Body fluid osmolality and tonicity in preterm infants

In adult patients, a recent physiological approach for the osmoregulatory system based on body fluid tonicity (the so-called effective osmolality) seems to provide better information on water movements than does the classical body fluid osmolality. To evaluate whether plasma or urinary tonicities could give a better assessment of osmoregulation than plasma and urine osmolalities in sick preterm infants cared for in a NICU. A prospective study was conducted in 30 preterm infants (BW=1284+/-377 g; GA=28.8+/-1.7 weeks). Fifteen consecutive 8-h urine collections were performed for each infant from the 8th h of life (450 periods). A plasma sample was obtained at the end of each urine collection. Sodium, potassium, creatinine, osmolality and tonicity were measured or calculated in urine and blood samples as often as possible. Hypernatremia (PNa=146-149 mmol/l) was observed in seven infants (23.3%) and in 5.9% of the periods. Fifty-three percent of the infants and 20.4% of the periods presented with plasma hyperosmolality (>300 mosmol/kg H2O). The relationship between Posm and PNa was significant, but the clinical relevance was weak (r(2)=0.411; P<0.001). Plasma osmolality (Posm) positively correlated with urine osmolality (Uosm), but did not correlate significantly with CH2O/100 ml GFR. Plasma tonicity (2x(PNa+PK)) positively correlated with both urine tonicity (2x(UNa+UK)) and effective water clearance (EWC/100 ml GFR). On an individual basis, the linear relationship between urine and plasma osmolalities was significantly weaker than the relationship between urine and plasma tonicities. This study suggests that the calculation of plasma and urine tonicities allows a better assessment of water movements in body fluid compartments than plasma and urine osmolalities.