OOCL
NT SOUACE
:
:
Unversity of California at Berkeley
The Bancroft Library/The Unversity Archives, Berkeley CA
RECCADS SEAIES TITLE
Kenneth
Gargan
ce
Kepers
BANC ROFTSUARC ID NO.
Sco/ Bi
COPY
CARTON NO.
i
FOLOER NAME
Be pacats
NOTES
gil £13
FOUND BY/D4TE FOUND
Keres
195? -{7# 6
fe LL2 /94
RED CELL AND BLOOD VOLUME IN HEALTHY MEN: Cm
consideration are revealed. That these may be obscured when calculated body surface area is used
as the basis of prediction was pointed out many
‘vears ago by Rountree and Brown (1) and by
Gibson and Evans'{2). For example, two heavy
men are found to the far right of Figure 1. Although noticeably different trom the rest of the
population when located'on the scatter diagram.
they cannot be distinguished irom the other large
individuals in our series when the data for Vrbe
are plotted against body suriace area,-as shown
in Figure 4.
.
Regression of \Vwb to height and weight accounted for only 36 per cent of the-eriginal variability of our data.
The coefficient of variation
1075
on blood volume. This is probably because the
blood content of fat tissue is not insignificant (5,
15).
Hicks, Hope, Turnbull and Verel (14)
found that consideration of fat thickness and girth
in addition to height and weight lowered the
coefficient of variation for Vwb from 9.6 to 7.6
per cent. Vwb, Vrbe and Vpl have been found
to be directly related to body density (5, 15). and
Muldowney (17) found a surprisingly good correlation between Vrbc and lean body mass, con-
sidering that both were determined indirectly.
On the other hand. Inkley, Brooks and Kreiger
(16) found no better correlation between Vpl and.
iean body mass, determined by densitometry or
by antipyrine space, than benveen Vpl and total
(8.1 per cent) after this regression is of about the — body mass. and Huff and Feller (15) were unsame inagnitude as has been found in other stuthes able to improve the prediction of blood volume
after regression or correlation of blood volume -to by determining body density and calculating the
tively more muscular and less obese than theirs.
Most of the varia-
tion to be accounted for after consideration of
height and weight is biologic rather than metho-
dologic, as shown by our repeat studies and by
similar experiences reported by workers using
a variety of methods, for example, P** (7, 12, 13).
Cr? (10), 2" (15), and T-1824 (5, 13, 16).
Within the population available to us for study.
we have found that age, somatotype and habits of
physical activity influence Vrbe and Vwb only
slightly when considered independently from
height and weight. This mav be because each of
these characteristics is strongly linked in its effect
on blood volume with body size and composition.
The effect of each characteristic is of course mini-
mized when examined by analysis of residuals
about the height x weight regression. planes.
Available evidence suggests that individual differences with respect to the content of fat and
lean tissue in the body have only a modest effect
In most other respects, our results are.in re-_
markably good agreement with those of the early -
workers who used the-dye'methods. The'pasitive:
aan FY
correlation found betweenresiduals fot Virbe and:
Vpl (r = + 0.42) m
at a man ehgseVibe =
is larger than average_for_his height and Geighe -.
will also tend to have~a-térge Vol. ETBIFis ig -
contrast to the inverse relation beuveen ¥r¥c find.
Vpl found in certain pathologic states,?.g:, anemia
and polycythemia (38), and suggests that the
normal biologic determinants of tttod-volume may
be concerned with circulatory volume rather than
with total circulating hemoglobin.
The apparent absence of age effect in the third
to fifth decades is in agreement with several reports (6.8.13). The suggested tendencyfor the
volumes per unit size to decrease in the sixth
decade corroborates the findings of Gibson and
Evans (2) and of Baker, Kozoll and Meyer (20).
The question of age, however, requires further
ab feapes,
persons, but only a few examples of extremely
The only apparent explanation: for
the difference is that our large subjects were rela-
a! We
curvature.
linear or obese body build.
Cectal
body surface area were flattened at the top. We
have beerr-enable to corroborate this finding; tone
of our regressian lines or planes show ecidance of -
geneity, in that it included men from several raciai
origins whose ages extended to above 50 years
and included physically active as well as sedentary
Ci
curvesféfating volumes to height. weight and
Vee
average American medical students and considerably higher than ours in a group of subjects
selected for extremes of somatotype (3). Our
sample of adult males was of intermediate homo-
The data of Gibson and Evans (2) appeared to
show that blood volume per unit body ‘size was.
less in large than in small subjects, so that thé
Foay
It was comparatively low in a group of
‘hen Geer,
tion.
expected blood content of fat and lean tissues
separately.
the
The mag-
nitude of the coefficient depends on the homogeneity and size of the population under investiga-
Yee.
weight and height or to surface area,