Dermatology: Practical and Conceptual Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 1 The Accuracy of Clinical Diagnosis in 2135 Lesions on the Face. A Retrospective Analysis of Histopathological Records Anna Czaplicka1, Magdalena Misiak- Gałązka2, Joanna Czuwara 2, Adam Gałązka3, Barbara Górnicka4, Lidia Rudnicka2 1 Department of Dermatology, Postgraduate Medical Education Center/Central Clinical Hospital of the Ministry of Interior and Administration in Warsaw, Warsaw, Poland 2 Department of Dermatology, Medical University of Warsaw, Warsaw, Poland 3 Department of Head and Neck Cancers, The Maria Sklodowska-Curie National Research Institute of Oncology, Warsaw, Poland. 4 Department of Pathology, Medical University of Warsaw, Poland. Key words: dermatopathology, cancer, standardized skin surface biopsy, face, inflammatory diseases Citation: Czaplicka A, Misiak- Gałązka M, Czuwara J, Gałązka A, Górnicka B, Rudnicka L. The accuracy of clinical diagnosis in 2135 lesions on the face. A retrospective analysis of histopathological records. Dermatol Pract Concept. 2022;12(4):e2022159. DOI: https://doi.org/10.5826/dpc.1204a159 Accepted: February 5, 2022; Published: October 2022 Copyright: ©2022 Czaplicka et al. This is an open-access article distributed under the terms of the Creative Commons Attribution- NonCommercial License (BY-NC-4.0), https://creativecommons.org/licenses/by-nc/4.0/, which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original authors and source are credited. Funding: None. Competing interests: None. Authorship: All authors have contributed significantly to this publication. Introduction: Biopsy of facial skin lesions is an important supplement to dermatological diagnostics, especially in doubtful cases or suspected of being malignant. Objectives: The aim of the retrospective study of 2135 histopathological records of lesions on the face was to: establish the most common indications for a skin biopsy in patients with facial lesions, establish the frequency of histopathological diagnoses, evaluate how often clinically suspected inflam- matory lesions are identified as tumors in histopathology, evaluate the accuracy of clinical diagnoses of the most common skin tumors and dermatoses. Methods: It was a retrospective study. Histopathological records from the lesions on the face from years 2010-2017 were analyzed. Results: The mean age of patients was 69.3 [7-98]. Fifty-eight percent of the patients were women. Among 2135 clinical diagnoses skin tumors were suspected in 1905 cases. Among 2169 obtained his- topathological results (34 biopsies showed 2 diseases), we identified skin tumors in 1940 cases, with 1388 confirmed as malignant. The clinical diagnosis of a specific benign or malignant skin tumor was accurate in 1013/1634 subjects, in comparison to inflammatory lesions, which were correct in 67/148 cases, (P = 0.0001). Among all preliminary inflammatory diagnoses, 33/204 lesions were identified as skin tumors in histopathology. ABSTRACT 2 Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 Corresponding author: Magdalena Misiak-Galazka MD, PhD. Department of Dermatology Medical University of Warsaw, Koszykowa 82A, 02-008 Warsaw, Poland, Tel/fax: 0048228242200, E-mail: magdamisiak@o2.pl Introduction Face is one of the special locations in dermatology with skin lesions easily visible having a negative impact on self-esteem and decreasing the quality of life. Many dermatoses on the face are easily diagnosed based on clinical examination and der- moscopy, and no skin biopsy is needed. Dermoscopic criteria of skin tumors on the face are well established and used in every- day clinical practice [1]. Also dermoscopic patterns of common facial inflammatory lesions were shown by Lallas et al [2]. However, in doubtful cases, when malignancy is sus- pected, a histopathological examination is mandatory, de- spite the fact, that skin biopsy may result in scar formation. Objectives The aim of the study was to analyze the histopathological records from lesions on the face, to define the most common indications for a skin biopsy, to establish the order of the common histopathological diagnoses, to evaluate how often skin biopsies suspected as inflammatory lesions turn out to be skin tumors, to evaluate the accuracy (the same clinical and histopathological diagnoses) of the most common skin tumors and dermatosis on the face. Methods It was a retrospective study performed at the Department of Dermatology, Medical University of Warsaw, Poland. Histopathological records from the lesions on the face from years 2010-2017 were analyzed. Patients for skin biopsies (punch, shave or excisional) were referred from our depart- ment as well as from many outpatients clinics in Warsaw area. The patients presented with skin phototypes I-III. The mean age of patients was 69.3 (range 7-98 years of age), and 58.28% of patients were women. Biopsies taken from the scalp and mucous membranes were excluded. A total num- ber of 89 cases were nondiagnostic and were excluded from the study. In total 2135 biopsy records were evaluated, 750 from cheeks with zygomatic area, 718 from noses, 434 from foreheads, 110 from ears, 71 from lips and 52 from chins. Because in 34 cases 2 final diagnoses were established a total number of 2169 of final diagnoses were analyzed. Statistical Analysis The results were computed with Statistica 13.1 software (StatSoft Incorporated) licensed to the Medical University of Warsaw. Chi 2 test was used for assessing binary variables. A P value below 0.05 was considered statistically significant. Results Among 2135 clinical diagnoses, the skin tumors were sus- pected in 1905 (89.23%) cases, and in 1609 (75.36%) cases malignant skin tumor or precancerous lesion were considered. Table 1. The most common clinical diagnoses (≥ 3 cases) among 2135 skin biopsies. In some cases, more than one clinical diagnosis was given Total number of cases % CHEEK CHIN EAR FOREHEAD LIPS NOSE Basal cell carcinoma 984 46.09% 313 18 51 196 27 379 Actinic keratosis 424 19.86% 149 5 14 102 4 150 Seborrheic keratosis 108 5.06% 40 0 6 30 0 32 Fibroma 105 4.92% 49 5 3 11 6 31 Squamous cell carcinoma, keratoacanthoma 96 4,5% 22 4 17 12 10 31 Lupus erythematosus 88 4.12% 58 1 0 15 0 14 Conclusions: In conclusion in most cases of skin tumors the clinical diagnosis is confirmed by histo- pathological examination. In case of facial inflammatory lesions, the accuracy of clinical diagnosis is lower, with a significant number of facial lesions appearing inflammatory in clinical evaluation but being diagnosed as skin cancers in pathology. Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 3 Table 2. The most common histopathological diagnoses (≥3 cases) among 2169 skin biopsies. The number of the correct diagnoses in the most common skin lesions † clinical diagnosis was confirmed by histopathological examination; SCC, squamous cell carcinoma Clinical diagnoses Number of the correct diagnoses (%)a Number of the correct diagnoses in total (%)a P = 0.0001 Number of cases with a skin tumor in histopathological diagnosis (%) Number of cases with a skin tumor in histopathological diagnosis in total (%) P <0 .0001 SKIN TUMORS Basal cell carcinoma 629/984 (63.92) 1013/1634 (61.99) 342/355 (96.33) 583/621 (93.88)Actinic keratosis 282/424 (66.51) 121/142 (85.21) Squamous cell carcinoma in situ, Bowen type 2/6 (33.33) 4/4 (100) Squamous cell carcinoma 29/74 (39.19) 42/45 (93.33) Melanocytic nevus 22/38 (57.89) 16/16 (100) Seborrheic keratosis 49/108 (45.37) 58/59 (98.3) INFLAMMATORY DERMATOSES Facial granuloma 6/13 (46.15) 67/148 (45.27) 2/7 (28.57) 19/81 (23.46)Lupus erythematosus 49/88 (55.68) 12/39 (30.77) Rosacea 5/10 (50.00)] 1/5 (20) Sarcoidosis 7/37 (18.92) 4/30 (13.33) Total number of cases % CHEEK CHIN EAR FOREHEAD LIPS NOSE Cutaneous horn 52 2.44% 23 3 4 9 2 11 Melanocytic nevus, blue nevus 38 1.78% 10 7 0 10 0 11 Sarcoidosis 37 1.73% 13 0 0 13 0 11 Skin tumor 32 1.50% 9 0 3 9 3 8 Pyogenic granuloma 23 1.08% 6 1 0 0 7 9 Verruca vulgaris 23 1.08% 10 0 2 5 0 6 Solar lentigo 14 0.66% 8 0 0 0 0 6 Facial granuloma 13 0.61% 3 0 0 4 0 6 Rosacea 10 0.47% 3 1 0 2 0 4 Eczema 8 0.37% 4 0 3 1 0 0 Lentigo maligna 8 0.37% 6 0 0 2 0 0 Lymphoma 7 0.33% 4 0 0 3 0 0 Granulomatous cheilitis 6 0.28% 0 0 0 0 6 0 Squamous cell carcinoma in situ. Bowen type 6 0.28% 4 0 0 2 0 0 Annular granuloma 6 0.28% 3 0 0 3 0 0 Sebaceous hyperplasia 5 0.23% 0 0 0 2 0 3 Cyst 4 0.19% 2 0 0 0 0 2 Lichen planus 4 0.19% 2 0 0 0 2 0 Lymphocytoma 4 0.19% 4 0 0 0 0 0 Dermatitis (superficial dermal inflammation) 3 0.14% 2 1 0 0 0 0 Morphea 3 0.14% 2 0 1 0 0 0 4 Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 The most common clinical diagnosis was basal cell carci- noma (BCC), followed by actinic keratosis, seborrheic kera- tosis, fibroma, and cutaneous lupus erythematosus (Table 1). Table 2 presents the most common clinical neoplasm and inflammatory diagnoses with the percentages of correct cases confirmed by histopathological examination. Clinical diag- noses of skin tumors were accurate in 1013/1634 (61.99%) cases, in comparison to inflammatory lesions, which were correct in 67/148 (45.27%) cases (P = 0.0001). Among 621 inaccurate clinical diagnoses of the most common skin tumors, 583 cases turned out to be other skin neoplasms (93.88%), and among 81 inaccurate clinical diagnoses of the most common inflammatory dermatoses 19 turned out to be skin neoplasms (23.46%)(P < 0.0001). Among all prelimi- nary inflammatory diagnoses, 33/204 (16.17%) cases turned out to be skin malignancies, such as basal and squamous cell carcinomas. Among 2169 histopathological diagnoses, skin tumors were diagnosed in 1940 (89.44%) cases, and malignant skin tumors were diagnosed in 1388 (63.99%) cases. The most common final diagnosis was basal cell carcinoma, followed by actinic keratosis, squamous cell carcinoma, fibroma, seborrheic keratosis, and cutaneous lupus erythematosus (Table 3). Majority of basal cell carcinomas were present on the nose (306 cases from 737; 41.5%), followed by cheeks (200; 27,13%). The most common locations of actinic keratosis constituted cheeks (171 out of 447; 38,2%) and nose (161; 36%), whilst squamous cell carcinomas were distrib- uted on the cheeks (49 out of 146; 33.6%), nose (41; 28%) and forehead (23; 15.8%) (Table 3). The most common diagnosed inflammatory dermatoses included cutaneous lupus erythematosus on the first place, followed by rosacea, demodicosis, facial granuloma, and sarcoidosis. Among patients older than 65 years, 74.83% (1124/1502) of cases were malignant skin tumors in com- parison to incidence of 39.58% (264/667) among patients aged ≤ 65 years (P < 0.00001). As far as regions of the face are concerned, no statisti- cal difference between numbers of correct diagnoses on the ear, nose, lips, and other regions on the face (cheeks, chin, forehead) was found (Table 4). Conclusions Face, which is constantly exposed to solar radiation is a com- mon location for skin tumors. According to study of Ferreira et al on topographic distribution of basal cell carcinomas head and neck area was the most frequent location of the tumor (75.55%), followed by trunk (10.5%) [3]. In another study, over 60% of nonmelanoma skin tumors presented in the head area [4]. In our study, as suspected, the most common clinical and histopathological diagnosis was a tumor: basal and squa- mous cell carcinoma, actinic keratosis, seborrheic keratosis, and fibroma. The most common distributions of squamous cell carcinoma were the cheeks, nose and forehead, and for basal cell carcinoma were the nose and cheeks. The results were in line with the study of Kato et al [5]. It was shown in a group of 106 Japanese patients, that the cheek (54.2%) and forehead (25.5%) were the most common facial distri- bution of squamous cell carcinoma [5]. BCChas correctly resulted as the most frequent skin le- sion on the face. Among the common skin lesions, the most incorrect diagnoses occurred in the case of  squamous cell carcinoma (only 33-39% of cases were correctly diagnosed). Squamous cell carcinoma was misdiagnosed with (in order of the most common): BCC, actinic keratosis. Another clin- ically misdiagnosed pairs of lesions are: for BCC - actinic keratosis, squamous cell carcinoma, seborrheic keratosis, epidermal nevus, for actinic keratosis - BCC, squamous cell carcinoma, seborrheic keratosis, morbus Bowen, papilloma, for seborrheic keratosis - papilloma, actinic keratosis, BCC, verruca vulgaris and for fibroma - verruca vulgaris, sebor- rheic keratosis, inverted follicular keratosis, cutaneous horn. What is interesting, the diagnosis of a skin tumor was correct in only 61.99% of cases. This means, that in many cases of a clinically suspicious lesion the histopathological diagnosis was a benign tumor. The results showed that ap- proach with taking a diagnostic biopsy before performing surgical procedure is reasonable and the result influences the need and extent of surgery. In case of benign diagnosis, the remnants of the lesion can be left and no surgical procedure is required. Among inflammatory dermatoses (10.56% of cases), the most common were in order cutaneous lupus erythe- matosus, rosacea, demodicosis, facial granuloma, and sar- coidosis. In the study only 45.27% of clinical diagnosis of inflammatory lesions were consistent with histopathological results. Among cases with suspicion of inflammatory der- matoses skin tumors were diagnosed. These results showed that in doubtful inflammatory lesions on the face the biopsy is recommended for two reasons. First of all, to exclude a malignant tumor. The second reason is to establish a final diagnosis. It should be underlined, that among inflammatory dermatoses such as cutaneous lupus erythematosus and sar- coidosis, systemic involvement can occur, and patients with such diseases need additional laboratory and imaging tests, therefore the correct diagnosis has special importance and requires medical follow-up. The study has some limitations. It was a single-centre study. Another restriction of our study is the fact that some cases of skin tumors are not biopsied due to clear clinical and dermoscopic aspects. It includes all cases where these Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 5 Ta b le 3 . T he n um be r of t he c or re ct d ia gn os es in t he m os t co m m on s ki n le si on s. T he m os t co m m on h is to pa th ol og ic al d ia gn os es ( > 10 c as es ) am on g 21 69 s ki n bi op si es To ta l n u m b er o f ca se s % C H EE K C H IN EA R FO R EH EA D LI PS N O SE A g e < = 65 a A g e > 65 a Sk in n eo pl as m s in t ot al a P < 0 .0 00 1 19 40 /2 16 9 89 .4 4 66 4/ 76 9 43 /5 6 99 /1 05 38 8/ 44 3 58 /7 1 68 8/ 72 5 51 5/ 66 7 (7 7. 21 % ) 14 25 /1 50 2 (9 4. 87 % ) M al ig na nt s ki n tu m or s in t ot al a P < 0 .0 00 01 13 88 /2 16 9 63 .9 9 45 0/ 76 9 21 /5 6 83 /1 05 27 3/ 44 3 40 /7 1 52 1/ 72 5 26 4/ 66 7 (3 9. 58 % ) 11 24 /1 50 2 (7 4. 83 % ) B as al c el l c ar ci no m a 73 7 33 .9 8 20 0 18 40 15 1 22 30 6 16 0 57 7 A ct in ic k er at os is 44 7 20 .6 1 17 1 0 14 92 9 16 1 77 37 0 Sq ua m ou s ce ll ca rc in om a 14 6 6. 73 49 2 22 23 9 41 22 12 4 Fi br om a 12 5 5. 76 49 5 3 32 4 32 64 63 Se bo rr he ic k er at os is 10 8 4. 98 45 0 5 39 0 19 35 73 C ut an eo us lu pu s er yt he m at os us 57 2. 63 37 1 0 7 0 12 50 7 M el an oc yt ic n ev us 56 2. 58 12 9 1 13 1 20 46 18 Sq ua m ou s ce ll ca rc in om a in si tu , B ow en t yp e 46 2. 12 24 0 6 4 0 12 2 44 R os ac ea 37 1. 71 20 0 0 11 0 6 26 11 In ve rt ed f ol lic ul ar k er at os is 35 1. 61 27 0 2 2 0 4 9 26 K er at oa ca nt ho m a 29 1. 34 14 2 0 0 0 13 9 20 V er ru ca v ul ga ri s 27 1. 24 10 4 0 3 1 9 18 9 C ut an eo us h or n 26 1. 20 10 0 3 5 0 8 3 23 D em od ic os is 22 1. 01 10 0 0 7 0 5 9 13 Py og en ic g ra nu lo m a 21 0. 97 9 0 0 1 5 6 16 5 E pi de rm al c ys t 16 0. 74 8 0 0 0 1 7 4 12 T ri ch ile m m om a 16 0. 74 6 0 0 0 0 10 4 12 Fa ci al g ra nu lo m a 14 0. 65 3 0 0 5 0 6 12 2 Sa rc oi do si s 13 0. 60 6 0 0 3 0 4 9 4 Se ba ce ou s hy pe rp la si a 11 0. 51 5 0 0 0 0 6 5 6 E cz em a 11 0. 51 4 2 1 3 1 0 7 4 a P < 0. 00 01 6 Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 diagnosis is confirmed by histopathological examination. In the case of inflammatory facial lesions, the accuracy of clin- ical diagnosis is lower, with a significant number of facial lesions appearing inflammatory in clinical evaluation but being diagnosed as skin cancers in pathology. References 1. Thomas L, Phan A, Pralong P, Poulalhon N, Debarbieux S, Dalle S. Special locations dermoscopy: facial, acral, and nail. Dermatol Clin. 2013;31(4):615-624. DOI: 10.1016/j.det.2013.06.006. PMID: 24075549. 2. Lallas A, Argenziano G, Apalla Z, et al. Dermoscopic patterns of common facial inflammatory skin diseases. J Eur Acad Dermatol Venereol. 2014;28(5):609-614. DOI: 10.1111/jdv.12146. PMID: 23489377. 3. Ferreira FR, Pevide Bda C, Rodrigues RF, Nascimento LF, Lira ML. Differences in age and topographic distribution of the different histological subtypes of basal cell carcinoma, Taubaté (SP), Brazil. An Bras Dermatol. 2013;88(5):726-730. DOI: 10.1590/abd1806-4841.20132145. PMID: 24173177. PMCID: PMC3798348. lesions are treated with imiquimod, cryotherapy or photody- namic therapy. Moreover, the limitation of the study was the lack of dermoscopic descriptions and both clinical images and clinical diagnoses were based on experience of the refer- ring physician. On the other hand, the patients were referred for punch, shave, or excisional biopsy from numerous public and private clinics. So, the analysis reflects a need for skin biopsy of the face in everyday life of dermatological practice and was not limited only to the academic settings. It is important to underline the fact, that even in modern setting, with a standard help of dermoscopy and without easy access to confocal microscopy (which allows to reduce the number of diagnostic biopsies), there is still a high percentage of false-positive and negative cases that may be treated incorrectly in case of lack of collaboration with a dermatopathologist. A suspicion of skin malignancy is the most common indi- cation for biopsy from the lesions on the face. Because 10% of cases turned out to be inflammatory dermatoses, inflam- matory entities should be also included in the differential di- agnosis of the lesions on the face, especially in patients under 65 years of age. In most cases of skin tumors, the clinical Table 4. The percentage of the correct diagnoses on the different regions of the face Region of the face Number of the correct diagnoses (%) P value with the comparison to the other regions Ear 56/105 (53.33) P = 0.734 Nose 417/725 (57.52) P = 0.285 Lips 44/71 (61.97) P = 0.253 Other regions (cheeks, forehead, chin) 698/1268 (55.05) Original Article | Dermatol Pract Concept. 2022;12(4):e2022159 7 4. Katalinic A, Kunze U, Schäfer T. Epidemiology of cutaneous mel- anoma and non-melanoma skin cancer in Schleswig- Holstein, Germany: incidence, clinical subtypes, tumour stages and localization (epidemiology of skin cancer). Br J Dermatol. 2003;149(6):1200- 1206. DOI: 10.1111/j.1365-2133.2003.05554.x. PMID: 14674897. 5. Kato H, Oda T, Watanabe S, Morita A. Facial distribution of squamous cell carcinoma in Japanese. Exp Dermatol. 2019;28 Suppl 1:72-74. DOI: 10.1111/exd.13830. PMID: 30698883.