Human and ovine amniotic fluid composition differences: implications for fluid dynamics.

Abstract

OBJECTIVE: The ovine model is frequently utilized to extrapolate data regarding fetal and amniotic fluid dynamics to human pregnancy. The ovine amnion is highly vascularized, facilitating intramembranous exchange of water and solutes between the amniotic fluid and fetal plasma. In comparison, the relatively avascular human amniotic membrane may have a reduced potential for intramembranous absorption. In view of these anatomical differences, we hypothesized that comparison of human and ovine amniotic fluid composition would provide insight into differences in the mechanisms of amniotic fluid exchange. METHODS: Amniotic fluid was sampled from 43 patients upon hospital admission, and from 27 ovine ewes at five days following amniotic fluid catheter placement. Both human (32 to 39 weeks' gestation) and ovine pregnancies (125 to 136 days' gestation) were sampled during the last 20% of gestation. Samples were analyzed for osmolality and sodium, potassium and chloride concentrations. The contribution of electrolytes to amniotic fluid osmolality and changes in osmolality and electrolyte composition versus gestational age were assessed by regression and covariance analysis. RESULTS: Mean (+/-SEM) amniotic fluid sodium concentration (134.6+/-1.9 vs. 127.1+/-2.0 mEq/1) was greater and potassium (4.6+/-0.1 vs. 6.1+/-0.6 mEq/l) and osmolality (263.9+/-3.7 vs. 285.1+/-1.6 mOsm/kg) less in human than sheep. The range of amniotic fluid osmolality was greater in human (223 to 336 mOsm/kg) than in sheep (274 to 298 mOsm/kg). Human amniotic fluid osmolality was highly correlated with amniotic fluid sodium (r = 0.97) and chloride (r = 0.96) while ovine amniotic fluid osmolality was only weakly correlated with amniotic fluid sodium (r = 0.75) and chloride (r = 0.51). The slope of the regression line of amniotic fluid sodium and osmolality was greater for human than for sheep amniotic fluid (P < 0.0001). The percent of amniotic fluid osmolality accounted for by sodium, chloride and potassium concentrations was greater for human (97%) than for sheep (86%; P < 0.0001). CONCLUSIONS: The results suggest that human amniotic fluid osmolality is comprised almost entirely of the major electrolytes while alternative solutes (e.g., fructose) contribute to ovine amniotic fluid osmolality. Extrapolation of fetal and amniotic fluid dynamics from ovine models to humans should incorporate differences in amniotic fluid osmolality and electrolyte composition.