Time and Concentration Components of Indicator-DilutionCurves Recorded Following Central Injections of Dye in

Normal Human SubjectsBy STEFAN A. CARTER, MI.D., M.Sc., II. .J. SWAN, M.l{, PH.D.,

AND EARL II. WtOOD, M1.1). 1ii.D.

Indicator-dilution curves were recorded by oximiieters at tlme ears anld the right radialartery following the injection of Evans blue (T 1824) into the superior vena cava andpulmonary arteries of 37 subjects who had no evidence of cardiovascular disease. Thevariability and ranges of various time and concentration components of these dilutioncurves are presented. These values can be used as standards of reference in the inter-pretation of abnormal dilution curves. Some of the factors responsible for the varia-bility of these values in healthy subjects are assessed and discussed.

T HE RECORDING of concentration-timecurves at a peripheral arterial site fol-

lowing injection of an indicator into the heartand great vessels provides a valuable toolfor the study of normal and abnormal circu-lation. Its increasing use in diagnostic andresearch laboratories attests to its efficacy.When this technic is used, it is desirable tohave a normal standard of reference withwhich the values obtained from individualsubjects can be compared. The present com-inunication reports the values and the varia-bility of various time and concentration coin-ponents derived from indicator-dilution curvesrecorded during catheterization of the rightside of the heart in a series of healthy human1)ei Imgs.

METHODS AND SUBJECTSThe right side of the heart was catheterized by

the methods previously described,"'2with the sub-jects resting in the supine position. They had alight meal prior to the study and were given aspremedication 30 lug. of codeine sulfate and 100ing. of secobarbital sodium at the beginning ofthe procedure. Evans blue (T 1824) *5 was usedas the indicator. Ten milligrams of the dye in 2iml. of solution was injected into the superior venaeava, main pulmonary artery, right pulmonaryartery, or left pulmonary artery. The subjectsFrom the Mayo Clinic and the Mayo FoundatioIi,

Rochester, Miin. (Dr. Carter is a Medical ResearchFellow of the National Research Council, Canada.)The Mayo Foundation, Rochester, Minnesota, is a

part of the Graduate School of the University ofYinnesota.

were breathing 100 per cent oxygen during therecording of the dilution curves, in order to avoidinterference due to fluctuations in the oxygensaturation of arterial blood.

Dye-dilution curves were recorded photographi-cally by means of ear oximneters placed on one orboth ears and a euvet oximueter connected to a,20-gage needle in the right radial artery. Thesensitivity of the system was such that dye con-centrations of 1 ing. per liter gave a deflectionof 0.4 to 0.7 cmn. for the radial artery curves and0.2 to 0.8 em. for the ear oximeter curves.Time components of the curves recorded by the

euvet oximneter were corrected for the volume ofthe instrument between the tip of the arterialneedle and the middle of the detecting photocell.The time taken for dyed blood to travel from theneedle tip in the artery to the detecting elementwas calculated fromm the volume of this "deadspace" and the flow rate of blood withdrawnthrough the cuvet s stem. This time correctionwas then subtracted from the appropriate timecomponents of the dilution curve. The dead spaceand physical dimensions of the oxiineters werespecified in a recent comamunication.3 Each milli-liter of flow through the oximieter was signaled omIthe photographic record. The flow rates averagedapproximately 23 il. per minute.The cardiac output was estimated by the Fick

imiethod in 24 subjects while they were breathing100 per cent oxygen. The average interval be-tween this procedure and the recording of dye-dilution curves after injection into the main pul-monary artery was 7.6 minutes, with a range of1 to 23 minutes. This interval exceeded 12 minutesin 4 subjects.

*Supplied through time courtesy of the Warner-Chileott Laboratories, Morris Plaimns, N.J.

Circulation, Volume XIX, March 1959430

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

TIME AND CONCENTRATION COMPONENTS

The concentration and time components of thedilution curves that are the subject matter of thispresentation are illustrated in figure 1. Cardiacoutput, central blood volume, and mean transittime were calculated from the radial artery curvesby the Stewart-Hamilton method.4 In additionthe portion of the central blood volume related tothe curve proper was calculated from the ratioof the amount of dye injected in milligrams (I)to the peak concentration in milligrams per liter(Cp) as described by Keys and co-workers.5 Thiscalculation was made only for the curves recordedfollowing injection into the main pulmonary artery.Except for the ratio of least concentration torecirculation concentration obtained from deflec-tions in centimeters, values for concentration comi-ponents were not measured from the earpiececurves owing to the difficulty of establishing anaccurate quantitative relationship between deflec-tion and blood-dye concentration for this instru-mnent.6' 7

A group of 37 subjects was studied. It con-sisted of 17 patients referred to the laboratoryto exclude a cardiac lesion as a diagnostic possi-bility, and 20 healthy physicians. All patientsincluded in this report were considered to havenormal cardiovascular systems on the basis of thefinal evaluation of the clinical and catheterizationdata. The number of dye curves recorded in thesesubjects varied from 1 to 4. Intervals betweenconsecutive injections varied from 3 to 57 minutes,with an average of 16 minutes. Only 3 pairs ofconsecutive curves were separated by more than30 minutes.For reasons that will be evident fromn the results,

subjects were divided into subgroups of 23 adultitiales, 7 adult females, and 7 teen-age subjects(2 females and 5 males). The average ages were31 (range 26 to 41), 34 (range 22 to 47), and 16(range 14 to 19) years respectively for the 3subgroups.

In order to assess the variability of the com-lponents of the dye-dilution curves that may occurduring the course of the procedure, a to 11 con-secutive curves recorded in 6 additional restingnormal subjects after injection of indicator intothe same site (main pulmonary artery or superiorvena cava) were analyzed. The interval from thefirst to the last injection varied from 24 to 60minutes in these subjects.

RESULTS

Table 1 shows the mean and standard devia-tion of various time components of the curvesrecorded at the radial artery following injec-tion into the superior vena cava or pulmonaryarteries in 37 subjects who constituted the

20 r

10

E

0I-

0

U

2

4i

cl

5

0.5

0.1

PCT fibk R T W

I I \ CR

I CL\\

11 , ,, 1,0 10 20 30 40

, a T -1 B T 1 D T

W PT -Onset of Dye Injection

0.13 mg./kg.

FIG. 1. Time and concentration components of anindicator-dilution curve with extrapolation of thedeclining slope of concentration, according to themethod of Hamilton, to eliminate the estimated effectof recirculated indicator. Time in seconds is repre-sented on the abscissa while the logarithm of concen-tration is shown on the ordinate. This dilution curvewas recorded for the radial artery following injectionof Evans blue into the pulmonary artery of a normalsubject.AT = appearance time, BT = build-up time, CG. =

least concentration, CP = peak concentration, CR =recirculation concentration, DT = disappearance time,LCT = least concentration time, PCT = peak con-centration time, PT = passage time, RT = recireula-tion time.

main experimental group. Considerable vari-ability is evident. The averages of the var-ious time components of the curve obtainedafter injection into the superior vena cavaother than the recirculation time and the leastconcentration time were all greater than thoseobtained following injection into the mainpulmonary artery or its branches. Pairedcomparisons revealed that the average differ-ences between the time components of thecurves recorded following injection into thesuperior vena cava and the main pulmonaryartery were 1.4, 0.9, 3.8, and 2.8 secondsrespectively for the appearance time, build-up time, disappearance time, and mean transit

-~~~ ~~ . j_----___

431

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CARTER, SWAN, WOOD

TABLE 1. Averages and Variability of Time Components of Dye-I)ilition Curves Recordedat the Radial Artery in Normal Subjects*

Time components (seconds)Site

Sub- Sub- of in- AT BT PCT DT PT MTT LCT RTgroup jects jection Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D. Mean S.D.

11 iRPA 10.9 1.8 6.1 0.9 17.0 2.5 19.8 4.2 25.9 5.0 19.1 2.9 13.1 2.5 21.8 2.7Adult 11 LPA 10.0 2.3 5.8 1.0 15.8 3.0 17.1 4.3 22.9 4.9 17.7 3.9 12.1 2.5 20.2 2.1male 16 M1PA 11.7 2.2 6.1 1.0 17.8 3.0 20.3 4.4 26.4 5.2 19.5 3.3 12.8 2.3 21.4 3.4

9 SYC 13.2 3.0 7.0 0.9 20.4 3.5 23.8 4.6 30.8 5.1 22.7 4.3 12.6 1.5 21.5 1.9

Adultfe- 5 MPA 7.0t 1.6 4.8t 0.6 11.9t 2.1 14.8+ 2.3 19.61 2.7 13.1t 2.1 9.71 1.1 16.1t 1.7

male 6 SYC 8.4t 1.8 5.3t 1.0 13.7t 2.5 16.81 3.2 22.0t 4.0 15.0t 2.6 9.9t 1.2 17.01 1.5

Teen- 6 MPA 7.5 1.3 4.8t 0.9 12.31 1.9 15.0+ 1.7 19.9t 2.6 13.5t 1.8 9.Ot 1.0 16.2t 2.3age 7 SYC 9.3t 1.5 5.9t 1.1 15.2t 2.2 19.1 4.4 25.0 ,5.3 16.9t 2.6 10.3t 2.1 17.3t 2.8

*Abbreviations: AT = appearance timne, BT = build-up time, PCT = peak concentrationtime, DT = disappearance time, PT = passage time, MTT = mean transit time, LCT = leastconcentration time, RT = recirculation time, RPA = right pulmonary artery, LPA = leftpulmonary artery, MPA = main pulmonary artery, SYC = superior vena cava, and S.D. =standard deviation.

tp < 0.01 for difference from male subgroup.tp < 0.05 for difference from male subgroup.

time. The p values for all these differenceswere less than 0.01.

Considerable variability of the time com-ponents following injection at any one siteis also evident in table 1. An attempt wasmade to relate this variability to differencesill cardiac output, blood volume, and bodysize encountered in these subjects. Figure 2shows the relationship of the appearance time,build-up time, and recirculation time from thecuryes recorded at the radial artery afterinjection into the main pulmonary artery tothe cardiac output as estimated by the Fickmethod, the part of the central blood volumerelated to the curve proper (I/CG), and thesurface area. Thie appearance time increasedwith the blood volume (I/Cp) and the surfacearea, but showed no correlation with cardiacoutput. Build-up time and reQirculation time,on the other hand, correlated with cardiacoutput, and buildtup time also correlated wellwith surfgce area. Build-up time and recir-culation time also correlated well with cardiacindex (p < 0.01), whereas no definite correla-tion of appear&uee time with cardiac indexcould be demonstrated. Most of the timecomponents were significantly shorter in theadult female &nd.tWen-age subgroups than in

the adult male subgroup. However, the sur-face area was significantly larger in the adultmales than in the adult females and the teen-age subjects. Correction of the appearancetime, build-up time, and recirculation timefor the surface area by dividing these timecomponents in seconds by the surface areain square meters eliminated the correlationwith surface area. However, the averagevalues for the corrected appearance time andrecirculation time, but not for the build-uptime, were significantly greater in the adultmale subgroup than in the other 2 subgroups.The over-all average values for the correctedtime components were 5.4, 3.1, and 10.5 sec-onds per square meter for the appearancethine, build-up time and recirculation time,and the standard deviations of these valueswere 1.4, 0.5, and 1.9 seconds per squaremeter respectively.The data on cardiac index, central blood

volume, disappearance slope, and various con-centration components of the same curves areshown in table 2. Paired comparisons re-vealed that the central blood volume, the leastconcentration, and the ratio of the least con-centration to the recirculation concentrationwere significantly larger, whereas the peak

432)

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

TIME AND CONCENTRATION COMPONENTS

CARDIAC OUTPUT1 1 1-1 1-1~~~~~~~~~~~~~

P>0.5 -

* 0

0 *0 0

5X X

*. X 0

.X X0 0

IA I

l S 7 9 11 1l

I/C BLOOD VOLUME

15 _P <0.001 * -

13* 0_

13 * * _

9_x *

.0 0

io o ~/+ -

S . I~~~~~~~~~~1 I

0.4 0.6 1.2

8 6 P<0.I -0

7 _

1 -

00

6 _, 0 0

K ash

.XX X 0

0.4 0.8 1.2

SURFACE AREA

15

13

9

7

5

6

7

6

5

41

-PsCo.00I -

.* */-* /-

0 0 0

X -

XX

.4 1.6 2.2

- ~P<0.02 -

-<X X

0i.4 1.6 2.2

P< 0.05

_ 0

C. a

* X XX

X Xx^ 'I 1

33P>0.2

29 V

25

21

17

x

X

.X X

**

S0

33

29p

25h

2 1 .0 _[ xX .1517[-0 xx xC

13 13, 3I-l

3 5 7 9 11 13 0.4 0.8 1.2 1.4 1.6

0 - Admit Malesbiters / minute) (liters )

- Adult Females (meter 2)Cardiac Output Value Derived from Dye-Dilution Curve and Femoales

FIG. 2. Relationship of the appearance, build-up and recirculation times to the cardiacoutput (estimated by the Fick method) blood volume (I/CP) and surface area for dilutioncurves recorded at the radial artery following the injection into the main pulmonary artery ofhealthy persons. p values of the regression coefficient are given at the top of each panel.Note: 1. Correlation of the appearance time with the blood volume (I/C,.) and the surface area

a ad absence of correlation wvith cardiac output. 2. Correlation of the build-up and recirculationtimes with the cardiac output and of the build-up time with the surface area.

concentration and the disappearance slopewere significantly smaller in the curves ob-tained after injection into the superior vena

cava than after injection into the main pul-monary artery. On the other hand, pairedcomparisons of cardiac index calculated fromthe dye curves from various sites did notshow ally systematic differences. The differ-ences in the various components shown illtable 2 among the 3 subgroups of subjects did

not attain statistical significance, except forthe difference in the disappearance slope of thecurves recorded after injections into the su-

perior vena cava in the male subgroup as com-

pared with the female subgroup, which mightbe related, at least in part, to the higher mean

(ardiac index in the latter subgroup.The values for various time components anad

the ratio of the least concentration to the re-

circulation concentration for the curves re-

Li

I-

Liu

0

LliCLC-

15

_ 13

C 11II0

CQ)U) 9

-

433

5

8Li

CL

-J

'3:

C)

,o0

U)C)

7

6

5

4

33

291.

Liu

00

JLiU:

Ui)1 25c:0

Q 21U)

17

P<0.I

2.2

~3

.

3

0

a0 0

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CARTER, SWAN, WOOD

TABLE 2.-Averaye.s and Variability of Conecentration Components andc of Other ParametersDerived from Dye-Dilution Carres Recorded at the Radial Artery in, Normal Subjects*

Cardiac Central Disap-Site index blood volume pearance

of in- (L. miniM.2) (ml/Kg.) slopetjection Mean S.D. Mean S.D. Mean S.D.

RPA 3.0 0.6 24.9 4.7 0.289 0.094LPA 3.4 0.9 -24.9 4.5 0.323 0.105MPA 3.3 0.8 27.0 7.0 0.301 0.091SVC 3.1 0.6 29.0 8.4 (0.242 0.053

Concentration (mg./L./mg. /Kg.)Cl Cl, Cic

Mean S.D. Mean S.D. Mean S.D.

99.0 18.7 7.0 1.6 18.8 4.497.1 22.5 6.0 2.6 16.7 3.89.5.7 21.4 (6.8 2.6 16.7 3.789.3 28.8 9.4 2.6 1 7.9 2.3

MrPA 3.4 0.9 22.9 3.9 0.365 0.047 103.1 17.1 7.9 1.9 18.4 2.6SVC 3.8 0.9 25.5 4.6 0.349* 0.083 98.3 1 8.9 9.0 2.2 19.4 5.3

Cl,/CRMean S.D.

0.42 0.120.36 0.140.40 0.090.56 0.14

0.42 0.060.47 0.16

3 MXPA 3.5 0.6 21.7 3.9 0.384 0.05.7 112.7 21.0 8.7 2.9 1 8.8 7.2 0.47 0.08SVC1 3.4 0.8 26.0 4.5 0.310 0.142 97.2 18.5 9.7 5.1 17.1 6.2' 0.55 0.14

*Abbreviations: Ci. = peak concentrationi, Cl, = least concentration, CR = recirculation con-centration, anid S.D. - standard deviatioii. Other abbreviations as in tal)le 1.

tDisappeance slopei loge Cl log2 C2 where Cl a1nd 2C are 2 cone(niitr-ationis closenti t2

aitr oiol the disapp)earance SloI)e before t1e occurrence of recirculation, atiil ti ald(l t2the correspondling times in secon'ds.

tp < 0.01 for differeneije from m1ales1ulbgroup.

corded at the left ear are 1)reseilted in table3. The differences between subgroups andilljection sites were similar to those obtainedfrom radial artery curves.

The appearance time was longer in the(urves recorded from the radial artery thanthose recorded at the ear, while no significantsystematic difference was evident in the othertime components. Furthermore, as illustratedin figure 3, the difference between the appear-

anice times of the curves recorded at the radialartery and at the ear increased with the abso-lute value of the appearance time.

Comparison of the timne components re-

corded at the right and left ears showed no

systematic differences. The mean differencesdid not exceed 0.1 second.The variability, with time, of repeated de-

terminations of appearance time, build-uptime and recirculation time of the dye-dilu-tion curves recorded at the left ear and rightradial artery following successive injectionsinto the same site in 6 subjects is illustratedin figure 4. Ini some subjects these comuponentsshowed no systematic change with time whilein others there was some progressive prolonga-tion or shortening during the period of obser-vation. however, the appearance time andbuild-up time remained relatively stable and

usually did miot vary by) mnore thai 2 secoldIs.whereas the recireulatiotn time frequentlyshowed wider variation.

DISCUSSION

The variability of such I)aramfleters as thosethat form the basis of the present study cati

be due to technical or biologic factors. Tech-nical factors involved ill the recordingr andalialysis of dye-dilution curves include poten-tial errors in tlme measurement of records or

signlaling of the exact imomenit of dye injec-tiOll, correction for the "dead spac.e" of thecuvet-oximeter systemii between the artery andphotocell, calibration of the instruments intermns of dye conceittrat ion, p)ossible differ-ences il dynlamic response aniollg various

oximneteris used, and variation ill the same

iimstruiuemmt with time. These factois ordinar-

ily could account for only a smnall part of thevariability that has been demonstrated all(l

that can be exempl)hified by a ralnge of 5.8 to15.3 seconds in. the case of appearance timerecorded at the radial artery following the ii-

jection of dye into the main pulmonary artery.The greatest part of this variability is mnost

likely due to the biologic inltra-individual andinterindividual differences. Many compolienltsof indicator-dilution curves vary in the same

Sub- Sub-group jects

11.Adult 1.1,,. e 1 6

9Adult -

fe-nmale 6

Teeni- Cage 7

434

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

TIME AND CONCENTRATION C(O )MIO)NENTS

TABLE 3. Averages and Variability of Time Components and of the Ratic of Least Con-centration to Recirculation Concentration in Centimeters Deflection, fro n Dye-DilutionCurves Recorded at the Left Ear in Normal Sub jects*

Site ofSubjects injection

1 6 MPA9 Svc

.) MPA(6 SVC

6 PA7 SVC

Time components (seconds)

AT BT PCT RTCl //CR

Mean S.D. Mean S.D. Mean S.D. Wean S.D. Mean S.D.

7.(6 1.2 6.0 (.9 13.6 1.8 21.5 3.5 0.34 0.119.1 1 .5 6.5) 0.6 15.8 2.0 '21.0 1.6 0.44 0.17

5.9t 0.8 .5.()0 0.7 1 0.9t 1.4 1 (;.4 1.8 0.46 0.067.4* 1.0 5.4t 0.9 12.8t 1.6 17.Ot 0.6 0.45 0.04

6.2) 1.1 5.2 1.() 11.4t 2.1 16.4t 5.2 0.53 0.178.1 1.5 6.4 0.9 1 4.5 2.3 18.0 2.8 (.61 0.09

*Abbreviationls the same as in tables 1 and 2.tp < 0.01 for difference from male subgroup.*p < 0.05 for difference from mumale subgroup.

individual when the injection sites or sam-

pling sites are changed. Differences observedin the present study between the componentsof dilution curves recorded after injectioninto the superior vena cava and those recordedafter injection into the main pulmonary ar-

tery and its branches confirm the observationsof Hetzel, Swan, and Wood,8 who found thatas the injection site was moved peripherally,the time components other than the recircula-tion time were prolonged, peak concentrationwas reduced, and central blood volume was ill-creased. However, their report was basedmainly oil a group of patients with cardiacabnormalities. The present study, which isbased oin subjects without cardiac disease,demonstrates in addition that the least con-

(entratiot1 time was also independent of theinjection site, similar to the finding in respectto recirculation time. These authors ascribedtheir findings to the longer anatomic path be-tween injection and sanmpling sites which istraversed by the dye and to increased longi-tudinal dispersion of the dye as it passes fromthe more peripheral to the central injectionsite. Similarly, appearance tinmes in curves

recorded at the radial artery that were longerthan those recorded at the ear probably reflectthe longer anatomic pathways to the radialsampling site. The finding that this differenceincreased with the absolute value for the ap-

pearainc e time is consistent with this explana-

tion. The absence of systematic differences inthe value for cardiac output obtained by thedilution technic following injection of indica-tor into the various sites, which was also re-

ported by Hetzel and co-workers,8 allows esti-mnation of cardiac output from dye-dilutioncurves recor(led after injection at these vari-ous sites.

W\\hen the injection sites and the samplingsites are the samc, intra-individual variationsapparently still occur probably because ofchanges in the hemodynaimic status of the sub-ject. The present study showed that the ap-

pearance time and build-up time changedrelatively little and were more stable than therecirculation time in a series of repeated dye-dilution curves. Sniall intra-individual vari-ability of various circulation times9' 10 andmore recently of the appearance time and thepeak concentration time"1 has been reportedpreviously. These considerations are impor-tant with respect to the validity of comparingcomponents of dye-dilution curves recordedsuccessively following injection into differentsites. Such comparisons may be of practicalimportance; for example, comparison of ap-

pearance times following injection into thevenae cavae, bramiches of the pulmonary ar-

tery and pulmonary veins is of value in thediagnosis of anomalous pulmonary venous

drainage.12 In adult and teen-age patientsdifferences in the appearance time of less than

Sub-group

Adultnale

Adultfe-

maleTeen-age

43.5

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CARTER, SWAN, WOOD

8

c-

UN 2

No

N

,,x z 6-

0

W C

1-i 4.E 4

U '

eq 2Q: _

oqW,,,*,z, o

0

00

-1 : 0

1 -i --T

0*0 m

X 0., X- - Aaw. Moles

0 - Ados f Fem0lesX - 10-20 yeo. mole$ fed -oes

l7 9 11 13 5 /7

RADIAL ARTERY APPEARANCE TIME seconds

FIG. 3. Relationship of the difference between theappearance times from curves recorded at the rightradial artery and the left ear to the actual value forthe appearance time at the radial artery. Data ob-tained from curves recorded after the injection intothe main pulmonary artery of healthy persons. Notethe rather close positive correlation between the fastestcirculation time from the pulmonary to the radialartery (i.e., the appearance time) and the differencebetween the fastest circulation times to the ear andthe radial artery at the wrist. The symbols are thesame as those used in figure 2.

2 seconds between any two curves may not besignificant. Care must be exercised in com-

paring time components of curves recorded atlonger time intervals, such as 30 or 40 min-utes, in view of the tendency exhibited bysome subjects to progressive prolongation or

shortening of the time components with time.Assessment of interindividual variation re-

veals lack of definite correlation of the appear-

ance time with cardiac output, but good cor-

relation with surface area and blood volume(I/Cp). Direct comparison with the "centralblood volume" was not carried out becausethe values for appearance time and also build-up time are included in the calculation of"central blood volume." The blood volume(I/Cp) that was correlated with the time com-

ponents is calculated independently of any

measurement of time components of the dilu-tion curve and is a measure of the dispersionof the injected indicator during its passage

from the injection site to the sampling site.5Dispersion of the indicator in turn is thoughtto be dependent in part upon the volume ofblood between the injection site and the sam-

pling site. Good correlation of appearance

tine with blood volume (I/Cp) and surface

area, which is in turn correlated with totalblood volume of the subject, suggests stronglythat a large part of the interindividual vari-ability for appearance time is due to the dif-ferences in central blood volume.A good negative correlation of build-up

time and recirculation time with cardiac out-put is probably due to the lesser dispersion ofthe indicator in subjects with high rates offlow. Furthermore, in subjects with the sametotal blood volume an inverse relationship be-tween cardiac output and the time requiredfor dyed blood to make a complete circuit ofthe vascular system (systemic recirculationtime) would be expected. Positive correlationof build-up time with surface area could beexplained again by greater dispersion of theindicator in subjects with greater blood vol-ume, as indicated by greater surface area.Lack of correlation of build-up time and re-circulation time with blood volume (I/Cp) isdifficult to explain. Metabolic rate and heartrate are known to be related to the velocityof the circulation,13 and undoubtedly severalunexplored parameters of the circulation andbody size in addition to those that have beendiscussed enter into the determination of in-terindividual variability. Because of the goodcorrelation of appearance time and build-uptime with surface area and a trend in that di-rection in the case of recirculation time, theexpression of various circulation times as cir-c'ulation "indexes " in seconds per squaremeter may allow a more valid comparison ofthe time components of the indicator-dilutioncurves in various patients.

All the time components in the adult maleswere longer than those in the adult femalesand the teen-age subgroup (mainly male).Blumgart and Weiss9 previouisly 'reportedshort circulation times in teen-age subjects,and Dees and co-workers' found shorter cir-culation times in females than in "males.However, the body size of our male subgiroupvas significantly" reater than in the other 2subgroups, which probably accounts, at leastin part, for'th longer time components in 'thieadult males. 0Correction for hody size by di-viding the appearance time, build-up timne

436

.

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

TIME AND CONCENTRATION COMPONENTS

APPEARANCE TIME BUILD -UP TIMECurves Recorded at Right Radial

0. 7 seconds

0 20 40 60 0 20 40 60

Curves Recorded at Left ElI

0.8 seconds

0 Al0 20 40 60

minutes

0.8 seconds

12

8

4

01 ' ' '0 20 40 60

minutes

RECIRCULATION TIME

-or

36

32

28

24

20

I'- 2.4 seconds

_ t ~~~~~~~~~~~~~_6

0 20 40 60minutes

FIG. 4. Intra-individual variability in appearance, build-up and recirculation times of dilu-tion curves recorded at the radial artery and ear of 6 persons following repeated injectionsof the indicator into the same site on the right side of the heart. Numbers identify timecomponents of each person in various panels. Variability expressed as the square root ofthe average variance of all persons is indicated at the top of each panel in seconds. Noterelatively small changes of the appearance

recirculation times.

aild recirculation time by the surface area didnot abolish significant differences in the ap-

pearance time and recirculation time betweenthe male subgroup and the other 2 subgroups.This means either that factors other than dif-ferences in body size are responsible for partof the observed difference or that the correc-

tion for body size on the basis of surface area

was not complete. Dees and co-workers"tfound, further, that the time components were

longer in the subjects more than 40 years ofage. Only a few subjects were in this oldergroup in the present study. However, no

systematic difference could be demonstratedbetween males more than 30 years of age andthose less than 30 years of age.

and build-up times and wider variations of the

Although considerable variability of vari-ous components of indicator-dilution curves

exists in normal subjects, gross changes occur

in conditions such as congestive heart failure,thyrotoxicosis, and others,l3, 14 and result incomponents outside the range encountered innormal man. Deviations from normal are

particularly evident and of practical impor-tance in patients with congenital heart defectsor valvular heart disease.15 Values that ex-

ceed 2 times the standard deviation from themean as set forth in the present study shouldbe considered abnormal.A special consideration has been given to

the ratio of the least concentration to the re-

(irculation concentration of the curves re-

In(t2c:0

Q.)(1)

16

v)0 i

00

U)

4

437

4

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CARTER, SWAN, WOOD

corded at the radial artery, since it has beenshown to be increased in patients with pre-dominant valvular regurgitation16 and in thosewith left-to-right shunt.7 All the values re-corded ill the present group of normal subjectswere less than 0.60 for the curves recordedafter injection into the pulmonary artery andless than 0.85 after injection into the superiorvena cava. The finding of larger values shouldtherefore arouse suspicion that one of theselesions exists. Curves recorded by ear oxiiii-eters appear to be lea;s reliable for this pur-pose.16

SUMMARYIndicator-dilution curves were recorded by

oximeters at the radial artery and at the ears,following injection of Evans blue (T-1824)into the superior vena cava and the pulmonaryarteries of 37 subjects who had no evidenceof cardiovascular disease.The variability and range of various com-

ponents of these curves are reported. Curvesrecorded following injection into the superiorvena cava showed significantly longer timecomponents, other than the recirculation timeand the least concentration time, than thecurves recorded following injection into themain pulmonary artery. Also, the "centralblood volume," the least concentration, andthe ratio of the least concentration to the re-circulation concentration were larger, whereasthe peak concentration and the disappearanceslope were smaller in the curves recorded afterinjection into the superior vena cava. Thevalues for cardiac output calculated from thecurves following injection into the superiorvena cava showed no systematic differencefrom those following injection into the pul-monary arteries.The appearance time was significantly

shorter in the curves recorded at the earsthan in the curves recorded at the radial ar-tery, whereas other time components showedno systematic difference.The time components of the curves recorded

at the right and left ears were practicallyidentical.The appearance time and the build-up time

remained relatively stable when injection of

the indicator into the same site ill the samesubject was repeated several times over a pe-riod of 30 to 60 minutes, whereas the reeireu-lation time showed wider variations.

Significantly longer time components of thecurves recorded in the adult males as coin-pared to the adult females and teen-age sub-jects are thought to be related chiefly to thelarger body size of the adult males.Rather large interindividual variability of

various components of indicator-dilutioncurves recorded in normal subjects could beaccounted for in large part by the differencesin body size, blood volume, and cardiac out-put.

ACKNOWLEDGMENTThis study was mnade possible by the unstinting

cooperation of many professional and technicalcolleagues. The authors are particularly indebtedto Miss M. Koelsch, Mrs. J. Frank, Miss L. Cro-nin, and Mr. W. F. Sutterer for their help andinterest in this study.

SUMMARIO IN INTERLINGUA(iurvas del dilution de colorante indicatori

esseva registrate per oxymetros al arteria ra-dial e al aures post le injection de blau deEvans (T-1824) ill le vena cave superior e learterias pulmonar de 37 subjectos qui exhi-biva nulle signo de mnorbo cardiovascular.Le variabilitate de diverse componentes de

iste curvas es reportate. Curvas registratepost injectiones inl le vena cave superior ex-hibiva significativenemnte plus lorge compo-nentes temporal-a parte le tempore de recir-(ulationl e le tempore del concentration mini-inal-que le curvas registrate post injectionesiii le principal arteria pulmonar. In plus, levolumine de sanguine central, " le concentra-

tion minimal, e le proportion inter le concen-tration minimal e le concentration de recir-(tilation esseva plus grande, durante que leconcentration maximal e le inclino de dispari-tion esseva plus miere in le curvas registratepost injection in le vena cave superior. LeXvalores del rendimento cardiac, calculate able curvas post injections in le vena cave su-perior, non differeva systematicainente ab levalores calculate post injectiones in le arteriapulmonar.

438

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

TIME AND CONCENTRATION COMPONENTS

lie tempore de apparition esseva significa-tivemente plus breve in le curvas registrateal aures que in le curvas registrate al arteriaradial, durante que le altere componentestemporal non differeva systematicamente.Le componentes temporal del curvas regis-

trate al aure dextere e al aure sinistre essevapracticamente identic.Le tempore del apparition e le tempore de

accumulation remaneva relativemente stabilequando le injection del indicator in le mesmesito in le mesme subjecto esseva repetite plurevices intra un periodo de 30 a 60 minutas. Letempore del recirculation monstrava plusgranide variationes.Le constatation de significativemente plus

longe compoilentes temporal del curvas regis-trate in masculos adulte in comparation con leconstatationes in adulte femininas e in ado-lescentes es interpretate como primarimenteun effecto del plus grande dimensiones corpo-ree ill masculos adulte.Un satis grande variabilitate interindividu-

al de varie componentes del curvas del dilu-tion de colorante indicatori in subjectosnormal esseva explicabile in grande parte perdifferentias del dimensiones corporee, del voiu-mine de sanguine, e del rendimento cardiac.

REFERENCES1. VVOOD, E. H.: Special technics of value in

cardiac catheterization laboratory. Proc.Staff Meet., Mayo Clin. 28: 58, 1953.

2. BARRATT-BOYES, B. G., AND WOOD, E. H.:Cardiac output and related measurementsand pressure values in the right heart andassociated vessels, together with an analysisof the hemodynamic response to the inha-lation of high oxygen mixtures in healthysubjects. J. Lab. & Clin. Med. 51: 72, 1958.

3. Fox, I. J., SUTTERER, W. F., AND WOOD, E.H.: Dynamic response characteristics ofsystems for continuous recording of con-centration changes in a flowing liquid (forexample, indicator-dilution curves). J. Appl.Physiol. 11: 390, 1957.

4. HAMILTON, W. F., MOORE, J. W., KINSMAN,J. M., AND SPURLING, R. G.: Studies oncirculation; further analysis of injectionmethod, and of changes in hemodynamicsunder physiological and pathological con-ditions. Am. J. Physiol. 99: 534, 1932.

o. KEYS, J. R., HETZEL, P. S., AND \WOOD, E. H.:Revised equations for calculation of bloodflow and central blood volume from indi-cator-dilution curves. J. Appl. Physiol. 11:385, 1957.

6. BEARD, E. F., AND WOODD, E. H.: Estihnationiof cardiac output by the dye dilutionmethod with ear oximeter. J. Appl. Phys-iol. 4: 177, 1951.

7. WARNER, H. R., AND WOOD, E. H.: Simpli-fied calculation of cardiac output fromt dyedilution curves recorded by oxinieter. J.,Appl. Physiol. 5: 111, 1952.

8. HETZEL, P. S., SWAN, H. J. C., ANI) WvOOD,E. H.: Influence of injection site on arte-rial dilution curves of T-1824. J. Appl.Physiol. 7: 66, 1954.

9. BLUJMGART, H. L., AND WEISS, S.: Studies on1velocity of blood flow; velocity of bloodflow in normal resting individuals, and acritique of method used. J. Clin. Invest. 4:15, 1927.

10. -, AND -: Studies on velocity of blood flow;pulmonary circulation time in normnal rest-ing individuals. J. Clin. Invest. 4: 399,1927.

11. DEES, T. M., RUMSFELD, J. A., MILLER, W.F., AND CHAPMAN, C. B.: Clinical measure-ment of circulation time. A comparison ofmagnesiumi sulphate and Evans blue dye innormal subjects. J. Appl. Physiol. 10: 451,1957.

12. SWAN, H. J. C., BURCHELIJ, H. B., AND) WOOD,E. H.: Differential diagnosis at cardiaccatheterization of anomalous pulmonary ve-nous drainage related to atrial septal de-fects or abnormal venous connections. Proe.Staff Meet., Mayo Clin. 28: 452, 1953.

13. BLUMGART, H. L.: Velocity of blood flow inhealth and disease; velocity of blood flowin man and its relation to other measure-ments of circulation. Medicine 10: 1, 1931.

14. TARR, L., OPPENHEIMER, B. S., AND SAGER, R.V.: Circulation timne in various clinicalconditions determined by use of sodiumdehydrocholate. AIm. Heart. J. 8: 766, 1933.

15. WOOD, E. Hl., SWAN, H. J. C., AND HELM-HOLZ, H. F., JR.: Recording and basic pat-terns of dilution curves. Norinal and ab-normal. Proc. Staff Meet., Mayo Clin. 32:464, 1957.

16. -, AND WOODWARD, E., JR.: A simple methodfor differentiating mitral regurgitation fromtmitral stenosis by means of indicator-dilu-tion curves. Proc. Staff Meet., Mayo Clini.32: 536, 1957.

17. CARTER, S. A., BAJEC, D., AND WOOD, E. H.:Unpublished data.

439

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

STEFAN A. CARTER, H. J. C. SWAN and EARL H. WOODFollowing Central Injections of Dye in Normal Human Subjects

Time and Concentration Components of Indicator-Dilution Curves Recorded

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1959 American Heart Association, Inc. All rights reserved.

75231is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TXCirculation

doi: 10.1161/01.CIR.19.3.4301959;19:430-439Circulation. 

http://circ.ahajournals.org/content/19/3/430located on the World Wide Web at:

The online version of this article, along with updated information and services, is

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer

of the Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle columnClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from