A clinical analysis of erythrocytapheresis for the treatment of

polycythemia
Huasheng Liu
a
, Haibo Liu
a
, Jingzhi Shen
b
, Chunhong Sun
a
, Caili Guo
a
, Huan Lin
a
,
Jianna Yao
a
, Rui Ma
a
, Mei Zhang
a,
a
Department of Hematology, The First Afliated Hospital of Xian Jiaotong University, Xian 710061, China
b
Department of Hematology, The First Afliated Hospital of Dalian Medical University, China
a r t i c l e i n f o
Article history:
Received 15 May 2012
Received in revised form 31 October 2012
Accepted 10 January 2013
Keywords:
Erythrocytapheresis
ECP
Polycythemia
Hematocrit
Hemoglobin
a b s t r a c t
Purpose: To evaluate the efcacy and safety of erythrocytap heresis (ECP) in the treatment
of polycythemia.
Methods: Patients diagn osed with polycythemia were included in this retrospective anal-
ysis and treated with ECP (n = 20) or conventional treatments (exsanguination; n = 20).
Blood laboratory values and adverse effects were recorded.
Results: In ECP-treated patients mean red blood cell (RBC) collection time was
25.7 4.5 min (range: 1937 min), with a mean collection volume of 773.5 129.3 mL
(range: 6001002 mL). From baseline, ECP reduced the mean number of RBCs
(0.6 10
12
/L [7.6%]), mean hemoglobin (31.1 g/L [14.8%]), and mean hematocrit (13.1%
[20.2%]) (P < 0.001 for each). After ECP, a marked reduction in symptoms associated wit h
polycythemia was also observed.
Conclusions: Treatment of patients with polycythemia using ECP reduces RBC count, hemo-
globin, and hematocrit. The advantages associated with ECP over conventional therapy
should be considered when choosing a treatment plan for patients with polycythemia.
2013 Published by Elsevier Ltd.
1. Introduction
Polycythem ia is a disease state characteri zed by an in-
crease in the proportion of blood volume that is occupied
by red blood cells [1]. Primary polycythem ia, also known
as polycythemi a vera (PV), is frequent ly found in patients
aged 5060 years and has been characterized as a clonal
neoplastic disorder. Primary polycythemi a occurs when
excess red blood cells are produced as a result of an abnor-
mality of the bone marrow [1]. Excess white blood cell and
platelets can also be produced in primary polycythemi a.
Secondary polycythem ia is caused by either natural or arti-
cial increases in the production of erythropoi etin, hence
an increased production of erythrocytes. Although primary
polycythemi a is different from secondary polycythemi a in
etiology, they share common clinical manifestations such
as thrombos is and bleeding secondary to increases in
blood viscosity.
Traditional treatments for polycythemia include venous
exsanguinati on and pharmacotherap y with aspirin, bone
marrow suppressant s (i.e., homoharrin gtonine, cyclophos-
phamide, and melphalan), radioactive
32
P, hydroxyu rea,
and interferon [2]. However, venous exsanguinati ons can-
not inhibit the proliferation of red blood cells and reduce
the white blood cell count and platelet count. Exsanguina-
tion is an initial step in the treatment of PV because exsan-
guination can rapidly reduce RBC count, total blood
volume, and improve symptoms [3]. Exsanguinati ons are
commonly performed 12 times per week. However, the
loss of antithrombi n as a result of exsanguinations may
cause a hypercoagul ative state, and exsanguinations
1473-0502/$ - see front matter 2013 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.transci.2013.01.011

Corresponding author. Tel: +86 029 85324035; fax: +86 029

85252580.
E-mail address: zhangmei@medmail.com.cn (M. Zhang).
Transfusion and Apheresis Science 48 (2013) 229233
Contents lists availabl e at SciVerse ScienceDi rect
Transfus ion and Apher esis Science
j our nal homepage: www. el sevi er. com/ l ocat e/ t ransci
cannot inhibit the prolifera tion of RBCs and reduce white
blood cells and platelets. As a result, enhanced haemopoi-
esis may subsequently increase the RBCs and platelets,
increasing the risk for bleeding and thrombosis [4]. For
older patients or those with a history of thrombos is, re-
peated exsanguinati ons may increase the risk for thrombo-
sis, iron loss, and elevate the risk for bone marrow brosis
[2].
Erythrocyta pheresis (ECP) is an extracorporeal blood
separation method whereby whole blood is extracted from
a donor or patient, the red blood cells are separated, and
the remaining blood is returned to circulation [4]. Clinical
benets of ECP include decreasing hematoc rit level with-
out volume overload and the short duration of the proce-
dure. ECP can be used in patients with sickle cell disease
(acute chest syndrome, acute stroke, multiorgan failure),
stroke prophylaxis , and the prevention of transfusion ac-
quired iron overload.
The use of ECP as a therapeutic option has advantag es
over traditional methods, in particular bloodletting (or
phlebotomy ), which is considered the standard effective
therapy that prevents progression of tissue damage in
polycythemi a. It has been suggested that 45 exsangui n-
ations would be necessary to match the effectiven ess of a
single cycle of ECP. More specically, the efcacy of ECP
once (>1100 mL of circulating blood) is equivalent to that
of manual bloodletting (200300 mL in single bloodletting)
[4,5]. ECP could therefore have clinical and economic
benets.
We currently report the ndings of a retrospective anal-
ysis of 40 patients with polycythemia treated with or with-
out ECP. The goal of treatment was to reduce blood volume
and RBC count to a near normal level and to inhibit the
hematopoieti c function of bone marrow, both of which
may attenuate the symptoms and complications associated
with polycythemi a. The efcacy of ECP in combination
with pharmacotherap y in the treatment of polycythem ia
was also evaluated.
2. Design and methods
2.1. Patients
A total of 40 patients diagnosed with polycythemi a (pri-
mary [n = 20]; secondary [n = 20]) were included in the
current retrospective analysis. Patients were diagnose d
according to published guidelines [6]. Detection of the
JAK2 V617F mutation as outlined in the WHO (2008) crite-
ria for diagnosis of polycythemi a [7], was not utilized in
the current study due to limitations around available labo-
ratory techniques at the time of patient treatment in this
retrospective analysis. Patients received ECP (n = 20 [14
men and 6 women]) or conventional therapy (n = 20 [15
men and 5 women]) and were treated from December
2008 to June 2011. The rst 20 consecutive patients who
chose to receive ECP were included in the analysis. The
control group was age-, gender- and syndrome-mat ched
to the ECP treatment group. Treatment protocols were ap-
proved by the institutiona l review board of the rst afli-
ated hospital of Xian Jiaotong University.
2.2. Study treatment
ECP was performed using a COBE

Spectra Apheresi s
System 6.1 (CaridianBCT, CO, USA) and disposab le closed
tubes. Prior to ECP, patients were not required to be fasted
or anesthetized. All patients were supplemented with cal-
cium (oral or via intravenous [IV] injection). Following ECP,
normal saline or 6% hydroxye thyl starch or equal volume
was re-perfused . Patients with elevated hemoglobin levels
(>200 g/L) received multiple cycles of ECP separated by 1
2 days. Some patients required additional medication fol-
lowing ECP. All patients, both in the ECP-treated group
and control group received the same pharmacolo gical ther-
apy, such as administrat ion of interferon- a (3 10
6
U/m
2
administere d 23 times per week and adjusted based on
patient response and the presence of side effects) and
hydroxyure a (1025 mg/kg/d). Hydroxyurea in the blood
was routinely measured and the dose adjusted as needed.
For patients receiving initial treatment, hydroxyu rea in
combination with interfero n was administered. Mainte-
nance therapy was performed with interferon when blood
laboratory values returned to normal.
The patients in the control group underwent conven-
tional therapy, consisting of exsangui nation performed to
achieve a normal Hct (4045%) followed by administrat ion
of hydroxyure a. The dose of hydroxyurea was adjusted
according to each patients hemoglobin level and body
weight (1015 mg/kg/d administere d orally), respectively .
For patients resistant or non-respon sive to hydroxy-
urea, treatment with interferon- a was initiated at
3 10
6
U three times per week. Oral allopurinol (300 mg)
was administere d daily in patients diagnosed with hyper-
uricemia. Itching was treated with antihistamines. Treat-
ment side effects, changes in symptoms, and results from
once weekly blood tests were recorded in both groups.
2.3. Statistica l analyses
Demograph ic data are presented as median with inter-
quartiles (Q1, Q3) and compare d using the MannWhitney
U test for age. Categorical data are presente d as n (%) and
were compared using the Pearson Chi-square test or Fish-
ers exact test in instances where n is less than ve. Labo-
ratory analyses are presented as mean standard
deviation (SD) and were compared between groups using
a paired t-test. Results were considered statistically signif-
icant with a P-value < 0.05. All statistical analyses were
performed using SPSS 18.0 statistics software (SPSS Inc.,
Chicago, IL, USA).
3. Results
Patient baseline demographics and clinical characteris-
tics were similar between treatment groups (Table 1). Age,
sex, polycythemi a, and most clinical symptoms were not
signicantly different between ECP-treated patients and
controls. At baseline, patients treated with ECP had an ele-
vated white blood cell count compared with patients in the
control group (15% vs 50%, P = 0.041), respectivel y.
230 H. Liu et al. / Transfusion and Apheresis Science 48 (2013) 229233
A total of 28 ECP procedures were performed in 20 pa-
tients; one patient received 3 treatments, 6 patients re-
ceived 2 treatments , and 13 patients received only 1
treatment. In cases of multiple ECP procedures, one to
two days separated procedures. In the treatment group,
the mean RBC collection time was 25.7 4.5 min (range:
1937 min), with a mean collection volume of
773.5 129.3 mL (range: 6001002 mL). Among the pa-
tients in the control group (n = 20), 4 newborn patients
(20%) received distinct treatments for their respective pri-
mary diseases. An 18-day-ol d patient whose mother had
complicated gestational kidney disease; an 11-day-old pa-
tient had neonatal hyperbilirubi nemia; a 13-day-old pa-
tient was full-term low birth weight infant; a 12-day-old
patient had neonatal aspiration pneumonia. Furthermore,
of the patients in the control group, 9 patients (45%) re-
ceived exsanguination plus hydroxyure a treatments; and
7 patients (35%) received treatment for primary disease
plus exsanguinations and hydroxyure a.
Blood laboratory values, RBC count, HGB count, and HCT
level, were signicantly reduced (P < 0.001 for each) in
patients treated with ECP (pre-treatment vs post-treat-
ment; Table 2). For the patients in ECP group, RBC, Hb,
and Ht levels were high at admission, which were sus-
tained prior to ECP. The pre-treatment levels of RBC, Hb,
and Ht can be characterized as the admission levels.
ECP patients are commonly discharged approximately 2 h
after ECP; therefore, for patients in the ECP group, post-
treatment levels of RBC, Hb, and Ht were similar to the dis-
charge levels of the control group. No signicant differ-
ences in blood laboratory values were observed in
patients in the control group.
After ECP, symptom s associated with polycythem ia
such as discolored/purp le face and palms, mucosal conges-
tion, fatigue, dizziness , and headache were immediatel y
and dramatically relieved. Hypocalc aemia was identied
in 2.76% (9 out of 326) of treated patients and was associ-
ated with numbness around the mouth, ngers, and toes.
For prevention, all patients received oral or IV administra-
tion of 10% calcium gluconate solution (20 mL) before ECP.
In addition, splenomega ly (n = 16), hypertension (n = 15),
elevation of white blood cells (n = 13), increase of platelets
Table 1
Patient demograp hics and clinical characteristics by group.
Variables ECP group (n = 20) Control group (n = 20) P-value
Age, median (Q1, Q3), years 61.5 (52.3, 69.5) 52 (40.3, 65.5) 0.093
Gender 1.000
Males 14 (70%) 15 (75%)
Females 6 (30%) 5 (25%)
Syndrome
Polycythemia vera 11 (55%) 9 (45%) 0.527
Secondary polycythemia 9 (45%) 11 (55%) 0.527
Clinical symptom
Splenomegaly 11 (55%) 5 (25%) 0.053
Numbness/swelling and pain of limbs and face 3 (15%) 6 (30%) 0.451
Itchy skin 1 (5%) 0 (0%) NA
Hypertension 7 (35%) 8 (40%) 0.744
Cerebral infarction 2 (10%) 0 (0%) NA
Dysuria 1 (5%) 0 (0%) NA
Elevated platelet count 5 (25%) 7 (35%) 0.490
Elevated white blood cell count 3 (15%) 10 (50%) 0.041
a
Therapy
ECP 20 (100%) ND
Distinct treatments for primary disorders ND 4 (20%)
Exsanguination + hydroxyurea ND 9 (45%)
Treatment of primary disease + exsanguination + hydroxyurea ND 7 (35%)
Data were presented as median (Q1, Q3) and compared using MannWhitney U test for age; Categorical data are presented as n (%) and compared using
Pearson Chi-square test or Fishers exact test if the number was less than ve.
ECP, Erythrocytapheresis; NA, not assessed; ND, not derived.
a
P < 0.05, indicated signicantly different between groups.
Table 2
Comparison of blood examination for both groups.
Variables ECP group (n = 20) Control group (n = 20)
Pre-treatment Post-treatment P-value Admission Discharge P-value
RBC (10
12
/L) 7.9 1.2 7.2 1.2 <0.001
a
6.5 1.4 6.2 1.5 0.14
HGB (g/L) 210.4 15.2 179.3 15.8 <0.001
a
193.2 33.4 183.4 28.6 0.092
HCT (%) 64.9 5.4 51.8 5.7 <0.001
a
56.6 8.8 55.5 8.1 0.423
Data presented as mean SD and comparisons between groups was performed using a paired t-test. The pre-treatment levels of RBC, HGB, and HCT in the
ECP group are equivalent to the admission levels of RBC, HGB, and HCT in the control group. Similarly, the post-treatment levels in the ECP group are
equivalent to the discharge levels in the control group.
ECP, erythrocytapheresis; HCT, hematocrit; HGB, hemoglobin; RBCs, red blood cells; SD, standard deviation.
a
P < 0.001, indicated signicantly different between pre-treatment and post-treatment.
H. Liu et al. / Transfusion and Apheresis Science 48 (2013) 229233 231
(n = 12), numbness or swelling/pain of limbs and face
(n = 9), cerebral infarction (n = 2), skin itching (n = 1), and
dysuria (n = 1) were reported. No patients develope d
symptoms of hypovolemia.
After ECP, pharmacotherap y was performed using
hydroxyure a, interferon, enteric-coated aspirin, and allo-
purinol. Consequentl y, factors associated with secondar y
erythrocyto sis were removed. Hb and Hct levels returned
to normal levels immediatel y after combination therapy
was initiated.
4. Discussion
The current retrospective analysis demonst rates ECP
treatment signicantly reduced the mean number of RBCs
(7.6%), mean Hb (14.8%), and mean Hct (20.2%) in patients
with polycythemi a. The symptoms related to hyperviscos-
ity were improved dramatically after ECP. The RBC count
was generally reduced, associated with reduced loss of
white blood cells and platelets. The time required for ECP
was under 30 min and was associated with few adverse
side effects. Overall, ECP was considered to be a safe option
in the treatment armamentariu m of polycythem ia.
It has been demonstrated that ECP is associate d with
cost savings resulting from a lower number of treatment
procedures and lower productivity loss in ECP-treated pa-
tients [8]. Rombout- Sestrienkova et al. reported the total
number and duration of treatments were reduced approx-
imately 70% in patients treated with ECP vs phlebotomy .
Although the per procedure costs were higher using ECP,
the total costs for ECP treatment were comparable or less
expensive than phlebotomy [9]. In total, cost per ECP is
2200 Chinese Renminbi (approximately $350 US), which
includes the indwelling needle, special catheter and corre-
sponding drugs. Comparativ ely, exsangui nations can cost
approximat ely up to 165 Renminb i (approximately $25
US), which includes costs associated with blood collection.
Rombout-Ses trienkova et al. has also recently demon-
strated no signicant difference in total treatment costs
between ECP-treated patients compared with those trea-
ted with phlebotom y. However, the costs resulting from
the number of hours absent from work were signicantly
lower for the ECP group with an estimate d cost savings
of over 1400 euros (approximately $1800 US) [10].
Bloodletting is often used as an initial step in the treat-
ment of PV because it can rapidly reduce RBC count and
improve symptom s [11], and has been shown to slightly
reduce RBC count, HCT, and blood volume to normal levels
and improve clinical symptom s. Bloodletting may lead to
reduced antithrombi n and a hypercoagulati ve state; in
addition, increased platelets increase the risk of bleeding
and thrombosis [12]. Repeated bloodletting may cause iron
deciency and promote secondar y brosis of bone
marrow.
Therapeutic ECP has the advantag es of bloodlett ing
without plasma protein and coagulation factor loss. It has
been demonstrated that ECP is a highly effective treatment
to reduce iron overload [8,10]. Since the current study was
retrospective by design, iron overload data was not avail-
able we consider this a limitatio n of the current analysis.
ECP may also selectively remove excess platelets and white
blood cells. Colloid and crystalloid solutions of equal vol-
ume are also typically administere d in conjunction with
ECP, a treatment that can dilute blood as red blood cell vol-
ume decrease s. Taken together, ECP has been shown to im-
prove hyperviscosity state and microcirc ulation, thereby
increasing oxygen supply, attenuating adverse clinical
symptoms, and reducing the overall risk for thrombos is.
Furthermore, if continuous-ow blood cell separation is
performed in a closed environm ent intra-oper ative compli-
cations are generally reduced [13,14].
The use of ECP over phlebotomy has also been exam-
ined in the literature. Vecchio et al. reported less frequent
treatments in patients treated with ECP (range: 2 months
to 7 months) compared with patients in a phlebotomy -
only group (range: 20 days to 2 months). When given the
choice, patients switched from phlebotomy-on ly treat-
ment to ECP treatment and demonstrated a considerably
prolonged interval between treatments [15].
Severe side effects were not observed in the patients
currently examined. These patients were tolerant to ECP
despite being older (age range: 40.365.5 years) and some
presenting with concomitan t cardiovascular and cerebro-
vascular diseases and obvious symptoms of hyperviscos ity.
In this patient population, Hct and Hb levels were high
prior to ECP, and blood volume was also signicantly
increased.
The rapid effectivenes s of ECP is critical to control
symptoms associate d with polycythemia and other disor-
ders. The rapid response will allow for accelerated initia-
tion of pharmacother apy. In clinical practice,
pharmacother apy does not consistently offer a rapid efca-
cious response; however , ECP prior to pharmacotherap y
has the potential to markedly increase pharmacotherap eu-
tic efcacy and to shorten the duration of hospital admis-
sions [5].
The efcacy of ECP has been demonst rated in a variety
disorders. Celikdemir et al. reported the rapid reduction
of carboxyhemogl obin levels in patients with acute car-
bon-dioxide poisoning who were treated with ECP (27%
of patients signicantly improved within 1 h of ECP treat-
ment). It has been demonstrat ed that ECP can effectively
reducing mortality and morbidity in carbon monoxide poi-
soning [16], reverse the symptom s of myocardial or cere-
bral ischemia, pulmona ry embolism, and intestina l
bleeding [17], and prevent placental infarction and fetal
death in pregnant patients with polycythemi a.
ECP can rapidly reduce RBC count, Hb, and Hct, and im-
prove clinical symptoms. ECP in combinati on with phar-
macotherapy may offer additional benets. Considering
the indications for ECP, including PV, hereditary polycythe-
mia, hemoglobin disease, and secondary polycythem ia, the
therapeutic implications for ECP are widespread . In total,
ECP has distinct advantages over conventional therapy
and should be considered when choosing a treatment plan
for patients with polycythemia.
References
[1] Hoffman R, Xu M, Finazzi G, Barbui T. The polycythemias. In:
Hoffman R, Benz EJ, Shattil SJ, et al., editors. Hoffman Hematology:
232 H. Liu et al. / Transfusion and Apheresis Science 48 (2013) 229233
Basic Principles and Practice. 5th ed. Philadelphia (Pa): Churchill
Livingstone Elsevier; 2008 [chapter 68].
[2] Yan S, Tian ZS. Polycythemia and exsanguination. Chin J Blood
Transfus 2004;3:21820 [in Chinese].
[3] Wang CY, Hao SX, Zhang W. Toxic effects of cyclophosphamide on
the sensitive period of teratogenesis in rats. China Occup Med
2010;3:2302 [in Chinese].
[4] Wang YZ, Liu MX, Gao XP, Hun SY, You M, Liu FR. Application of
erythropheresis in patients with polycythemia vera. China J Blood
Transfus 2010;16(4):2489.
[5] Lv YL, Ma HL, Sun XL, et al. Clinical efcacy of ECP in combination
with hydroxyurea and a-interferon in the treatment of polycythemia
vera. China J Blood Transfus 2007;20(2):145.
[6] Zhang ZN, Shen T. Criteria for Diagnosis and Therapeutic Efcacy of
Hematological Diseases 2007. 3rd edition. Beijing: Beijing Science
Press; 2007. p. 10622,15761.
[7] Thiele J, Kvasnicka HM. The 2008 WHO diagnostic criteria for
polycythemia vera, essential thrombocythemia, and primary
myelobrosis. Curr Hematol Malig Rep 2009 Jan;4(1):3340.
[8] Adams PC, Barton JC. How I treat hemochromatosis. Blood
2010;116:31725.
[9] Rombout-Sestrienkova E, van Noord PA, van Deursen CT, Sybesma BJ,
Nillesen-Meertens AE, Koek GH. Therapeutic erythrocytapheresis
versus phlebotomy in the initial treatment of hereditary
hemochromatosis - a pilot study. Transfus Apher Sci
2007;36(3):2617.
[10] Rombout-Sestrienkova E, Nieman FH, Essers BA, van Noord PA,
Janssen MC, van Deursen CT, et al. Erythrocytapheresis versus
phlebotomy in the initial treatment of HFE hemochromatosis
patients: results from a randomized trial. Transfusion 2012
Mar;52(3):4707.
[11] Zhang JH, Huang P. Modern hemopathic therapy, 1st edition.
Peoples Military Medical Press; 1997. [in Chinese. Chapter 7.
Section 2.]
[12] Wang AM et al. Erythropheresis treatment for polycythemia vera.
Shandong Medl J 1999;39(3):32.
[13] Kaboth U, Rumpf KW, Liersch T, Vehmeyer K, Krieter D, Kaboth W.
Advantages of isovolemic large-volume erythrocytapheresis as a
rapidly effective and long lasting treatment modality for red blood
cell depletion in patients with polycythemia vera. Ther Apher
1997;1(2):1314.
[14] Kaboth U, Rumpf KW, Lipp T, Bigge J, Nauck M, Beyer JH, et al.
Treatment of polycythemia vera by isovolemic large-volume
erythrocytapheresis. Klin Wochenschr 1990;68(1):1825.
[15] Vecchio S, Leonardo P, Musuraca V, DEttoris AR, Geremicca W. A
comparison of the results obtained with traditional phlebotomy and
with therapeutic erythrocytapheresis in patients with
erythrocytosis. Blood Transfus 2007;5(1):203.
[16] Celikdemir A, Gokel Y, Guvenc B, Tekinturan F. Treatment of acute
carbon-monoxide poisoning with therapeutic erythrocytapheresis:
clinical effects and results in 17 victims. Transfus Apher Sci
2010;43(3):3279.
[17] Tabuchi T, Ubukata H, Saniabadi AR, Soma T. Granulocyte apheresis
as a possible new approach in cancer therapy: a pilot study involving
two cases. Cancer Detect Prev 1999;23(5):417.
H. Liu et al. / Transfusion and Apheresis Science 48 (2013) 229233 233