Do abnormal Starling forces cause fetal hydrops In red blood cell alloimmunization? Kenneth J. Moise, Jr., MD, Robert J. Carpenter, Jr., MD, and Diane E. Hesketh, RN Houston, Texas OBJECTIVE: The purpose of the current investigation was to ascertain whether derangements in umbilical venous pressure and plasma colloid osmotic pressure are involved in the pathophysiologic condition of immune hydrops fetalis. STUDY DESIGN: Umbilical venous pressure (corrected for ambient amniotic fluid pressure) and colloid osmotic pressure were measured during intravascular transfusion. Fetal hydrops was defined as the presence of ascites by ultrasonography. The Mann-Whitney test was used for comparison of groups; a value of p < 0.05 was considered statistically significant. RESULTS: Fifteen hydropic fetuses were matched for gestational age with 15 nonhydropic fetuses also undergoing intrauterine transfusion for anemia. On comparison with their nonhydropic counterparts, hydropic fetuses had a statistically lower colloid osmotic pressure. Umbilical venous pressure was higher and the colloid osmotic pressure - umbilical venous pressure gradient was lower in association with fetal hydrops although these differences did not achieve statistical significance. CONCLUSION: Mild abnormalities of intravascular Starling forces may playa role in the formation of hydrops in anemic fetuses. (AM J OSSTET GVNECOL 1992;167:907-12.) Key words: Rhesus alloimmunization, hydrops fetalis, umbilical venous pressure, colloid osmotic pressure In 1896, Starling' described the various forces that affect fluid movement at the level of the capillary. He proposed that colloid osmotic pressure of interstitial proteins and capillary hydrostatic pressure were the major factors affecting fluid movement from the in- travascular compartment to the interstitial space. The balancing forces that caused fluid to shift back into the intravascular space from the interstitium were the hy- drostatic pressure of the interstitial fluid and colloid osmotic pressure of plasma proteins. In spite of the widespread use of rhesus immune globulin, red blood cell alloimmunization continues to contribute to perinatal morbidity and mortality.2 With the recent advent of ultrasonography and diagnostic in utero umbilical blood sampling, fetal hydrops has been found to occur only when severe anemia results from the transplacental transfer of maternal red blood cell antibodies." In 1932, utilizing the various permuta- tions of Starling forces that would produce a net move- ment of fluid into interstitial spaces, Diamond et al.I proposed three mechanisms for the development of hydrops fetalis. The first hypothesis conjectured that extramedullary hematopoiesis in the liver could lead to From the Division of Maternal-Fetal Medicine, Department of Ob- stetrics and Gynecology, Baylor College of Medicine. Presented at the Twelfth Annual Meeting of the Society of Perinatal Obstetricians, Orlando, Florida, February 3-8, 1992. Reprint requests.' Kenneth J. Moise, J r .• MD, Department of Obstetrics and Gynecology, One Baylor Plaza, Houston, TX 77030. 616139689 depressed production of serum proteins. This would in turn lead to a decrease in plasma colloid osmotic pressure and a loss of fluids from the intravascular compartment. In the second proposed mechanism, congestive heart failure in the fetus would result in elevated hydrostatic pressure in the capillaries. Finally, severe tissue hypoxia could lead to capillary damage and enhanced permeability to fluids and proteins. More recently, Bowman' suggested that liver congestion in association with extramedullary hematopoiesis causes portal and umbilical venous hypertension leading to ascites and eventual anasarca. To date, animal and human neonatal research has not validated any of the previously proposed mecha- nisms for immune hydrops fetalis. The purpose of the current investigation was to ascertain if derangements in umbilical venous pressure and colloid osmotic pres- sure are involved in the pathophysiologic hydrops for- mation in the anemic human fetus. Material and methods The study was approved by the human investiga- tional review committees of Baylor College of Medicine and St. Luke's Episcopal Hospital. Between December 1987 and December 1991, 126 transfusions were performed on 46 fetuses at Baylor College of Medicine. Just before each procedure, a complete ultrasonographic examination was per- formed (Acuson 128, Mountain View, Calif.) with a 3.5 907