Relevant English language articles published from 1966 to 1995 regarding ethanol therapy in the treatment of thyroid and parathyroid diseases were identified through a MEDLINE search and manual searches of identified articles. The sclerosing properties of ethanol have been recognized for many years and have offered interventional possibilities in the management of various benign as well as malignant lesions. The mechanism of action of ethanol appears to be related to a direct coagulative necrosis and local partial or complete small vessel thrombosis. Ultrasound-guided percutaneous ethanol injection therapy (PEIT) is rapid and performed on an out-patient basis and has now gained wide acceptance due to the accumulating evidence of the efficacy and safety of this therapeutic tool. Yet, there is a lack of prospective, randomized clinical trials comparing PEIT with 131I therapy or surgery with regard to its effects, especially long-term ones and it should therefore still be considered an experimental procedure. In benign endocrine diseases, PEIT has shown promising results in the treatment of autonomous thyroid nodules, benign solitary cold solid as well as cystic thyroid nodules and parathyroid tumours. Its use in pretoxic and toxic thyroid nodules has been evaluated in several uncontrolled studies, all demonstrating a high success rate in spite of the large number of treatments needed. So far efficacy and cost-effectiveness seem inferior to 131I and surgery. Short-term results of PEIT in benign cystic thyroid nodules are convincing with a high cure rate, but no controlled studies with long-term results are available. Preliminary results suggest that PEIT could become an alternative to surgical excision or levothyroxine therapy in the symptomatic solid cold benign thyroid nodule. Ultrasound-guided PEIT of parathyroid tumours has proven to be a useful method in highly selected patients in whom surgery has been found non-attractive and medical treatment ineffective. However, no prospective randomized trials have been published comparing the results of PEIT in parathyroid tumours with conventional surgical and medical treatments. PEIT has never been tested against standard therapy, but seems inferior to 131I and surgery. Side-effects caused by ethanol injection are generally few and transient and are related to the injection into solid nodules rather than cysts. Ethanol injection into solid profund nodules may seriously jeopardize subsequent surgery because of perinodular fibrosis. As an experimental procedure, not yet evaluated sufficiently, it should be reserved for patients who cannot or will not undergo standard therapy. Caution in routine use is advisable.