- Email: [email protected]
Dermoscopic insight into skin microcirculation – Burn depth assessment
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Dermoscopic insight into skin microcirculation – Burn depth assessment Kyomi Mihara a,b,*, Tomoko Nomiyama c, Koji Masuda c, Hajime Shindo a,b, Maki Yasumi c, Takahiro Sawada c, Kotaro Nagasaki a, Norito Katoh c a
Burns Unit, Nagasaki Hospital, 3-11, Yokogawa-Shinmachi, Nishi-ku, Hiroshima City 733-0013, Japan Department of Dermatology, Nagasaki Hospital, 3-11, Yokogawa-Shinmachi, Nishi-ku, Hiroshima City 733-0013, Japan c Department of Dermatology, Kyoto Prefectural University of Medicine, Graduate School of Medical Science, 465, Kajii-Cho, Kamigyo-ku, Kyoto City 602-8566, Japan b
article info
abstract
Article history:
To investigate the effectiveness of dermoscopic observation of skin microcirculation, the
Accepted 26 August 2015
dermal capillary integrity of burn wounds was evaluated by dermoscopy according to a
Keywords:
burns: SDB, and deep dermal burns: DDB. As the gold standard for comparison, two widely
proposed algorithm that is designed to distinguish burn wounds between superficial dermal Burn depth assessment
accepted endpoints of primary healing within 21 days (SDB) or over 21 days after injury
Dermoscopy
(DDB) were used. A number of dermatologists conducted diagnostic imaging by dermo-
Skin ulcer
scopy. Comparison among polarized noncontact dermoscopy (PNCD), polarized contact
Skin microcirculation
dermoscopy (PCD) and nonpolarized contact dermoscopy (NPD) was also conducted. Images from the three modalities were evaluated for color, pattern and qualitative differences among them. The results of dermoscopy measurements according to the proposed algorithm showed accuracy of 96.7%, sensitivity of 100.0% and specificity of 94.4%. Dermoscopy measurements were significantly more accurate than clinical assessment ( p < 0.05). The recognition of dots increased for NPD, vessels were most clearly observed under PCD and colours tended to be more distinctly recognized under polarized light. Dermoscopy is a useful and simple tool to evaluate not only epidermal and superficial dermal skin components but also the skin microcirculation. # 2015 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
Skin microcirculation is vital to the functioning not only of skin but also of the whole body. Many approaches have been conducted to investigate skin microcirculation from various points of views [1–4]. However, they have not become familiar
technology because of their cost, cumbersome procedure, training requirements and so on. Dermoscopy is now thought to be an essential tool in daily clinical practice for dermatologists. At first, melanin chromospheres in pigmented skin lesions (PSL) were the main interest of dermoscopy studies [5–9]. A spate of reports followed on various skin tumors, and then vascular structures of skin
* Corresponding author at: Department of Dermatology, Nagasaki Hospital, 3-11, Yokogawa-Shinmachi, Nishi-ku, Hiroshima City 733-0013, Japan. Tel.: +81 82 208 5801; fax: +81 82 208 5821. E-mail address: [email protected] (K. Mihara). http://dx.doi.org/10.1016/j.burns.2015.08.032 0305-4179/# 2015 Elsevier Ltd and ISBI. All rights reserved.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
2
burns xxx (2015) xxx–xxx
tumors such as amelanotic melanoma [10], pyogenic granuloma [11], Kaposi’s sarcoma [12], dermatofibrosarcoma protuberans [13], common inflammatory skin diseases [14], psoriasis, seborrheic dermatitis, [15,16] and pityriasis rosea [17], as well as connective tissue diseases [18] were reported to be evaluated effectively by dermoscopy. Dermoscopy has been proved to be an effective tool to evaluate the vascular structure of the superficial dermis. In both skin tumors and inflammatory skin diseases, the epidermis is normally located above the superficial dermis. Therefore, it is unavoidable that dermoscopic images of such conditions are interfered with by the existence of the epidermal layer. In skin ulcers, such as in intermediate-depth burns, the epidermal layer is injured and lacking, so the vascular structure of the superficial dermis can be observed without the interference by the epidermal layer. The burn healing potential is reported to be a function of blood flow to the wound. In the daily medical practice of burn care, burn wound depth is critically important because it determines not only therapeutic strategy but also a patient’s prognosis, including morbidity. Burn depth is effectively assessed by evaluating the dermal capillary integrity [19–24]. Nonpolarized contact (fluid immersion) dermoscopy (NPD) has been a conventional approach and many investigations on dermoscopy have been reported using NPD. Fluid immersion decreases the amount of light reflected at the stratum corneum and visualizes deeper epidermal and dermal structures. Subsequently, advanced techniques enable new dermoscopic observation, such as polarized dermoscopy (PD), which enables dermoscopic observation of a deeper layer than NPD without fluid immersion. Therefore, contact and noncontact dermoscopy with polarization is now available. Deeper structures such as the vasculature can be better observed by PD. Some studies reported the differences in observation by NPD, polarized noncontact dermoscopy (PNCD) and polarized contact dermoscopy (PCD) [25–29]. Additionally, dermoscopy is proved to be useful in the assessment of burn wounds and scars [30,31]. The aim of this study was to evaluate the usefulness of dermoscopy in the evaluation of dermal capillary integrity through burn depth assessment, and to examine the differences in the observation of the vasculature without interference of the epidermis by NPD, PCD and PNCD.
2.
Materials and methods
2.1.
Subjects
This investigation was conducted at the Burns Unit, Nagasaki Hospital. The hospital’s ethical committee approved this study. Patients who presented at the Burns Unit of Nagasaki Hospital with intermediate-depth burns, the depth of which was difficult to assess with the naked eye, were prospectively assessed clinically and dermoscopically. The identification of intermediate-depth burns was conducted by the first author (KM), who was an experienced dermatologist with more than 15 years’ experience of burn care. Patients with burn wounds of a size greater than or equal to 1% and less than 10% of the total body surface area (TBSA), older than 15 years old and who
gave written informed consent were included in the study. Patients with a concomitant illness such as diabetes and other known vascular problems, as well as psychiatric disease, were excluded from the study.
2.2.
Dermoscopy
A handheld dermoscope, ONDEKO DERMOSCOPE EPILITEx8 (Ondeko Corporation, Tokyo, Japan), with a slide switch between non-polarization and polarization was used to conduct dermoscopy measurement over the entire wound because of its simplicity and convenience. Derma9500 (Derma Medical, Inc., Tokyo, Japan), which can be used both with and without a polarization filter, composed of a close-up adapter unit mounted on a Canon Powershot G11 digital camera (Canon, Inc., Tokyo, Japan), was used to take dermoscopic images. At first, the wound was cleaned well under running water, photographed clinically and assessed for its depth clinically, and then covered with thin transparent polyvinylidene chloride film, Kurewrap (Kureha Corporation, Tokyo, Japan). The film was laid on the wound with care not to compress it to avoid the possibility of transmitting infection through the dermoscopy and to alleviate the patients’ haphalgesia by the procedure. The first author (KM), who is experienced in clinical burn practice and burn depth assessment clinically and dermoscopically, took clinical digital photos of burn wounds with a digital camera (Nikon D40; Nikon, Inc., Tokyo, Japan) mounted with a micro-lens (AF-s Micro NIKKOR 60 mm f/ 2.8G ED; Nikon, Inc., Tokyo, Japan) and an electric flash (Sigma EM-140DG; Sigma Corporation, Kanagawa, Japan), and assessed their depth clinically according to the classification of burns by ISBI/WHO [32]. After that, KM conducted dermoscopy measurement over the entire wound through investigation of the dermal capillary integrity. KM carefully investigated the area that was thought to indicate the overall burn depth and took digital photos of the dermoscopic image. At first, the close-up adapter unit with a polarization filter was mounted on a Canon Powershot. Two digital photos were taken while avoiding contacting the wound (PNCD) using an automatic focus. The second or later photos were taken with a dot marked on the polyvinylidene chloride films above the wounds with an oil-based marker pen (MO-120-MC-BK; Zebra Co., Ltd., Tokyo, Japan) to fix the area used in the investigation. The first photo (PNCD) without a marked dot was used for burn depth assessment. The second or later photos (PNCD, PCD, and NPD) with a marked dot were used to evaluate the difference between the images with or without polarization and contact between glass plate and wounds. PCD and NPD images were taken using programmed focus, which was associated with the glass faceplate of the adapter (Fig. 1). After the assessment, each patient received conservative treatment for 21 days after injury. Conservative treatment provided in this study was as follows: after thorough cleaning under running water, topical administration of 0.25% tretinoin tocoferil cream (KYORIN Rimedio Co., Ltd., Ishikawa, Japan), white petrolatum-impregnated gauze and dry gauze.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
3
Fig. 1 – Dermoscopic images via different modalities. 1-1: PNCD, PCD and NPD.
In this study, two widely accepted endpoints of primary healing within 21 days (superficial dermal burn: SDB) or over 21 days after injury (deep dermal burn: DDB) [19,20] were adopted. The wounds with no available healing with conservative treatment on 21 days, indication of excision and skin graft were considered. Besides the first author (KM), the other observers were blinded to the study hypothesis (difference between PNCD, PCD and NPD) and clinical information of the patient in order to avoid possible observer bias and ensure the independence of dermoscopic assessment.
2.3.
Burn depth assessment
As soon as after the investigation of burn wounds, the first photo (PNCD without a dot) was sent to the second and third authors (TN and KMa) as a JPEG attachment separately by KM. At that time, both of them were provided only with the photo and blinded to the clinical information of the patient. TN and KMa were experienced in dermatology, dermoscopy assessment in daily clinical practice and burn depth assessment clinically. KM, TN and KMa reviewed and assessed the image separately according to the proposed algorithm for dermoscopy measurement (Fig. 2). The dermoscopic patterns recognized by all investigators were recorded directly. When there was any disagreement in the recognized patterns, the patterns recognized by two observers were used. When the observers recognized three different patterns, a consensus about the pattern was reached among the three observers and was used. Finally, all patterns were approved by the three authors (KM, TN and KMa).
2.4. Examination of effects by experience as a dermatologist An experienced dermatologist (HS) and two residents (MY and TS) were provided with the NCPD without dot photos in JPEG format, and conducted the burn depth assessment according to the algorithm. The accuracies of dermoscopic assessments by the three experienced dermatologists (TN, KMa and HS) and residents (MY and TS) were examined.
2.5.
Comparison between PNCD, PCD and NPD
Three images (PNCD, PCD and NPD) of every patient were randomly arranged by using RAND function (Microsoft Excel 2010, Microsoft Corporation, Redmond, WA, USA) in one image of a Powerpoint presentation without information on the dermoscopic mode (PNCD, PCD and NPD), and provided to the second, third and fourth authors (TN, KMa and HS) separately (Figs. 3 and 4). They were all experienced in dermatology and dermoscopy assessment in daily clinical practice. The provided images were evaluated for the presence or absence of each dermoscopic feature described in the algorithm of this study. Subsequently, they evaluated the qualitative difference between PNCD, PCD and NPD described in the list of dermoscopic features. The items of the list were extracted and modified from the literature [28] for this study and are provided below. 1. 2. 3. 4. 5. 6.
Vessels’ visibility. More red. More white. Dots partially disappear or become blurred. Reticular becomes blurred. Brown structure looks dark.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
4
burns xxx (2015) xxx–xxx
Intermediate-depth burns 29(18/11) Diffuse hemorrhage(-)
30 (18/12)* 1 (0/1) Diffuse hemorrhage(+)
6 (6/0)
23 (12/11) Dots(-)
Dots(+)
4(0/4) Reticular(-)
19(12/7) Reticular(+)
11(11/0) Fine reticular
8 (1/7) Coarse reticular
17 (17/0) Superficial partial thickness
13(1/12) Deep partial thickness
Fig. 2 – Proposed algorithm and results of assessment by dermoscopy according to the proposed algorithm. *Number of cases (SDB/DDB: clinical outcome).
Fig. 3 – Three images of every patient were randomly arranged in one image: SDB. The microvascular structure is clearly observed.
They ranked the images in order of the visibility of each item.
2.6.
Dermoscopic algorithm used in this study
The algorithm used in this study is based on the microanatomy of skin microcirculation and physiological activity. A dot is
denoted as the top of a capillary loop, and reticular area is denoted as the superficial plexus. The activity determines the function of the area supplied. Therefore, a ‘diffuse hemorrhage’ pattern was set for the first gate because it is noticeable and makes it difficult to detect the background pattern. Moreover, it indicates the rupture of the vessel wall and severe damage to the dermal capillary. A wound with diffuse hemorrhage is defined as a deep partial-thickness burn. The
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
5
Fig. 4 – Three images of every patient were randomly arranged in one image: DDB. The microvascular structure is damaged and blurred.
top of a capillary loop is seen as a ‘dot’ pattern dermoscopically and is the shallowest component of the dermal capillary. A recognizable dot pattern indicates the activity of the capillary loop, and a wound with a dot pattern is defined as a superficial-thickness burn. Under the capillary loop, the superficial plexus was observed as a ‘reticular’ pattern dermoscopically. Capillary loop rises up from the superficial plexus. A vigorous, fine reticular pattern indicates the activity of the superficial plexus. On the other hand, a damaged, coarse reticular pattern indicates severe damage of the superficial plexus. The wounds with a fine reticular pattern and with a coarse reticular pattern are defined as superficial and deep partial-thickness burns,
respectively. Significant heat energy can destroy the whole physiological structure of the microvascular system, so wounds with no particular pattern are deep partial-thickness burns.
2.7.
Fisher’s exact test was used for comparisons among groups. A p-value of less than 0.05 was defined as significant. The interobserver agreements were evaluated by Kappa coefficient. Data analysis was performed using JMP statistical software, version 9 (SAS Institute, Inc., Cary, NC, USA).
3. Table 1 – Patient profile.
Age, years TBSA, % Assessment time, h Clinical outcome Location Lower extremity Upper extremity Trunk Cause of injury Scald Flame Contact Steam
Mean SD
Range
46.4 15.7 2.0 1.7 67.1 46.7 SDB 18
15–75 1.0–9.0 4.5–166.5 DDB 12 14 10 6 18 7 3 2
Statistical analysis
Results
The characteristics of these patients are shown in Table 1. Thirty burn wounds and a total of 120 images were included in the study (30 PNCD without dot images for burn depth assessment; and 30 PNCD, 30 PCD and 30 NPD with dot images for comparison among PNCD, PCD and NPD). This study was conducted from 2012 to 2013; the duration of follow-up was approximately 5.5 months. No subjects dropped out of the study. In previous clinical studies of burn depth assessment, inclusion of approximately 30 cases has been thought to be appropriate [19,20].
3.1.
Burn depth assessment
The burns assessed as diffuse hemorrhage included one DDB, ‘dots’ included 6 SDB, ‘fine reticular’ included 11 SDB, ‘coarse
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
6
burns xxx (2015) xxx–xxx
Table 2 – Results of assessment by dermoscopy according to the proposed algorithm. Dermoscopy
Case
Clinical assessment
Accurate
Inaccurate
Accurate
29
1
23
Accuracy, % Sensitivity, % Specificity, %
Inaccurate 7
96.7 100.0 94.4
0.0227
76.7 85.7 73.9
KM Accurate
p-Value
TN
KMa
Inaccurate
Accurate
Inaccurate
Accurate
Inaccurate
1
21
9
22
8
29
Interobserver agreements
SE
k-Value
Burn depth KM and KMa KM and TN KMa and TN Each category KM and KMa KM and TN KMa and TN
0.389 0.403 0.247
0.166 0.174 0.168
0.421 0.450 0.331
0.119 0.121 0.118
Effects by experience as a dermatologist Experienced TN
KMa
Accurate 21
Resident HS
Inaccurate
Accurate
Inaccurate
9
22
8
21/9
Accurate
MY (resident1)
Inaccurate
Accurate
20 10 Average (accurate/inaccurate)
22
TS (resident2)
Inaccurate
Accurate
Inaccurate
8
18 20/10
12
Accuracy, % = number of burns that were assessed as DDB and failed to heal within 21 days and number of burns that were assessed as SDB and healed within 21 days/total number of burns assessed in this study (30). Sensitivity, % = number of burns that were assessed as DDB and failed to heal within 21 days after injury/number of burns that failed to heal within 21 days after injury; specificity, % = number of burns that were assessed as SDB and healed within 21 days after injury/number of burns that healed within 21 days after injury.
reticular’ included 1 SDB and 7 DDB, and the burns that did not show a ‘reticular’ pattern included 4 DDB (Fig. 2). The results of measurements according to the proposed algorithm showed an accuracy of 96.7% (29 accurate vs. 1 inaccurate). The sensitivity for DDB was 100.0% and its specificity was 94.4%. The results of clinical assessment showed an accuracy of 76.7% (23 accurate vs. 7 inaccurate). The accuracies differ significantly between the dermoscopic and clinical assessment. The interobserver agreements on the burn depth assessment (SDB or DDB) and on each category of the algorithm of KM and KMa, KM and TN, and KMa and TN are shown in Table 2, which showed fair to moderate agreements [33].
3.2.
Effects by experience as a dermatologist
The assessment by experienced dermatologists (TN, KMa and HS) and residents (MY and TS) according to the algorithm showed accuracies of 70.0% (21 accurate vs. 9 inaccurate), 73.3% (22 accurate vs. 8 inaccurate), 66.7% (20 accurate vs. 10 inaccurate), 73.3% (22 accurate vs. 8 inaccurate), and 60.0% ((18 accurate vs. 12 inaccurate), respectively. The average accuracies of each group (three experienced and two residents) were 70.0% and 66.7%, respectively (Table 2).
3.3.
Comparison among PNCD, PCD and NPD
Dots were most observed under NPD. Reticular area was observed equally under the three modalities. Hemorrhage was the most observed under PNCD and NPD, and less under PCD (Table 3). To show the comprehensive characteristics of each modality, the ranks and the total scores are shown in Table 3. The methodology for calculating the scores was that the modality that was ranked first was given 2 points, that ranked as second was given 1 point and that ranked as third was given no score. Vessels’ visibility: PCD images were ranked first the most and also got the highest total score. More red: PNCD images were ranked first the most and PCD images got the highest total score. More white: PNCD images were ranked first the most and NPD images got the highest total score. Dots partially disappear or become blurred: PNCD images were ranked first the most and got the highest total score. Reticular becomes blurred: PNCD images were ranked first the most and got the highest total score. Brown structure looks dark: NPD images were ranked first the most and PCD images got the highest total score. The interobserver agreements on each category of KMa and SH, KMa and TN, and SH and TN are shown in Table 3, which showed slight to moderate agreements [33].
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
Table 3 – Results of the comparison between PNCD, PCD and NPD.
4.
7
Discussion
The presence or absence of each dermoscopic feature
Dot Reticular Hemorrhage
PNCD
PCD
NPD
26/90 61/90 31/90
28/90 61/90 29/90
35/90 61/90 31/90
Qualitative differences: ranks
Vessels’ visibility
More red
More white
Dots partially disappear or become blurred Reticular become blurred
Brown structure look dark
Rank
PNCD
PCD
NPD
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
23 21 46 36 22 32 38 16 36 38 27 20 44 22 22 6 8 11
34 39 17 26 43 21 21 41 28 19 39 27 18 39 31 9 10 6
33 30 27 28 25 37 31 33 26 28 19 38 26 27 35 10 7 8
Qualitative differences: total scores
Vessels’ visibility More red More white Dots partially disappear or become blurred Reticular become blurred Brown structure look dark
Interobserver agreements
PNCD
PCD
NPD
67 94 92 103
107 95 83 77
96 81 95 75
110
75
79
20
28
k-Value
Vessels’ visibility 0.239 KMa and SH 0.182 TN and SH 0.252 KMa and TN More red 0.135 KMa and SH 0.170 TN and SH 0.316 KMa and TN More white 0.312 KMa and SH 0.270 TN and SH 0.547 KMa and TN Dots partially disappear or become blurred 0.040 KMa and SH 0.034 TN and SH 0.044 KMa and TN Reticular become blurred 0.169 KMa and SH 0.191 TN and SH KMa and TN 0.170 Brown structure look dark 0.400 KMa and SH TN and SH 0.226 0.346 KMa and TN
27
SE
0.094 0.108 0.092 0.084 0.105 0.104 0.104 0.103 0.110 0.085 0.086 0.083 0.097 0.112 0.092 0.112 0.087 0.069
Burn depth assessment is the main task at the beginning of burn care. The strategy of burn care and the prognosis of the patient are determined by burn depth. Although clinical assessment is the most widely conducted, its accuracy is reportedly limited up to about 75%. Thus, whether blood flow into the burn wound, especially into the superficial dermis, is active or not has been used to assess the burn depth more accurately. A number of procedures are used to evaluate dermal capillary integrity [19–24] however, they are not as versatile as dermoscopy. In this study, intermediate burn depth that is difficult to assess by the naked eye was significantly accurately assessed by dermoscopy. The accuracy of dermoscopic burn depth assessment by the consensus of three experienced dermatologists (KM, TN and KMa was 96.7%). The accuracy of dermoscopic burn depth assessment by KM was 96.7%, by TN was 70.0% and by KMa was 73.3%. Kittler et al. [9] reported that the diagnostic accuracy of melanoma with dermoscopy significantly depended on the experience of the examiners, and the diagnostic performance of dermoscopy improved by making a consensus of a group of examiners. The same tendency was found in this study; KM, who is experienced in burn wound practice and burn depth assessment clinically and dermoscopically, could assess the burn depth with higher accuracy than the other examiners, and the diagnostic performance of dermoscopy improved upon a consensus being reached by the three examiners. TN and KMa are experienced in dermatology and dermoscopic observation in daily clinical practice; however, they are not experienced in burn depth assessment by dermoscopy. The diagnostic accuracies of dermoscopic burn depth assessment were thought to be equivalent among the two groups whose experience as dermatologists differed (experienced and residents). Therefore, the diagnostic procedure in this study is thought to be simple even for a non-experienced observer. At the time of introduction of dermoscopy, only NPD was available. Then, PD was introduced into clinical use, so it became possible to observe deeper structures of the skin without the use of either a liquid interface or direct contact with the instrument. Therefore, three different modalities (PNCD, PCD and NPD) are now available in daily clinical practice. It has been reported that there were some differences among the images of these three modalities [25–29]. Benvenuto-Andrade et al. [28] evaluated the differences among them. They reported that most dermoscopic structures and patterns, except milialike cyst and peppering, were consistently recognized for the three different modalities. PD blocked the superficial component of light, so the visualization of superficial components such as milialike cysts was attenuated. They also noticed significant changes in quantitative assessment. Most lesions appeared darker under polarized light, and a more reddish appearance was recognized under PNCD compared with NPD. They concluded that the capabilities of the three different modalities were not equivalent but complementary. In this study, some differences among the three different modalities were also recognized. The recognition of dots
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
8
burns xxx (2015) xxx–xxx
increased for NPD compared with PNCD and PCD. This is thought to be because dots denoted the top of a capillary loop, which is the shallowest component of the capillary. Light from the superficial component was blocked on PD, so dots were less frequently recognized with polarized light. On the other hand, recognitions of reticular areas and hemorrhage were almost consistent among the three imaging modalities. Vessels were the most clearly observed under PCD. In the literature [26,28,29], it is reported that the vasculature was better visualized by PD than by NPD because deeper structures were better visualized by polarized light. The dermoscopic observation in this study enabled the vascular structure of the superficial dermis to be observed without interference by the epidermal layer; however, polarized light even made a difference in the visibility of vessels. This is partly thought to be because of the differences in the observation of colors and shapes. A more reddish appearance is reported to be recognized under PNCD than under NPD [28]. In this study, colors also tended to be more distinctly recognized under polarized light. Light colors, such as red and white, were more distinctly recognized under PNCD and PCD (PNCD images were ranked first the most in these two recognitions and PCD images got the highest total score in the recognition of red). NPD images were ranked first the most in the recognition of a darker color, brown. NPD offers increased illumination and resolution compared with PD, and colors under non-polarized light are seen sharper and less distorted than with polarized light [34]. The lesion edge seems jagged and not so sharp under polarized light, and the evaluation of shape asymmetry prevails under nonpolarized light [27]. Blurred recognitions of dots and reticular patterns were the most clearly observed under PNCD in this study. This study has the limitation that biopsy was not conducted to validate the dermoscopic findings. Biopsy has been considered as the ‘gold standard’; however, biopsy of burn wound provides only a snapshot of the level of microvascular injury and may leave an additional scar [21]. The best outcome of burn wounds is clinical wound healing, as demonstrated by the present and previous studies [19,20]. Dermoscopy opened the window through which simple and reliable inspection into the microanatomy of skin structures became possible, even in daily clinical dermatology. With this tiny and simple device, not only epidermis and superficial dermis but also deeper structures of the skin could be observed under certain states, such as intermediate-depth burns. It is to be remembered that dermoscopic inspection could be affected by its modalities, and the difference among modalities should be borne in mind.
5.
Conclusion
Dermoscopy is a useful and simple tool to evaluate not only epidermal and superficial dermal skin components such as melanin chromospheres, horny cells and interstitium, but also skin microcirculation. In evaluating the microcirculation, differences among the three different dermoscopic modalities should be borne in mind.
Acknowledgement We are indebted to Prof. Junko Tanaka, Ph.D. and Dr. Tomoyuki Akita, Ph.D., Department of Epidemiology, Infectious Disease Control and Prevention, Hiroshima University Graduate School of Biomedical Sciences, for their useful advices on the interpretation of data and statistical analysis used in this study.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. burns.2015.08.032.
references
[1] Awan ZA, Wester T, Kvernebo K. Human microvascular imaging: a review of skin and tongue videomicroscopy techniques and analyzing variables. Clin Physiol Funct Imaging 2010;30:79–88. [2] Mahe´ G, Humeau-Heurtier A, Durand S, Leftheriotis G, Abraham P. Assessment of skin microvascular function and dysfunction with laser speckle contrast imaging. Circ Cardiovasc Imaging 2012;5:155–63. [3] Chao CY, Cheing GL. Microvascular dysfunction in diabetic foot disease and ulceration. Diabetes Metab Res Rev 2009;25:604–14. [4] De Backer D, Ospina-Tascon G, Salgado D, Favory R, Creteur J, Vincent JL. Monitoring the microcirculation in the critically ill patient: current methods and future approaches. Intensive Care Med 2010;36:1813–25. [5] Pehamberger H, Steiner A, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. I. Pattern analysis of pigmented skin lesions. J Am Acad Dermatol 1987;17:571–83. [6] Steiner A, Pehamberger H, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. II. Diagnosis of small pigmented skin lesions and early detection of malignant melanoma. J Am Acad Dermatol 1987;17:584–91. [7] Westerhoff K, McCarthy WH, Menzies SW. Increase in the sensitivity for melanoma diagnosis by primary care physicians using skin surface microscopy. Br J Dermatol 2000;143:1016–20. [8] Argenziano G, Soyer HP. Dermoscopy of pigmented skin lesions – a valuable tool for early diagnosis of melanoma. Lancet Oncol 2001;2:443–9. [9] Kittler H, Pehamberger H, Wolff K, Binder M. Diagnostic accuracy of dermoscopy. Lancet Oncol 2002;3:159–65. [10] Menzies SW, Kreusch J, Byth K, Pizzichetta MA, Marghoob A, Braun R, et al. Dermoscopic evaluation of amelanotic and hypomelanotic melanoma. Arch Dermatol 2008;144:1120–7. [11] Zaballos P, Carulla M, Ozdemir F, Zalaudek I, Ban˜uls J, Llamgbrich A, et al. Dermoscopy of pyogenic granuloma: a morphological study. Br J Dermatol 2010;163:1229–37. [12] Cheng ST, Ke CL, Lee CH, Wu CS, Chen GS, Hu SC. Rainbow pattern in Kaposi’s sarcoma under polarized dermoscopy: a dermoscopic pathological study. Br J Dermatol 2009;160:801–9. [13] Bernard J, Poulalhon N, Argenziano G, Debarbieux S, Dalle S, Thomas L. Dermoscopy of dermatofibrosarcoma
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
[14]
[15]
[16]
[17]
[18] [19]
[20]
[21]
[22]
[23]
[24]
protuberans: a study of 15 cases. Br J Dermatol 2013;169: 85–90. Lallas A, Argenziano G, Apalla Z, Gourhant JY, Zaballos P, Di Lernia V, et al. Dermoscopic patterns of common facial inflammatory skin diseases. J Eur Acad Dermatol Venereol 2014;28:609–14. Va´zquez-Lo´pez F, Manjo´n-Haces JA, Maldonado-Seral C, Raya-Aguado C, Pe´rez-Oliva N, Marghoob AA. Dermoscopic features of plaque psoriasis and lichen planus: new observations. Dermatology 2003;207:151–6. Kim GW, Jung HJ, Ko HC, Kim MB, Lee WJ, Lee SJ, et al. Dermoscopy can be useful in differentiating scalp psoriasis from seborrheic dermatitis. Br J Dermatol 2011;164:652–6. Lallas A, Kyrgidis A, Tzellos TG, Apalla Z, Karakyriou E, Karatolias A, et al. Accuracy of dermoscopic criteria for the diagnosis of psoriasis, dermatitis, lichen planus and pityriasis rosea. Br J Dermatol 2012;166:1198–205. Hasegawa M. Dermoscopy findings of nail fold capillaries in connective tissue diseases. J Dermatol 2011;38:66–70. McGill DJ, Sørensen K, MacKay IR, Taggart I, Watson SB. Assessment of burn depth: a prospective, blinded comparison of laser Doppler imaging and videomicroscopy. Burns 2007;33:833–42. Hoeksema H, Van de Sijpe K, Tondu T, Hamdi M, Van Landuyt K, Blondeel. et al. Accuracy of early burn depth assessment by laser Doppler imaging on different days post burn. Burns 2009;35:36–45. Monstrey S, Hoeksema H, Verbelen J, Pirayesh A, Blondeel P. Assessment of burn depth and burn wound healing potential. Burns 2008;34:761–9. Jaskille AD, Shupp JW, Jordan MH, Jeng JC. Critical review of burn depth assessment techniques: Part I. Historical review. J Burn Care Res 2009;30:937–47. Mihara K, Shindo H, Ohtani M, Nagasaki K, Nakashima R, Katoh N, et al. Early burn depth assessment of local burns by videomicroscopy: 24 h after injury is a critical time point. Burns 2011;37:986–93. Mihara K, Shindo H, Mihara H, Ohtani M, Nagasaki K, Katoh N. Early burn depth assessment of local burns by videomicroscopy: a novel proposed classification. Burns 2012;38:371–7.
[25] Pan Y, Gareau DS, Scope A, Rajadhyaksha M, Mullani NA, Marghoob AA. Polarized and nonpolarized dermoscopy: the explanation for the observed differences. Arch Dermatol 2008;144:828–9. [26] Agero AL, Taliercio S, Dusza SW, Salaro C, Chu P, Marghoob AA. Conventional and polarized dermoscopy features of dermatofibroma. Arch Dermatol 2006;142:1431–7. [27] Pellacani G, Seidenari S. Comparison between morphological parameters in pigmented skin lesion images acquired by means of epiluminescence surface microscopy and polarized-light videomicroscopy. Clin Dermatol 2002;20:222–7. [28] Benvenuto-Andrade C, Dusza SW, Agero AL, Scope A, Rajadhyaksha M, Halpern AC, et al. Differences between polarized light dermoscopy and immersion contact dermoscopy for the evaluation of skin lesions. Arch Dermatol 2007;143:329–38. [29] Liebman TN, Jaimes-Lopez N, Balagula Y, Rabinovitz HS, Wang SQ, Dusza SW, et al. Dermoscopic features of basal cell carcinomas: differences in appearance under nonpolarized and polarized light. Dermatol Surg 2012;38:392–9. [30] Campanati A, Savelli A, Sandroni L, Marconi B, Giuliano A, Giuliodori K, et al. Effect of allium cepa-allantoinpentaglycan gel on skin hypertrophic scars: clinical and video-capillaroscopic results of an open-label, controlled, nonrandomized clinical trial. Dermatol Surg 2010;36:1439–44. [31] Campanati A, De Blasio S, Giuliano A, Ganzetti G, Giuliodori K, Pecora T, et al. Topical ozonated oil versus hyaluronic gel for the treatment of partial- to full-thickness seconddegree burns: a prospective, comparative, single-blind, non-randomised, controlled clinical trial. Burns 2013;39:1178–83. [32] Latarjet J. A simple guide to burn treatment. International Society for Burn Injuries in collaboration with the World Health Organization. Burns 1995;21:221–5. [33] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [34] Marghoob AA, Swindle LD, Moricz CZ, Sanchez Negron FA, Slue B, Halpern AC, et al. Instruments and new technologies for the in vivo diagnosis of melanoma. J Am Acad Dermatol 2003;49:777–97.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
9
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/burns
Dermoscopic insight into skin microcirculation – Burn depth assessment Kyomi Mihara a,b,*, Tomoko Nomiyama c, Koji Masuda c, Hajime Shindo a,b, Maki Yasumi c, Takahiro Sawada c, Kotaro Nagasaki a, Norito Katoh c a
Burns Unit, Nagasaki Hospital, 3-11, Yokogawa-Shinmachi, Nishi-ku, Hiroshima City 733-0013, Japan Department of Dermatology, Nagasaki Hospital, 3-11, Yokogawa-Shinmachi, Nishi-ku, Hiroshima City 733-0013, Japan c Department of Dermatology, Kyoto Prefectural University of Medicine, Graduate School of Medical Science, 465, Kajii-Cho, Kamigyo-ku, Kyoto City 602-8566, Japan b
article info
abstract
Article history:
To investigate the effectiveness of dermoscopic observation of skin microcirculation, the
Accepted 26 August 2015
dermal capillary integrity of burn wounds was evaluated by dermoscopy according to a
Keywords:
burns: SDB, and deep dermal burns: DDB. As the gold standard for comparison, two widely
proposed algorithm that is designed to distinguish burn wounds between superficial dermal Burn depth assessment
accepted endpoints of primary healing within 21 days (SDB) or over 21 days after injury
Dermoscopy
(DDB) were used. A number of dermatologists conducted diagnostic imaging by dermo-
Skin ulcer
scopy. Comparison among polarized noncontact dermoscopy (PNCD), polarized contact
Skin microcirculation
dermoscopy (PCD) and nonpolarized contact dermoscopy (NPD) was also conducted. Images from the three modalities were evaluated for color, pattern and qualitative differences among them. The results of dermoscopy measurements according to the proposed algorithm showed accuracy of 96.7%, sensitivity of 100.0% and specificity of 94.4%. Dermoscopy measurements were significantly more accurate than clinical assessment ( p < 0.05). The recognition of dots increased for NPD, vessels were most clearly observed under PCD and colours tended to be more distinctly recognized under polarized light. Dermoscopy is a useful and simple tool to evaluate not only epidermal and superficial dermal skin components but also the skin microcirculation. # 2015 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
Skin microcirculation is vital to the functioning not only of skin but also of the whole body. Many approaches have been conducted to investigate skin microcirculation from various points of views [1–4]. However, they have not become familiar
technology because of their cost, cumbersome procedure, training requirements and so on. Dermoscopy is now thought to be an essential tool in daily clinical practice for dermatologists. At first, melanin chromospheres in pigmented skin lesions (PSL) were the main interest of dermoscopy studies [5–9]. A spate of reports followed on various skin tumors, and then vascular structures of skin
* Corresponding author at: Department of Dermatology, Nagasaki Hospital, 3-11, Yokogawa-Shinmachi, Nishi-ku, Hiroshima City 733-0013, Japan. Tel.: +81 82 208 5801; fax: +81 82 208 5821. E-mail address: [email protected] (K. Mihara). http://dx.doi.org/10.1016/j.burns.2015.08.032 0305-4179/# 2015 Elsevier Ltd and ISBI. All rights reserved.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
2
burns xxx (2015) xxx–xxx
tumors such as amelanotic melanoma [10], pyogenic granuloma [11], Kaposi’s sarcoma [12], dermatofibrosarcoma protuberans [13], common inflammatory skin diseases [14], psoriasis, seborrheic dermatitis, [15,16] and pityriasis rosea [17], as well as connective tissue diseases [18] were reported to be evaluated effectively by dermoscopy. Dermoscopy has been proved to be an effective tool to evaluate the vascular structure of the superficial dermis. In both skin tumors and inflammatory skin diseases, the epidermis is normally located above the superficial dermis. Therefore, it is unavoidable that dermoscopic images of such conditions are interfered with by the existence of the epidermal layer. In skin ulcers, such as in intermediate-depth burns, the epidermal layer is injured and lacking, so the vascular structure of the superficial dermis can be observed without the interference by the epidermal layer. The burn healing potential is reported to be a function of blood flow to the wound. In the daily medical practice of burn care, burn wound depth is critically important because it determines not only therapeutic strategy but also a patient’s prognosis, including morbidity. Burn depth is effectively assessed by evaluating the dermal capillary integrity [19–24]. Nonpolarized contact (fluid immersion) dermoscopy (NPD) has been a conventional approach and many investigations on dermoscopy have been reported using NPD. Fluid immersion decreases the amount of light reflected at the stratum corneum and visualizes deeper epidermal and dermal structures. Subsequently, advanced techniques enable new dermoscopic observation, such as polarized dermoscopy (PD), which enables dermoscopic observation of a deeper layer than NPD without fluid immersion. Therefore, contact and noncontact dermoscopy with polarization is now available. Deeper structures such as the vasculature can be better observed by PD. Some studies reported the differences in observation by NPD, polarized noncontact dermoscopy (PNCD) and polarized contact dermoscopy (PCD) [25–29]. Additionally, dermoscopy is proved to be useful in the assessment of burn wounds and scars [30,31]. The aim of this study was to evaluate the usefulness of dermoscopy in the evaluation of dermal capillary integrity through burn depth assessment, and to examine the differences in the observation of the vasculature without interference of the epidermis by NPD, PCD and PNCD.
2.
Materials and methods
2.1.
Subjects
This investigation was conducted at the Burns Unit, Nagasaki Hospital. The hospital’s ethical committee approved this study. Patients who presented at the Burns Unit of Nagasaki Hospital with intermediate-depth burns, the depth of which was difficult to assess with the naked eye, were prospectively assessed clinically and dermoscopically. The identification of intermediate-depth burns was conducted by the first author (KM), who was an experienced dermatologist with more than 15 years’ experience of burn care. Patients with burn wounds of a size greater than or equal to 1% and less than 10% of the total body surface area (TBSA), older than 15 years old and who
gave written informed consent were included in the study. Patients with a concomitant illness such as diabetes and other known vascular problems, as well as psychiatric disease, were excluded from the study.
2.2.
Dermoscopy
A handheld dermoscope, ONDEKO DERMOSCOPE EPILITEx8 (Ondeko Corporation, Tokyo, Japan), with a slide switch between non-polarization and polarization was used to conduct dermoscopy measurement over the entire wound because of its simplicity and convenience. Derma9500 (Derma Medical, Inc., Tokyo, Japan), which can be used both with and without a polarization filter, composed of a close-up adapter unit mounted on a Canon Powershot G11 digital camera (Canon, Inc., Tokyo, Japan), was used to take dermoscopic images. At first, the wound was cleaned well under running water, photographed clinically and assessed for its depth clinically, and then covered with thin transparent polyvinylidene chloride film, Kurewrap (Kureha Corporation, Tokyo, Japan). The film was laid on the wound with care not to compress it to avoid the possibility of transmitting infection through the dermoscopy and to alleviate the patients’ haphalgesia by the procedure. The first author (KM), who is experienced in clinical burn practice and burn depth assessment clinically and dermoscopically, took clinical digital photos of burn wounds with a digital camera (Nikon D40; Nikon, Inc., Tokyo, Japan) mounted with a micro-lens (AF-s Micro NIKKOR 60 mm f/ 2.8G ED; Nikon, Inc., Tokyo, Japan) and an electric flash (Sigma EM-140DG; Sigma Corporation, Kanagawa, Japan), and assessed their depth clinically according to the classification of burns by ISBI/WHO [32]. After that, KM conducted dermoscopy measurement over the entire wound through investigation of the dermal capillary integrity. KM carefully investigated the area that was thought to indicate the overall burn depth and took digital photos of the dermoscopic image. At first, the close-up adapter unit with a polarization filter was mounted on a Canon Powershot. Two digital photos were taken while avoiding contacting the wound (PNCD) using an automatic focus. The second or later photos were taken with a dot marked on the polyvinylidene chloride films above the wounds with an oil-based marker pen (MO-120-MC-BK; Zebra Co., Ltd., Tokyo, Japan) to fix the area used in the investigation. The first photo (PNCD) without a marked dot was used for burn depth assessment. The second or later photos (PNCD, PCD, and NPD) with a marked dot were used to evaluate the difference between the images with or without polarization and contact between glass plate and wounds. PCD and NPD images were taken using programmed focus, which was associated with the glass faceplate of the adapter (Fig. 1). After the assessment, each patient received conservative treatment for 21 days after injury. Conservative treatment provided in this study was as follows: after thorough cleaning under running water, topical administration of 0.25% tretinoin tocoferil cream (KYORIN Rimedio Co., Ltd., Ishikawa, Japan), white petrolatum-impregnated gauze and dry gauze.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
3
Fig. 1 – Dermoscopic images via different modalities. 1-1: PNCD, PCD and NPD.
In this study, two widely accepted endpoints of primary healing within 21 days (superficial dermal burn: SDB) or over 21 days after injury (deep dermal burn: DDB) [19,20] were adopted. The wounds with no available healing with conservative treatment on 21 days, indication of excision and skin graft were considered. Besides the first author (KM), the other observers were blinded to the study hypothesis (difference between PNCD, PCD and NPD) and clinical information of the patient in order to avoid possible observer bias and ensure the independence of dermoscopic assessment.
2.3.
Burn depth assessment
As soon as after the investigation of burn wounds, the first photo (PNCD without a dot) was sent to the second and third authors (TN and KMa) as a JPEG attachment separately by KM. At that time, both of them were provided only with the photo and blinded to the clinical information of the patient. TN and KMa were experienced in dermatology, dermoscopy assessment in daily clinical practice and burn depth assessment clinically. KM, TN and KMa reviewed and assessed the image separately according to the proposed algorithm for dermoscopy measurement (Fig. 2). The dermoscopic patterns recognized by all investigators were recorded directly. When there was any disagreement in the recognized patterns, the patterns recognized by two observers were used. When the observers recognized three different patterns, a consensus about the pattern was reached among the three observers and was used. Finally, all patterns were approved by the three authors (KM, TN and KMa).
2.4. Examination of effects by experience as a dermatologist An experienced dermatologist (HS) and two residents (MY and TS) were provided with the NCPD without dot photos in JPEG format, and conducted the burn depth assessment according to the algorithm. The accuracies of dermoscopic assessments by the three experienced dermatologists (TN, KMa and HS) and residents (MY and TS) were examined.
2.5.
Comparison between PNCD, PCD and NPD
Three images (PNCD, PCD and NPD) of every patient were randomly arranged by using RAND function (Microsoft Excel 2010, Microsoft Corporation, Redmond, WA, USA) in one image of a Powerpoint presentation without information on the dermoscopic mode (PNCD, PCD and NPD), and provided to the second, third and fourth authors (TN, KMa and HS) separately (Figs. 3 and 4). They were all experienced in dermatology and dermoscopy assessment in daily clinical practice. The provided images were evaluated for the presence or absence of each dermoscopic feature described in the algorithm of this study. Subsequently, they evaluated the qualitative difference between PNCD, PCD and NPD described in the list of dermoscopic features. The items of the list were extracted and modified from the literature [28] for this study and are provided below. 1. 2. 3. 4. 5. 6.
Vessels’ visibility. More red. More white. Dots partially disappear or become blurred. Reticular becomes blurred. Brown structure looks dark.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
4
burns xxx (2015) xxx–xxx
Intermediate-depth burns 29(18/11) Diffuse hemorrhage(-)
30 (18/12)* 1 (0/1) Diffuse hemorrhage(+)
6 (6/0)
23 (12/11) Dots(-)
Dots(+)
4(0/4) Reticular(-)
19(12/7) Reticular(+)
11(11/0) Fine reticular
8 (1/7) Coarse reticular
17 (17/0) Superficial partial thickness
13(1/12) Deep partial thickness
Fig. 2 – Proposed algorithm and results of assessment by dermoscopy according to the proposed algorithm. *Number of cases (SDB/DDB: clinical outcome).
Fig. 3 – Three images of every patient were randomly arranged in one image: SDB. The microvascular structure is clearly observed.
They ranked the images in order of the visibility of each item.
2.6.
Dermoscopic algorithm used in this study
The algorithm used in this study is based on the microanatomy of skin microcirculation and physiological activity. A dot is
denoted as the top of a capillary loop, and reticular area is denoted as the superficial plexus. The activity determines the function of the area supplied. Therefore, a ‘diffuse hemorrhage’ pattern was set for the first gate because it is noticeable and makes it difficult to detect the background pattern. Moreover, it indicates the rupture of the vessel wall and severe damage to the dermal capillary. A wound with diffuse hemorrhage is defined as a deep partial-thickness burn. The
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
5
Fig. 4 – Three images of every patient were randomly arranged in one image: DDB. The microvascular structure is damaged and blurred.
top of a capillary loop is seen as a ‘dot’ pattern dermoscopically and is the shallowest component of the dermal capillary. A recognizable dot pattern indicates the activity of the capillary loop, and a wound with a dot pattern is defined as a superficial-thickness burn. Under the capillary loop, the superficial plexus was observed as a ‘reticular’ pattern dermoscopically. Capillary loop rises up from the superficial plexus. A vigorous, fine reticular pattern indicates the activity of the superficial plexus. On the other hand, a damaged, coarse reticular pattern indicates severe damage of the superficial plexus. The wounds with a fine reticular pattern and with a coarse reticular pattern are defined as superficial and deep partial-thickness burns,
respectively. Significant heat energy can destroy the whole physiological structure of the microvascular system, so wounds with no particular pattern are deep partial-thickness burns.
2.7.
Fisher’s exact test was used for comparisons among groups. A p-value of less than 0.05 was defined as significant. The interobserver agreements were evaluated by Kappa coefficient. Data analysis was performed using JMP statistical software, version 9 (SAS Institute, Inc., Cary, NC, USA).
3. Table 1 – Patient profile.
Age, years TBSA, % Assessment time, h Clinical outcome Location Lower extremity Upper extremity Trunk Cause of injury Scald Flame Contact Steam
Mean SD
Range
46.4 15.7 2.0 1.7 67.1 46.7 SDB 18
15–75 1.0–9.0 4.5–166.5 DDB 12 14 10 6 18 7 3 2
Statistical analysis
Results
The characteristics of these patients are shown in Table 1. Thirty burn wounds and a total of 120 images were included in the study (30 PNCD without dot images for burn depth assessment; and 30 PNCD, 30 PCD and 30 NPD with dot images for comparison among PNCD, PCD and NPD). This study was conducted from 2012 to 2013; the duration of follow-up was approximately 5.5 months. No subjects dropped out of the study. In previous clinical studies of burn depth assessment, inclusion of approximately 30 cases has been thought to be appropriate [19,20].
3.1.
Burn depth assessment
The burns assessed as diffuse hemorrhage included one DDB, ‘dots’ included 6 SDB, ‘fine reticular’ included 11 SDB, ‘coarse
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
6
burns xxx (2015) xxx–xxx
Table 2 – Results of assessment by dermoscopy according to the proposed algorithm. Dermoscopy
Case
Clinical assessment
Accurate
Inaccurate
Accurate
29
1
23
Accuracy, % Sensitivity, % Specificity, %
Inaccurate 7
96.7 100.0 94.4
0.0227
76.7 85.7 73.9
KM Accurate
p-Value
TN
KMa
Inaccurate
Accurate
Inaccurate
Accurate
Inaccurate
1
21
9
22
8
29
Interobserver agreements
SE
k-Value
Burn depth KM and KMa KM and TN KMa and TN Each category KM and KMa KM and TN KMa and TN
0.389 0.403 0.247
0.166 0.174 0.168
0.421 0.450 0.331
0.119 0.121 0.118
Effects by experience as a dermatologist Experienced TN
KMa
Accurate 21
Resident HS
Inaccurate
Accurate
Inaccurate
9
22
8
21/9
Accurate
MY (resident1)
Inaccurate
Accurate
20 10 Average (accurate/inaccurate)
22
TS (resident2)
Inaccurate
Accurate
Inaccurate
8
18 20/10
12
Accuracy, % = number of burns that were assessed as DDB and failed to heal within 21 days and number of burns that were assessed as SDB and healed within 21 days/total number of burns assessed in this study (30). Sensitivity, % = number of burns that were assessed as DDB and failed to heal within 21 days after injury/number of burns that failed to heal within 21 days after injury; specificity, % = number of burns that were assessed as SDB and healed within 21 days after injury/number of burns that healed within 21 days after injury.
reticular’ included 1 SDB and 7 DDB, and the burns that did not show a ‘reticular’ pattern included 4 DDB (Fig. 2). The results of measurements according to the proposed algorithm showed an accuracy of 96.7% (29 accurate vs. 1 inaccurate). The sensitivity for DDB was 100.0% and its specificity was 94.4%. The results of clinical assessment showed an accuracy of 76.7% (23 accurate vs. 7 inaccurate). The accuracies differ significantly between the dermoscopic and clinical assessment. The interobserver agreements on the burn depth assessment (SDB or DDB) and on each category of the algorithm of KM and KMa, KM and TN, and KMa and TN are shown in Table 2, which showed fair to moderate agreements [33].
3.2.
Effects by experience as a dermatologist
The assessment by experienced dermatologists (TN, KMa and HS) and residents (MY and TS) according to the algorithm showed accuracies of 70.0% (21 accurate vs. 9 inaccurate), 73.3% (22 accurate vs. 8 inaccurate), 66.7% (20 accurate vs. 10 inaccurate), 73.3% (22 accurate vs. 8 inaccurate), and 60.0% ((18 accurate vs. 12 inaccurate), respectively. The average accuracies of each group (three experienced and two residents) were 70.0% and 66.7%, respectively (Table 2).
3.3.
Comparison among PNCD, PCD and NPD
Dots were most observed under NPD. Reticular area was observed equally under the three modalities. Hemorrhage was the most observed under PNCD and NPD, and less under PCD (Table 3). To show the comprehensive characteristics of each modality, the ranks and the total scores are shown in Table 3. The methodology for calculating the scores was that the modality that was ranked first was given 2 points, that ranked as second was given 1 point and that ranked as third was given no score. Vessels’ visibility: PCD images were ranked first the most and also got the highest total score. More red: PNCD images were ranked first the most and PCD images got the highest total score. More white: PNCD images were ranked first the most and NPD images got the highest total score. Dots partially disappear or become blurred: PNCD images were ranked first the most and got the highest total score. Reticular becomes blurred: PNCD images were ranked first the most and got the highest total score. Brown structure looks dark: NPD images were ranked first the most and PCD images got the highest total score. The interobserver agreements on each category of KMa and SH, KMa and TN, and SH and TN are shown in Table 3, which showed slight to moderate agreements [33].
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
Table 3 – Results of the comparison between PNCD, PCD and NPD.
4.
7
Discussion
The presence or absence of each dermoscopic feature
Dot Reticular Hemorrhage
PNCD
PCD
NPD
26/90 61/90 31/90
28/90 61/90 29/90
35/90 61/90 31/90
Qualitative differences: ranks
Vessels’ visibility
More red
More white
Dots partially disappear or become blurred Reticular become blurred
Brown structure look dark
Rank
PNCD
PCD
NPD
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
23 21 46 36 22 32 38 16 36 38 27 20 44 22 22 6 8 11
34 39 17 26 43 21 21 41 28 19 39 27 18 39 31 9 10 6
33 30 27 28 25 37 31 33 26 28 19 38 26 27 35 10 7 8
Qualitative differences: total scores
Vessels’ visibility More red More white Dots partially disappear or become blurred Reticular become blurred Brown structure look dark
Interobserver agreements
PNCD
PCD
NPD
67 94 92 103
107 95 83 77
96 81 95 75
110
75
79
20
28
k-Value
Vessels’ visibility 0.239 KMa and SH 0.182 TN and SH 0.252 KMa and TN More red 0.135 KMa and SH 0.170 TN and SH 0.316 KMa and TN More white 0.312 KMa and SH 0.270 TN and SH 0.547 KMa and TN Dots partially disappear or become blurred 0.040 KMa and SH 0.034 TN and SH 0.044 KMa and TN Reticular become blurred 0.169 KMa and SH 0.191 TN and SH KMa and TN 0.170 Brown structure look dark 0.400 KMa and SH TN and SH 0.226 0.346 KMa and TN
27
SE
0.094 0.108 0.092 0.084 0.105 0.104 0.104 0.103 0.110 0.085 0.086 0.083 0.097 0.112 0.092 0.112 0.087 0.069
Burn depth assessment is the main task at the beginning of burn care. The strategy of burn care and the prognosis of the patient are determined by burn depth. Although clinical assessment is the most widely conducted, its accuracy is reportedly limited up to about 75%. Thus, whether blood flow into the burn wound, especially into the superficial dermis, is active or not has been used to assess the burn depth more accurately. A number of procedures are used to evaluate dermal capillary integrity [19–24] however, they are not as versatile as dermoscopy. In this study, intermediate burn depth that is difficult to assess by the naked eye was significantly accurately assessed by dermoscopy. The accuracy of dermoscopic burn depth assessment by the consensus of three experienced dermatologists (KM, TN and KMa was 96.7%). The accuracy of dermoscopic burn depth assessment by KM was 96.7%, by TN was 70.0% and by KMa was 73.3%. Kittler et al. [9] reported that the diagnostic accuracy of melanoma with dermoscopy significantly depended on the experience of the examiners, and the diagnostic performance of dermoscopy improved by making a consensus of a group of examiners. The same tendency was found in this study; KM, who is experienced in burn wound practice and burn depth assessment clinically and dermoscopically, could assess the burn depth with higher accuracy than the other examiners, and the diagnostic performance of dermoscopy improved upon a consensus being reached by the three examiners. TN and KMa are experienced in dermatology and dermoscopic observation in daily clinical practice; however, they are not experienced in burn depth assessment by dermoscopy. The diagnostic accuracies of dermoscopic burn depth assessment were thought to be equivalent among the two groups whose experience as dermatologists differed (experienced and residents). Therefore, the diagnostic procedure in this study is thought to be simple even for a non-experienced observer. At the time of introduction of dermoscopy, only NPD was available. Then, PD was introduced into clinical use, so it became possible to observe deeper structures of the skin without the use of either a liquid interface or direct contact with the instrument. Therefore, three different modalities (PNCD, PCD and NPD) are now available in daily clinical practice. It has been reported that there were some differences among the images of these three modalities [25–29]. Benvenuto-Andrade et al. [28] evaluated the differences among them. They reported that most dermoscopic structures and patterns, except milialike cyst and peppering, were consistently recognized for the three different modalities. PD blocked the superficial component of light, so the visualization of superficial components such as milialike cysts was attenuated. They also noticed significant changes in quantitative assessment. Most lesions appeared darker under polarized light, and a more reddish appearance was recognized under PNCD compared with NPD. They concluded that the capabilities of the three different modalities were not equivalent but complementary. In this study, some differences among the three different modalities were also recognized. The recognition of dots
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9
8
burns xxx (2015) xxx–xxx
increased for NPD compared with PNCD and PCD. This is thought to be because dots denoted the top of a capillary loop, which is the shallowest component of the capillary. Light from the superficial component was blocked on PD, so dots were less frequently recognized with polarized light. On the other hand, recognitions of reticular areas and hemorrhage were almost consistent among the three imaging modalities. Vessels were the most clearly observed under PCD. In the literature [26,28,29], it is reported that the vasculature was better visualized by PD than by NPD because deeper structures were better visualized by polarized light. The dermoscopic observation in this study enabled the vascular structure of the superficial dermis to be observed without interference by the epidermal layer; however, polarized light even made a difference in the visibility of vessels. This is partly thought to be because of the differences in the observation of colors and shapes. A more reddish appearance is reported to be recognized under PNCD than under NPD [28]. In this study, colors also tended to be more distinctly recognized under polarized light. Light colors, such as red and white, were more distinctly recognized under PNCD and PCD (PNCD images were ranked first the most in these two recognitions and PCD images got the highest total score in the recognition of red). NPD images were ranked first the most in the recognition of a darker color, brown. NPD offers increased illumination and resolution compared with PD, and colors under non-polarized light are seen sharper and less distorted than with polarized light [34]. The lesion edge seems jagged and not so sharp under polarized light, and the evaluation of shape asymmetry prevails under nonpolarized light [27]. Blurred recognitions of dots and reticular patterns were the most clearly observed under PNCD in this study. This study has the limitation that biopsy was not conducted to validate the dermoscopic findings. Biopsy has been considered as the ‘gold standard’; however, biopsy of burn wound provides only a snapshot of the level of microvascular injury and may leave an additional scar [21]. The best outcome of burn wounds is clinical wound healing, as demonstrated by the present and previous studies [19,20]. Dermoscopy opened the window through which simple and reliable inspection into the microanatomy of skin structures became possible, even in daily clinical dermatology. With this tiny and simple device, not only epidermis and superficial dermis but also deeper structures of the skin could be observed under certain states, such as intermediate-depth burns. It is to be remembered that dermoscopic inspection could be affected by its modalities, and the difference among modalities should be borne in mind.
5.
Conclusion
Dermoscopy is a useful and simple tool to evaluate not only epidermal and superficial dermal skin components such as melanin chromospheres, horny cells and interstitium, but also skin microcirculation. In evaluating the microcirculation, differences among the three different dermoscopic modalities should be borne in mind.
Acknowledgement We are indebted to Prof. Junko Tanaka, Ph.D. and Dr. Tomoyuki Akita, Ph.D., Department of Epidemiology, Infectious Disease Control and Prevention, Hiroshima University Graduate School of Biomedical Sciences, for their useful advices on the interpretation of data and statistical analysis used in this study.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. burns.2015.08.032.
references
[1] Awan ZA, Wester T, Kvernebo K. Human microvascular imaging: a review of skin and tongue videomicroscopy techniques and analyzing variables. Clin Physiol Funct Imaging 2010;30:79–88. [2] Mahe´ G, Humeau-Heurtier A, Durand S, Leftheriotis G, Abraham P. Assessment of skin microvascular function and dysfunction with laser speckle contrast imaging. Circ Cardiovasc Imaging 2012;5:155–63. [3] Chao CY, Cheing GL. Microvascular dysfunction in diabetic foot disease and ulceration. Diabetes Metab Res Rev 2009;25:604–14. [4] De Backer D, Ospina-Tascon G, Salgado D, Favory R, Creteur J, Vincent JL. Monitoring the microcirculation in the critically ill patient: current methods and future approaches. Intensive Care Med 2010;36:1813–25. [5] Pehamberger H, Steiner A, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. I. Pattern analysis of pigmented skin lesions. J Am Acad Dermatol 1987;17:571–83. [6] Steiner A, Pehamberger H, Wolff K. In vivo epiluminescence microscopy of pigmented skin lesions. II. Diagnosis of small pigmented skin lesions and early detection of malignant melanoma. J Am Acad Dermatol 1987;17:584–91. [7] Westerhoff K, McCarthy WH, Menzies SW. Increase in the sensitivity for melanoma diagnosis by primary care physicians using skin surface microscopy. Br J Dermatol 2000;143:1016–20. [8] Argenziano G, Soyer HP. Dermoscopy of pigmented skin lesions – a valuable tool for early diagnosis of melanoma. Lancet Oncol 2001;2:443–9. [9] Kittler H, Pehamberger H, Wolff K, Binder M. Diagnostic accuracy of dermoscopy. Lancet Oncol 2002;3:159–65. [10] Menzies SW, Kreusch J, Byth K, Pizzichetta MA, Marghoob A, Braun R, et al. Dermoscopic evaluation of amelanotic and hypomelanotic melanoma. Arch Dermatol 2008;144:1120–7. [11] Zaballos P, Carulla M, Ozdemir F, Zalaudek I, Ban˜uls J, Llamgbrich A, et al. Dermoscopy of pyogenic granuloma: a morphological study. Br J Dermatol 2010;163:1229–37. [12] Cheng ST, Ke CL, Lee CH, Wu CS, Chen GS, Hu SC. Rainbow pattern in Kaposi’s sarcoma under polarized dermoscopy: a dermoscopic pathological study. Br J Dermatol 2009;160:801–9. [13] Bernard J, Poulalhon N, Argenziano G, Debarbieux S, Dalle S, Thomas L. Dermoscopy of dermatofibrosarcoma
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
JBUR-4744; No. of Pages 9 burns xxx (2015) xxx–xxx
[14]
[15]
[16]
[17]
[18] [19]
[20]
[21]
[22]
[23]
[24]
protuberans: a study of 15 cases. Br J Dermatol 2013;169: 85–90. Lallas A, Argenziano G, Apalla Z, Gourhant JY, Zaballos P, Di Lernia V, et al. Dermoscopic patterns of common facial inflammatory skin diseases. J Eur Acad Dermatol Venereol 2014;28:609–14. Va´zquez-Lo´pez F, Manjo´n-Haces JA, Maldonado-Seral C, Raya-Aguado C, Pe´rez-Oliva N, Marghoob AA. Dermoscopic features of plaque psoriasis and lichen planus: new observations. Dermatology 2003;207:151–6. Kim GW, Jung HJ, Ko HC, Kim MB, Lee WJ, Lee SJ, et al. Dermoscopy can be useful in differentiating scalp psoriasis from seborrheic dermatitis. Br J Dermatol 2011;164:652–6. Lallas A, Kyrgidis A, Tzellos TG, Apalla Z, Karakyriou E, Karatolias A, et al. Accuracy of dermoscopic criteria for the diagnosis of psoriasis, dermatitis, lichen planus and pityriasis rosea. Br J Dermatol 2012;166:1198–205. Hasegawa M. Dermoscopy findings of nail fold capillaries in connective tissue diseases. J Dermatol 2011;38:66–70. McGill DJ, Sørensen K, MacKay IR, Taggart I, Watson SB. Assessment of burn depth: a prospective, blinded comparison of laser Doppler imaging and videomicroscopy. Burns 2007;33:833–42. Hoeksema H, Van de Sijpe K, Tondu T, Hamdi M, Van Landuyt K, Blondeel. et al. Accuracy of early burn depth assessment by laser Doppler imaging on different days post burn. Burns 2009;35:36–45. Monstrey S, Hoeksema H, Verbelen J, Pirayesh A, Blondeel P. Assessment of burn depth and burn wound healing potential. Burns 2008;34:761–9. Jaskille AD, Shupp JW, Jordan MH, Jeng JC. Critical review of burn depth assessment techniques: Part I. Historical review. J Burn Care Res 2009;30:937–47. Mihara K, Shindo H, Ohtani M, Nagasaki K, Nakashima R, Katoh N, et al. Early burn depth assessment of local burns by videomicroscopy: 24 h after injury is a critical time point. Burns 2011;37:986–93. Mihara K, Shindo H, Mihara H, Ohtani M, Nagasaki K, Katoh N. Early burn depth assessment of local burns by videomicroscopy: a novel proposed classification. Burns 2012;38:371–7.
[25] Pan Y, Gareau DS, Scope A, Rajadhyaksha M, Mullani NA, Marghoob AA. Polarized and nonpolarized dermoscopy: the explanation for the observed differences. Arch Dermatol 2008;144:828–9. [26] Agero AL, Taliercio S, Dusza SW, Salaro C, Chu P, Marghoob AA. Conventional and polarized dermoscopy features of dermatofibroma. Arch Dermatol 2006;142:1431–7. [27] Pellacani G, Seidenari S. Comparison between morphological parameters in pigmented skin lesion images acquired by means of epiluminescence surface microscopy and polarized-light videomicroscopy. Clin Dermatol 2002;20:222–7. [28] Benvenuto-Andrade C, Dusza SW, Agero AL, Scope A, Rajadhyaksha M, Halpern AC, et al. Differences between polarized light dermoscopy and immersion contact dermoscopy for the evaluation of skin lesions. Arch Dermatol 2007;143:329–38. [29] Liebman TN, Jaimes-Lopez N, Balagula Y, Rabinovitz HS, Wang SQ, Dusza SW, et al. Dermoscopic features of basal cell carcinomas: differences in appearance under nonpolarized and polarized light. Dermatol Surg 2012;38:392–9. [30] Campanati A, Savelli A, Sandroni L, Marconi B, Giuliano A, Giuliodori K, et al. Effect of allium cepa-allantoinpentaglycan gel on skin hypertrophic scars: clinical and video-capillaroscopic results of an open-label, controlled, nonrandomized clinical trial. Dermatol Surg 2010;36:1439–44. [31] Campanati A, De Blasio S, Giuliano A, Ganzetti G, Giuliodori K, Pecora T, et al. Topical ozonated oil versus hyaluronic gel for the treatment of partial- to full-thickness seconddegree burns: a prospective, comparative, single-blind, non-randomised, controlled clinical trial. Burns 2013;39:1178–83. [32] Latarjet J. A simple guide to burn treatment. International Society for Burn Injuries in collaboration with the World Health Organization. Burns 1995;21:221–5. [33] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [34] Marghoob AA, Swindle LD, Moricz CZ, Sanchez Negron FA, Slue B, Halpern AC, et al. Instruments and new technologies for the in vivo diagnosis of melanoma. J Am Acad Dermatol 2003;49:777–97.
Please cite this article in press as: Mihara K, et al. Dermoscopic insight into skin microcirculation – Burn depth assessment. Burns (2015), http:// dx.doi.org/10.1016/j.burns.2015.08.032
9