ORIGINAL ARTICLES Modified Fine-Needle Aspiration Technique for Diagnosis of Granulomatous Skin Lesions With Special Reference to Leprosy and Cutaneous Tuberculosis Ghazala Mehdi, M.D., D.C.H.,1 Veena Maheshwari, Sonal Saxena, M.D.,1 and Rajeev Sharma, M.D.2 1 M.D., Skin infections are commonly assessed by slit skin or scrape methods. Fine-needle aspiration biopsy (FNAB) is highly effective especially with blanching of skin to ensure good yield and reduced bleeding. The aim of this study was to assess usefulness of cytology, especially modified FNAB technique, in diagnosis of leprosy and cutaneous tuberculosis and to identify specific cytological characteristics for diagnosis and classification. The study was conducted on 40 patients—25 cases of leprosy and 15 cases of cutaneous tuberculosis. Smears were prepared using modified FNAB technique, slit skin, and scrape methods (depending on type of lesion). Cytological diagnosis was confirmed by histopathology where the Ridley-Jopling system was used to classify cases of leprosy. A similar attempt was made for diagnosis and classification of leprosy on cytology. Diagnoses rendered by both modalities were compared to assess the efficacy of cytological examination. Cytological diagnosis was made in 23 cases of leprosy and 12 cases of cutaneous tuberculosis. The smears showed good cellularity. A broad division into tuberculoid and lepromatous leprosy could be made fairly accurately on cytology. Maximum agreement among clinical, cytological, and histopathological diagnosis was observed in cases of tuberculoid leprosy. Smears of cutaneous tuberculosis were characterized by epithelioid cell granulomas with caseation. Overall accuracy of diagnosis was 92% in leprosy and 80% in tuberculosis. FNAB is an inexpensive and accurate procedure for diagnosis of leprosy and cutaneous tuberculosis. The modified technique yields good results. However, clinical correla1 Department of Pathology, Jawaharlal Nehru Medical College and Hospital, Aligarh Muslim University, Aligarh, Uttar Pradesh, India 2 Bishen Skin Centre, Aligarh, Uttar Pradesh, India *Correspondence to: Hena A. Ansari, M.D., Bait-ul-Anwaar, Ahmad Gali Medical Road AMU, Aligarh 202002, Uttar Pradesh, India. E-mail: hena.jnmc@gmail.com Received 12 May 2009; Accepted 20 August 2009 DOI 10.1002/dc.21207 Published online 26 October 2009 in Wiley InterScience (www. interscience.wiley.com). ' 2009 WILEY-LISS, INC. Hena A. Ansari, 1 M.D., * tion, acid-fast staining, and culture are essential as they provide valuable supportive information. Diagn. Cytopathol. 2010;38:391–396. ' 2009 Wiley-Liss, Inc. Key Words: leprosy; cutaneous tuberculosis; modified FNAB technique; Ridley-Jopling system Cutaneous granulomatous infections encompass a wide spectrum of lesions. Their diagnosis can be based on cytological examination or by histopathological examination of biopsies. Cytological preparations can be of different types. Slit-skin smears and scrape preparations (for ulcerated lesions) are the most commonly used techniques, especially for leprosy and cutaneous tuberculosis. In addition to these conventional methods of sampling, we used fine-needle aspiration (FNA) as an additional diagnostic modality, especially for papular and nodular lesions. To improve cell yield, a modification of the FNA technique involves blanching of the skin by pinching it prior to aspiration.1–3 This also minimizes the presence of blood in the smears. FNA is also a more patient-friendly technique as there is no residual scarring, in contrast to the slit-skin technique. Cutaneous tuberculosis and Hansen’s disease (leprosy) are frequently encountered in the tropics. Leprosy usually spreads through droplet infection, but skin inoculation is not unknown.4 The widely accepted Ridley-Jopling classification5 is generally used to classify the disease on histopathology. Indeterminate leprosy and the histoid variant of leprosy are also well-recognized entities.6,7 Recently, it has been proposed that a new variant of Mycobacterium leprae (Mycobacterium lepromatosis) is the causative organism of diffuse lepromatous leprosy.8 Diagnostic Cytopathology, Vol 38, No 6 391 Diagnostic Cytopathology DOI 10.1002/dc MEHDI ET AL. Tuberculosis of the skin can either be a primary infection or a secondary/reinfection lesion.9,10 The modes of infection are variable. Subcategories of cutaneous tuberculosis include lupus vulgaris, tuberculosis verrucosa cutis, scrofuloderma, lichen scrofulosorum, papulonecrotic tuberculid, orificial tuberculosis, and disseminated cutaneous tuberculosis.10 The early diagnosis of granulomatous skin infections has gained even greater significance in the wake of the HIV pandemic. Surprisingly, the HIV infection has not affected the incidence/prevalence of leprosy.11,12 This study comprised 40 cases of clinically diagnosed cutaneous tuberculosis and leprosy. The aim was to assess the usefulness of cytology, especially by applying the modified FNA technique, in diagnosing these diseases, by identifying characteristic cytological features. An attempt was also made to categorize cases of leprosy according to the Ridley-Jopling classification5 on the basis of cytological features. Material and Methods Cytological samples were obtained from 40 patients, comprising 28 men and 12 women, with a mean age of 30.4 years. Material for cytological analysis was obtained by FNA as well as by slit skin and scrape methods. The sampling method chosen depended on the type of lesion. FNA was used for sampling papular and nodular lesions more than 5 mm in size, in both leprosy and tuberculosis. The procedure was explained to the patient. Using aseptic precautions, the skin in the region of the lesion was pinched to blanch it (modified technique). Three or four passes were made using a 23-gauge needle attached to a 10 mL disposable syringe. The aspirated material was transferred to clean slides and processed as per routine laboratory protocol. Slit-skin smears were prepared from macular lesions of leprosy. After cleaning the site, the skin was pinched and an incision made in the skin using a thin sterile blade. The incisions were approximately 6-7 mm long and 3 mm deep. Material was collected by scraping the exposed tissue and was spread on clean slides which were submitted for processing. The scrape method was used to sample ulcerated lesions in cases of tuberculosis. The surface of the lesion was scraped lightly to remove any exudates, followed by a more viorous scraping of the deeper tissue. The material obtained was spread on clean slides and processed accordingly. Smears were stained with the Papanicolaou, hematoxylin and eosin, and May–Grünwald–Giemsa (MGG) stains. Ziehl–Neelsen staining was performed for visualization of the mycobacteria. The bacterial index (BI) was also determined in all cases of leprosy. BI refers to the average number of bacilli per field and is calculated using an oil immersion lens.13 Skin biopsy was performed to confirm 392 Diagnostic Cytopathology, Vol 38, No 6 Table I. Distribution of Cases of Leprosy Diagnosis Polar tuberculoid leprosy (TT) Borderline tuberculoid (BT) Midborderline (BB) Borderline lepromatous (BL) Polar lepromatous leprosy (LL, including histoid leprosy) Indeterminate leprosy Total Clinical cases Diagnosis on cytology Diagnosis on histopathology 6 6 6 (TL) 4 (TL) 1 (BB) 3 BB 2 TL 1 LL 6 3 (BT) 2 (TT) 3 BB 2 BT 1 TT 2 4 1 BB 2 LL 1 BB 1 BB 1 LL 2 BB 1 (?leprosy) 25 — 21 — 21 6 cytological diagnoses by histopathology and to assess the degree of cytohistological correlation. The Ridley–Jopling classification5 was used for categorization of cases of leprosy on histopathology and a similar attempt was made for cytological diagnoses. Cases of cutaneous tuberculosis were not subclassified on cytology. Results In most of the cases where fine-needle aspiration biopsy (FNAB) was done, we obtained adequate material, which was comparable to the yield with slit-skin technique. Leprosy Of the 25 patients clinically diagnosed as leprosy, an accurate cytopathological diagnosis of leprosy was made in 23 cases (92%) where adequate material was available on cytology preparations. Biopsies were not available in an additional two cases, leaving 21 cases in which all three parameters were available for comparison (Table I). Of these, a concordant cytohistological diagnosis was available in 19 cases. Cytological Features Smears from clinically diagnosed borderline tuberculoid and polar tuberculoid leprosy were cellular, with cohesive epithelioid cell granulomas. A diffuse lymphocytic cell population was also present but granulomas were not infiltrated by lymphocytes. The cases were broadly classified as ‘‘tuberculoid leprosy’’ as similar cytological features were seen in borderline tuberculoid leprosy and polar tuberculoid leprosy (Fig. 1). In midborderline leprosy, the granulomas were poorly cohesive with a mixture of both epithelioid cells and macrophages. A few lymphocytes were observed infiltrating the granulomas (Fig. 2). In the cases diagnosed on cytology as lepromatous leprosy, granulomas were absent. Numerous foamy macrophages were noted against a fatty background interspersed with a few lymphocytes (Fig. 3a). On stain- Diagnostic Cytopathology DOI 10.1002/dc MODIFIED FNAB TECHNIQUE Fig. 2. (a) Midborderline leprosy (BB): smear showing loose epithelioid cell granuloma infiltrated by few lymphocytes (H&E, 3250). (b) Midborderline leprosy (BB): diffuse epithelioid cell granuloma containing scanty lymphocytes, with clear subepidermal zone (H&E, 3125). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] Fig. 1. (a) Tuberculoid leprosy (TL): smear showing cohesive epithelioid cell granuloma surrounded by lymphocytes and necrosis (H&E, 3250). (b) Tuberculoid leprosy (TT): caseating granuloma with necrosis, epithelioid cells, giant cells, and lymphocytes (H&E, 3125). (c) Borderline tuberculoid leprosy (BT): epithelioid cell granuloma infiltrated by lymphocytes, extending upto epidermis (H&E, 3125). [Color figure can be viewed in the online issue, which is available at www. interscience.wiley.com.] ing with MGG, numerous negative bacterial images, both within the macrophages and lying free in the background were observed (Fig. 3b). Smears of histoid leprosy, in addition, showed spindle-shaped macrophages (Fig. 3c). In only one case was lepromatous leprosy confirmed on histopathology. A definite variation in the bacterial index was observed in cytology smears in the various lesions. The BI ranged from 0 to 2+ at the tuberculoid end of the spectrum to 5+ in cases of lepromatous leprosy and 6+ in histoid leprosy. We found a maximum clinico cytological concordance in cases of tuberculoid leprosy. Cases of lepromatous and borderline lepromatous leprosy were small in number which may nave contributed to the poor clinical, cytopathological, and hostopathological concordance in these cases. Tuberculosis Fifteen cases of cutaneous tuberculosis were included in the study. Three main clinical groups were identified— Diagnostic Cytopathology, Vol 38, No 6 393 Diagnostic Cytopathology DOI 10.1002/dc MEHDI ET AL. Table II. Distribution of Cases of Cutaneous Tuberculosis Clinical cases Diagnosis on cytology Diagnosis on histopathology Lupus vulgaris Scrofuloderma 9 3 5 3 Tuberculosis verrucosa cutis Pyogenic granuloma Total 2 — 5 2 (scrofuloderma) 1 LV 2 1 15 — 11 1 (TVC) 11 Diagnosis lupus vulgaris, tuberculosis verrucosa cutis, and scrofuloderma (Table II). Lupus vulgaris comprised the majority of cases. Adequate material was obtained for cytological evaluation in 12 cases. Biopsy was possible in 11 cases; thus all three parameters were available for assessment in these cases. The cytological diagnosis concurred broadly with the clinical and histological finding, although subclassification was not possible on cytology. Cytological Features Tuberculosis of skin, irrespective of the clinical or histopathological subtype, showed epithelioid cell granulomas against a background of mononuclear inflammatory infiltrate (lymphocytes) and caseous necrosis. Acute inflammatory cells were also present, particularly in scrofuloderma. All cases were diagnosed as cutaneous tuberculosis and no specific subtyping was carried out (Fig. 4). Discussion Fig. 3. (a) Lepromatous leprosy (LL): smear showing foamy macrophages (lepra cells) on a fatty background (May–Grünwald–Giemsa, 3500). (b) Histoid leprosy: lepra cells laden with bacilli which appear as negative images, both intra- and extracellularly (May–Grünwald– Giemsa, 3500). (c) Histoid leprosy: spindle shaped and foamy macrophages (:). The spindle-shaped macrophages show slight cytoplasmic vacuolation (H&E, 3250). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] 394 Diagnostic Cytopathology, Vol 38, No 6 Histopathology of skin biopsy specimens is the gold standard for the diagnosis of both non-neoplastic and neoplastic skin lesions. Cytopathology using the slit skin and scrape techniques are widely used diagnostic modalities for infectious diseases such as leprosy and tuberculosis. However, the role of FNAB is less well defined.14 FNAB can be extremely useful in inflammatory diseases, especially where the modified1–3 technique is used to ensure a relatively bloodless aspirate. The cytomorphology of granulomatous skin lesions has been studied in detail.15 Although skin lesions in leprosy are usually analyzed by slit-skin smears, studies have been conducted to assess the utility of FNAC in lepromatous leprosy,16 histoid leprosy,17 reactional and nonreactional states,3,18,19 and pure neuritic leprosy.20 We achieved a high degree of accuracy in diagnosing leprosy on cytology and the cytological features reported in this study correlated well with the published literature. Other authors have also achieved similar results, especially in cases of tuberculoid and lepromatous leprosy.19,21 Prasad et al.3 also reported a high degree of correlation between clinical features and FNA diagnosis as compared to clinical and histopathological diagnosis, especially in tuberculoid leprosy. Exact classification Diagnostic Cytopathology DOI 10.1002/dc MODIFIED FNAB TECHNIQUE Fig. 4. (a) Lupus vulgaris (LV): smear showing clusters of epithelioid cells with copious cytoplasm and vesicular nuclei, surrounded by lymphocytes and necrosis (Papanicolaou, 3250). (b) Lupus vulgaris (LV): epithelioid cell granulomas with slight necrosis, Langhan’s giant cells, and lymphocytes (H&E, 3250). (c) Scrofuloderma: epithelioid cell cluster surrounded by acute inflammatory cells (Papanicolaou, 3250). [Color figure can be viewed in the online issue, which is available at www. interscience.wiley.com.] along the Ridley–Jopling spectrum is difficult, as demonstrated in this study. A recent study has also reported slightly varying results and the authors of this study also feel that cytology is not sensitive enough to classify leprosy.3 However, a broad division can be made between polar tuberculoid and lepromatous leprosy. As the disease progresses from the tuberculoid end of the spectrum to the lepromatous form, there is demonstrable change in the morphology of the granulomas, with decreased cohesion, lymphocytic infiltration of the granulomas, and increase in number of macrophages. However, these features may overlap considerably, especially among the borderline and midborderline cases, which is why it is difficult to accurately apply the Ridley-Jopling system to cytology. Also these features may be present in other skin lesions as well; therefore, clinical correlation, acid-fast staining, and culture are essential for a final diagnosis. The main cytodiagnostic feature in cutaneous tuberculosis is the presence of caseating epithelioid cell granulomas with lymphocytic infiltrate. Kathuria et al.22 have reported similar findings. It is difficult to distinguish the various subtypes of cutaneous tuberculosis on the basis of cytology. Clinical correlation is essential to arrive at a final diagnosis. As the sampling technique used depended on the type of lesion, a comparison of the efficacy of different techniques was not done. However, a good cellular yield was obtained using the modified FNA method. Apart from cutaneous tuberculosis and leprosy, epithelioid cell granulomas are also a feature of sarcoidosis and secondary syphilis.15 The lesions of sarcoidosis usually show little lymphocytic infiltrate. Also, the presence of caseation necrosis in tuberculosis is a helpful pointer to the diagnosis. A polymorphous exudate may be present in filariasis, cysticercosis, and actinomycosis, apart from scrofuloderma.15 Histiocytic granulomas can be observed in cryptococcosis and leishmaniasis.15 Granulomas can also have a noninfectious etiology, as in foreign body granulomas, drug reactions, sarcoidosis, granuloma annulare, and rheumatoid nodules.23 Therefore, Ziehl–Neelsen staining for acid-fast bacilli is a simple way of differentiating between these conditions. Skin lesions are common in patients of AIDS and therefore, cytology, including FNA, has an important role to play in the rapid diagnosis of such cases. An interesting phenomenon has been reported in patients on highly active antiretroviral therapy, known as ‘‘immune re-constitution inflammatory syndrome or IRIS.’’24,25 Further studies are needed to assess the role of cytology in the diagnosis of these lesions. In recent times, efforts have been made to develop rapid, sensitive, and specific tests for leprosy and cutaneous tuberculosis26,27 Until such tests come into Diagnostic Cytopathology, Vol 38, No 6 395 Diagnostic Cytopathology DOI 10.1002/dc MEHDI ET AL. widespread use, the diagnosis must rest on clinical examination and tissue microscopy. Cytology is a rapid, cost-effective, and fairly accurate method for the diagnosis of mycobacterial skin lesions. The modified FNA technique combines the utility of conventional aspiration cytology with the efficacy of the slitskin smear method. We feel that its use can be extended to other skin lesions, apart from leprosy and tuberculosis. Thus, the role of cytology can be expanded in the evaluation of granulomatous infections of the skin, in conjunction with clinical history and special stains for identification of microorganisms. References 1. King M. A Medical laboratory for developing countries. Norwich, Great Britain: Oxford Medical Publications; 1976. 3.23, 11.1, 11.11b, 11.11d. 2. Singh N, Bhatia A, Gupta K, Raman M. Cytomorphology of leprosy across the Ridley-Jopling spectrum. Acta Cytol 1996;40:719–723. 3. Prasad PVS, George RV, Kaviarasan PK, Viswanathan P, Tippoo R, Anandhi C. Fine needle aspiration cytology in leprosy. Indian J Dermatol Venereol Leprol 2008;74:352–356. 4. Machin M. The mode of transmission of human leprosy. In: Ryan TJ, McDougall AC, editors. Essays on leprosy. Oxford: St. Francis Leprosy Guild; 1988. p 1–29. 5. Ridley DS, Jopling WH. Classification of leprosy according to immunity: A five group system. Int J Lepr 1966;34:255–273. 6. Ridley DS. Histological classification and the immunological spectrum of leprosy. Bull World Health Organization 1974;51:451–465. 7. Wade HW. The histoid variety of lepromatous leprosy. Int J lepr 1963;31:129–142. 8. Han XY, Seo Y-H, Sizer KC, et al. A new mycobacterium species causing diffuse lepromatous leprosy. Am J Clin Pathol 2008;130: 856–864. 9. Ramesh V, Mishra RS, Jain RK. Secondary tuberculosis of the skin. Clinical features and problems in laboratory diagnosis. Int J Dermatol 1987;26:578–581. 10. Brown FS, Anderson RH, Burnett JW. Cutaneous tuberculosis. J Am Acad Dermatol 1982;6:101–106. 11. Lucas SB. Human immunodeficiency virus and leprosy. Lepr Rev 1993;64:97–103. 12. Ustianowski AP, Lawn SD, Lockwood DN. Interactions between HIV infection and leprosy: A paradox. Lancet Infect Dis 2006;6: 350–360. 396 Diagnostic Cytopathology, Vol 38, No 6 13. Ridley DS. Bacterial indices. In: Cochrane RG, Davey TF, editors. Leprosy in theory and practice. 2nd ed. Bristol: John Wright and Sons; 1964. p 620–622. 14. Daskalopoulou D, Maounis N, Kokalis G, Liodandonaki P, Belezini E, Markibou S. The role of fine-needle aspiration cytology in the diagnosis of primary skin tumours. Arch Anat Cytol Pathol 1993; 41:75–81. 15. Bhatia A, Singh N, Arora VK, Bhattacharya SN. Diagnosing granulomatous inflammation of the skin. A cytomorphologic approach based on evaluation of cellular reaction patterns. Acta Cytol 1999; 43:761–766. 16. Singh N, Bhatia A, Arora VK, Bhattacharya SN. Fine needle aspiration cytology of lepromatous leprosy. Lepr Rev 1998;69:145–150. 17. Jena S, Surain JP, Mishra S. FNAC study of histoid nodule: An early tool for diagnosis. Indian J Lepr 2005;77:135–139. 18. Malik A, Bhatia A, Singh N, Bhattacharya SN, Arora VK. Fine needle aspiration cytology of reactions in leprosy. Acta Cytol 1999;43: 771–776. 19. Nigam PK, Kumar P, Pathak N, Mittal S. Fine needle aspiration cytology in reactional and non-reactional leprosy. Indian J Dermatol Venereol Leprol 2007;73:247–249. 20. Siddaraju N, Yaranal PJ. Use of fine needle aspiration cytology in leprotic lesions: A report of 4 cases. Acta Cytol 2007;51:235–238. 21. Rao IS, Singh MK, Gupta SD, Pandhi RK, Kapila K. Utility of fine needle aspiration cytology in the classification of leprosy. Diagn Cytopathol 2001;24:317–321. 22. Kathuria P, Agarwal K, Koranne RV. Role of fine needle aspiration cytology and Ziehl – Neelsen staining in the diagnosis of cutaneous tuberculosis. Diagn Cytopathol 2006;34:826–829. 23. Ko CJ, Glusac EJ, Shapiro PE. Non-infectious granulomas. In: Elder DE, Elenitsas R, Johnson BL Jr., Murphy GF, Xu X, editors. Lever’s histopathology of the skin. 10th ed. (South Asian ed.). New Delhi: Lippincott Williams and Wilkins; 2009. p 361–387. 24. Shelburne SA, Hamill RJ, Rodriguez-Barradas MC, et al. Immune reconstitution inflammatory syndrome: Emergence of a unique syndrome during highly active antiretroviral therapy. Medicine (Baltimore) 2002;81:213–227. 25. Lawn SD, Wood C, Lockwood DN. Borderline tuberculoid leprosy: An immune reconstitution phenomenon in human immunodeficiency virus-infected persons. Clin Infect Dis 2003;36:(suppl)e5–e6. 26. Martinez AN, Britto CFPC, Nery JAC, et al. Evaluation of realtime and conventional PCR targeting complex 85 genes for detection of mycobacterium leprae DNA in skin biopsy samples from patients diagnosed with leprosy. J Clin Microbiol 2006;44:3154– 3159. 27. Abdalla CMZ, De Oliveira ZNP, Sotto MN, Leite KRM, Canavez FC, De Carvalho CM. Polymerase chain reaction compared to other laboratory findings and to clinical evaluation in the diagnosis of cutaneous tuberculosis and atypical mycobacteria skin infection. Int J Dermatol 2009;48:27–35.