Incomplete global cerebral ischemia alters
platelet biology in neonatal and adult sheep
MARGUERITE T. LITTLETON-KEARNEY, PATRICIA D. HURN,
THOMAS S. KICKLER, AND RICHARD J. TRAYSTMAN
Department of Anesthesiology/Critical Care Medicine, The Johns Hopkins Hospital,
Baltimore, Maryland 21287-2725
platelet aggregation; platelet entrapment; chronological age
PLATELETS HAVE BEEN IMPLICATED etiologically as agents
in cerebral ischemia associated with thromboembolic
vessel occlusion and as mediators of neuronal injury
following an ischemic insult (14, 24, 38). Considerable
evidence indicates that platelets aggregate within the
brain vasculature after focal (23, 34) or global (8, 20)
cerebral ischemia or ischemic stroke (35). Platelet
aggregation contributes to hypoperfusion and microcirculatory dysfunction during recirculation after prolonged global cerebral ischemia (20) and as part of the
pathophysiology of stroke (16, 26, 31). Furthermore,
platelet activation triggers release of dense granule
constituents including adenine nucleotides, thromboxane A2, calcium, and serotonin, leading to functional
alterations within the vascular endothelium (41) and
significant vasoconstrictive consequences. Clinical studies demonstrate increased dense granule secretion (24),
platelet release of microparticles (32), platelet activation in subtypes of ischemic stroke (22), increase in
mean platelet volume (36), and platelet cytosolic Ca21
efflux (11, 25, 28) after stroke, suggesting altered
platelet biology. Little information exists regarding the
effects of transient ischemia on platelet function in the
newborn. Platelets of newborns are known to be less
responsive to aggregants such as collagen, thrombin,
and ADP compared with those of adults (3, 33, 37).
Therefore, it is possible that differences in platelet
function associated with immaturity may confer some
degree of protection during transient cerebral ischemia
in the young. The purpose of the present study was to
determine whether incomplete global cerebral ischemiareperfusion directly and rapidly alters platelet biology
and whether this effect is age dependent in newborn
versus adult animals. We also sought to determine
whether platelets remain sequestered in the brain
during early reperfusion.
METHODS
Animals. Nineteen neonatal lambs ranging in age from 2 to
5 days (mean age 3 days) and 20 adult sheep from 1 to 5 yr of
age (mean age 1.5 yr) were used in these studies. Groups I
(n 5 8 lambs) and II (n 5 9 sheep) were used for the ischemia
and platelet aggregation studies; groups III (n 5 4 lambs) and
IV (n 5 5 sheep) served as their respective nonischemic
controls. For the platelet labeling studies, four lambs and four
sheep comprised the ischemic groups (groups V and VI,
respectively), whereas three lambs and two sheep (groups VII
and VIII) formed the respective nonischemic groups.
Animal preparation. All the protocols for this study were
approved by the Johns Hopkins Animal Care and Use Committee. Anesthesia was induced by external jugular vein injection of pentobarbital sodium (25–30 mg/kg). As previously
described (30), either a tracheostomy (sheep) or endotracheal
intubation (lambs) was performed, and all animals were
mechanically ventilated to maintain normal blood gases.
Both axillary arteries were cannulated with polypropylene
catheters (PE-120). The left axillary catheter was advanced
into the left ventricle for microsphere injection, and the right
catheter (advanced to the aorta) was used for reference
sample withdrawal and arterial blood gas sampling. Additional catheters were inserted into a femoral artery and both
femoral veins for hemodynamic monitoring and drug and
fluid administration.
The animals were repositioned prone to permit cannulation of the superior sagittal sinus. Temporalis skin and
muscle were retracted, and a burr hole was drilled midline
between the lambdoidal and coronal sutures. A polypropylene
catheter (PE-90) for cerebral venous blood samples was
threaded into the sagittal sinus. A Silastic multiple-port
ventricular catheter (model 901302, Cordis, Miami, FL) was
inserted into the lateral ventricle through another burr hole
in the skull for monitoring of intracranial pressure (ICP) and
infusion of artificial cerebrospinal fluid (CSF). A thermistor
was inserted between the bone and dura to monitor epidural
temperature, and wound edges were approximated to minimize temperature losses from the open cranial areas. After
surgery, pancuronium (0.1 mg/kg) for muscle paralysis was
administered with continuous intravenous pentobarbital infusion (3 mg/kg). Each animal was warmed with heating lamps
and water blankets to maintain an epidural temperature of
38–39°C.
Measurements. Arterial blood pressure and ICP were continuously recorded via a Gould-Brush polygraph. Regional
0363-6135/98 $5.00 Copyright r 1998 the American Physiological Society
H1293
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
Littleton-Kearney, Marguerite T., Patricia D. Hurn,
Thomas S. Kickler, and Richard J. Traystman. Incomplete global cerebral ischemia alters platelet biology in neonatal and adult sheep. Am. J. Physiol. 274 (Heart Circ. Physiol.
43): H1293–H1300, 1998.—Platelets are implicated as etiologic agents in cerebral ischemia and as modulators of neural
injury following an ischemic insult. We examined the effects
of severe, transient global ischemia on platelet aggregation
during 45-min ischemia and 30-, 60-, and 120-min reperfusion in adult and neonatal lambs. We also examined postischemic platelet deposition in brain and other tissues (120-min
reperfusion) using indium-111-labeled platelets. Ischemic
cerebral blood flow fell to 5 6 1 and 5 6 2 ml · min21 · 100 g21 in
lambs and sheep, respectively. During ischemia, platelet
counts fell to 47.5 6 5.1% of control (P , 0.05) in lambs and
59 6 4.9% of control in sheep (P , 0.05). Ischemia depressed
platelet aggregation response (P , 0.01) to 4 µg collagen in
lambs and sheep (20.4 6 29.2 and 26 6 44.7% of control,
respectively). Marked platelet deposition occurred in brain
and spleen in sheep, whereas significant platelet entrapment
occurred only in brain in lambs. Our findings suggest that
ischemia causes platelet activation and deposition in brain
and noncerebral tissues.
H1294
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
pended in ACD-saline (2 ml) containing 850 µCi of indium-111tropolone solution prepared as described (5). This mixture
was incubated for 30 min (25°C) and then centrifuged (3,200
rpm, 10 min). The supernatant was discarded, and the
labeled platelets were resuspended in ACD-saline (2 ml) with
subsequent determination of radioactivity. Thirty minutes
before initiation of experimental protocols, animals were
injected with 200–300 µCi of autologous labeled platelets.
Forty-five minutes of incomplete global cerebral ischemia
were induced, followed by 120 min of reperfusion. At 120 min
of reperfusion, venous and sagittal sinus blood were obtained.
The animal was killed, and samples of brain, skin, muscle,
lung, heart, liver, spleen, and kidney were harvested. The
radioactivity of blood and tissue was measured by autogamma spectrometer as for microsphere studies, but with a
preset window for photo peaks of 171, 245, and 426. All tissue
samples were normalized to radioactive counts per gram of
weight, and the ratio of tissue to blood (per ml) was calculated.
Experimental protocol. Baseline measurements were obtained for arterial pressure, ICP, epidural temperatures, and
CBF. Sagittal sinus (3 ml) and arterial (3 ml) blood samples
were collected at baseline, 45 min of ischemia, and 30, 60, and
120 min of reperfusion for analysis of pH, PCO2, hematocrit,
O2 saturation, O2 content, and platelet aggregation. Incomplete global ischemia was produced by infusion of warmed
artificial CSF (in mM: 151 Na1, 3 K1, 2.5 Ca21, 1.2 Mg21, 134
Cl2, and 25 HCO2
3 and 6 meq/l urea) for 45 min through the
Silastic multiple-port ventricular catheter, which was inserted into the lateral ventricle through a burr hole. ICP was
controlled by CSF infusion, resulting in a cerebral perfusion
pressure (CPP) of 0–5 mmHg (21). The arterial pressure was
permitted to vary spontaneously, and ICP was adjusted
accordingly. To initiate the 120-min reperfusion, the infusion
was halted and the ICP was allowed to fall to normal values
spontaneously. At end reperfusion, the animals received an
overdose of pentobarbital and were killed by KCl injection.
Statistical analysis. Data are expressed as means 6 SE. All
repeated measurements were evaluated by two-way analysis
of variance (ANOVA). If the F statistic for group treatment or
the interaction between group and time was significant (P ,
0.05), means among groups at individual time points were
compared by a Newman-Keuls multiple-comparison test. If
the F statistic for time effects or interactions between group
and time were significant (P , 0.05), a one-way ANOVA with
repeated measures was performed individually on each group
to determine in which group the time effect was significant.
Dunnett’s test was then used to determine which time periods
differed from baseline values. For the platelet labeling studies, a t-test for independent samples was used to test differences between counts per gram and counts per milliliter of
blood in the various organs compared with control values. A P
value of ,0.05 was considered significant.
RESULTS
Arterial blood gases and hematocrit are summarized
in Table 1. Ischemic arterial oxygen tension (PaO2) was
higher in sheep compared with lambs, reflecting additional inspired oxygen administered to support cardiac
function. Baseline arterial pH of the lambs was lower
than that of the adult sheep but fell by the same
amount during ischemia. By 120 min of reperfusion,
arterial pH recovered in sheep but not lambs (7.4 6 0.0
vs. 7.25 6 0.03).
No differences in CPP were observed between lambs
and sheep during ischemia or reperfusion (Fig. 1). ICP
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
cerebral blood flow (CBF) was measured at baseline, 45 min of
ischemia, and 30, 60, and 120 min of reperfusion, using the
radiolabeled microsphere technique (15 6 0.5-µm diameter
spheres) (18). Briefly, a dose of ,1.5 3 106 microspheres
labeled with 153Gd, 114In, 113Sn, 95Nb, or 45Sc (DuPont NEN,
Boston, MA) was injected into the left ventricle as previously
described (1, 30), followed by a 3-ml saline flush. A reference
sample was withdrawn from the right aortic catheter at a rate
of 2.5 ml/min. After harvesting and fixation, we dissected the
brain into cerebellum, medulla, pons, midbrain, hippocampus, caudate nucleus, and cerebrum. Radioactivity of the
brain samples and the blood reference samples was counted
on an autogamma scintillation spectrometer (model 5530,
Packard Instruments, Downers Grove, IL). Blood flows were
determined using the reference organ technique as previously
described (18, 30).
Arterial and sagittal sinus blood gas samples were analyzed for pH, PO2, and PCO2 with a Radiometer ABL30
electrode system (Radiometer, Copenhagen, Denmark). Oxygen content, saturation, and hemoglobin were measured by a
CO-oximeter. Glucose and lactate were measured with a
Yellow Springs glucose analyzer. Cerebral oxygen consumption (CMRO2) was calculated as the product of CBF and the
arterial-sagittal sinus O2 content difference. Fractional O2
extraction was calculated as the ratio of cerebral arteriovenous O2 content difference to arterial O2 content.
Platelet studies. Platelet aggregation and ATP secretion
were measured on a whole blood, dual-channel lumiaggregometer (Chrono-Log, model 500VS; Havertown, PA) using established methods (2). Briefly, 2.7 ml of blood were collected from
the sagittal sinus and arterial catheters at baseline, 45 min of
ischemia, and 30, 60, and 120 min of reperfusion, and gently
mixed with 0.3 ml of 3.8% citrate. Platelet counts were
performed by phase-contrast microscopy using a Neubauer
hemocytometer before 1:1 sample dilution with isotonic saline. Aliquots of the diluted samples (950 µl) were pipetted
into cuvettes, and 50 µl luciferase-luciferin (Chrono-lume;
Chrono-Log no. 395) were added. The samples were warmed
to 37°C and stirred with a Teflon-coated stir bar for 2 min
before the addition of collagen. Collagen is classified as a
strong aggregant because of its interaction with the membrane receptor, which triggers platelet adhesion, shape change,
activation, aggregation, arachidonate release, and secretion
(19). Therefore, all samples were individually tested with 1, 2,
and 4 µg of collagen (final concn). Maximum aggregation
response was determined as the maximum amplitude of the
impedance curve at 6 min after collagen addition. Platelet
secretion of ATP was evaluated by peak amplitude of collageninduced ATP release compared with a preset 20 nM ATP
standard. Platelet aggregation response and ATP secretion
were quantified by computer-assisted data reduction system
(810/DR Aggrolink, Chrono-Log). To test for possible effects of
platelet number on aggregation, whole blood (60 ml) was
collected in 3.8% citrate via jugular venous puncture in adult
sheep and prepared as above, then diluted to 55% of baseline
values with autologous sheep platelet-rich plasma (PRP).
Aggregation was quantified by paired samples from the same
animal (n 5 7).
For the isotopic platelet labeling studies, 60 ml (sheep) or
30 ml (lambs) of blood were gently aspirated from a femoral
catheter into an acid-citrate-dextrose (ACD) solution (6:1
ratio), mixed thoroughly, and centrifuged [1,800 revolutions/
min (rpm), 25°C] for 5 min. The PRP supernatant was
separated and centrifuged again at 2,300 rpm (25°C) for 10
min to form a platelet pellet. The platelet was isolated and
gently washed with ACD-saline (2 ml) and centrifuged (2,300
rpm, 25°C) for 10 min. The platelet pellet was then resus-
H1295
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
Table 1. Physiological data during incomplete global ischemia and reperfusion
Reperfusion, min
Control
30
60
120
27 6 2
28 6 2
30 6 2
27 6 1
32 6 5
27 6 1
32 6 2
29 6 1
35 6 2
27 6 1
2.9 6 0.3
2.1 6 0.3
1.6 6 0.9
1.0 6 0.5
3.0 6 0.7
2.2 6 0.4
2.5 6 0.3
1.9 6 0.3
2.5 6 0.4
1.6 6 0.3
50 6 2
34 6 2
50 6 9
36 6 7
23 6 7
21 6 5
28 6 6
17 6 3
22 6 2
15 6 3
134 6 10
135 6 7
260 6 60‡
149 6 18*
156 6 30
109 6 9
150 6 9
125 6 11
155 6 14
116 6 6
35 6 1
36 6 1
39 6 1
41 6 3
39 6 4
42 6 3
37 6 2
38 6 3
37 6 1
39 6 1
7.46 6 0
7.37 6 0.01†
7.374 6 0.2‡
7.28 6 0.02†
7.35 6 0.03‡
7.27 6 0.09
7.45 6 0.02
7.31 6 0.03†
7.4 6 0.01
7.25 6 0.03†‡
Values are means 6 SE. Hct, hematocrit; CMRO2 , cerebral oxygen consumption; O2 Ext, oxygen extraction. Significant difference between
sheep and lambs: * P , 0.05, † P , 0.01; significant difference from baseline control: ‡ P , 0.05.
elevation reduced CBF during the ischemic period from
40 6 6 to 5 6 1 ml · min21 · 100 g21 in lambs and from
43 6 3 to 5 6 2 ml · min21 · 100 g21 in sheep (Fig. 2).
During reperfusion, there were no age-related differences in CBF recovery. However, hyperemia occurred in
sheep at 30 (90 6 9 ml · min21 · 100 g21; P , 0.001) and
60 (70 6 12 ml · min21 · 100 g21; P , 0.05) min of
reperfusion, whereas lambs evidenced hyperemia only at
60 min (120 6 32 ml · min21 · 100 g21; P , 0.05). In both
groups, CBF returned to baseline values by 120 min of
reperfusion. Ischemia reduced lamb CMRO2 to 1.0 6 0.5
ml · min21 · 100 g21 and sheep CMRO2 to 1.6 6 0.7 ml ·
min21 · 100 g21, with return to baseline values at 30 min
of reperfusion (Table 1). At 45 min of ischemia, circulating platelet counts fell in both lambs (271,000 6
50,489/µl; P , 0.05) and sheep (368,000 6 29,510/µl;
P , 0.05) compared with nonischemic controls (lambs,
525,500 6 66,400/µl; sheep, 436,000 6 39,900/µl).
Figure 3 shows that this fall in platelet numbers was
47.5 6 5.1% of baseline values in lambs and 59 6 4.9%
of baseline values in sheep. Furthermore, platelet
counts remained similarly depressed throughout reperfusion (P , 0.05; Fig. 3).
To evaluate the effects of severe global ischemia on
platelet function, we simultaneously tested platelet
aggregation and platelet dense granule secretory response (ATP release) to three concentrations of collagen
using whole blood lumiaggregometry. After ischemia,
both lambs and sheep demonstrated a depressed doseresponse curve for all collagen doses, as measured by
platelet impedance aggregometry (Figs. 4 and 5). Both
the magnitude (Fig. 6) and the slope (Fig. 7) of platelet
aggregation were depressed during ischemia and remained depressed throughout reperfusion (P , 0.01) in
Fig. 1. Time course of changes in cerebral perfusion pressure during
45 min of ischemia and 120 min of reperfusion. Ischemic sheep (IS,
n 5 9) and lambs (IL, n 5 8) are compared with saline-infused,
nonischemic sheep (NIS, n 5 5) and lambs (NIL, n 5 4). All data are
reported as means 6 SE.
Fig. 2. Time course of changes in cerebral blood flow during 45 min of
ischemia and 120 min of reperfusion. IS (n 5 9) and IL (n 5 8) are
compared with NIS (n 5 5) and NIL (n 5 4). All data are reported as
means 6 SE. * P , 0.05, *** P , 0.001 from control.
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
Hct, %
Sheep
Lambs
CMRO2 , ml · min21 · 100 g21
Sheep
Lambs
O2 Ext, %
Sheep
Lambs
PO2 , mmHg
Sheep
Lambs
PCO2 , mmHg
Sheep
Lambs
pH
Sheep
Lambs
Ischemia
H1296
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
both lambs and sheep. Stimulation with 4 µg of collagen after 45 min of incomplete global ischemia resulted
in reduced platelet aggregation in sheep and lambs
(20.4 6 29.2 and 26.8 6 44.7% of baseline, respectively;
Fig. 6). Similar reductions were observed in response to
stimulation with 2 and 1 µg of collagen. Depressed
platelet responsiveness was not observed in nonischemic animals. After adjustment for the depression of
circulating platelet numbers, ATP secretion was not
different from baseline values in either lambs or sheep.
However, at 45 min of ischemia, six of nine sheep and
three of six lambs demonstrated a marked increase in
platelet secretion of ATP (.100% of control) in response
to 4 µg of collagen. This heightened ATP release was
observed in only one of five nonischemic sheep and was
not detected in any nonischemic lambs.
To determine whether quantitative changes in aggregation accompanied the fall in platelet numbers, we
evaluated the effects on aggregation by dilution of
Fig. 4. Mean dose-response curve in sheep for platelet response to
stimulation with 1, 2, and 4 µg of collagen as measured by amplitude
(magnitude in ohms) of aggregation. Values are shown at baseline
(C), 45 min of ischemia (ISC), and 120 min of reperfusion (RP).
Fig. 5. Mean dose-response curve in lambs for platelet response to
stimulation with 1, 2, and 4 µg of collagen as measured by amplitude
(magnitude) of aggregation. Values are shown at baseline (C), 45 min
of ischemia (ISC), and 120 min of reperfusion (RP).
whole blood with autologous sheep PRP. A reduction of
platelet counts to 55 and 40% of baseline attenuated
aggregation to collagen (P , 0.01) in a dose-dependent
manner.
We examined the effects of ischemia-reperfusion on
indium-111-labeled platelet deposition in brain, muscle,
skin, heart, lung, liver, spleen, and kidney. Figure 8
demonstrates that sheep (P 5 0.045) and lamb (P 5
0.013) brain sequestered significant platelet numbers
compared with nonischemic brain. The greatest plateletassociated radioactivity was detected in the spleen of
sheep (P 5 0.03; Fig. 9). No difference was found
between nonischemic and ischemic lamb (P 5 0.71)
spleen platelet entrapment. However, the spleen of
nonischemic lambs trapped large numbers of platelets,
probably subsequent to platelet injury during labeling.
Therefore, the difference between platelet deposition in
Fig. 6. Platelet response to stimulation with 4 µg of collagen expressed as percentage of baseline values. IS (n 5 9) and IL (n 5 8) are
compared with NIS (n 5 5) and NIL (n 5 4). All data are reported as
means 6 SE. At 45 min of ischemic and throughout reperfusion, all
platelet aggregations for IS and IL are different from baseline (# P ,
0.01). Both ischemic groups are different from time controls (* P ,
0.05).
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
Fig. 3. Fall in platelet counts expressed as percentage of control
during 45 min of ischemia and 120 min of reperfusion. IS (n 5 9) and
IL (n 5 8) are compared with NIS (n 5 5) and NIL (n 5 4). All data are
reported as means 6 SE. At 45 min of ischemia and throughout
reperfusion all platelet counts are depressed (* P , 0.01).
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
nonischemic versus ischemic lambs was reduced. None
of the other tissues, in either sheep or lambs, evidenced
significant platelet entrapment during ischemia.
DISCUSSION
This study demonstrates four major findings. First,
severe, transient global cerebral ischemia produces
marked reduction of circulating platelets in both adult
sheep and neonatal lambs. The decline in platelet count
persists throughout reperfusion, suggesting ongoing
platelet activation. Second, platelets are sequestered
during early reperfusion in brain, independent of age
and despite full restoration of CBF. The preponderance
of isotopic activity was found in the spleen, an organ
that is a common site for platelet sequestration and
platelet clearance. Third, platelets are clearly hyporeactive to a well-known and physiologically strong aggregant (collagen) throughout early reperfusion. Platelet
dense granule ATP secretion during ischemia is quantitatively inconsistent and independent of age. Finally,
postischemic CBF abnormalities such as hyperemia are
present in both newborn and adult sheep; however,
hyperemia occurs earlier and persists longer in the
mature brain. Nevertheless, prolonged derangement of
CBF does not correspond with greater platelet patho-
Fig. 8. Brain, muscle, and skin entrapment of indium-111-labeled
platelets at 120 min of reperfusion. IS (n 5 3) and IL (n 5 3) are
compared with NIS (n 5 2) and NIL (n 5 3). All data are reported as
means 6 SE (* P , 0.05).
Fig. 9. Organ entrapment of indium-111-labeled platelets at 120 min
of reperfusion. IS (n 5 3) and IL (n 5 3) are compared with NIS (n 5
2) and NIL (n 5 3). All data are reported as means 6 SE (* P , 0.05).
physiology, at least not in the short time window of our
observations. In light of these findings, we conclude
that severe, incomplete global ischemia alters platelet
biology equivalently in both newborn and adult brain.
We hypothesize that platelet aggregates are actively
formed in both newborn and mature brain vessels as a
consequence of exposure to an ischemic microvasculature bed subsequent to an early and evolving endothelial injury.
Platelets have been implicated both as etiologic
agents in cerebral ischemia associated with thromboembolic vessel occlusion and as mediators of neuronal
injury following an ischemic insult. Furthermore, platelet activation triggers release of dense granule constituents and mediators with significant vasoconstrictive
consequences. However, it is not known whether cerebral ischemia specifically induces abnormalities in
platelet function, thereby creating a vehicle for secondary brain injury during reperfusion or revascularization. Our experimental model allowed observation of
‘‘isolated’’ cerebral ischemia by selectively reducing
perfusion pressure only within the cerebral circulation.
We examined not only systemic aortic platelet samples
but platelets obtained from the sagittal sinus, which
reflects venous composition and drainage directly from
brain. The data suggest that severely reduced flow to
the brain induces significant thrombocytopenia that
persists during early reperfusion. Accounts of thrombocytopenia following clinical ischemic stroke are variable; however, our findings are generally consistent
with previous reports in experimental global, but not
focal, ischemia in animals (13, 17, 20). Several clinical
studies report moderate depression of circulating platelet numbers after stroke, ranging from 15 to 26%
compared with age- and gender-matched controls (10,
36, 40), whereas others report no differences (16, 24,
31). Because initial postischemia platelet counts have
not been reported in acute human stroke, it is difficult
to determine whether our animal data are consistent
with findings after clinical cerebral ischemia. However,
we detected a drastic fall in circulating platelets to as
low as 50% of control values in animals in which
ischemic CBF was most severely reduced (e.g., to 5
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
Fig. 7. Velocity of aggregation as measured by slope of aggregation
curve in response to stimulation with 4 µg of collagen during 45 min
of ischemia and 120 min of reperfusion. At 45 min of ischemia and
throughout reperfusion, all slopes are less than baseline (** P , 0.01).
H1297
H1298
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
eral carotid artery occlusion (7). However, clinical
studies examining platelet responsiveness to collagen
after ischemic stroke demonstrate either no change in
aggregation (24) or decreased magnitude of the aggregation response (14, 40). Our data indicate that circulating platelets in whole blood are hypoaggregable to
stimulation after global cerebral ischemia, and the
hyporeactivity persists throughout early reperfusion.
These data are novel in that we examined platelet
responsiveness to increasing concentrations of collagen
in whole blood sampled directly from the sagittal sinus
draining blood from brain regions where CBF was
homogeneously and reversibly reduced. It seems unlikely that the reduced number of platelets over the
experimental course resulted in a lowered threshold for
aggregation; others have demonstrated no correlation
between similar platelet numbers and aggregation
(40). Our dilutional studies did reveal attenuated platelet responsiveness when samples were diluted by 50%;
however, dilution with PRP diminished sample hematocrit, increasing the acellular volume fraction. Therefore, the final collagen concentration presented to the
platelet as a stimulus for aggregation was reduced.
At all doses of collagen, there was a significant
reduction in both the magnitude of platelet aggregation
and the rapidity of the response. These data do not
directly suggest a mechanism for the loss of reactivity,
but postischemic hypoaggregation is clearly independent of animal age. This finding is surprising given
previous reports that platelets in the normal human
newborn show less sensitivity to aggregants such as
ADP, collagen, thrombin, and epinephrine relative to
adult platelets (3, 4, 33, 37). We did not observe
baseline differences in platelet aggregation between
the newborn lambs and sheep. This may be related to
species differences or may indicate that an age of 3–5
days in lambs is not equivalent to the same age in
neonatal humans. However, ischemia-sensitized platelets harvested from both lambs and sheep responded in
a similar, depressed manner to collagen stimulation.
Although we detected higher ischemic PaO2 values in
sheep, it is unlikely that these differences had an effect
on platelet function, particularly because we observed
depressed platelet function in both lambs and sheep,
regardless of differences in PaO2. Ischemia in both
lambs and sheep elicited a profound Cushing response.
Sheep were unable to tolerate the stress on the myocardium incurred by the cardiovascular response, which
often raised mean arterial blood pressure to .225
mmHg, whereas lambs tolerated ischemia better, demonstrating a lower ischemic blood pressure. Therefore,
the differences in PaO2 reflect administration of higher
inspired oxygen concentrations required in sheep to
support cardiac function. Additionally, we observed
lower pH in lambs than in sheep throughout the study.
Normally, neonatal lambs have a lower pH than adults
because of an inability to regurgitate the cud. Furthermore, neonates possess less functional ability to adjust
serum bicarbonate during acidosis as a result of immature renal distal tubule function. However, ischemia
resulted in lowered pH in lambs proportionally as in
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
ml · min21 · 100 g21 ). The reduction could not be attributed to hemodilution, because hematocrit was unchanged in lambs and slightly more concentrated in
sheep over the study time course. Furthermore, newborn animals were not spared from ischemia-induced
thrombocytopenia; it appears to be an age-independent
response to ischemic brain insult.
Consequently, we examined both cerebral and noncerebral tissues to determine whether the location and
magnitude of platelet deposition was also age independent. We found platelets widely dispersed throughout
the reperfused brain within 2 h of recirculation in both
the neonatal and mature brain. This finding is similar
in time frame to that observed after prolonged global
ischemia in adult cat (20) and photochemical carotid
injury in rat (6), effectively within hours of the acute
injury. Similar to the reports by others (6, 20), platelet
deposits formed in visceral tissues such as spleen
during reperfusion, to a lesser extent in the newborn in
which splenic entrapment was not elevated relative to
nonischemic lambs. It is possible that, in the adult,
platelets become temporarily adherent and are subsequently released to be removed later by the spleen. In
contrast to others (6, 20), we found platelet deposition
only in the spleen, but not liver, lung, or kidney. Most
likely these differences may be attributed to either
species or technique differences rather than perfusion
changes in these tissues. In the current study it is
unclear whether cerebral ischemia resulted in blood
flow reduction in peripheral organs. However, earlier
studies from our laboratory and from others indicate
that cerebral ischemia does not result in diminished
blood flow to skin, intestine, kidney, muscle, lung, or
liver (20, 30). On the basis of these data, it is unlikely
that reduction in splenic blood flow occurs, either.
However, both liver and spleen are known sites for
removal of damaged platelets, and our findings most
likely represent enhanced normal splenic function
rather than ischemic endothelial injury. Elgjo and
Hovig (9) demonstrated extensive platelet sequestration in spleen sinusoids as well as phagocytic removal
of damaged platelets. Others show that normal platelet
loads can be cleared by the liver, but increased platelet
load results in heightened platelet localization in spleen
and lung (29). These data are consistent with reports of
platelet entrapment in liver, lung, and spleen in patients with inflammatory disorders such as sepsis.
Therefore, it is not surprising that we observed increased platelet entrapment in spleen in the adult
sheep. The lack of significant sequestration in the
newborn spleen may be caused by its relative immaturity in the 2- to 5-day-old lamb (12). Therefore, it is
likely that our data reflect augmented splenic removal
of platelets damaged as a consequence of cerebral
ischemia.
The effect of global cerebral ischemia on platelet
biology has not been extensively investigated. Therefore, we sought to determine whether exposure to a
large segment of the cerebral vascular bed made acutely
ischemic triggers platelet hyperreactivity. Platelet aggregates increase in jugular venous blood after unilat-
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
accompanying ischemia could intensify neuronal injury
and negatively affect neurological outcome.
This work was supported by grants from the National Institutes of
Health (NR-03816, NR-03521, and NS-20020) and the Maryland
Affiliate of the American Heart Association (MDG-20595).
Address for reprint requests: M. T. Littleton-Kearney, Dept. of
Anesthesiology/Critical Care Medicine, Blalock 1404, The Johns
Hopkins Hosp., 600 N. Wolfe St., Baltimore, MD 21287-2725 (E-mail:
[email protected]).
Received 12 September 1997; accepted in final form 19 December
1997.
REFERENCES
1. Borel, C. O., J. E. Backofen, R. C. Koehler, M. D. Jones, Jr.,
and R. J. Traystman. Cerebral blood flow autoregulation
during intracranial hypertension in hypoxic lambs. Am. J.
Physiol. 253 (Heart Circ. Physiol. 22): H1342–H1348, 1987.
2. Cardinal, P. C., and R. Flower. The electronic aggregometer: a
novel device for assessing platelet behavior in whole blood. J.
Pharmacol. Methods 3: 135–158, 1980.
3. Corby, D. G., and T. F. Zuck. Newborn platelet dysfunction: a
storage pool and release defect. Thromb. Haemost. 36: 200–207,
1976.
4. Corby, D. G., and T. P. O’Barr. Decreased alpha-adrenergic
receptors in newborn platelets: cause of abnormal response to
epinephrine. Dev. Pharmacol. Ther. 2: 215–225, 1981.
5. Dewanjee, M. K., S. A. Rao, and P. Didisheim. Indium-111
tropolone, a new high-affinity platelet label: preparation and
evaluation of labeling parameters. J. Nucl. Med. 22: 981–987,
1981.
6. Dietrich, W. D., S. Dewanjee, R. Prado, B. D. Watson, and
M. K. Dewanjee. Transient platelet accumulation in the rat
brain after common carotid artery thrombosis. An 111In-labeled
platelet study. Stroke 24: 1534–1540, 1993.
7. Dougherty, J. H., D. E. Levy, and B. B. Weksler. Platelet
activation in acute cerebral ischemia. Serial measurements of
platelet function in cerebrovascular disease. Lancet 1: 821–824,
1977.
8. Dougherty, J. H., D. E. Levy, and B. B. Weksler. Experimental cerebral ischemia produces platelet aggregates. Neurology 29:
1460–1465, 1979.
9. Elgjo, R. F., and T. Hovig. Ultrastructural studies of platelet
‘‘activation’’ and aggregation in normal spleen from different
species. Scand. J. Haematol. 9: 587–602, 1972.
10. Farkkila, M. A., V. Rasi, R. S. Tilvis, E. Ikkala, L. Viinikka,
O. Ylikorkala, A. M. Farkkila, and T. A. Miettinen. Low
platelet arachidonic acid in young patients with brain infarction.
Thromb. Res. 48: 721–727, 1987.
11. Fisher, M., and R. Francis. Altered coagulation in cerebral
ischemia: platelet, thrombin, and plasmin activity. Arch. Neurol.
47: 1075–1079, 1990.
12. Freedman, R. M., D. Johnston, M. J. Mahoney, and H. A.
Pearson. Development of splenic reticuloendothelial dysfunction in neonates. J. Pediatr. 96: 466–468, 1980.
13. Furlow, T. W., and J. M. Hallenbeck. Indomethacin prevents
impaired perfusion of the dog’s brain after global ischemia.
Stroke 9: 591–594, 1978.
14. Galante, A., A. Pietroiusti, M. Silvestrini, P. Stanzione, B.
Domenici, and G. Bernardi. Leukocyte and platelet aggregation in patients with ischemic stroke: a case-control study. Eur.
Neurol. 33: 442–445, 1993.
15. Gawaz, M. P., J. C. Loftus, M. L. Bajt, M. M. Frojmovic, E. F.
Plow, and M. H. Ginsberg. Ligand bridging mediates integrin
aIIbb3 (platelet GPIIb-IIIa ) dependent homotypic and heterotypic
cell-cell interactions. J. Clin. Invest. 88: 1128–1134, 1991.
16. Grau, A. J., R. Sigmund, and W. Hacke. Modification of
platelet aggregation by leukocytes in acute ischemic stroke.
Stroke 25: 2149–2152, 1994.
17. Gross, C. E., D. B. Howard, R. H. Dooley, S. J. Raymond, S.
Fuller, and M. M. Bednar. TGF-b1 post-treatment in a rabbit
model of cerebral ischaemia. Neurol. Res. 16: 465–470, 1994.
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
sheep, making it unlikely to significantly depress platelet function.
Ischemic exposure could alter platelet biology by
potentiating either primary (reversible) or secondary
(irreversible) aggregation via alteration of membranebound surface receptors, depletion of granular constituents, or increasing the fraction of exhausted, disaggregated platelets in the reperfused brain. Platelet
aggregation requires receptor binding of fibrinogen
(15), and in vitro studies suggest a correlation between
time-dependent loss of platelet aggregation and internalization of surface-bound fibrinogen (42). Postischemic redistribution of bound fibrinogen to inaccessible intracellular storage sites could explain the platelet
hyporeactivity we noted. Furthermore, clinical ischemic
stroke (26) induces a-granule depletion, indicating
increased platelet secretion after cerebrovascular insult. A reduction in granular constituents may denote
primary aggregation and cause attenuation of the
aggregation response. Loss of a-granule constituents
could also account for the marked hyporesponsiveness
in both adult and newborn sheep.
Enhanced ATP secretion from dense granules (14,
40), increased platelet calcium release (24, 27), and
liberation of a-granule and b-thromboglobulin (11, 39)
and thromboxane from activated platelets (31) have
been noted in patients after stroke. Because ATP
production is platelet dependent, whole blood platelet
numbers affect the measurement accuracy. Therefore,
we normalized data for ATP secretion per 100,000
platelets/µl to determine whether ischemia alters ATP
release. Consequently, we did not detect a statistically
significant change in ATP secretion. We found extreme
variability in platelet ATP release postischemia likely
caused, in part, by combinations of partially activated
and circulating platelet aggregates within our samples.
On the basis of their findings of poststroke enhanced
ATP release and depressed aggregation, Joseph and
colleagues (24) suggest that dense granular activation
may be independent of aggregation. Our observations
would support this hypothesis.
In conclusion, global cerebral ischemia triggers a
drastic fall in circulating platelet numbers in both the
newborn and adult. Whole blood platelet aggregation
studies with collagen indicate that postischemic platelets are refractory to strong platelet agonists, possibly
as a consequence of preformed platelet aggregates,
which are no longer responsive to physiological levels of
agonists, sequestration of platelet-bound fibrinogen, or
diminished granular constituents. During reperfusion,
platelet aggregates are sequestered in the brain equivalently in the young and the adult animal. It seems
likely that platelet aggregates are actively formed in
both newborn and mature brain vessels as a direct
consequence of exposure to an ischemic microvascular
bed with progressive endothelial injury. It is unclear
what impact this has on the outcome of cerebrovascular
injury. In addition to endothelial denudation, the mass
of accumulated platelets may contribute to microvascular obstruction and the release of platelet secretory
products, potentiating cerebrovascular vasospasm. The
H1299
H1300
CEREBRAL ISCHEMIA ALTERS PLATELET BIOLOGY
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
Am. J. Physiol. 265 (Heart Circ. Physiol. 34): H1557–H1563,
1993.
Koudstaal, P. J., G. Ciabattoni, J. van Gijn, K. Nieuwenhuis, P. G. de Groot, J. J. Sixma, and C. Patrono. Increased
thromboxane biosynthesis in patients with acute cerebral ischemia. Stroke 24: 219–223, 1993.
Lee, Y. J., W. Jy, L. L. Horstman, J. Janania, Y. Reyes, R. E.
Kelley, and Y. S. Ahn. Elevated platelet microparticles in
transient ischemic attacks, lacunar infarcts, and multiinfarct
dementias. Thromb. Res. 72: 295–304, 1993.
Mull, M. M., and W. E. Hathaway. Altered platelet function in
newborns. Pediatr. Res. 4: 229–237, 1970.
Obrenovitch, T. P., and J. M. Hallenbeck. Platelet accumulation in regions of low blood flow during the postischemic period.
Stroke 16: 224–233, 1985.
Olah, L., M. Misz, D. Bereczki, I. Fekete, J. E. Bordanne,
and E. I. Takacs. Low doses of acetylsalicylic acid effectively
inhibits thrombocyte aggregation after ischemic stroke. Orv.
Hetil. 137: 455–459, 1996.
O’Malley, T., P. Langhorne, R. A. Elton, and C. Stewart.
Platelet size in stroke patients. Stroke 26: 995–999, 1995.
Rajasekhar, D., A. S. Kestin, F. J. Bednarek, P. A. Ellis,
M. R. Barnard, and A. D. Michelson. Neonatal platelets are
less reactive than adult platelets to physiological agonists in
whole blood. Thromb. Haemost. 72: 957–963, 1994.
Satoh, S., Y. Suzuki, T. Harada, I. Ikegaki, T. Asano, M.
Shubuya, K. Sugita, and A. Saito. The role of platelets in the
development of cerebral vasospasm. Brain Res. Bull. 27: 663–
668, 1991.
Shah, A. B., N. Beamer, and B. M. Coull. Enhanced in vivo
platelet activation in subtypes of ischemic stroke. Stroke 16:
643–647, 1985.
Tohgi, H., H. Suzuki, K. Tamura, and B. Kimura. Platelet
volume, aggregation, and adenosine triphosphate release in
cerebral thrombosis. Stroke 22: 17–21, 191.
Ware, J. A., and D. D. Heistad. Platelet-endothelium interactions. N. Engl. J. Med. 328: 628–635, 1993.
Wencel-Drake, J. D., C. Boudignon-Proudhon, M. G. Dieter, A. B. Criss, and L. V. Parise. Internalization of bound
fibrinogen modulates platelet aggregation. Blood 87: 602–612,
1996.
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.3 on June 15, 2017
18. Heymann, M. A., B. D. Payne, J. I. E. Hoffman, and A. M.
Rudolph. Blood flow measurements with radionuclide-labeled
particles. Prog. Cardiovasc. Dis. 20: 55–79, 1997.
19. Holmsen, H. Platelet secretion and energy metabolism. In:
Hemostasis and Thrombosis: Basic Principles and Clinical Practice, edited by R. W. Coleman, J. Hirsh, V. J. Marder, and E. W.
Salzman. Philadelphia, PA: Lippincott, 1994, p. 527.
20. Hossmann, V., K. A. Hossmann, and S. Takagi. Effect of
intravascular platelet aggregation on blood recirculation following prolonged ischemia of the cat brain. J. Neurol. 222: 159–170,
1980.
21. Hurn, P. D., R. C. Koehler, S. E. Norris, K. K. Blizzard, and
R. J. Traystman. Dependence of cerebral energy phosphate and
evoked potential recovery on end-ischemic pH. Am. J. Physiol.
260 (Heart Circ. Physiol. 29): H532–H541, 1991.
22. Iwamoto, T., H. Kubo, and M. Takasaki. Platelet activation in
the cerebral circulation in different subtypes of ischemic stroke
and Binswanger’s disease. Stroke 26: 52–56, 1995.
23. Jafar, J. J., R. Menoti, H. Feinberg, G. LeBreton, and R.
Crowell. Selective platelet deposition during focal cerebral
ischemia in cats. Stroke 20: 664–667, 1989.
24. Joseph, R., G. D’Andrea, S. B. Oster, and K. M. A. Welch.
Whole blood platelet function in acute ischemic stroke. Stroke 20:
38–44, 1988.
25. Joseph, R., E. Han, C. Tsering, S. Grunfeld, and K. M. A.
Welch. Platelet activity and stroke severity. J. Neurol. Sci. 108:
1–6, 1992.
26. Joseph, R., J. M. Riddle, K. M. A. Welch, and G. D’Andrea.
Platelet ultrastructure and secretion in acute ischemic stroke.
Stroke 20: 1316–1319, 1989.
27. Joseph, R., K. M. A. Welch, S. Grunfield, S. B. Oster, and G.
D’Andrea. Baseline and activated cytoplasmic ionized calcium
in acute ischemic stroke. Stroke 19: 1234–1238, 1988.
28. Joseph, R., K. M. A. Welch, S. B. Oster, S. Grunfeld, and G.
D’Andrea. A plasmic factor may cause platelet activation in
acute stroke. Circ. Res. 65: 1679–1687, 1989.
29. Kaplan, J. E., and T. M. Saba. Platelet removal from the
circulation by liver and spleen. Am. J. Physiol. 235 (Heart Circ.
Physiol. 4): H314–H320, 1978.
30. Kearney, M. L., J. E. Backofen, R. C. Koehler, M. D. Jones,
Jr., and R. C. Traystman. Effect of hypoxemia on the cardiovascular response to intracranial hypertension in postnatal lambs.