OBJECTIVE Reference values are usually defined based on blood samples from healthy men or nonpregnant women. This is not optimal as many biological markers changes during pregnancy and adequate reference values are of importance for correct clinical decisions. There are only few studies on the variations of laboratory tests during normal pregnancies, especially during the first two trimesters. It is thus a need to establish such reference values. METHODS Longitudinal study of laboratory markers in normal pregnancies. METHODS Uppsala University Hospital, Sweden. METHODS Healthy pregnant females. METHODS We have studied 25 frequently used laboratory tests during 52 normal pregnancies. Each woman was sampled up to nine times and the samples were divided according to collection time into the following groups: gestational week 7-17; week 17-24; week 24- 28; week 28-31; week 31-34; week 34-38; predelivery (0-2 weeks before delivery) and postpartum (> 6 weeks after delivery). The 2.5 and 97.5 percentiles for these markers were calculated according to the recommendations of the International Federation of Clinical Chemistry on the statistical treatment of reference values. RESULTS Reference intervals are reported for plasma alanine aminotransferase, albumin, alkaline phosphatase, pancreas amylase, apolipoprotein A1, apolipoprotein B, aspartate aminotransferase, bilirubin, calcium, chloride, creatinine, cystatin C, ferritin, gamma-glutamyltransferase, iron, lactate dehydrogenase, magnesium, phosphate, potassium, sodium, transferrin, triglycerides, thyroid-stimulating hormone, urate and urea during these pregnancy periods. CONCLUSIONS Most of the analytes change during normal pregnancy. It is thus of importance to use special reference values during pregnancy.