VOLUME OF BLOOD 



^7 



predicted values. Reliable data are scattered because 

 of the need for multiple sampling at short intervals 

 over a considerable period. Most investigators have 

 been concerned only with obtaining evidence that 

 mixing is complete after a liberal time has been al- 

 lowed. 



The Observed Mixing Time for Labeled Cells 



In the earliest studies with Fe^'-tagged cells in 

 dogs, the conclusion was reached that cell mixing is 

 complete in 4 to 5 min, since no decline in sample 

 activity was observed thereafter (106). The conclusion 

 appears to have been based on inadequate data, how- 

 ever, as it was subsequently found that injections of 

 epinephrine even later than this produced a drop in 

 circulating iron activity (loi). In barbitalized dogs, 

 whose splenomegaly and muscular inactivity would 

 be expected to prolong mixing, most investigators 

 allow 10 min. But if the spleen is removed 30 min 

 after injection of Cr^'-labeled cells, it is found to con- 

 tain cells, unmixed with label, equivalent to 4 to 5 

 per cent of the total cell xolume (131)- By using 

 equation 2 to distinguish mixing from tag elution, it 

 has been possible to show that P''--labelcd cells may 

 require as much as 40 min for mixing in intact 

 barbitalized dogs (143). 



In human studies, samples are usually not taken at 

 sufficiently short intervals to determine when cell 

 mixing is actually complete. Early data with carbon 

 monoxide suggested a 20-min mixing requirement 

 (54), but much of this is now attributed to an expand- 

 ing extravascular distribution. When cells labeled 

 with K''- and with P'^- are injected in normal men, the 

 initial fast disappearance phase is terminated for both 

 labels in 7.5 to 10 min (23). The mixing period for 

 ?'' cells in man appears to be less than 20 min, since 

 their distribution space is the same at 20 and at 30 

 min (194). Another study in man suggests that the 

 period is less than 15 min, since vigorous exercise at 

 that time caused no change in circulating P^' activity 

 (172). It may be prolonged somewhat, however, by 

 dependent pooling of blood in the erect posture (173), 

 and in cases of splenomegaly, where cell-mixing may 

 require as much as 30 min (204), a mixing time ap- 

 proaching that in the dog with spleen enlarged by 

 barbiturates. Very long mixing times, up to i hour, 

 have been reported for Cr^' cells in battlefield casual- 

 ties with lacerating wounds (184). This greatly exceeds 

 the prolongation of mixing for plasma labels which 

 has been reported for men in states of shock (i 70). 



In small animals, such as rats and mice, groups are 



usually sampled at difl'erent times to avoid volume 

 depletion. In studies of this kind on normal unanes- 

 ihetized rats, Cr*' cells have been found with 

 essentially the same distribution space at 15, 30, and 

 60 min (180). Similar data have been obtained in 

 mice with Fe^' cells (74). Cell-mixing times were not 

 reported in recent studies on horses (126) and swine 

 (30). It is unfortunate that more precise data are not 

 available for animals of extreme size. 



The Observed Mixing Time for Labeled Plasma 



The predicted minimal mixing time would under 

 most conditions be somewhat greater for labeled 

 plasma than for labeled cells. As is shown in a later 

 section, the ratio cells: plasma in the blood pumped by 

 the heart (central circulation) is usually higher than 

 the mean ratio for the whole circulatory system. As a 

 consequence, a larger fraction of the total cell volume 

 than of the total plasma volume passes through the 

 heart in any interval of time. It is apparent, however, 

 that mixing time cannot be predicted from these 

 simplest parameters, since most of the available data 

 fail to reveal a significant difference in the mixing 

 time for cell and for plasma labels. In man, the initial 

 steep disappearance slope is terminated at the same 

 time for radio-iodinated serum albumin and for the 

 two cell labels K''- and P^- (23). In normal subjects 

 with arterial blood continuously monitored for 2 

 hours, the flattening of the disappearance slope for 

 I''^' albumin occurs at 4 to 15 min, with an average of 

 9 min (185). Even in congestive heart failure, with 

 mixing time for both cell and plasma labels prolonged 

 to 10 to 15 min, no consistent difference between the 

 two has been observed (110, 159, 207). No difference 

 has been observed in the mixing time of the two 

 plasma labels T-1824 and I'^' albumin (170, 185). 



Barbitalized dogs in our laboratory often have an 

 appreciably longer mixing time for P^- cells than for 

 the plasma label T-1824. I" a series of 24 dogs sampled 

 at 5-min intervals for 1 hour, the average cell mixing 

 time was 20.5 min, while simultaneously injected 

 T-1824 had art average time of 14 min ( fig. 2). In 

 eight acutely splenectomized dogs, removal of the 

 spleen appeared to have abolished the difference, cells 

 mixing in 14.4 min, and plasma in 13.8 min (Lawson, 

 unpublished data). 



Only limited data are available for other species. 

 In the cow, serial samples have been drawn for i hour 

 after injection of T-1824, and the initial rapid disap- 

 pearance phase found to terminate at 5 to 10 min 

 (195). That plasma mixing in an animal of this large 



