Granulomatous dermatitis is a distinctive histopathologic cutaneous reaction pattern against various infectious and noninfectious agents. Cytologically, granulomatous dermatitis shows granulomas and multinucleated giant cells. Various etiologic agents of granulomatous diseases can also be identified.

We aimed to investigate Tzanck smear findings in granulomatous skin diseases.

Patients who had granulomas and/or multinucleated giant cells of Langhans, foreign body- and/or Touton type in Tzanck smear tests were included in the study. In these patients, Tzanck preparations were then further evaluated for additional cytologic findings. Samples stained with May-Grünwald-Giemsa stain were evaluated by the same dermatologist throughout the study. In some patients, methylene blue, Gram and/or Erlich-Ziehl-Nielsen stains were also performed. In all of the study cases, the final diagnosis was established after the evaluation of clinical and laboratory findings (including, when appropriate, potassium hydroxide examination; bacterial, leishmanial, and fungal cultures; histopathology; tuberculosis and leishmania polymerase chain reaction). We also calculated the sensitivity and specificity of the Leishman-Donovan body for cutaneous leishmaniasis.

Over a 2-year period, 94 of 950 patients (9.9%) in whom Tzanck smear tests were performed had cytologic findings consistent with a granulomatous reaction. In 74 (78.7%) and 20 (21.3%) patients, the granulomatous reaction was due to infectious and noninfectious causes, respectively. Infectious causes included cutaneous leishmaniasis in 65 patients (87.8%), candidal granuloma in two patients, botyromycosis in two patients, and aspergillosis, blastomycosis, mucormycosis, leprosy, and cutaneous tuberculosis in one patient each. In 58 of 74 patients (78.4%) with infectious granulomatous dermatitis, the causes of the granulomas were identified. Noninfectious granulomatous reactions were due to granuloma annulare in 7 patients, sarcoidosis in 5 patients, a foreign body in 4 patients, necrobiosis lipoidica in 2 patients, and juvenile xanthogranuloma in 2 patients. In 17 of 20 patients (85%) with noninfectious granulomatous reactions, the cytologic findings were characteristic of the final diagnoses. The sensitivity and specificity of Leishman-Donovan bodies for cutaneous leishmaniasis were 76.9% and 100%, respectively.

All of the samples were evaluated by the same dermatologist throughout the study; therefore no comment could be made regarding the reliability of the Tzanck smear test. In addition, the sensitivity and specificity of Tzanck smear test findings for diseases other than cutaneous leishmaniasis could not be calculated because of an insufficient number of patients.

The Tzanck smear test may be a useful diagnostic tool for certain granulomatous skin diseases.