34 



HANDBOOK OF PHYSIOLOGY 



CIRCULATION 



right atrium monitored, the time-concentration curve 

 for returning label is so complex that mean transit 

 time for the system as a whole cannot be estimated 

 (Lawson, unpublished data). It may be significant, 

 however, that in the dog, within 20 to 40 sec after in- 

 travenous injection of labeled cells and plasma, 

 samples drawn from the aorta show plasma label to 

 have decreased its concentration in blood by about 20 

 per cent more than cell label (234). Figure i shows a 

 confirmation of this observation by our laboratory, 

 and demonstrates, in addition, that the excessive 

 distribution space for plasma label persists for at least 

 4 min. 



The method devised by Bradley and his colleagues 

 (26) for estimating regional vascular volumes com- 

 putes the quantity of label distributed within a region 

 from the volume of flow and the accumulated differ- 

 ence in label activity of inflowing and outflowing 

 blood until the two activities have become equal. If 

 label activity throughout the region at the instant of 

 equilibration is assumed to be the same as that of the 

 blood flowing in and out at that instant, volume may 

 be calculated from equation i . By using cell and 

 plasma labels simultaneously, and making separate 

 calculations for each, this procedure has been used 

 in man to estimate the hematocrit of the splanchnic 

 vascular bed as about 91 per cent of the central hema- 

 tocrit (136). The mean transit times, calculated in 

 these studies as volume/flow, were 36 sec for tagged 

 cells, and 42 sec for tagged plasma. Since the spleen 

 was included in the circuits studied, the data are 

 difficult to reconcile with data on the splenic circuits 

 alone in barbitalized dogs, which show mean splenic 

 transit times of 19 min for cells, and of 3.75 min for 

 plasma (218). The discrepancy seems too great to be 

 attributed to diff"erences, however large, in the cell- 

 storage function of the spleen in man and in the 

 barbitalized dog. The ratio between the mean circu- 

 latory hematocrit and that of drawn blood is reported 

 to be unchanged in splenectomized dogs by complete 

 evisceration (91). After the spleen has been removed 

 from the splanchnic area, accordingly, the hematocrit 

 of the region appears to be typical of the circulatory 

 system as a whole. 



Attempts to withdraw or express blood from the 

 minute vessels for direct determination of its hema- 

 tocrit have given equivocal results. Ebert & Stead 

 (62) reported that blood stripped from the upper ex- 

 tremity in man by means of an elastic bandage had a 

 lower hemoglobin content than blood drawn from the 

 otiier arm. Since the procedure produces circulatory 

 stasis and local asphyxia, however, it is not at all cer- 



tain that the expressed fluid represents the normal 

 contents of the small vessels. It also seems unlikely 

 that a representative sample could be obtained 

 through a catheter wedged distally in an artery, since 

 any significant volume drawn through such a catheter 

 must have come from collaterals. This probably ex- 

 plains a recent finding that blood drawn through 0.5 

 mm tubes wedged distally in the dog's foreleg arteries 

 has the same hematocrit as blood drawn from large 

 vessels (99). A report that external compression of the 

 kidney expels blood through the renal vein with the 

 same hematocrit as large vessel blood is more difficult 

 to evaluate (163). It appears to be in conflict with the 

 finding that cell-poor blood drains spontaneously 

 from the kidney after the renal artery has been 

 clamped (230). If a peripheral layer of adherent, 

 stagnant plasma is maintained in vessels despite the 

 viscous drag of blood flow through them, it seems im- 

 probable that it could be displaced by any of these 

 procedures. 



Excess Plasma 



In the original study of Smith el al. (222), which 

 first demonstrated the error in CVu, it was estimated 

 that a volume of cell-free plasma equivalent to g per 

 cent of the total intravascular space could explain the 

 error. Although this amount of cell-free plasma does 

 not seem preposterous, it is obviously an utider- 

 estimate if there are also cell-rich areas, such as the 

 spleen. Since equal volumes of cell-poor and com- 

 parably cell-rich blood would cancel out in the 

 calculation, any estimate of cell-free plasma repre- 

 sents not its total volume, but only its excess over 

 comparably cell-rich blood. 



When plasma volume and cell volume are measured 

 independently and at the same time, the virtual vol- 

 ume of cell-free plasma can readily be calculated. 

 This is the difference in the plasma volume measured 

 from the distribution of plasma label, and the volume 

 which would be required to suspend the total cell 

 volume in plasma at the hematocrit found for the 

 central circulation. For example, Huggins and his 

 colleagues (personal communication) have found 

 average values for 100 morphinized-nembutalized 

 dogs as follows: Cell volume (by Fe»', Cr»', or P^^) = 

 33.5 ml per kg; plasma volume (by T-1824 or io- 

 dinated albumin) 50.2 ml per kg; venous hematocrit 

 0.452. If cells were associated with plasma throughout 

 their distribution space in the same ratio as in venous 

 blood, there would be a plasma volume equal to 33.5 

 X .548 '.452 = 40.6 ml per kg. The excess plasma is 



