Organic nitrates are well established in the treatment of a wide variety of cardiovascular disorders, most notably angina pectoris and congestive heart failure. However, attenuation of, or tolerance to, haemodynamic and anti-ischaemic effects may occur with all long-acting nitrate formulations. In the majority of patients continuous administration of long acting nitrates tends to promote the development of attenuation, while intermittent administration avoids it. Likewise, higher-doses appear to induce attenuation to a greater degree than lower doses. Attenuation of haemodynamic effects and exercise tolerance in heart failure patients, and of clinical end-points in angina patients, appears to be less than attenuation of exercise testing end-points in angina patients. While the use of intermittent therapy avoids the development of attenuation, it may expose the patient to an as yet undefined risk of silent and/or symptomatic anginal episodes occurring during the nitrate-free interval. Likewise, the role of concomitant therapy in avoiding this potential risk remains to be defined. Means of avoiding attenuation may include the coadministration of sulfhydryl donors such as N-acetylcysteine. Alternatively, angiotension-converting enzyme (ACE) inhibitors such as captopril may block renin-angiotensin system-induced reflex sympathetic stimulation. Attenuation may occur to a greater or lesser degree in individual patients. The proportion of attenuators vs non-attenuators remains to be defined, as does a means of identifying such patients prospectively by clinical and/or laboratory parameters. Conflicting results among smaller studies may reflect variable proportions of attenuators vs non-attenuators. However, conflicting results among larger studies may reflect differences in patient selection criteria, such as selecting patients with positive and reproducible stress tests and little in the way of spontaneously occurring angina versus selecting patients with positive but variable stress tests and frequent episodes of spontaneously induced angina. The former group may reflect pure fixed coronary artery disease with little in the way of vasospasm, or change in vasomotor tone, while the latter group may reflect greater variability in vasomotor tone and/or more in the way of plaque instability. The clinical efficacy of long acting nitrates might therefore be expected to be greatest in those patients with larger numbers of spontaneously occurring angina episodes. Recent data suggest that nitrates may have important direct effects on coronary vessels including dilating eccentric coronary stenoses, dilating intercoronary collateral channels and having greater dilating effects on more diseased segments as opposed to less diseased coronary segments.(ABSTRACT TRUNCATED AT 400 WORDS)