JOCUMENT SOURCE . . Uniwverstty of California at Berkeley The Bancroft Library/The University.Archives, Berkelay CA RECORDS SEAIES TITLE Kénne th Gardan Scott Happs BANCROFTAJARC ID NO. FO / BI ) | CARTON NO. COPY —_ FOLDER NAME SG§ FZ -~/GGS Renciats NOTES! 2. 2473 NO Agen "Holmea FouND BY/DATE F wae £0 f/f 12 £9 F 1073 RED CELL AND BLOOD VOLUME [IN HEALTHY MEN: Cr tered. or c) methodologic factors. Influence of the latter seems unlikely since the within-month eect was as great at the end of the study as at the beginning am~. since the personnel. procedures and methods were unchanged throughout. E) Blood donations. Fiity-three blood donors were included in the series. Their residuals of Vrbe and Vopl are plotted in Figure 3 according to the interval of time between donation and ex- periment. The points are distributed evenly around the zero line which represents the pre- dicted Vrbe or Vpl of each individualcalculated irom his height and weight. Most of the residuals were within plus or minus one standard deviation Ineasurements enable us to estimate the amount of variation due to measurement errors and short- tern: temporal changes, The estimated standard deviation of such variation was 64 ml. for Vrbc and 211 mi. for Vwb. The standard deviations about the regression planes described by Equations 3 and 11, Table II, are reduced only modestly by taking into account the degree of variance shown by the repeat measurements, which includes the errors of measurement and the changes in blood volume of individual subjects from time to time? Thus. for Vrbe almost all of the predic- tive error seems to relate to intrinsic differences berween subjects rather than to errors of meas- urement or temporal changes within subjects. of the mean predicted volume so that the vanatten is no greater in this group than in the population as_. For Vwb this is so to a lesser degree; this volume awhole. The residuals showed no tendency tq in- crease or decrease progressively with time, between 42 and 180 davs. For all donors, the mean is either less stable or less accurately measured. or both. DICUSSION residuals (ml.}) were as follows: Vrbce = + 17 1S.E. 27), Vpi= — 17 (S.E. 33), Vwb = + 02 13.E. 50). Only the observed venous hematocrit values were used in our calculations. It is recognized 2. Repeat determinations culated-trom- the Cr°! space and observed heinatocrit becauseof plasma trapping in the hematocrit eae that the true \rbe is slightly smaller than that cal- In 15 men the blood volume was re-estimated one or more times after intervals of three to 31 weeks. The mean of the 20 differences for Vrbc was 73 mi. (range, 10 to 160 ml.) and for Vwb, 222 ml. (range, 10 to 710 ml.). These repeat +500 ° ~ +300 a from 370 mb to ¥(370)* — (211)'= 304 mil. e oomesouat |j Asran rea sacomaar Tee a.0eecours -0 Se Oo ~~ Rs 7. 7 2°? —z ° ae = 7 poe ee GT rr tet ce en ce et errr nee 2 oe meSD Ase +200. 2. yg Oe ee ee ee ee pr 8B ces . ° ap 4 2 00! i 2 “es 3 x o | s, a: H-100“ the 5D. for Vwb (Equation 11, Table II) is reduced omeaetiee nl asran a ses ocees pemarew +400 . >The SEr-for Vrbe (Equation 3, Table IL) is re- duced from 190 mi to ¥(190)'— (64)'=179 ml. and 3, o * o ° “op 8 > eZ ~ . . . >} af, . - = o . . —OT ma a we e ~Ret z = . s. , nik o Dy Ao _ c= “27 [mine i.e‘ ‘ te T OS 2 E eo? te : — - He Pee ee Re Be em ,ti ae -r99-- “sor 8° pee ae pottcese wee ee 8Tm mre eter rt eer em eae wna ~~ SB. - tgers”.. 7a00* ° -500~ 4 a0 - & i oa Ro 2 FR?oT f% wo 2 7m ° @ e. ol LE - TAaae¢ - - 7 m a F OL EE. “Lo 1 65006UWOlUCSCOO 4 4 i COC‘ sOIIO ad, «120 2 1 1 «130 140 180 1 160 GAYS BETWEEN BLOOD DOMATION AND MEASUMEMENT OF «170 180 Vrbx n : FEFeusog Tr. Cc xc fe = La eo ¢ - ee - - PESEQEL- Fig. 3. Resipvats or Vasc ano Vet ron 46 Men wo Rap Given fixe Broo © 2 = = DJOMNATION AND Seven wHo AAD Donated Twich, ARRANGED ACCORDING TOTES Baack Bo0b Se cls