Presented here is a stable isotope dilution technique for determining cortisol production rate (CPR). The method involves extraction and derivatization of cortisol isoforms from serum (0.5 ml), separation of derivatives by gas chromatography, and detection by using negative ion chemical ionization mass spectrometry. This method provides 50-100-fold greater sensitivity than positive ion mass spectrometry and allows for estimations of cortisol production rate with the use of small amounts of pooled serum, even in the presence of high concentrations of lipophilic contaminants. The area under the curve for the total selected ion chromatogram of fluoroacyl derivatives of cortisol (d0, m/z 782) and deuterated cortisol (d3, m/z 785) were used to determine the isotopic dilution ratio in three types of samples: 1) standards: d0/d3 ratios ranging from 1 to 8%; 2) controls: d3-cortisol added to serum with known cortisol concentration; 3) subjects: 24-h pooled serum samples (q 30 min over 24 h) from healthy children (male 10-13 years; female 7-11 years) receiving continuous infusions of d3-cortisol at 2-4% of their estimated CPR. Recovery after the solid phase extraction and derivatization process was >90%, as determined by thin-layer chromatography. Expected versus measured ratios for d3/d0 in standards and serum controls were highly correlated (r2(standard) = 0.99; r2(control) = 0.99) over a wide range of d3-cortisol enrichment (1.0-10.0%). Mean 24-h CPRs were 4.8 +/- 0.6 mg/m2/24 h (mean +/- SEM, n = 7) in male children and 4.4 +/- 0.5 mg/m2/24 h in female children (n = 4). These CPR values are lower than those derived by radio tracer methods, but are in agreement with previous isotopic dilution studies. This technique is an important tool for assessing CPRs in a wide range of disease states affecting cortisol production.