Commercial sources for neuropeptide radioimmunoassays have made this sensitive tool available to clinical investigators for monitoring the potential involvement of neuropeptides in pain modulation. We measured substance P-like immunoreactivity in the plasma, saliva, and pericardial fluid of subjects with and without pain (chronic and acute) to determine if substance P levels are altered. Some recent studies have suggested that substance P in various body fluids may be a correlate of chronic pain. To test this correlation it is important to ensure that the assay is measuring what it was designed to measure. Therefore, the influence of three tachykinins on the analysis of substance P concentrations was assessed with a commercially available radioimmunoassay kit. A small (approximately 2 to 6%), apparently nonspecific elevation in measured substance P was found when alpha-neurokinin, beta-neurokinin, or eledoisin was incubated with substance P and its antibody. Our results also indicate an apparent specific affinity of the substance P antibody for alpha-neurokinin (above 1,000 pg/ml) and beta-neurokinin (above 5,000 pg/ml). Substance P levels in the body fluids we tested ranged from 0.47 to 62.88 pg/mg protein (47.4 to 230.8 pg/ml). Levels of the tested tachykinins have not been determined in body fluids. If alpha-neurokinin or beta-neurokinin is found to be present in high concentrations in these fluids, this commercially available substance P kit may overestimate substance P levels. The concentrations of tachykinins necessary to interfere specifically with the assay are 10- to 100-fold higher than substance P in body fluids.(ABSTRACT TRUNCATED AT 250 WORDS)