Background: Skin adnexal tumours are a morphologically heterogeneous group of neoplasms arising from cutaneous appendages. Their clinical similarity to common benign lesions renders preoperative diagnosis challenging, and systematic concordance data from Indian tertiary institutions remain limited. Objectives: The objectives of this study are to characterise the histopathological spectrum and clinicopathological associations of skin adnexal tumours and to quantify clinical-histopathological concordance. Methods: A cross-sectional, descriptive-analytical study was conducted at a tertiary care teaching hospital over six years using consecutive sampling. Data were analysed using descriptive statistics, Fisher's exact test, Chi-square test, Mann-Whitney U test, and Cohen's Kappa coefficient. Results: Forty-seven cases were studied (M:F = 1.47:1; mean age 42.32 ± 15.37 years). Eccrine differentiation predominated (66.0%), followed by follicular (23.4%) and sebaceous (10.6%) lineages. The consolidated Hidradenoma group was the most frequent individual diagnosis (17.0%), followed by pilomatricoma (10.6%). Benign tumours constituted 83.0%, whereas malignant tumours constituted 17%; the head and neck was the predominant site (44.7%). No clinicopathological variable was significantly associated with biological behaviour. The overall concordance rate was 21.3% (Cohen's κ = −0.575; 95% CI: −0.792 to −0.357; poor agreement). Differentiation type (χ² = 11.655, p = 0.003) and anatomical site (χ² = 7.881, p = 0.049) were the only significant concordance predictors; sebaceous tumours (80.0%; OR = 28.0) and head and neck location (38.1%) showed the highest rates, while extremity tumours were uniformly discordant (0%). Conclusion: A 78.7% discordance rate confirms the indispensability of histopathological examination for all excised adnexal lesions. Sebaceous differentiation and head-and-neck location predict better clinical recognition; extremity lesions require heightened diagnostic vigilance.
Priyadharshini et al. (Sun,) studied this question.