Osmotic pressure measurements of serum and ascitic fluid in patients with cirrhosis of the liver.

Numerous contributions regarding the pathogenesis of ascites and edema have been made since Starling's first description of the balance between serum and tissue hydrostatic and osmotic forces. The original concept (1) proposed that, at equilibrium, serum protein osmotic pressure minus tissue fluid osmotic pressure equals capillary hydrostatic pressure minus tissue hydrostatic pressure. Direct application of this relationship to the problem of ascites received relatively little attention until recently when Mankin and Lowell (2) and James (3) employed this approach for the approximation of the degree of elevation of portal pressure. Further clarification of the applications and limitations of this principle to the pressure relationships in the formation of ascites represented the primary purpose of the present study. The importance of the serum protein osmotic pressure was emphasized in early studies suggesting that a critical level existed, approximately 240 to 270 mm. H2O, below which edema would occur (4-8) This view was questioned by Kylin, however, who performed plasmapheresis experiments in rabbits and was able to obtain very low serum osmotic pressure values without edema (9). However, when he added salt to the diet, the animals became edematous even at serum osmotic pressure levels near normal. The clinical counterpart of these experiments appears in the cases described in the literature (10) in which edema was absent at serum osmotic pressures below 200 mm. A similar concept has developed concerning the relationship of the serum osmotic pressure to the formation of ascites. Butt, Snell, and Keys (11) found that ascites occurred most frequently when the serum osmotic pressure was below 240 mm. Bjorneboe, Brun, and Raaschou (12) emphasized this critical level and concluded that a reduced serum osmotic pressure was the principal factor in the pathogenesis of ascites and that portal hypertension played a subordinate role. As they pointed out, the data were obtained from patients with cirrhosis following hepatitis. Other investigators (13) reached similar conclusions stating that ascites was probably not due to liver disease when the serum osmotic pressure was greater than 300 mm. H20. Armstrong (14) suggested a critical level for the "effective" osmotic pressure (serum osmotic pressure minus ascitic osmotic pressure) below which ascites occurs. Indirect evidence against the concept of a strict relationship between serum osmotic pressure and ascites formation has been brought forth by several investigators (15, 16). Evidence was obtained in the present study indicating that the type of cirrhosis is an important factor to be considered in evaluating the relationship between ascites and serum osmotic pressure levels.