Vol. 89 • No. 2
BRIEF SCIENTIFIC REPORTS
14. Neufeld EJ, Wilson DB, Sprecher H, Majerus PW. High affinity
esterification of eicosanoid precursor fatty acids by platelets. J
Clin Invest 1983;72:214-220.
15. Pareti FI, Mannucci PM, D'Angelo A, Smith JB, Sautebin L, Galli
G. Congenital deficiency of thromboxane and prostacyclin.
Lancet 1980;1:898-901.
16. Rak K., Boda Z. Haemostatic balance in congenital deficiency of
platelet cyclo-oxygenase. Lancet 1980;2:44.
17. Rao, AK, Holmsen H. Congenital disorders of platelet function.
SeminHematol 1986;23:102-118.
221
18. Remuzzi G, Benigni A, Dodesini P, et al. Reduced platelet
thromboxane formation in uremia. J Clin Invest 1983,71:762768.
19. Stampfer MJ, Jakubowski JA, Deykin D, Schafer AL, Willett WC,
Hennekens CH. Effect of alternate-day regular and entericcoated aspirin on platelet aggregation, bleeding time, and
thromboxane A2 levels in bleeding-time blood. Am J Med
1986;81:400-404.
20. Weiss HJ, Lages BA. Possible congenital defect in platelet thromboxane synthetase. Lancet 1977;1:760.
Hemolysis in Sickle Cell Disease as Measured by
Endogenous Carbon Monoxide Production
A Preliminary Report
DILIP L. SOLANKI, M.D., PAUL R. McCURDY, M.D., FRANK F. CUTTITTA, PH.D.,
AND GERALDINE P. SCHECHTER, M.D.
To detect and quantitate temporal variations of the hemolytic
rate in sickle cell disease, the authors measured endogenous
carbon monoxide (CO) production in five normal subjects, nine
patients with sickle cell anemia (SS) in steady clinical state,
and two patients with sickle cell-hemoglobin C(SC) disease in
and out of pain crises. The red blood cell life span calculated
from these data (RCLSC0) ranged from 81.2 to 102.9 days
(mean ± standard deviation [SD] 88.0 ± 9.2, coefficient of
variation [CV] 10.2%) for the normal subjects and 8.0-24.7
days (mean ± SD 12.1 ± 5.1, CV 42.1%) for those with SS.
Although the individual figures for RCLSC0 for the normal
subjects and those with SS fell within the range previously
obtained by radioisotopic techniques for the respective groups,
the mean values calculated from the CO technique were
slightly (though not significantly) shorter for the normal sub-^
jects and about 25% shorter for the subjects with SS (P
< 0.01). Repetitive studies were performed in four subjects
with SS who were clinically stable; the temporal variability in
the calculated hemolytic rate differed considerably from patient to patient (CV 3.6%, 7.0%, 17.0%, 28.0%). In two patients, concurrent RCLS studies were performed by the CO
Received November 11, 1986; received revised manuscript and accepted for publication July 28, 1987.
Supported by grants from the National Heart, Lung and Blood Institute (HL 14131) and the Veterans Administration.
Dr. McCurdy's current address is Division of Blood Diseases and
Resources, National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, Maryland.
Address reprint requests to Dr. Solanki: Veterans Administration
Medical Center (111J), 921 NE 13th Street, Oklahoma City, Oklahoma 73104.
Department of Medicine, Hematology Section, University of
Oklahoma College of Medicine and the Veterans
Administration Medical Center, Oklahoma City, Oklahoma,
Department of Medicine, Division of Hematology,
Georgetown University School of Medicine, and Department
of Medicine, Hematology Section, George Washington
University School of Medicine, and the Veterans
Administration Medical Center, Washington, D.C.
technique and 5lCr tagged red blood cells. In one patient, the
RCLS was similar by the two techniques, in the other, a two
exponent 5lCr curve did not permit calculation of RCLS. In the
two patients with SC disease there was no difference in RCLSTO
during and after recovery from pain crisis. Although the CO
technique may overestimate the turnover of circulating heme
mass, especially in the presence of hemolysis, the results of
serial studies in a small number of patients with SS suggest but
do not prove temporal variations in hemolytic rate in SS. (Key
words: Sickle cell disease; Hemolysis; Endogenous CO production) Am J Clin Pathol 1988;89:221-225
W H E N D E O X Y G E N A T E D , hemoglobin S polymerizes and forms molecular aggregates that distort the red
blood cell into the familiar sickle shape. Initially this
process is completely reversible upon reoxygenation,
but after repeated anoxic cycles the cell becomes irreversibly sickled. 17 The sickled forms regularly found on
routine air-dried blood smears from patients with sickle
cell anemia are irreversibly sickled cells (ISCs) and ap-
222
SOLANKI ET AL.
pear to result from membrane damage because the intracellular hemoglobin S in these cells still depolymerizes when oxygenated.5 Their number, which presumably reflects the magnitude of intravascular sickling, has
been strongly correlated with the rate of hemolysis in
sickle cell anemia.15,20 Rigid sickled cells are also believed to cause obstruction in the microcirculation, resulting in vasoocclusive crises and tissue infarctions. Because the rate of hemolysis is strongly and directly correlated with the magnitude of sickling and can be readily
quantitated, its measurement, most commonly with the
use of 5lCr tagged autologous red blood cells, has often
been used to monitor effects of various antisickling
agents.'"•' 3 ' 6 Because sickling undoubtedly varies with
time, it seems likely that hemolysis also varies with
time.18 However, the commonly used radioisotopic
techniques for measuring red blood cell life span are
insensitive to such daily or hourly fluctuations. Furthermore, the studies take days or weeks to complete and
involve radiation exposure; hence, they are unsuitable
for repeated studies. The use of 51Cr, as a red blood cell
label, also carries the problem of a variable elution
rate,1415 not only from one subject to another, but also
in the same person. The varying rate of elution from red
blood cells is believed to be the cause of the frequent
two-exponent 51Cr red blood cell survival curves in patients with sickle cell anemia.15 This obscures the Tl/2
and could falsely suggest improvement from a therapeutic maneuver. A technique that overcomes these shortcomings uses the rate of endogenous carbon monoxide
(CO) production as a measure of hemolysis8 based on
the fact that CO is quantitatively liberated during the
initial step in heme catabolism. It is sensitive to small
changes in hemolytic rate,7 requires only a few hours to
complete, and can be readily repeated. Nevertheless, it
has not been widely applied to the study of sickle cell
disease.
We describe the use of this technique to detect variation in hemolytic rate and thus sickling in patients with
sickle cell disease in a steady clinical state and during
pain crises.
Subjects and Methods
Five normal volunteers, nine patients with sickle cell
anemia in steady state, and two patients with sickle cell
hemoglobin C disease during pain crises were selected
from our hematology clinic and studied after an informed consent was obtained. The study was approved
by the Human Experimentation Committees at the institutions involved. The diagnosis of sickle cell disease
was based upon a combination of a single band in the S
position or bands in S and C positions and the absence
of hemoglobin A from the hemolysate by electrophoresis on cellulose acetate or starch gel at an alkaline pH
A.J.C.P. • February 1988
and agar gel at an acid pH, clinical evaluation, hemoglobin solubility, and family studies when possible.
Endogenous CO production was measured essentially
as described by Coburn and colleagues.8 Briefly, the
subject to be studied was placed in a rebreathing system
consisting of a cylindric translucent plastic hood fitted
over the subject's head and sealed at the neck with a
silicone lubricated rubber diaphragm. Oxygen was
added as necessary to maintain 20-25% proportion with
the use of an automatic monitor controller. Air was
circulated within the system by a pump to ensure proper
mixing. Subjects had not smoked for at least 12 hours
before the procedure.12 After a period of equilibration,
serial venous blood samples were drawn every 15-30
minutes (minimum six samples for each subject)
through an indwelling needle and analyzed for CO content as carboxyhemoglobin (COHB) by a gas chromatographic technique.9 The variability of the replicate measurements of COHb was less than 1.0% for normal subjects and less than 3% for patients with sickle cell
disease. The CO content of the blood was plotted against
time and from the slope; the red blood cell life span
derived from the CO production (RCLSC0) was calculated by the formula adapted from Coltman and Dudley10
RCLSCo
8.334
Slope X normal red blood cell life span (120 days)
with the use of the figure 75% as the proportion of CO
production coming from the circulating heme breakdown4 (Fig. 1). The other 25% comes from catabolism of
other heme compounds in the body (e.g., myoglobin).
Results
For the five normal subjects, RCLSco ranged from
81.2 to 102.9 days (mean ± standard deviation [SD],
88.0 ± 9.2, coefficient of variation [CV] 10.4%) (Table
1). Although slightly shorter than the usually quoted
100-120 days, it is not outside the range obtained by us
and others using the DF32P technique.6
The mean RCLSC0 for the nine subjects with sickle cell
anemia was 12.1 days (SD 5.1, CV 42%), with a range of
8.0-24.7 days (Table 2). This mean value is about
three-fourths of that determined by the DF32P technique
(17.3 days) previously obtained in our laboratory in a
different group of patients with sickle cell anemia. Although, the individual figures for this group of patients
fell within the same range as obtained by the radioisotopic method,15 the difference between the means is significant (P < 0.01).
Four patients were studied repetitively: one six times
(five of six studies over a 38-day period), a second four
times, a third three times, and a fourth, twice (Table 3).
BRIEF SCIENTIFIC REPORTS
Vol. 89 • No. 2
*
2.0
223
8.0
JF cf
Slope 1.125 x 10 '
KB Cf SS
Slope 8.516 x 10"'
r = 0.93
r = 0.98
RCLS ' 8 2 . 3 days
FIG. 1. Carbon monoxide content of the blood plotted against
time in a normal subject (subject 4)
and a patient with sickle cell anemia (patient 3), showing a higher
basal level and a more rapid increase in carboxyhemoglobin level
in the patient with sickle cell anemia.
RCLS • 10.9 days
1.5
6.0
z
5
o
1.0
o
o
4.0
X
2.0
w
2
w
>-
0.5
X
O
m
tr
<
o
20
40
60
100
80
20
120
40
60
80
100
120
TIME (minutes)
For patient 1, a 48-year-old man with sickle cell anemia,
severe sickling retinopathy, and moderate renal insufficiency, the average RCLSC0 was 8.0 days (SD 2.1, CV
28.0%), with a range of 5.6-10.6 days. The variability
was less striking in the other three. We could ascertain
no clinical changes during these studies that might correlate with the variability in RCLSC0.
Two patients with sickle cell-hemoglobin C disease
studied during pain crisis showed no difference in the
RCLSCo whether in or out of crisis; one of these patients
was studied about 36 hours after the onset of the crisis
(Table 4).
In two patients, RCLSC0 was determined twice in a
two-week period while they were undergoing red blood
cell life span (RCLS) study using 51Cr-tagged autologous
red blood cells (RCLSCr). In one, the RCLS was similar
with both techniques; in the other, a two-exponent 5'Cr
curve did not permit calculation of RCLSCrDiscussion
Temporal variations in the hemolytic rate in individual patients with sickle cell disease in a steady clinical
state, although long suspected, have been difficult to
detect and quantitate because the commonly used techniques involving radioisotopic red blood cell tags are
insensitive and poorly suited for the repetitive determinations of RCLS necessary for this purpose. Our study
shows that the CO technique to measure the rate of
hemolysis is a feasible alternative. Its speed and sensitivity are well established,2'7 and, in our experience, it is
well accepted by most patients. The rebreathing system
can be taken to the bedside, thus allowing study of patients who are unable to be moved to the laboratory. It
was difficult, but not impossible, to study patients during pain crisis.
The mean RCLSC0 of 88 days for normal subjects is
slightly shorter than the usually quoted 100-120 days.
Although this value falls within the range obtained with
the DF32P technique,6 it is possible that the CO technique overestimates the turnover of circulating heme
mass. This possibility is even more strongly suggested in
our patients with sickle cell anemia whose mean RCLS
of 12.3 days is about three-fourths of the RCLS determined in our laboratory by the DF32P technique (17.3
days),15 a difference that is significant. Nevertheless, the
individual figures for this group of patients fell within
the same range as that obtained by the radioisotopic
technique, and each study was done at a time different
from the other. Hence, the interpretation must be cautious. On the other hand, it is possible that a greater
fraction of CO than the 25% assumed in our calculations
Table 1. CO Generation: Normal Subjects
Subject
Sex
1
2
3
4
5
F
M
M
M
M
Mean
±SD
CV(%)
Hct
(L/L)
Slope
(X10"3)
RCLS
(days)
0.37
0.45
0.44
0.40
0.40
0.41
0.03
0.908
1.019
1.140
1.125
1.189
1.062
0.099
9.3
102.9
90.9
81.2
82.3
82.7
88.0
9.2
10.2
Table 2. CO Generation: Patients with SS
Patient
Sex
1
2
3
4
5
6
7
8
9
M
F
M
M
F
M
M
M
M
Mean
±SD
CV(%)
* Mean of six studies.
t Mean of four studies.
% Mean of three studies.
§ Mean of two studies.
Hct
L/L
MCV
(fL)
Slope
(X10~3)
RCLS
(days)
0.17
0.21
0.19
0.20
0.23
0.19
0.26
0.21
0.23
0.21
0.03
95
93
89
100
96
89
99
86
98
93.9
5.0
12.340
8.699
8.516
8.227
6.716
7.117
3.749
11.747
9.0311
8.4602
2.5821
8.0*
10.5f
10.4*
11.3§
13.8
13.0
24.7
7.9
10.3
12.1
5.1
42.1
SOLANKI ET AL.
224
Table 3. CO Generation: Repetitive Studies in Four
Patients with SS
No. of
Studies
Patient 1
RCLS
(days)
Patient 2
RCLS
(days)
Patient 3
RCLS
(days)
Patient 4
RCLS
(days)
1
2
3
4
5
6
Mean
SD
CV(%)
9.2
5.8
10.6
5.6
8.7
7.8
8.0
1.97
28
11.3
12.6
8.7
9.2
10.9
9.6
10.7
10.9
11.5
10.5
1.8
17
10.4
0.7
7
11.3
0.4
3.6
came from sources other than circulating red blood
cells. Berk and associates4 found a larger fraction of bilirubin to come from such other sources when there was
brisk hemolysis than when red blood cells survived normally. They suggested that the excess bilirubin was of
erythropoietic origin, possibly from small amounts of
cytoplasm remaining on the extruded erythroblast nuclei and/or from an absolute increase in ineffective
erythropoiesis. Alternatively, other factors such as diet,
varying catabolism of other heme enzymes, or minor
alterations in the fraction of ineffective erythropoiesis
may account for the variability. Additional study is
needed to address these questions. Such information is
of obvious importance in the study of patients with
sickle cell disease.
Of particular interest are the results of the repetitive
studies in four patients. The variability in the results of
six studies in patient 1 (CV 28%) was only slightly less
than it was among the group of nine separate persons
(CV 42%), suggesting an inherent error of the technique
as its cause. However, the considerably less variability in
repeated studies in patients 2, 3, and 4 raises other possibilities. The most pertinent likelihood is that the degree
of variability in hemolytic rate over time is an individual
attribute. Our previous data3 from use of the radioisotopic technique support this interpretation. In ten studies
in five patients with sickle cell anemia with the use of
51
Cr-tagged red blood cells, the CVs of Tl/2 5lCr were 3,
Table 4. CO Generation During Pain Crisis in Two
Patients with SC Disease
Slope (XIO"3)
8, 18, 9, and 11%; in the first three of these patients the
corresponding CVs of RCLS with the use of the DF32P
red blood cell tag were 5, 3, and 35%. If that is the case,
any study involving measurement of changes in hemolytic rate would require that variability in each patient
be established in order that changes in the rate of hemolysis from a therapeutic maneuver be properly interpreted. Alternative possibilities of influence of diet or
clinical events were considered but could not be detected in our patients during the period of the study.
Whether the presence of moderate renal insufficiency
and chronic heart failure in patient 1 contributed in
some way to the variability is not known.
Although our data are limited, the apparent lack of
change in endogenous CO production in and out of pain
crises is difficult to understand because one might expect vasoocclusion by sickled cells to result in destruction of those cells and an increase in CO production.
Possibly, the change is too small to detect in the background variation. In this regard, the observations of
Rodgers and colleagues19 are of interest. Using the technique of laser-Doppler velocimetry, they noted marked
local oscillations in bloodflowin patients with sickle cell
disease in contrast to normal control subjects, although
the mean blood flow was similar in both groups. Such
oscillations may limit vasoocclusion and also create
"bolus flow," which may dislodge the blockade by sickled cells while the cells are still viable. On the other
hand, our timing of study may be incorrect. Additional
studies of endogenous CO production in and out of
crises would be of great interest.
We have shown that repeated measurements of endogenous CO production are feasible in patients with
sickle cell disease both during steady clinical state as well
as during pain crisis. The CO technique seems well
suited for the study of hemolysis in sickle cell disease
and thus sickling in vivo under a variety of circumstances. However, additional studies on a larger number
of patients during steady clinical state as well as pain
crisis are needed to confirm the preliminary results of
our study. Also, concurrent studies of RCLS using
DF32P, should this agent become available again in the
future, would be necessary to determine to what extent,
if any, the variability in the heme catabolism in individual patients by the CO technique is accounted for by
changes in the catabolism of heme from sources other
than the circulating red blood cells.
References
RCLS (days)
No Crisis
Crisis
No Crisis
Crisis
3.744*
4.686
3.759
5.404
25.5*
19.8
24.6
14.5
•Mean of two studies.
A.J.C.P. •February 1988
1. Alter BP, Kan YW, Nathan DG. Reticulocyte survival in sickle
cell anemia: effect of cyanate. Blood 1972;40:733-739.
2. Bensinger TA, Maiseles MJ, Mahmood L, McCurdy PR, Conrad
ME. Effect of intravenous urea in invert sugar on heme catabolism in sickle cell anemia. N Engl J Med 1971;285:995-997.
3. Bensinger TA, Mahmood L, Conrad ME, McCurdy PR. The effect
Vol. 89 • No. 2
4.
5.
6.
7.
8.
9.
10.
11.
225
BRIEF SCIENTIFIC REPORTS
of oral urea administration on red cell survival in sickle cell
disease. Am J Med Sci 1972;264:283-287.
Berk PD, Blaschke TF, Scharschmidt BF, Waggoner JG, Berlin
NI. A new approach to quantitation of the various sources of
bilirubin in man. J Lab Clin Med 1976;87:767-780.
Bertles JF, Dobler J. Reversible and irreversible sickling: a distinction by electron microscopy. Blood 1969;33:884-898.
Cline MJ, Berlin NI. An evaluation of DF32P and 5,Cr as methods
of measuring red cell lifespan in man. Blood 1963;22:459-465.
Coburn RF, Williams WJ, Forster RE. Effect of erythrocyte destruction on carbon monoxide production in man. J Clin Invest 1964;43:1098-1103.
Coburn RF, Williams WF, Kahn SB. Endogenous carbon monoxide production in patients with hemolytic anemia. J Clin
Invest 1966;45:460-468.
Collison HA, Rodkey FL, O'Neal JD. Determination of carbon
monoxide in blood by gas chromatography. Clin Chem
1968;14:162-171.
Coltman CA Jr, Dudley GM III. The relationship between endogenous carbon monoxide production and total heme mass in
normal and abnormal subjects. Am J Med Sci 1969,238:374385.
Gillette PN, Manning JM, Cerami A. Increased survival of sickle
cell erythrocytes after treatment in vitro with sodium cyanate.
Proc Natl Acad Sci USA 1971;68:2791-2793.
12. Kambam JR, Chen LH, Hyman SA. Effect of short-term smoking
halt on carbonxyhemoglobin levels and Pj 0 values. Anesth
Analg 1986;65:1186-1188.
13. May A, Bellingham AJ, Huehns ER, Beaven GH. Effect of cyanate
on sickling. Lancet 1972;1:658-661.
14. McCurdy PR. DF32P and 51Cr for measurement of red cell life
span in abnormal hemoglobin syndromes. Blood
1969;33:214-222.
15. McCurdy PR, Sherman AS. Irreversibly sickled cells and red cell
survival in sickle cell anemia: a study with both DF32P and "Cr.
Am J Med 1978;64:253-258.
16. Milner PF, Charache S. Life span of carbamylated red cells in
sickle cell anemia. J Clin Invest 1973;52:3161 -3171.
17. Padilla F, Bromberg PA, Jensen WN. The sickle-unsickle cycle: a
cause of cell fragmentation leading to permanently deformed
cells. Blood 1973;41:653-660.
18. Pearson HA, Noyes WD. Failure of phenothiazimes in sickle cell
anemia. JAMA 1967;199:91-92.
19. Rodgers GP, Schechter AN, Noguchi CT, Klein HG, Nienhuis
AW, Bonner RF. Periodic microcirculatory flow in patients
with sickle cell disease. N Engl J Med 1984;311:1534-1538.
20. Serjeant GR, Serjeant BF, Milner PF. The irreversibly sickled cell:
a determinant of hemolysis in sickle cell anemia. Br J Haematol
1969;17:527-533.
Neutrophil Lactoferrin Content in Viral Infections
ROY D. BAYNES, M.MED, WERNER R. BEZWODA, PH.D., AND NAZMA MANSOOR, B.Sc.
In an attempt to elucidate the previously observed decrease in
plasma lactoferrin-neutrophil ratio in subjects with acute viral
infections, a study of the neutrophil lactoferrin content in such
infections was undertaken. With the use of an immunoperoxidase stain for lactoferrin, neutrophils in viral illness were
shown to have reduced lactoferrin content (mean score 97.9
± SD (38.0] per 100 neutrophils) as compared with normal
subjects (mean score 196.4 ± SD [3.6] per 100 neutrophils) (t
= 7.69; P < 0.0005). This suggests an acquired defect of neutrophil lactoferrin synthesis in viral infection. Thisfindingis of
importance when seen against the well-recognized increased
risk of bacterial superinfection in subjects who have recently
had a viral infection. (Key words: Neutrophil; Lactoferrin;
Viral illness) Am J Clin Pathol 1988;89:225-228
DATA PREVIOUSLY REPORTED from this laboratory have demonstrated that the plasma lactoferrin concentration is reduced in subjects with acute viral illnesses.2 The reduced plasma concentration of lactofer-
Department of Hematology and Oncology and Department of
Medicine, University of the Witwatersrand Medical School,
Johannesburg, South Africa
rin in patients with viral infection occurred despite the
fact that total neutrophil counts in these subjects were
comparable to those of normal persons. There was thus
a significant reduction in the lactoferrin-neutrophil
ratio in subjects with viral illness. It was speculated that
the mechanisms responsible might include an acquired
defect of lactoferrin synthesis by neutrophils, defective
or inhibited degranulation, or increased lactoferrin
clearance. The current study was undertaken in an attempt to clarify the answer to this question.
Subjects and Methods
Received March 11, 1987; received revised manuscript and accepted
for publication June 8, 1987.
Supported by a grant from the Medical Research Council of South
Africa.
Address reprint requests to Dr. Bezwoda: Department of Medicine,
University of the Witwatersrand Medical School, York Road, Parktown, 2193, Johannesburg, South Africa.
Twenty-six young adults with viral illnesses were studied together with nine normal young adults. The viral
diagnoses were based on clinical presentation and positive serologic test results. They included chickenpox (11
subjects), measles (3 subjects), rubella (8 subjects), Ebstein-Barr virus infection (2 subjects), and 1 subject each