Abstract
Objective To define the metabolic profile of postmenopausal hormone replacement therapies when delivered through gels, patches, implants or other non‐oral routes. Such information may be useful in the absence of reliable clinical data on the effects of these therapies on the risk of cardiovascular disease. Design and methods Selective literature review. Patients Postmenopausal women. Results Non‐oral oestrogen therapies fail to invoke the hepatic response associated with oral therapy. Changes in hepatic protein synthesis are minimal and so plasma levels of binding globulins and other proteins tend to be normal. Many of the perturbations of the haemostatic system seen with oral therapy are avoided. In the absence of hepatic over‐synthesis of apolipoproteins, plasma lipoprotein levels are unchanged or reduced. The direct effects of oestrogen on vascular function are apparent when the hormone is administered non‐orally. Conclusions The net effect of non‐oral oestrogen therapies on the risk of cardiovascular disease is difficult to predict on the basis of current data. Some changes in plasma lipoprotein levels, such as the reduced fasting levels of triglycerides, would be considered desirable, but the cardioprotective increase in levels of high‐density lipoproteins is absent. The differential effect on haemostasis markers is promising, but preliminary data relating to transdermal patches fail to support the idea that non‐oral therapies will avoid the increased risk of venous thromboembolism associated with oral therapy. The ability of non‐oral therapies to improve vascular function implies that they will offer postmenopausal women at least some of the cardiovascular protection seen with oral therapy.