Clinical Science and Molecular Medicine (1975) 49, 115-120.
Determination of platelet and fibrinogen half-life
with ~sSe]selenomethionine:studies in normal and
in diabetic subjects
J. C. FERGUSON, N. MACKAY, J. A. D . P H I L I P A N D D. J. SUMNER
Department of Materia Medica, University of Glasgow, and Department of Clinical Physics and
Bio-Engineering, West of Scotland Health Boar&
(Received 24 December 1974)
vascular thrombosis, diminished lysis or to a combination of these factors. The simultaneous measurement of platelet and fibrinogen survival offers a
potentially useful assessment of the dynamic state
of the haemostatic mechanism and this study was
carried out to observe if there was any difference
between diabetic and non-diabetic subjects with
respect to these indices.
SummarY
1. Simultaneous platelet and fibrinogen survival
studies were carried out on a group of seven normal
persons and eleven diabetic patients.
2. Survival time was calculated: (a) by measuring
the interval between 50% of the maximum radioactivity in the anabolic phase and 50% of the
maximum radioactivity in the catabolic phase, and
(b) by fitting an exponential function to the decay
phase of the curve.
3. With the first method, the normal group had a
mean platelet survival which did not differ significantly from that in the diabetic group. With the
second method, however, the platelet half-life in the
diabetic patients was significantly shorter than in the
normal subjects.
4. Fibrinogen survival was significantly shorter
in the diabetic group with either method.
5. It is concluded that there is an increased utilization of platelets and fibrinogen in diabetic subjects.
Subjects
Key words : fibrinogen, platelet, [75Se]selenomethionine.
Introduction
Occlusive vascular disease is a major complication of
diabetes mellitus. This may reflect intrinsic disease
in the vessel walls, an increased tendency to intraCorrespondence: Dr J. Campbell Ferguson, General
Medicine, Ballochmyle Hospital, Mauchline, Ayrshire
KA5 6LQ, Scotland.
Seven normal, healthy subjects, six male and one
female, aged from 26 to 46 years, were studied and
compared with eleven diabetic patients (Table 1). All
had given informed consent; the normal subjects
were members of the staff of the Materia Medica
Department. In three of the diabetic subjects, the
platelet survival was not calculated as inadequate
numbers of blood samples were obtained. None of
the diabetic subjects had overt peripheral vascular
complicationsalthough one had a moderate diabetic
retinopathy. This was thought to be important as
the existence of major arterial disease might have in
itself influenced the results; for example, the peripheral utilization of platelets may have been increased
by the diseased vessels. None of the subjects had
proteinuria or a raised blood urea. The diabetic
subjects were all insulin-dependent and the approximate duration of their diabetes is indicated in Table
1. There is an obvious difference in age distribution
between the two groups but the mean values were
not significantly different.
115
J. C. Ferguson et al.
116
TABLE
1 . Age of subjects and duration of
diabetes
Subject
1
2
3
4
5
6
7
8
9
10
11
Diabetic
Normal
Age
Duration
(years) of diabetes
(years)
Age
(years)
20
58
28
26
68"'
30
24
60"'
67
55
22"'
35
2
30
10
< I
30
30
5
43+18
-
1
33
34
32
46
26
28
31
<1
< I
Mean
fSD
(')
33k6
Not included in platelet survival.
Methods
A dose (50 pCi) of [75Se]selenomethionine(radiation
dose 0.4 rad) was injected intravenously and blood
removed by venepuncture (19 gauge needle into a
plastic syringe) at 1 h, 5 h, and daily for 1 week and
on alternate days thereafter. The sample (18 ml)
of whole blood was placed in 2 ml of 3.8% sodium
citrate and a whole-blood platelet count carried out.
In seven subjects (three diabetic patients and four
normal subjects) the platelet counts were done
visually (Dacie & Lewis, 1968); in the remaining
subjects a Coulter Thrombocounter was used. The
blood was centrifuged at 1000 r.p.m. (250 g) for 10
min and 5 ml of platelet-rich plasma removed. A
platelet button was obtained from this which, after
washing, was counted in a Wallac automatic gamma
counter. In order to correct for changes in the total
numbers of platelets sampled each time, the c.p.m.
were corrected by using the equation:
centrifugation of the whole blood and fibrinogen
assayed by the method of Ratnoff & Menzies
(1951). The fibrinogen clots were washed and the
radioactivity was counted as for platelets. The counts
were again corrected to the maximum fibrinogen
concentration observed and expressed as a percentage
of the highest clot radioactivity.
Platelet and fibrinogen survival times were
estimated by two methods, that of Brodsky, Siegel,
Kahn, Ross & Petkov (1970), who took the interval
between 50% of the maximum radioactivity on the
anabolic phase of the survival curve and 50% of the
maximum radioactivity on the catabolic phase, and
also by calculating a half-life by fitting the decay
phase of the curve to an exponential function. This
was done by plotting the points after the maximum
on log linear paper and fitting them to a straight
line by a least-squares fitting programme on a
Hewlett Packard 9100B Programmable Calculator.
In general the decay phases of both platelet and
fibrinogen survival curves were good fits to single
exponential functions and there was a high correlation between log activity and time during this phase.
The mean correlation coefficient was 0.93, with a
range from 0.8 to 0.99.
Results
A group of typical curves for normal platelet activity
are shown in Fig. 1 and a group of those obtained
from diabetic subjects in Fig. 2.
1007
90 80 -
70 -
60 OIo
50 -
Corrected platelet button activity (%) =
Platelet button activity (c.p.m.)
-)1
platelet count
platelet count
100
maximum platelet button activlty
Platelet-poor plasma was obtained by further
Time (days 1
FIG. 1 . Platelet survival curves for1three normal subjects.
Platelet andfibrinogen hav-life
A
2 4 6 8 10 12 14 16 18
0
0
Time (days)
117
2
4
8
6
10
12
14
16
18
Time (days)
FIG. 2. Platelet survival curves for three diabetic subjects.
FIG.4. Fibrinogen survival curves for three diabetic subjects.
The shape of the fibrinogen curve differs from that
of the platelets as the anabolic phase is so rapid. A
group of curves for normal fibrinogen activity are
shown in Fig. 3 and those obtained from diabetic
subjects in Fig. 4.
Platelet survival for the normal group by the
Brodsky et al. (1970) method was 12.4f2.5 days
(meanfl SD), which was comparable with the
normal values obtained by Brodsky of 10.6 f 3.3
days. This did not differ significantly from the values
for the diabetic group, which were 10.2 f 2.0 days
(tI3=1-91,P10.05) (Fig. 5).
The time taken to reach maximum platelet button
radioactivity was also not significantly different
between normal subjects (7.3 f1.7 days) and
diabetic patients (8.9 3.2 days, t 1 3 = 1.2, P 0.2)
(Fig. 6). The platelet half-life was, however, significantly shorter in diabetic patients, being 5.2 f2-3
days compared with 11.7f5-4 days in normal
subjects (tI3=3*1, P<O.Ol). This is illustrated in
Fig. 7. In the diabetic group the fibrinogen survival
(4.0 +_ 2.4 days) was significantly shorter than that
of the normal group (6-9k2.1 days, t16=3*36,
P c OWS), and this difference was also reflected in
the fibrinogen half-life, which for diabetic patients
was 6.7 52.5 days and for normal subjects was
Diabetic
14
0
2
4
6
8 10 12
Time (days)
14
16
18
FlG.[3. Fibrinogen survival curves for three normal subjects.
FIG. 5. Platelet life-time in normal and diabetic subjects;
results obtained with the method of Brodsky et al. (1970).
118
J. C. Ferguson et al.
9.9 k 2.3 days ( f I 6 =2.8, P < 0.02). This is illustrated
in Fig. 8 and Fig. 9. The mean platelet counts and
mean fibrinogen concentrations of all samples
showed no significant difference between the two
groups (Table 2).
Diabetic
10-
Normal
I
0-
..-.
:
73
6-
I
.-E 4 t-
Diabetic
Normal
20-
FIG.8. Fibrinogen survival obtained with the method of
Brodsky et al. (1970).
12
c
.E
I-
Diabetic
q
8
6
Normal
‘t
2
FIG.6. Time taken to reach maximum platelet button
activity. Results for diabetic and normal subjects are shown.
I
2
FIG.9. Fibrinogen half-life in diabetic and normal subjects.
0
l
Normal
o
FIG.7. Platelet half-life in diabetic and normal subjects.
TABLE
2. Mean daily platelet counts and
fibrinogen concentrations in normal and
diabetic subjects
Normal
Diabetics
t
Mean daily
platelet count
(number/mm3)
216 000
65 000
230 000
f 65 000
0.4
Mean daily
fibrinogen
(mg/100 ml)
403+51
415+89
0.3
Discussion
[ SeISelenomethionine, produced by substituting
75Se for sulphur in methionine, was shown by
Cowie & Cohen (1957) to follow the same metabolic
pathways as methionine. The first clinical application
was by Blau & Manske (1961), who used it to visualize the pancreas by isotope scanning techniques.
Platelet and fibrinogen half-life
Cohen, Cooley & Gardner (1965) first used it to
label platelets, initially in dogs and subsequently in
humans. The technique was further developed by
Brodsky et al. (1970). The technique has the major
advantage of the label being incorporated into the
platelets during development and this eliminates the
risks to platelet function and viability associated with
labelling in oitro and re-injection methods. In addition, selenomethionineis incorporated into fibrinogen
and survival of this protein can also be measured.
The method of calculating platelet survival times
has been widely discussed. How the platelet survival
curves obtained in this study are interpreted will
depend on what is assumed about the mechanism of
platelet destruction. A great deal of conflicting
evidence on this point exists in the literature; the two
principal models that have been considered are:
(1) the platelets are utilized randomly irrespective of
age, and (2) 'senescence' is the major factor governing their removal from the circulation (Aas &
Gardner, 1958; Adelson, Rheingold & Crosby,
1957; Aster & Jandl, 1964; Cohen, Gardner &
Barnett, 1961a, b; Gardner & Cohen, 1966;
Ginsburg & Aster, 1969; Heyssel, 1961; Mustard,
Rowsell & Murphy, 1966). If we consider a
'senescence' model, the platelet survival curves
obtained with [75Se]selenomethionine will vary
considerably in appearance, depending on whether
the platelet life-time is greater or less than the
megakaryocyte maturation time, which is essentially
the time taken to reach maximum platelet radioactivity and is not necessarily related to the platelet
survival time (Harker & Finch, 1969).
If the platelet life-time is less than the megakaryocyte maturation time, at the point of maximum
radioactivity the labelled platelets present will have
a wide range of ages. Hence if the life-time is fairly
well defined the decay phase of the curve will be a
straight line; the intercept of this line on the time
axis, less the time taken to reach maximum activity,
will be equal to the platelet life-time. In this situation
the interval used by Brodsky is essentially equal not
to the platelet survival time but to the megakaryocyte maturation time.
If the platelet life-time is greater than the megakaryocyte maturation time, the decay phase of the
curve will mirror the ascending phase and the platelet
life-time will then be given by the interval used by
Brodsky et al. (1970). The appearance of the curve
will be further modified if the life-time has an appreciable standard deviation. With the present data
119
it is usually not possible to distinguish between the
different types of curve referred to above, owing
mainly to the small number of points on many of the
survival curves and to the experimental errors
involved.
If we assume that platelets are utilized randomly,
the decay phase of the survival curve will be exponential and it is then appropriate to express platelet
survival in terms of a half-life. It is evident that
analysis of the data with this model is simpler than
with the 'senescence'.
The fibrinogen data present less problems in view
of the very rapid anabolic phase, and the half-life
has again been estimated by fitting the points on the
decay phase to an exponential curve. By using both
this and the Brodsky methods of estimating fibrinogen half-life, there is a significant difference between
normal subjects and diabetic patients.
Intravascular thrombosis is the consequence of the
normal balance between blood coagulation and lysis
being altered in favour of fibrin deposition. Results in
the current study indicate increased utilization of
platelets and fibrinogen in diabetic subjects in
comparison with normal subjects. Abrahamsen
1968) demonstrated shortened platelet life-times in
diabetic patients with platelets labelled with "Cr,
and these results are in keeping with our own.
There is no evidence of increased fibrinolytic
activity being a prominent feature of diabetes
mellitus (MacKay & Hume, 1964), and increased
utilization of platelets and fibrinogen demonstrated
in the present study cannot be explained on the
basis of an overactive fibrinolytic system.
Acknowledgments
The authors thank Professor A. Goldberg for his help
and support in this study, and Miss Katherine
O'Connor and Mr Charles McNeill for their able
technical assistance.
References
AAS, K.A. & GARDNER,F.H. (1958) Survival of blood
platelets labelled with chromium. Journal of Clinical
Investigation, 37, 1257-1268.
ABRAHAMSEN,
A.F. (1968) Platelet survival studies in man
(with special reference to thrombosis and atherosclerosis).
Scandinavian Journal of Haematology, Suppl. 3.
ADELSON,E., RHEINGOLD,
J.J. & CROSBY,W.H. (1957)
Studies of platelet survival by tagging in uioo with "P.
Journal of Laboratory and Clinical Medicine, 50, 570.
ASTER,R.H. & JANDL, J.H. (1964) Platelet sequestration in
120
J. C. Ferguson et al.
man. I. Methods. Journal of Clinical Investigation, 43,
843-855.
BLAU,M. & MANSKE,R.F. (1961) Pancreas specificity of
75Se-selenomethionine.Journal of Nuclear Medicine, 2,102.
BRODSKY,
I., SIEGEL,N.H., KAHN,S.B., Ross, E.M. &
PETKOV,G. (1970) Simultaneous fibrinogen and platelet
survival with [75Se]selenomethionine in man. British
Journal of Haematology, 18, 341.
COHEN,
P., COOLEY,
M.H. & GARDNER,
F.H. (1965) The use
of selenomethionine (Se7') as a label for canine and
human platelets. Journal of Clinical Investigation, 44,
1036-1037.
COHEN,P., GARDNER,F.H. & BARNETT,G.O. (1961a)
Reclassification of the thrombocytopenias by the 51Cr
labelling method for measuring platelet life span. New
England Journal of Medicine, 264, 1294-1299.
COHEN,P., GARDNER,F.H. & BARNETT,G.O. (1961b)
Reclassification of the thrombocytopenias by the 51Cr
labelling method for measuring platelet life span (Concluded). New England Journal of Medicine, 264,1350-1 355.
Cowre, D.B. & COHEN, G.N. (1957) Biosynthesis by
Escherichia coli of active altered proteins containing
selenium instead of sulfur. Biochimica et Biophysica
Acta, 26, 252.
DACIE,J.V. & LEWIS,S.M. (1968) Practical Haematology,
4th edn, p. 70. Churchill, London.
GARDNER,
F.H. & COHEN,P. (1966) Platelet life span.
Transfusion, 6,23.
GINSBURG,
A.D. & ASTER,R.H. (1969) Kinetic studies with
'chromium-labelled platelet cohorts in rats. Journal
of Laboratory and Clinical Medicine, 14, 138-144.
HARKER,L.A. & FINCH,C.A. (1969) Thrombokinetics in
man. Journal of Clinical Investigation, 48,963-974.
HEYSSEL,R.M. (1961) Determination of human platelet
survival utilizing 14C labelled serotonin. Journal of
Clinical Investigation, 40, 2134-2142.
MACKAY,N. & H u m , R. (1964) Fibrinolytic activity in
diabetes mellitus. Scottish Medical Journal, 9, 359-364.
MUSTARD,
J.F., ROWSELL,
H.C. & MURPHY,E.A. (1966)
Platelet economy (platelet survival and turnover). British
Journal of Haematology, 12, I .
RATNOFF,
O.D. & MENZIES,C. (1951) A new method for the
determination of fibrinogen in small samples of plasma.
Journal of Laboratory and Clinical Medicine, 31, 3 16.