OBJECTIVE: In cross-sectional studies, the variability between women may mask or deny gestational changes, related to the nitric oxide-cyclic GMP system. Therefore, we analyzed longitudinally as markers of this system, the urinary levels of nitrite+nitrate (NOx), cyclic guanosine 3',5' monophosphate (cGMP) and of the inhibitor of nitric oxide synthase, dimethylarginine (DMA). STUDY DESIGN: Late-afternoon urine samples were obtained from 20 women with uncomplicated pregnancies and nine non-pregnant women. Creatinine concentrations (mol) were determined with the Jaffé reagent and NOx (mmol) with the Griess reagent after reduction of nitrate. cGMP (micromol) was determined in an enzyme immunoassay and DMA (mmol) after solid-phase extraction and liquid chromatography. Trend analyses and (paired) t-tests were done for detection of time-related differences. RESULTS: The NOx/creatinine (mmol/mol) ratios of the non pregnant women (63.8+/-18.8, S.D.) did not differ from those of the pregnant women at the onset of pregnancy (70.5+/-36.4). Over the entire pregnancy period these ratios declined significantly (P<0.001) and lower values were found at the end of gestation and after birth (49.6+/-22.4). The cGMP/creatinine (micromol/mol) and DMA/creatinine ratios (mmol/mol) changed parabolically (P<0.001). The maxima of 68.0+/-19.9 and of 4.95+/-1.01 were found at week 20 and 16, respectively. These ratios declined to 45.2+/-17.7 and to 4.03+/-0.83 at the end of gestation but not further during parturition (39.6+/-17.2 and 4.01+/-1.90). The lowest cGMP/creatinine ratios occurred one month after birth (27.4+/-15.7) while in the non-pregnant women the value was 15.3+/-6.2 microM/M. The lowest DMA/creatinine ratios, measured one month after birth (3.41+/-1.28 mmol/mol) were similar to those of the non-pregnant women (3.75+/-0.39 mmol/mol). Positive instead of negative relationships were found between the DMA results and those of the cGMP (P<0.001) and NOx determinations (P<0.05). CONCLUSIONS: (1) The gestational changes of the urinary NOx/creatinine and especially of the GMP/creatinine ratio reflect most likely changes in vascular resistance. (2) Because of the variability of the results between but also within women, these ratios are useless to monitor supposed changes in NO production during parturition.