Reference values for clinical chemistry tests during normal pregnancySir, I read with interest the article written by Larsson et al. 1 on reference values for clinical chemistry tests during normal pregnancy. Their effort is to be applauded, first, in highlighting the importance of the effect of normal pregnancy on commonly measured biological markers and hence the need for caution not to interpret physiological changes in normal pregnancy as pathological; second, in being able to consolidate and define reference intervals for as many parameters as they have in their article.I, however, have some concerns with their article; they stated that 'Previous studies of normal variations during pregnancy are incomplete and often limited to the third trimester and focused on a single or a few analytes'; this, however, is not entirely accurate and furthermore the concept they are talking about cannot be claimed by them because it has been out there for years. 2 Davison et al. 3 in work spanning well over a decade, has demonstrated unambiguously, changes in osmoregulation function in the three trimesters of normal pregnancy, which we now know are physiological, but would have been interpreted as pathological in the context of nonpregnant reference values.They demonstrated that, plasma osmolality falls by 8-10 mOsm/kg in the first few weeks of pregnancy, a relative hypotonicity that is maintained until parturition. This decrease in osmolality is accompanied by parallel decreases in the main extracellular fluid electrolytes, mainly sodium and its attendant ion chloride, as well as urea. 3 Furthermore, this group demonstrated robustly that the explanation for the observed decreases was not due simply to haemodilution, a commonly held and accepted explanation for most of the observed physiological changes in normal pregnancy, but was due instead to the lowering of the osmotic thresholds for both vasopressin release and thirst to the observed lower osmolality values. 4 This explanation, they argued, was why normally pregnant women were able to continue to concentrate and dilute their urine normally, about their new lower osmolality levels instead of being in a state of continuous diuresis as one would have expected if it was simply due to haemodilution.Finally, I would suggest that the authors now do a thorough search of the literature in order to make their list of reference complete because in its current state, they have failed to acknowledge previous similar and important work that have been undertaken by other groups.