Bioimpedance spectroscopy (BIS) is a technique of interest when human pregnancy is studied since it can assess extracellular water (ECW), intracellular water (ICW) and total body water (TBW) as ECW plus ICW. The technique requires appropriate resistivity coefficients and has not been sufficiently evaluated during the reproductive cycle. Therefore, in a methodological study we estimated ECW, ICW and TBW, by means of BIS, and compared the results with the corresponding estimates obtained using reference methods. Furthermore, results obtained by means of population-specific resistivity coefficients were compared with results obtained by means of general resistivity coefficients. These comparisons were made before pregnancy, in gestational weeks 14 and 32, as well as 2 weeks post partum in 21 healthy women. The reference methods were isotope and bromide dilution. Average ICW, ECW and TBW, estimated by means of BIS, were in agreement with reference data before pregnancy, in gestational week 14 and post partum. The corresponding comparison in gestational week 32 showed good agreement for ICW, while estimates by means of BIS were significantly (p<0.001) lower than the corresponding reference values for ECW and TBW. Thus the BIS-technique, which was based on a model developed for the non-pregnant body, estimated increases in ICW accurately, while increases in ECW and TBW tended to be underestimated. Estimates obtained using population-specific and general resistivity coefficients were very similar. In conclusion, the results indicated that BIS is potentially useful for studies during pregnancy but that further work is needed before it can be generally applied in such studies.