Nitric oxide (NO) is a free radical gas and a short-lived messenger which has many paracrine functions. Direct assessment of NO production is very difficult in vivo. However, the paranasal cavities generate a high amount of NO which diffuses in the nasal cavity where it can be easily measured. Several studies have suggested alterations of the NO production in heart failure. Thus, we assessed nasal NO concentration in normal subjects and in heart failure patients. The nasal NO concentration averaged 227 +/- 10 ppb in the control group (n = 20), and 210 +/- 10, 198 +/- 20 and 159 +/- 54 ppb in New York Heart Association (NYHA) class II (n = 30), III (n = 28) and IV (n = 7) patients, respectively (mean +/- standard error [SE], not significant using analysis of variance [ANOVA]). Nasal NO level was not influenced by age, sex or etiology of the heart failure or by treatment with frusemide, angiotensin-converting enzyme inhibitor or digoxin. However, treatment with NO-releasing drugs (nitrates or molsidomine) significantly decreased the nasal NO level in heart failure patients. A two-way ANOVA revealed that treatment with a NO-releasing drug influenced nasal NO concentration (P = 0.0005), whereas NYHA class did not (P = 0.23), with a trend towards an interaction between the two parameters (P = 0.09): the inhibitory effect of NO-releasing drug on nasal NO concentration was more pronounced in severe heart failure. In an additional group of 12 patients (NYHA class II or III), the nasal NO concentration was 174 +/- 19 ppb during NO-releasing drug treatment and increased to 231 +/- 27 ppb 3 days after withdrawal of the nitrates (P = 0.0007 using paired t-test). Conversely, the nasal NO concentration in another group of seven patients (NYHA class II or III) was 219 +/- 32 ppb without nitrate treatment and decreased to 188 +/- 28 ppb 7 days after nitrate addition (P = 0.02 using paired t-test). In contrast, the nasal NO concentration in another group of ten ischemic patients without heart failure was 203 +/- 25 ppb without nitrate treatment and was similar (207 +/- 28 ppb) 7 days after nitrate addition (not significant using paired t-test). In conclusion, nasal NO production is normal in heart failure, except in patients receiving NO-releasing drugs. Nasal NO concentration could be useful for investigating the mechanism(s) by which exogenous NO donors decrease endogenous NO production.