discoloration of this skin phatics vessels in OCT images when using two different may occur and Methylene blue may cause permanent gold nanorods as contrast agents [26]. This was further- tattooing and photobleaching of the skin [31]. Thus, more supported by a recent study investigating vascular potential side effects may depend on choice of agent or networks and microcirculation in melanomas in mice using on whether the contrast agents are applied topically or OCT angiography and gold nanoprisms. Interestingly, the intradermally. study showed a 54% increase in visibility of microvascula- Although the use of contrast agents may enhance ture after injection of gold nanoprisms, thus highlighting OCT and CM in terms of highlighting some important RING ET AL. 7of9

FIGURE 1 Reflectance and fluorescence confocal microscopy of indocyanine green (ICG) topically applied to laser channels in normal skin. A,

Dermoscopy image from inner arm of a healthy volunteer showing the rectangular grit (black lines) from a fractional CO2 laser at 10 mJ, 10% density (UltraPulse; Lumenis Inc., Santa Clara, CA). B, Reflectance confocal microscopy image (RCM) showing the same grit pattern in epidermis (depth 35 μm) in an 8 × 8 mm image mosaic. C, Fluorescence confocal microscopy (FCM) images demonstrate immediate ICG fluorescence in a pattern similar to the laser grit detected in RCM at 10 minutes. D, FCM image 3.5 hours after application, fluorescence faded after 22 hours (image not shown). ICG concentration: 80 μg/mL kept under occlusion 60 minutes after application. FCM images obtained by VivaScope Multilaser 1500 in 658 nm mode (Mavig GmpH, Germany). Courtesy to Dr. Arne Meisner, Amsterdam, NL. The case study was performed in accordance to the ethical guidelines of Declaration of Helsinki skin structures in images, it is safe to say that the der- This systematic review suggests that some contrast matological imaging field has come a long way without agents may assist bedside OCT and CM diagnosis of skin use of contrast agents [1, 2, 32–34]. Clinical OCT and disease. In CM by enhancement of keratinocyte carcinomas CM studies have previously described important hall- (AlCl, ICG, acetic acid); enhancement of keratinocyte struc- marks in various skin diseases and moderate to high ture and dermal vessels in psoriasis (fluorescein) and diagnostic accuracy without the use of contrast agents. adnexal structures (ICG). In OCT by enhancement of the natural skin openings, hair follicles and sweat ducts (GNS). In OCT, several studies have demonstrated potential for Further studies are required to elucidate the clinical versatil- fast skin cancer margin delineation and tumor thickness ity and diagnostic accuracy of in vivo contrast agents for measurement [3, 4, 13, 35]. optical skin imaging. 8of9 RING ET AL.

FIGURE 2 Fluorescence confocal microscopy (FCM) of sodium fluorescein in laser and microneedle channels in normal skin. FCM images of sodium fluorescein in laser and microneedle channels obtained by Vivascope in 488 nm mode. The laser grit following fractional CO2 laser at 25 mJ, 5% density is recognized in the upper row. The skin channels after microneedling are randomly distributed as a 1 mm microneedle device was rolled over the same skin area twice. Sodium fluorescein (1% Minims, Bausch & Lomb, UK) was applied on top of laser channels immediately after laser and needle exposure. It is evident that fluorescence highlights the epidermal ridges and intercellular tissue in both methods. Refer to Figure 1 for technology. Courtesy to Dr. Daniel Thaysen-Petersen, Bispebjerg Hospital. The case study was performed in accordance to the ethical guidelines of Declaration of Helsinki

FIGURE 3 Optical coherence tomography (OCT) images of gold nanoshells in laser channels. A, Clinical photo from OCT probe. B, OCT image showing an en face representation of CO2 laser channels in skin at the epidermal level (25 mJ, 5% density). OCT scans performed by Vivosight Dx OCT-system (Michelson Diagnostics Ltd, Kent, UK). OCT visualizes the gold nanoshells inside laser channels. C, All through dermis hyperreflective vertical slender columns represents the laser channels (white arrows) and superficial gold nanoshell accumulation is indicated by red arrows

Furthermore, as utility of bedside optical imaging tech- ACKNOWLEDGMENT nologies increases, future studies with robust methodological The authors would like to acknowledge support from quality are required to implement contrast agents into derma- Innovationsfonden through the ShapeOCT grant tological practice. No. 4107-00011A. RING ET AL. 9of9

CONFLICT OF INTEREST [22] E. Sattler, R. Kastle, M. Arens-Corell, J. Welzel, Skin Res. Technol. 2012, 18, 370. The authors have no conflict of interest to report. [23] H. Skvara, H. Kittler, J. A. Schmid, U. Plut, C. Jonak, J. Biomed. Opt. 2011, 16, 096010. ORCID [24] Z. Tannous, A. Torres, S. Gonzalez, Dermatol. Surg. 2003, 29, 839. [25] I. Paladini, U. Menchini, R. Mencucci, Eur. J. Ophthalmol. 2010, Hans C. Ring https://orcid.org/0000-0002-6145-5549 20,776. [26] O. Liba, E. D. SoRelle, D. Sen, A. de la Zerda, Sci. Rep. 2016, 6, 23337. REFERENCES [27] P. Si, E. Yuan, O. Liba, Y. Winetraub, S. Yousefi, E. D. SoRelle, [1] M. Mogensen, L. Thrane, T. M. Joergensen, P. E. Andersen, G. B. Jemec, D. W. Yecies, R. Dutta, A. de la Zerda, ACS Nano 2018, 12, 11986. Semin. Cutan. Med. Surg. 2009, 28, 196. [28] D. Crisan, K. Scharffetter-Kochanek, M. Crisan, S. Schatz, H. Hainzl, [2] J. Welzel, E. Lankenau, R. Birngruber, R. Engelhardt, J. Am. Acad. L.Olenic,A.Filip,L.A.Schneider,A.Sindrilaru,Exp. Dermatol. 2018, 27, Dermatol. 1997, 37, 958. 1166. [3] M. Mogensen, H. A. Morsy, L. Thrane, G. B. Jemec, Dermatology 2008, [29] M. A. Safwat, G. M. Soliman, D. Sayed, M. A. Attia, Mol. Pharm. 2018, 217, 14. 15, 2194. [4] M. Mogensen, T. M. Joergensen, B. M. Nurnberg, H. A. Morsy, [30] T. H. Degett, H. S. Andersen, I. Gogenur, Langenbeck's Arch. Surg. 2016, J. B. Thomsen, L. Thrane, G. B. Jemec, Dermatol. Surg. 2009, 401, 767. 35, 965. [31] J. Welzel, R. Kastle, E. C. Sattler, Dermatol. Clin. 2016, 34, 527. [5] H. C. Ring, M. Mogensen, C. Banzhaf, L. Themstrup, G. B. Jemec, Arch. [32] J. Dinnes, J. J. Deeks, D. Saleh, N. Chuchu, S. E. Bayliss, L. Patel, Dermatol. Res. 2013, 305, 299. C. Davenport, Y. Takwoingi, K. Godfrey, R. N. Matin, R. Patalay, [6] H. C. Ring, M. Mogensen, A. A. Hussain, N. Steadman, C. Banzhaf, H. C. Williams, Cochrane Database Syst. Rev. 2018, (12), CD013191. L. Themstrup, G. B. Jemec, J. Eur. Acad. Dermatol. Venereol. 2015, 29,890. https://doi.org/10.1002/14651858.CD013191 [7] S. Gonzalez, M. Rajadhyaksha, R. R. Anderson, J. Invest. Dermatol. 1998, [33] J. Dinnes, J. J. Deeks, N. Chuchu, D. Saleh, S. E. Bayliss, Y. Takwoingi, 111, 538. C. Davenport, L. Patel, R. N. Matin, C. O'Sullivan, R. Patalay, [8] M. Rajadhyaksha, S. Gonzalez, J. M. Zavislan, R. R. Anderson, H. C. Williams, Cochrane Database Syst. Rev. 2018, (12), CD013190. R. H. Webb, J. Invest. Dermatol. 1999, 113, 293. https://doi.org/10.1002/14651858.CD013190 [9] S. A. Boppart, A. L. Oldenburg, C. Xu, D. L. Marks, J. Biomed. Opt. 2005, [34] C. Ruini, D. Hartmann, M. Bastian, T. Ruzicka, L. French, C. Berking, 10, 41208. T. V. Braunmuhl, J. Biophotonics 2019, e201800391. https://doi.org/10. [10] S. P. Mattison, W. Kim, J. Park, B. E. Applegate, Curr. Mol. Imaging 1002/jbio.201800391 2014, 3, 88. [35] S. A. Alawi, M. Kuck, C. Wahrlich, S. Batz, G. McKenzie, J. W. Fluhr, [11] M. Mogensen, S. Bojesen, N. M. Israelsen, M. Maria, M. Jensen, J. Lademann, M. Ulrich, Exp. Dermatol. 2013, 22, 547. A. Podoleanu, O. Bang, M. Haedersdal, J. Biophotonics 2018, 11, [36] H. Sierra, S. Damanpour, B. Hibler, K. Nehal, A. Rossi, M. Rajadhyaksha, e201700348. Lasers Surg. Med. 2016, 48, 133. [12] O. Babalola, A. Mamalis, H. Lev-Tov, J. Jagdeo, Arch. Dermatol. Res. [37] H. Ra, W. Piyawattanametha, E. Gonzalez-Gonzalez, M. J. Mandella, 2014, 306,1. G. S. Kino, O. Solgaard, D. Leake, R. L. Kaspar, A. Oro, C. H. Contag, [13] M. Ulrich, B. T. von, H. Kurzen, T. Dirschka, C. Kellner, E. Sattler, J. Invest. Dermatol. 2011, 131, 1061. C. Berking, J. Welzel, U. Reinhold, Br. J. Dermatol. 2015, 173, 428. [14] N. Khlebtsov, L. Dykman, Chem. Soc. Rev. 2011, 40, 1647. SUPPORTING INFORMATION [15] J. M. Bergen, H. A. von Recum, T. T. Goodman, A. P. Massey, S. H. Pun, Additional supporting information may be found online in Macromol. Biosci. 2006, 6, 506. [16] Y. Akiyama, T. Mori, Y. Katayama, T. Niidome, J. Control. Release 2009, the Supporting Information section at the end of this article. 139, 81. [17] Oxford Centre for Evidence-Based Medicine. http://www.cebm.net/index. aspx?o=5653. Published May 2016. Accessed 2 February 2019. How to cite this article: Ring HC, Israelsen NM, [18] E. S. Flores, M. Cordova, K. Kose, W. Phillips, A. Rossi, K. Nehal, Bang O, Haedersdal M, Mogensen M. Potential of M. Rajadhyaksha, J. Biomed. Opt. 2015, 20, 061103. [19] C. Jonak, H. Skvara, R. Kunstfeld, F. Trautinger, J. A. Schmid, PLoS One contrast agents to enhance in vivo confocal micros- 2011, 6, e23972. copy and optical coherence tomography in dermatol- [20] L. D. Swindle, S. G. Thomas, M. Freeman, P. M. Delaney, J. Invest. ogy: A review. J. Biophotonics. 2019;12:e201800462. Dermatol. 2003, 121, 706. [21] C. Suihko, J. Serup, Skin Res. Technol. 2012, 18, 397. https://doi.org/10.1002/jbio.201800462