36 



HANDBOOK OF PHYSIOLOGY 



CIRCULATION I 



the total blood volume which occupies the heart, the 

 minute vessel segments of the entire circulatory system 

 would contain i 7.4 per cent of the total blood volume. 

 Takinp; the 0.3 mm upper limit of table 5 as the most 

 liberal estimate, and assuming that the mean hema- 

 tocrit in all vessels smaller than 0.3 mm diameter is 

 one-half that of blood in the central circulation, room 

 can be found within tlie anatomical confines of the 

 circulatory system for excess plasma equal to no more 

 than 13 per cent of the total volume. Our concepts of 

 the quantitative anatomy of the cardiovascular system 

 need revision if 25 per cent of it is to be filled with cell- 

 free plasma in the rat (236), 19 per cent in the 

 splenectomized dog (Lawson, unpublished data), and 

 17 percent in the monkey (97). 



Recent studies raise serious doubts about the ability 

 of blood capillaries to keep even the largest macro- 

 molecular plasma labels within the confines of the 

 vascular system. The isolated perfused hind limb of 

 the dog under certain experimental conditions be- 

 haves as though the capillaries have pores of fairly 

 uniform radius in the range of 35 to 45 A (178). 

 Electron microscopy of a variety of capillaries, how- 

 ever, reveals no such uniformity. Capillaries in mus- 

 cle, lung, and the central nervous system have no 

 visible pores, whereas those in the kidneys, intestinal 

 villi, and some endocrine glands do. At the other 

 extreme, there are sizeable gaps between endothelial 

 cells in the liver and spleen (21). The appearance of 

 dextrans of graded molecular size in leg, cervical, and 

 hepatic lymph in the dog following their intravenous 

 injection suggests that, in addition to pores of rather 

 uniform radius, there are an even larger number of 



o 



leaks permitting bulk flow of particles up to 350 A 

 radius (98). Furthermore, the permeability to large 

 molecules seems to be increased in the dog by ex- 

 pansion of the plasma volume. The concentration in 

 lymph of previously injected radioiodinated albumin 

 or dextran is increased when plasma volume is in- 

 creased by infusing large volumes of albumin solution, 

 at the same time that the concentration in plasma is 

 lowered by dilution. The increase in lymph concen- 

 tration occurs for all dextrans up to mol wt 255,0(}0, 

 but seems to be more marked in the case of the larger 

 molecules (215). 



In the rabbit there appears to be a rapid flux of 

 large molecules such as antipneumococcus globulin 

 between plasma and extravascular regions. The 

 plasma titre falls to about one-half within 24 hours 

 after these globulins are injected, and rises, presum- 

 ably by exchange with extraxascular pools, for about 



the same length of time after normal plasma has been 

 infused (85). There is good reason to believe that in 

 the cat I"' albumin spreads to a considerable extra- 

 vascular distribution in the kidney within 5 min of its 

 injection. Lowering arterial pressure in the cat causes 

 an increase in both the hemoglobin content and the 

 f' content of the kidney, such that if both labels re- 

 mained intravascular, and their increase represented 

 an increase in cell and plasma volume, kidney weight 

 must increase by about 6.5 per cent. Since the weight 

 of the kidney not only failed to increase, but actually 

 decreased significantly, it was concluded that the in- 

 crease in I'^' content did not represent an increase in 

 plasma volume, but an increase in the distribution 

 space of this label beyond the vascular system (179). 

 This may be comparable to the extravascular spread 

 of I'^' albumin which has been reported for the small 

 intestine of the rabbit, where I'^' appears in intestinal 

 juice in presumably undamaged loops within a few 

 minutes of label injection {14). 



It is obviously not possible, on the basis of present 

 knowledge, to reconcile the operational definition 

 of plasma volume with its conceptual definition. 

 Whether the conceptual definition needs revision to 

 permit bulk flow of plasma beyond the confines of 

 the vascular system remains to be seen. Differences in 

 the distribution of individual plasma labels are dis- 

 cussed in the next section. 



PL.ASMA LABELS 



Barratt & Yorke (20) in 1909 injected a solution of 

 hemoglobin into rabbits, and calculated plasma vol- 

 ume from its concentration in plasma samples. This 

 appears to have been the first use of a plasma label 

 whose concentration, of necessity, was measured in 

 plasma rather than in whole blood. The following 

 year Abderhalden & Schmid ( i ) injected dextrin 

 into dogs and calculated plasma volume from plasma 

 concentrations. Regarding this simply as a technically 

 necessary step toward the calculation of lilood volume, 

 only the latter values were reported. From the data 

 given in their protocols, however, it is apparent that 

 the plasma volumes found in the three dogs of their 

 study were 66.6, 64.5, and 57.7 ml per kg, respectively. 

 The average plasma volume found for dogs with 

 T-1824 by modern techniques is about 48 ml per kg 

 (25). It seems probable, therefore, that fairly large 

 losses of de.xtrin occurred during the 5-minute period 

 which Abderhalden and Schmid allowed for mixing. 



