Background: Erythroderma is a morphologic presentation of a variety of cutaneous and systemic diseases. Diagnosing erythroderma is very challenging mainly as regards the differentiation between neoplastic and benign (inflammatory) causes. Aim of work: This work aimed at studying cases presenting with erythroderma, clinically, histopathologically and immunohistochemically in order to develop criteria that help in reaching a final diagnosis and in differentiating between benign (inflammatory) and neoplastic causes of this condition. Patients and methods: Thirty patients with erythroderma aged more than 12 years old were included. Each was subjected to complete medical history, thorough general and cutaneous examination as well as laboratory investigations. Accordingly, a provisional clinical diagnosis was reached. Two Skin biopsies were examined blindly to any clinical data and a histopathological diagnosis was defined. Cases with mismatched provisional clinical and histopathological diagnoses as well as histopathologically unsettled cases were subjected to further clinical correlation and immunohistochemical study of the expression of CD2, CD5, CD7, CD4 and CD8 to verify the diagnosis. Immunohistochemistry was also done for cases representing both inflammatory and neoplastic conditions in order to characterize their immunophenotypic profile. Results: Based on the constellation of clinical, histopathological and immunohistochemical data, the final diagnoses of studied cases included: 15 cases (50%) of erythroderma due to a pre-existing dermatosis namely: 12 cases of psoriasis (40%), 2 cases of atopic dermatitis (6.7 %), and 1 case of pemphigus foliaceus (3.3%). The remaining 15 cases were diagnosed as drug induced erythroderma in 7 cases (23.3%), 1 case (3.3%) of papuloerythroderma of Ofuji and 7 cases (23.3%) of erythrodermic CTCL which included 1 case (3.3%) of Sézary syndrome and 6 cases (20%) of erythrodermic MF. On clinical basis, a diagnosis of psoriasis was significantly associated with fiery red erythema, dry white silvery scales, absent oozing, palmoplantar scaling and erythema (P value =0.003, <0.0001, =0.031 and 0.007 respectively). A diagnosis of CTCL was significantly associated with dusky red erythema, fine branny scales, infiltrated areas, alopecia and palmoplantar keratoderma (P value =0.003, <0.0001, =0.043, =0.015 and 0.007 respectively). Eosinophilia was significantly seen in cases of drug induced erythroderma (P value = 0.015). On histopathological basis, psoriasiform hyperplasia, hypogranulosis and superficial perivascular infiltrate were significantly seen in cases of psoriasis (P value < 0.0001, = 0.011, =0.005 respectively). Irregular hyperplasia, epidermotropism, the presence of atypical lymphocytes in the infiltrate, band like infiltrate, dense infiltrate and wiry fibrosis in the papillary dermis were significantly seen in CTCL (P value = 0.001,= 0.002, <0.0001, =0.045, 0.027 and 0.001respectively). The presence of vacuolar degeneration along the dermoepidermal junction was significantly seen in cases of drug induced erythroderma (P value = 0.007). Immunohistochemical results showed variable expression of CD4 and CD8 in malignant cases with tendency towards loss of pan T cell markers and a CD8-ve profile while inflammatory conditions mainly showed a CD8+ve profile. Diagnosis of cases of erythroderma is a complex process that depends on a proper constellation of clinical, histopathological and immunohistochemical data. Multiple biopsies from different sites as well as repeated biopsies in chronic cases are highly recommended. Although immunohistochemistry is helpful, it has many limitations.