E u r J Hoeriiirtol 19Y3: 51: 125-131 C‘(1pWight 0 Murihguurd I Y Y 3

Prrrrred rrr Belgiurir - rill rights reserved EUROPEAN

JOURNAL OF HAEMATOLOGY
I S S N 0902-4441

Review article

Apparent polycythaemia: diagnosis,

pathogenesis and management
Messinezy M, Pearson TC. Apparent polycythaemia: diagnosis, Maria Messinezy and Thomas C. Pearson
pathogenesis and management. Division of Haematology, United Medical and
Eur J Haematol 1993: 51: 125-131. 0 Munksgaard 1993. Dental Schools, St Thomas’ Hospital, London
U.K.
Abstract: The term “apparent polycythaemia” is applied to a group of
patients who have a raised PCV (>0.51 in males, >0.48 in females) but a
normal red cell mass (less than 25% above their predicted mean normal
value). Some have additionally a marked reduction in plasma volume and
can be defined as a subgroup: relative polycythaemia. Smoking, hypertension
and to a lesser extent obesity, excessive alcohol, low-dose diuretic therapy
and hypoxaemia have all been associated with apparent polycythaemia but
the mechanism is both uncertain and likely to be complex. This group of
patients is unlikely to be uniform in pathogenesis and may well include some
normal individuals. Investigation requires exclusion of factors associated
with other types of polycythaemia. The possibility of an increased vascular Key words: apparent polycythaemia -
occlusive risk is uncertain in these patients except at the higher PCV venesection - vascular occlusion
values. Reduction of PCV by venesection is sensible at PCV> 0.54 or where Correspondence: Dr. M. Messinezy,
there is perceived to be an increased risk of vascular occlusion. The Department of Haernatology, St Thomas‘
remaining patients should be managed by regular observation to detect Hospital, Lambeth Palace Road, London, SEI
further rise in PCV or evolution to absolute polycythaemia (raised red cell 7EH, U.K.
mass). In some, the PCV returns to normal. Accepted for publication 22 March 1993

Introduction Measurement of PCV and RCM

About two-thirds of all patients referred for investi-
gation of polycythaemia (packed cell volume> 0.51
in males, > 0.48 in females) are found to have a red
cell mass (RCM) within the normal range (1); a
group now identified as having apparent poly-
cythaeniia. This term is generally agreed to have
replaced a number of rather nebulous terms used in
the past such as Geisbock’s syndrome, pseudopoly-
cythaemia, spurious or stress polycythaemia. These
were either not defined according to modern mea-
surements or implied an unsubstantiated aetiology.
From among these vague diagnostic labels, the term
apparent polycythaemia (with a subgroup of relative
polycythaemia having a significant reduction in
plasma volume) has been chosen to represent a con-
dition now more clearly defined by modern criteria.
Since the diagnosis of apparent polycythaemia de-
pends on accurate measurement of packed cell vol-
ume (PCV) and RCM, these aspects will be consid-
ered first.
PCV
The duration of venous occlusion significantly alters
PCV values therefore a diagnosis of polycythaemia
is only secure when the blood specimen is taken
easily with a minimum of venous occlusion and also
is confirmed by at least one further specimen taken
on a separate occasion. When red cell indices are
normal, electronic red cell counters give essentially
similar PCV values to the standard microhaemat-
ocrit method. But in the presence of iron deficiency
the PCV may be underestimated by some automated
counters which provide calculated results, and in
this situation as well as at initial evaluation the mi-
crohaematocrit method is recommended (2). The
normal upper limit of PCV (mean plus two standard
deviations) may be taken to be 0.51 for males and
0.48 for females (3,4); however, by definition, there
will be a small number of normal people who have
values above this level.

125
Messinezy and Pearson
Red cell mass and plasma volume (PV)
Standard methods for RCM and PV measurements
( 5 ) expressed these results in terms of ml/kg total
body weight. The error of this method of expression
in obese individuals has repeatedly been underlined.
This is because fat tissue, being relatively less vas-
cular, results in lower measured RCM expressed as
ml/kg for obese compared with lean individuals.
Similarly, if predicted normal values are only related
to body weight, inappropriately high predictions are
given for obese individuals. Expression of RCM in
relation to lean body mass, though ideal, is imprac-
tical because of difficulty in measuring the latter. At
present, results are best expressed in relation to pre-
dicted values for that patient’s height and weight.
Different formulae (6, 7) give satisfactory agreement
when used to predict RCM over a wide range of
height and weight combinations. The accepted nor-
mal range for these predictions, based on statistical
evaluation of the confidence limits, is 25% of the
mean value. Therefore the RCM in patients with
apparent polycythaemia will by definition be less
than 25% above their predicted mean normal value
(bearing in mind, however, that the precision of the
RCM measurement itself is about 5 % ) . Values
greater than 25% are termed an absolute increase
and occur in primary and secondary polycythaemia.
Plasma volume (PV) should no longer be ex-
pressed as ml/kg for the same reasons; however,
expression by comparison with predicted values de-
rived from height and weight formulae are less sat-
isfactory than for RCM because the results differ
according to the formula used. In our laboratory PV
is expressed according to the formula derived from
Nadler (7) and taking the lower limit of normal as
12.5 yo below the predicted mean normal value. This
admittedly somewhat arbitrary definition of low
plasma volume allows the separation of a subgroup
of patients from within the apparent polycythaemia
group; these patients with normal RCM and low
plasma volume are conveniently labelled “relative
polycythaemia”, though until the current difficulties
of defining a normal PV have been resolved the exact
limits of this sub-group remain indistinct and any
discussion of its pathogenesis less certain. Relative
polycythaemia defined in this way, was diagnosed in
only 18 % of male patients with PCV values between
0.500 and 0.599 (1).
Why is the PCV raised?
The concept of a raised PCV when RCM and PV are
normal is at first sight confusing, but further inspec-
tion reveals possible mechanisms. Given the diffi-
culties of measurement and of defining the normal
range, it is not surprising that some patients with a
126
RCM near the upper limit of normal would have a
raised PCV, in particular if this RCM is combined
with a PV in the lower part of the normal range
(possibly a consequence of homeostatic mechanisms
to maintain a normal blood volume). Indeed a num-
ber of “normal” subjects may demonstrate this situ-
ation, their raised PCV being at the extreme end of
the normal distribution. With increasing PCV Val-
ues, the likelihood of these mechanism being the
explanation becomes progressively less (1). In other
subjects with raised PCV there may be pathological
processes which tend to both increase the RCM
towards the upper end of the normal range and re-
duce the PV; or pathological states which eventually
would lead to an absolute polycythaemia (primary or
secondary polycythaemia) but are at an early stage,
not having yet resulted in a RCM greater than 25%
above the mean predicted value.
Apparent polycythaemia need not therefore repre-
sent a uniform group of patients where the underly-
ing mechanism for producing a raised PCV is iden-
tical in all. It could be seen as a haematological
syndrome rather than a disease. A few “patients”
may be normal subjects, others may be in an early
phase of progression to absolute polycythaemia while
in others reduction in plasma volume, perhaps com-
bined with some increase in RCM above the patient’s
own normal value but still within the normal range,
is the main mechanism. These possibilities need to
be taken into account when planning investigation
and management.
Clinical associations
The heterogeneity of patients with apparent poly-
cythaemia adds to the problem of identifying under-
lying causes, but a number of studies have attempted
to define this group of patients and have demon-
strated some clinical associations which may throw
light on the aetiology. The sub-group of relative poly-
cythaemia has not always been identified in these
studies and the value, if any, of maintaining this
subdivision is as yet uncertain.
Age and sex
Male predominance is marked. This could be partly
due to the high incidence of minor degrees of iron
deficiency in females, as well as a tendency for phy-
sicians to focus from habit on low Hb and also to
neglect the fact that the normal upper limit of PCV
for females is lower than for males. The mean age at
diagnosis varies but tends to be lower than in pri-
mary polycythaemia, consistent with its generally
non-neoplastic origin and, with the incidence of
clinically associated factors (see below) being wide-
spread throughout middle age. Certainly apparent

polycythaemia is commoner than primary poly-
cythaemia among younger adults.
Obesity
Earlier publications (8) using ml/kg RCM and PV
expressions have noted a higher incidence of obesity
in apparent polycythaemia though both definition
and the controls used are debatable. Studies using
RCM expressed as ml/kg are in any case suspect
(see above). A recent study (9) which defined over-
weight as body mass index (weight in kg/height2 in
m) higher than 25, did not confirm this finding,
though certain mechanisms (association with hy-
poventilation and hypoxia at night, or with hyper-
tension which is commoner in the obese) could be
relevant and provide an explanation for a raised
PCV.
Hypertension and cardiovascular disease
An association of hypertension with apparent poly-
cythaemia has been frequently reported even in pa-
tients not taking diuretics (9-15). Indeed, studies of
hypertensive populations have shown slightly higher
PCV than normotensive groups though the PCV is
still within the normal range in the majority (16). The
fact that both renal disease and sleep apnoea are
commoner in hypertension (17) could suggest an
erythropoietin-mediated mechanism. There is also
the possibility of reduced plasma volume though this
has been described only in some hypertensive pa-
tients (18). Increased venous tone could be the
mechanism (19-21). A reduced body:venous hae-
matocrit ratio suggesting a change in red cell distri-
bution may be relevant and has been described in
some hypertensive patients (13). A number of stud-
ies of cardiovascular disease have noted an increased
incidence in apparent polycythaemia not necessarily
associated with hypertension (10, 12, 13) but the
control groups have not been ideal and the mecha-
nisms not identified.
Smoking
A recent study (9) noted that 60% of patients with
apparent polycythaemia were smokers, compared
with 30% in a control group of similar age in the
U.K. This association has frequently been noted
elsewhere and is sufficiently strong to suggest a di-
rect causative link most obviously by reduction of
the arterial oxygen content via an increase in carbon
monoxide (22). The latter will additionally shift the
oxygen dissociation curve to the left, further reduc-
ing oxygen delivery to the tissues (and presumably
raising the erythropoietin level). In some patients
smoking may be linked to lung disease (and absolute
Apparent polycythaemia
polycythaemia in severe cases), obesity, hyperten-
sion, high alcohol intake or sleep apnoea, thus en-
tering the constellation of factors thought to be as-
sociated with apparent polycythaemia. Studies of
groups of smokers have shown a similar small rise
in PCV as found in hypertensives (23). Though this
is only occasionally enough to raise the PCV above
the upper limit of normal (24) it may be particularly
important if there is also some reduction of plasma
volume. Both raised venous tone (25) and possibly
an increase in capillary permeability (26) have been
reported and could explain reduction of plasma vol-
ume in some smokers.
Arterial oxygen desaturation is classically linked to
absolute polycythaemia (RCM above the normal
range) via activation of the renal sensor and increase
in erythropoietin. A recent prospective study (27)
has shown that 4 of 16 patients with apparent poly-
cythaemia had either nocturnal oxygen desaturation
or daytime desaturation when the supine position
( <92%). One may conclude that mild degrees of
hypoxaemia insufficient to cause secondary poly-
cythaemia (absolute increase in RCM) may never-
theless cause smaller increases in RCM still within
the normal range and hence apparent polycythaemia
in some cases. Two of the 4 patients showed the
decreased plasma volume of relative polycythaemia,
and lesser degrees of plasma volume reduction could
also be a contributing factor in some mildly hy-
poxaemic patients with apparent polycythaemia. In-
creased venous tone has been demonstrated in hy-
poxia (28).
Alcohol
An association of apparent polycythaemia with high
alcohol intake is only tenuous in published series
(9, 12) though individual cases which improved with
cessation of alcohol have been described (29,30).
Suggested though unproven mechanisms are by di-
uresis causing a decrease in P V or via mild stimu-
lation of erythropoiesis. The latter could occur by
increasing nocturnal oxygen desaturation in at-risk
individuals (31, 32) or via a hepatic erythropoietin
mechanism secondary to liver damage (33). In-
creased alcohol intake is itself positively associated
with obesity, hypertension and smoking.
Renal disease
Renal disease is well recognized as a cause of sec-
ondary (absolute) polycythaemia via the erythropoi-
etin mechanism, and it is therefore to be expected
that lesser degrees of increase in red cell mass (still
maintained within the normal range) may occur in
some situations (34). Patients with raised PCV Val-
ues following renal transplantation sometimes have
127
Messinezy and Pearson
a reduced PV rather than the expected increase in
RCM (35-37) and impairment of renal function
could be expected to challenge PV homeostasis.
Stress
Stress has historically (8) been linked with apparent
polycythaemia; the suggested mechanism being sym-
pathetic nervous stimulation with increased cate-
cholamine excretion. Adrenaline and noradrenaline
infusions have been noted to lead to an increase in
venous tone with mild reductions in PV and small
increases in PCV (38, 39), and in occasional patients
with phaeochromocytoma raised PCV may be found
in association with reduced PV (40,41). While men-
tal and physical stress does stimulate the autonomic
nervous system and increases catecholamine levels,
the evidence for an association with polycythaemia
is tenuous (42-46).
Diuretic therapy and fluid depletion
Transient apparent polycythaemia associated with
acute fluid depletion as in gastrointestinal loss, sep-
ticaemia (capillary leakage), severe induced diuresis
and reduced fluid intake is well recognized and eas-
ily understandable. Chronic low-dose diuretic thera-
py led to a small (5 %) reduction in PV in one series
(47) which could be enough to raise PCV levels at
the upper limit of normal into the abnormal range.
Summary of clinical associations
Smoking and hypertension probably have a stronger
association with apparent polycythaemia than obe-
sity, excessive alcohol or low dose diuretic therapy.
A recent series of 34 patients from our department
(9) showed that only 9% of the patients were nega-
tive for all five factors and that no factor was par-
ticularly associated with the relative polycythaemia
subgroup. The occasional association with hy-
poxaemia particularly at night has also been noted.
It is likely that at least in some patients the associa-
tions are cause and effect, and also that more than
one causative factor may operate in a single patient.
The fact that these factors have a tendency to asso-
ciate with each other as well as with polycythaemia
makes elucidation of causes still more complex.
Mechanisms believed to control plasma volume are
both numerous, complicated, and interactive. De-
tailed studies of individual patients would be re-
quired to establish the relative contribution of
changes in the different components of P V control
(renin/angiotensin/aldosterone axis, atrial natriuretic
peptide, antidiuretic hormone and vascular. tone,
particularly venous). The fact that some patients with
apparent polycythaemia are actually normal but at
128
the extreme of the PCV range, that others are on the
way to developing primary or secondary poly-
cythaemia and that the normal P V value is as yet an
uncertain quantity makes the question of pathogen-
esis in apparent polycythaemia still more complex.
Investigation
Once the diagnosis of apparent polycythaemia has
been established by PCV and blood volume studies,
further investigation should be directed at looking
for the possible associated factors described above,
including arterial oxygen saturation by day (and by
night if there are borderline levels when supine or
any suspicious symptoms of sleep apnoea such as
excessive snoring, or diurnal somnolence), renal ul-
trasound and liver function. In case the patient has
an early primary polycythaemia, splenic sizing and
erythropoietin levels would be theoretically useful
though perhaps not cost effective in most patients.
Vascular occlusive risk
The critical question in managing apparent poly-
cythaemia is whether the raised PCV matters and
this is not easy to answer. One would certainly ex-
pect the risk of occlusive vascular lesions to be
smaller than in primary polycythaemia because there
are no quantitative or qualitative platelet changes.
The only prospective randomized study to see
whether a reduction in PCV leads to a reduction in
vascular occlusive events in apparent polycythaemia
was initiated in the UK in 1985 but had to be aban-
doned after a few years because of inability to col-
lect the large number of patient entries required. Of
the two retrospective studies of vascular occlusion in
apparent polycythaemia one (1 1) showed a mortal-
ity six times greater than expected for a sex- and
age-matched population, but the number of patients
and length of follow-up were small with only 4 deaths
due to vascular occlusion and 2 other thrombotic
episodes. The second study (12) involved 47 pa-
tients, a few of whom were followed for up to 12
years, but the authors concluded that the major risk
factor for death was hypertension rather than the
raised PCV.
There are, however, other observations which sug-
gest, though indirectly, that a raised PCV in itself
predisposes to vascular occlusion. Firstly, studies in
patients with vascular occlusion (48,49) and experi-
ments using carotid artery occlusion in animals
(50-52) have shown larger areas of local ischaemia
at high compared with normal PCV values. Sec-
ondly, cerebral blood flow measurements in appar-
ent polycythaemia (53) (as well as in primary poly-
cythaemia (54)) have shown an inverse correlation
with PCV; the supposition being that reduced blood

flow predisposes to thrombosis. It has been sug-
gested that changes in cerebral blood flow with PCV
are purely a homeostatic attempt to maintain con-
stant cerebral oxygen transport. However, a study of
cerebral oxygen transport in 20 patients with raised
PCV and normal arterial oxygen saturation showed
an increase in oxygen transport in half of the patients
following venesection to reduce the PCV to normal
(55). Thirdly, a fall in thrombotic complications after
PCV reduction has been reported in idiopathic eryth-
rocytosis (56, 57) where platelet function should be
normal. Similarly patients with secondary poly-
cythaemia post renal transplant have an increased
risk of thrombosis (35). Finally, post-mortem stud-
ies in “normal” populations have shown more my-
ocardial and cerebral thromboses at high compared
with low normal PCV values (58, 59).
Perhaps the strongest argument for reducing the
PCV in apparent polycythaemia nevertheless comes
from studies in primary polycythaemia. Venesection
alone should not alter the abnormal platelet clone in
the latter, yet just lowering the PCV has been clearly
shown to reduce the incidence of vascular occlusion
which is otherwise manifestly high in primary poly-
cythaemia (60-62). On the other hand, the possibil-
ity that raised PCV and abnormal platelets act syn-
ergistically to increase the risk of thrombosis would
reduce the force of this argument, particularly at the
less extreme PCV increases most commonly seen in
apparent polycythaemia.
Management
The first stage of management, having identified any
factor believed to be associated with apparent poly-
cythaemia, is to reverse it where possible. Patients
should lose excessive weight, stop smoking, reduce
high alcohol intake, and have hypertension treated
(preferably not with a strongly diuretic regime). All
these manoeuvres have been reported to be effective
in some patients and would in any case be recom-
mended medical practice irrespective of the presence
of apparent polycythaemia. Our recent study (9)
showed that after 3 months of such management
(where the change of life style intentions were by no
means always achieved), the PCV of one-third of the
group fell into the normal range, one-third had only
intermittently raised values, and the remainder con-
tinued to have a raised PCV. Undoubtedly some
patients reverted to a stable normal PCV state with-
out any particular alteration in factors that have been
associated with apparent polycythaemia. This may
be purely a reflection of our limited knowledge of
causation, but it does nevertheless underline the
point of delaying specific treatment aimed at lower-
ing the PCV in most of these patients. Exceptions
indicating the need for early intervention to lower the
Apparent polycythaemia
PCV are rare but might include very high PCV val-
ues and sudden ischaemic symptoms or thrombotic
events.
The small group of patients who continue to have
a raised PCV need further consideration, bearing in
mind also that they could be progressing towards a
different form of polycythaemia and therefore in time
may need a repeat of earlier investigations. When the
PCV is consistently greater than 0.54 one would not
hesitate to recommend reduction. This is a some-
what arbitrary figure but it does represent a level
which is very unlikely to include any normal people,
and also one where by analogy with primary poly-
cythaemia the risk of vascular occlusion becomes
impossible to ignore. For patients with raised PCV,
but less than 0.54, the decision whether to lower the
PCV depends on the individual practitioner’s assess-
ment of the thrombotic risk. In patients with is-
chaemic symptoms, with a past history of occlusive
vascular events, or with other factors likely to in-
crease their thrombotic risk (hypertension, diabetes,
hypercholesterolaemia or strong family history), re-
duction of the PCV would seem the most reasonable
option.
Venesection remains the standard method when a
decision to lower the PCV has been taken. Dextran
infusions have been used to increase plasma volume
in relative polycythaemia but the effect is too brief to
be practical as a long-term measure, while low-dose
fludrocortisone has been ineffective (63). Blood vol-
ume studies on chronically venesected patients are
generally reassuring in demonstrating that the reduc-
tion in PCV is associated with a reduction in red cell
mass and an increase in PV without change in total
blood volume (64). It is, however, difficult to under-
stand how venesection could achieve a reversal of
the original PV-reducing mechanism. Perhaps the
mechanism is still active but stronger homeostatic
changes are brought into play to maintain a constant
blood volume by increasing the plasma volume.
Venesection of 450 ml of blood without volume
replacement is the standard procedure in all patients
apart from those with current ischaemic symptoms
or a recent history of thrombosis. In these, initial
slow venesections of 250-300 ml as well as concur-
rent saline replacement to maintain the blood vol-
ume, at least during the first few venesections, are
reassuring to the doctor. It is rare for any clinical
problem to develop even in patients with a history of
ischaemic heart disease and/or receiving treatment
for hypertension (65).
The target PCV value for the venesected patient is,
not surprisingly, uncertain. Recommendations are
based at present on the arguments that were used to
propose a target PCV of 0.45 or below in primary
polycythaemia, namely the incidence of vascular oc-
clusion found at post-mortem in the normal popu-
129
Messinezy and Pearson

lation, the incidence of occlusive vascular lesions at 12. WEINREBNJ, SHIH C-F. Spurious polycythaemia. Semin
Haematol 1975: 12: 397-407.
different PCV values in primary polycythaemia and 13. WATTEJ, LEWISSM. Spurious polycythaemia - a study of
the cerebral blood flow findings in patients with 35 patients. Scand J Haematol 1983: 31: 241-247.
raised PCV values (66). The fact that the PCV level 14. RUSSELL RP, CONLEYCL. Benign polycythaemia: Geisb-
proposed is well below the upper limit of normal ock‘s syndrome. Arch Int Med 1964: 114: 734-740.
means that patients entered into a venesection pro- 15. DAVIES SW, GLYNNE-JONES E, LEWISEP. Red face and
reduced plasma volume. J Clin Pathol 1974: 27: 109-112.
gramme might expect to go on requiring venesection 16. TIBBLING G, BERGENTZ S-E, BJUREJ , WILHELMSEN L.
for the rest of their lives. Because of the size of this Haematocrit, plasma protein, plasma volume and viscosity in
commitment as well as the knowledge that a pro- early hypertensive disease. Am Heart J 1966: 72: 165-176.
portion of patients with apparent polycythaemia re- 17. KALESA, CADIEUX RJ, SHAWLC, et al. Sleep apnoea in
hypertensive population. Lancet 1984: (ii): 1005-1008.
turn to a normal PCV level spontaneously with time, 18. KOBRINI, FROHLICH ED, VENTURA HO, et al. Stable red
the importance of clarifying the need for venesection cell mass despite contracted plasma volume in men with es-
in this condition becomes evident. sential hypertension. J Lab Clin Med 1984: 104: 11-14.
Meanwhile, unless there are clear indications to 19. OHLSON 0,HEMINGSEN NC, HOODB. Plasma volume and
the contrary, patients in whom a venesection policy plasma volume distribution at rest during muscular work,
cold pressure test and psychological stress in male offspring
has been undertaken should at a later date be reas- from families with heavy aggregation of hypertension. Acta
sessed and the possibility of stopping venesection Med Scand 1982: 212: 337-341.
should be considered and implemented if appropri- 20. BURKEMJ, STEKIEL WJ, LOMBARD JH. Reduced veno-
ate until there is evidence once more that the PCV constrictor reserve in spontaneously hypertensive rats sub-
jected to haemorrhage stress. Circ Shock 1984: 14: 25-37.
is rising into the polycythaemia range. Patients not 21. LONDON GM, SAFARME, WEISSYA, SIMONCHA. Total
being venesected should continue under surveillance effective compliance of the vascular bed in essential hyper-
for evidence of return of the PCV to normality, of tension. Am Heart J 1978: 95: 325-330.
further rise in PCV level which may presage a change 22. DOLL DC, GREENBERG BR. Cerebral thrombosis in
in diagnosis, or of clinical developments that could smokers’ polycythemia. Ann Int Med 1985: 102: 786-787.
23. ISACERH, HACERUPL. Relationship between cigarette
lead to a decision to venesect. smoking and high packed cell volume and haemoglobin lev-
els. Scand J Haematol 1971: 8: 241-244.
24. SPIERSASD, LEVINEM. Smokers’ polycythaemia. Lancet
Acknowledgement 1983: (i): 120.
25. VELASQUEZ MT, SCHECTER GP, MCFARLAND W, COHN
We are grateful to Monica Nestor for typing the manuscript.
JN. Relative polycythaemia: A state of high venous tone. Clin
Res 1974: 22: 409.
26. SIGGAARD-ANDERSON J, PETERSENFB, HANSENTI,
References MELLEMGAARD K. Plasma volume and vascular permeabil-
ity during hypoxia and carbon monoxide exposure. Scand J
1. PEARSON TC, BOTTERILL CA, GLASSUH, WETHERLEY- Clin Lab Invest (Supplement) 1968: 103: 39-48.
MEING. Interpretation ofmeasured red cell mass and plasma 27. MESSINEZY M, AUBRYS, O’CONNELL G, TREACHER DF,
volume in males with elevated PCV values. Scand J Haema- PEARSON TC. Oxygen desaturation in apparent and relative
to1 1984: 33: 68-74. polycythaemia. Br Med J 1991: 302: 216-217.
2. GUTHRIE DL, PEARSON TC. PCV measurement in the man- 28. ECKSTEIN JW, HORSLEY AW. Effects of hypoxia on periph-
agement of polycythaemia patients. Clin Lab Haematol 1982: eral venous tone in man. J Lab Clin Invest 1960: 56: 847-853.
4: 257-265. 29. SMITHJFB, LUCIEN P . Alcohol - a cause of stress eryth-
3. PEARSON TC. Clinical and laboratory studies in the poly- rocytosis. Lancet 1973: (i): 637.
cythaemias. M D Thesis 1977 (London University) 30. O’BRIENH, ELLIOTT PJ, AMESSJAL. Alcohol and relative
4. HALLR, MALIARG. Basic haematological practice. In: polycythaemia. Lancet 1981: (ii): 987.
Medical Laboratory Haematology. Oxford: Butterworth- 3 1. ISSAFG, SULLIVAN CE. Alcohol, snoring and sleep apnoea.
Heinemann Ltd., 1991: 84-112. J Neurol Neurosurg Psychiatry 1982: 45: 353-359.
5. International Committee for Standardization in Haematol- 32. MOORE-GILLON J, PEARSON TC. Smoking, drinking and
ogy. Recommended methods for measurement of red cell and polycythaemia. Br Med J 1986: 292: 1617-1618.
plasma volume. J Nucl Med 1980: 21: 793-800. 33. MIRANDEA, MURPHYGP. Erythropoietin alterations in
6. HURLEY PJ. Red cell and plasma volumes in normal adults. human liver disease. N Y State J Med 1971: 71: 860-864.
Br J Nuclear Medicine 1975: 16: 46-52. 34. HOPPINEC, DEPNERT, YAMUCHI H, HOPPERJ. Eryth-
7. NADLER SB, HIDALGO JU, BLOCHT. Prediction of blood rocytosis associated with diffuse parenchymal lesions of the
volume in normal human adults. Surgery 1962: 51: 224-232. kidney. Br J Haematol 1976: 32: 557-563.
8. LAWRENCE JH, BERLINNI. Relative polycythaemia - the 35. WICKRECG, NORMALDJ, BENNISON A, BARRYJM,
polycythaemia of stress. Yale J Biol Med 1952: 24: 498-505. BENNETTWM. Post renal transplant erythrocytosis: A re-
9. MESSINEZY M, PEARSON TC. A retrospective study of ap- view of 53 patients. Kidney Int 1983: 23: 731-737.
parent and relative polycythaemia: associated factors and 36. DAVISHP. Polycythaemia following renal transplantation. J
early outcome. Clin Lab Haematol 1990: 12: 121-129. R Soc Med 1987: 80: 475-476.
10. BROWNSM, GILBERT HS, KRAUSS S, WASSERMAN LR. 37. OBERMILLER LE, TZAMALOUKAS AH, AVASTHI PS, HALP-
Spurious (relative) polycythaemia. A non existent disease. ERN JA, STERLING WA. Decreased plasma volume in post-
Am J Med 1971: 50: 200-207. transplant erythrocytosis. Clin Nephrol 1985: 23: 213-
11. BURGEPS, JOHNSON WS, PRANKERD TAJ. Morbidity and 217.
mortality in pseudopolycythaemia. Lancet 1975: (i): 1266- 38. COHNJN. Relationship of plasma volume changes to resis-
1269.

130
 Apparent polycythaemia
tance and capacitance vessel effects of sympathomimetic DOSHIMA S, OMAET. Effects of haematocrit on brain meta-
amines and angiotensin in man. Clin Sci 1966: 30: 267-278. bolism in experimental induced cerebral ischaemia in spon-
39. HILSTEDJ, CHRISTENSEN NJ, LARSENS. Effect of cate- taneously hypertensive rats. Stroke 1985: 16: 835-840.
cholaniines and insulin on plasma volume and intravascular 53. HUMPHREY PRD, DUBOULAY GH, MARSHALL J, etal.
mass of albumin in man. Clin Sci 1989: 77: 149-155. Cerebral blood flow and viscosity in relative polycythaemia.
40. SJOERDSMA A, ENGELMAN K, WALDMANN TA, COOPER- Lancet 1979: (ii): 873-877.
MAN LH, HAMMOND WG. Phaeochromocytoma: current 54. THOMAS DJ, DUBOULAY GH, MARSHALL J, et al. Cerebral
concepts in diagnosis and treatment. Ann Intern Med 1966: blood flow in polycythaemia. Lancet 1977: (ii): 161-163.
65: 1302-1325. 5 5 . WADEJPH. Transport of oxygen to the brain in patients with
41. BATTLE JD, ALFIDIRJ, STRAFFON FA. Polycythaemia sec- elevated haematocrit values before and after venesection.
ondary to phaeochromocytoma. Report of a case. Cleveland Brain 1983: 106: 513-523.
Clinics Quarterly 1971: 38: 121-124. 56. NAJEANY, TRIEBELF, DRESCHC. Pure erythrocytosis:
42. JERNC, WADENVIK H, MARKH, HALLGREN J, JERNS. Reappraisal o f a study of 51 cases. Am J Hematol 1981: 10:
Haenlatological changes during acute mental stress. Br J 129- 136.
Haematol 1989: 71: 153-156. 57. PEARSON TC, WETHERLEY-MEIN G. The course and com-
43. BENITONE J, KLINGA. Polycythaemia of stress in psychi- plications of idiopathic erythrocytosis. Clin Lab Haematol
atric hospital populations. J Psychosom Res 1970: 14: 105- 1979: 1: 189-196.
108. 58. RHOADS GG, BLACKWELDER WC, STEMMERMANN GN,
44. MATHEW R, WILSONWH. Hematocrit and anxiety. J Psy- et al. Coronary risk factors and autopsy findings in Japanese-
chosom Res 1986: 30: 307-311. American mer1. Lab Invest 1978: 38: 304-311.
45. SCHUBERT DSP, ROHIRAL. Polycythemia of stress in a 59. TOHGIH, YAMANOUCHI H, MURAKAMI M, KAMEYAMA
psychiatric hospital population: failure to replicate. J Psycho- M. Importance of the hematocrit as a risk factor in cerebral
som Res 1981: 25: 109-110. infarction. Stroke 1978: 9: 369-374.
46. WARDMM, MEFFORD IN, PARKER SD, et al. Epinephrine 60. CHIEVITZ E, 'THIEDE T. Complications and causes of death
and norepinephrine responses in continuously collected in polycythaernia. Acta Med Scand 1962: 171: 513-523.
human plasma to a series of stressors. Psychosom Med 1983: 61. PEARSONTC, WETHERLEY-MEIN G. Vascular occlusive
45: 471-486. episodes and venous haematocrit in primary proliferative
47. TARAZIRC, DUSTANHP, FROHLICH ED. Long-term thi- polycythaemia. Lancet 1978: (ii): 1219-1222.
azide therapy in essential hypertension. Circulation 1970: 6 1: 62. NAJEANY, MUGNIERP, DRESCHC, RAIN J-D. Poly-
709-717. cythaemia Vera in young people: an analysis of 58 cases diag-
48. HARRISON MJG, KENDALL BE, POLLOCK S, MARSHALL nosed before 40 years. Br J Haematol 1987: 67: 285-291.
J. Effect of haematocrit on carotid stenosis and cerebral in- 63. HUMPHREY PRD, MICHAEL J, PEARSON TC. Management
farction. Lancet 1981: (ii): 114-115. of relative polycythaemia: studies of cerebral blood flow and
49. RING CP, PEARSONTC, SANDERS MD, WETHERLEY- viscosity. Br .lHaematol 1980: 46: 427-433.
MEING. Viscosity and retinal vein thrombosis. Br J Oph- 64. HUMPHREY PRD, MICHAELJ, PEARSON TC. Red cell
thalmol 1976: 60: 397-410. mass, plasma volume and blood volume before and after
50. SUNDTTM, WALTZAG, SAYREGP. Experimental cerebral venesection in relative polycythaemia. Br J Haematol 1980:
infarction: modification by treatment with haemodiluting, 46: 435-438.
haemoconcentrating and dehydrating agents. J Neurosurg 65. THOMAS SHL, WESTON-SMITH SG, SLATER NGP, PEAR-
1967: 26: 46-56. SON TC, TREACHER DF. The haemodynamic responses to
51, POLLOCK S, TSITSOPOULOS P, HARRISON MJG. The effect venesection and the effects of cardiovascular disease. Clin
of haematocrit on cerebral perfusion and clinical status fol- Lab Haematol 1992: 14: 201-208.
lowing carotid occlusion in the gerbil. Stroke 1982: 13: 167- 66. THOMASDJ, DUBOULAY GH, MARSHALL J, etal. The
170. effect of haematocrit on cerebral blood flow in man. Lancet
52. KIYOARA Y, FUJISHIMA M, ISHITSUKA T, TAMAKI K, SA- 1977: (ii): 941-943.

131