Increased plasma viscosity in plasma cell dyscrasia and whole blood viscosity in polycythemia vera

Abstract

BACKGROUND:

Although hyperviscosity syndrome in plasma cell dyscrasia (PCD) and thrombosis in myeloproliferative neoplasm (MPN) are major causes of morbidity and mortality, blood viscosity measurements are often underutilized.

OBJECTIVE:

This study aimed to characterize whether whole blood viscosity (WBV) or plasma viscosity (PV) could be predictive of hyperviscosity syndrome in PCD and could be elevated in subgroups of MPN.

METHODS:

A total of 75 patients with hematologic diseases: PCD (n = 26), MPN (n = 25) including polycythemia vera (P. vera) and lymphoma (n = 24) were enrolled along with 104 healthy controls. Both WBV and PV were measured using a capillary tube viscometer. Hyperviscosity syndrome was defined as having 2 or more hyperviscosity symptoms.

RESULTS:

Patients with PCD showed significantly higher PVs at high and low shear rates when compared to healthy controls, especially in those with hyperviscosity syndrome. The sensitivity and specificity of WBV and PV in detecting hyperviscosity syndrome were 28.6% and 94.1%, and 71.4% and 66.7%, respectively. Patients with P. vera exhibited high WBV and RBC counts compared to healthy controls.

CONCLUSION:

PV is predictive of hyperviscosity syndrome in PCD and WBV is elevated in patients with P. vera. It suggests that hemorheologic disturbances exist in patients with PCD and MPN and that tests of viscosity may be helpful in detecting hemorheological disturbances.

Keywords

Blood viscosity plasma cell dyscrasia myeloproliferative neoplasm hyperviscosity syndrome

1 Introduction

Testing for blood viscosity has been underutilized in medical practice, in part because of limitations in how to best apply data gleaned from commercial available blood viscosity kits to clinical decisions. Although rotational viscometers have been used in some studies, their clinical applicability is limited by several key drawbacks including the: i) need to clean the test section after each measurement and ii) potential risk for contact with contaminated blood [1, 2]. A scanning capillary tube viscometer has been developed which uses a disposable tube thus removing the need to clean [3]. Since the scanning capillary tube viscometer is easy to use, it is suitable for measuring viscosity in clinical practice.

In hematologic diseases such as plasma cell dyscrasia (PCD) and myeloproliferative neoplasm (MPN), rheological alteration has been suggested as a thrombogenic risk factor [4, 5]. In PCD, M-proteins contribute to a hyperviscosity syndrome which is characterized by neurologic symptoms (eg, headaches and dizziness, visual disturbances) and organ failure [6]. Although hyperviscosity syndrome is a major cause of morbidity and mortality in PCD [7], laboratory tests to identify/quantify hyperviscosity syndrome are not clinically applicable. Simple and accurate laboratory assessments would be beneficial because prompt reduction of viscosity by plasmapheresis would prevent hyperviscosity syndrome-related morbidity. Meanwhile, elevations in the circulating levels of RBC, WBC, and platelets in MPN are considered to be contributors to prothrombotic conditions [8, 9]. Therefore, it is worthwhile to evaluate whether high levels of RBC, WBC, and platelets in polycythemia vera (P. vera), chronic myeloid leukemia (CML) and essential thrombocythemia (ET) of MPN may increase blood viscosity.

Blood viscosity can be measured using whole blood or plasma. Since whole blood viscosity (WBV) has the characteristics of a non-Newtonian fluid, it depends on blood flow and blood cell composition [10]. Plasma viscosity (PV), however, acts as a Newtonian fluid and its viscosity does not depend on blood composition [11]. It is necessary to evaluate the performances of both WBV and PV in clinical practice, because blood characteristics vary depending on the disease.

It is hypothesized that viscosity levels are associated with the occurrence of hyperviscosity syndrome in PCD and that clinical practice and patient management would benefit from an effective test to measure viscosity. In this study, WBV and PV were both measured in patients with PCD and healthy controls by using a scanning capillary tube viscometer to explore whether either or both viscosity measurements could predict or identify hyperviscosity syndromes. In addition, both WBV and PV were measured in patients with MPN to investigate whether three blood cells (RBC, WBC and platelets) influence on both viscosity in three MPN subgroups (P. vera, CML and ET).

2 Methods

2.1 Study population

A total of 75 adult (age≥20) patients newly diagnosed with hematologic diseases including: i) PCD (n = 26), ii) MPN (n = 25) and iii) lymphoma (n = 24) were included in this study. Among the 26 patients with PCD, there were 24 with multiple myeloma, one with monoclonal gammopathy of undetermined significance, and one with Waldenström macroglobulinemia. Patients with lymphoma (all of them had no evidence of bone marrow involvement) were used as a disease control group. A total of 104 adults were included as healthy (ie, not been hospitalized during the last 6 months) controls. Adults with a history of vascular diseases (coronary artery disease, stroke, peripheral vessel occlusive disease, autoimmune disease, thrombosis), anemia (male <12 g/dL, female <11 g/dL) or a history of transfusions within a month were excluded. Patients with hematologic diseases under investigation and controls with: i) congenital predisposition to thrombosis, ii) hemorrhagic history (eg, hemophilia, history of surgery during the past 2 months, taking oral anticoagulants or anti-platelet agents) were also excluded. Written informed consent was obtained under approval of the Institutional Review Board of Seoul National University Hospital.

2.2 Survey of hyperviscosity syndrome

The relationship between blood viscosity and hyperviscosity syndromes in patients with PCD were investigated. Questionnaire was administered to identify patients with hyperviscosity symptoms (ie, fatigue, headache, dizziness, tinnitus, visual disturbance, neurologic symptoms including ataxia, drowsy and seizure, coma, and sexual dysfunction) [7]. Patients were given one point for each hyperviscosity symptom they were experiencing. For each subject, the total hyperviscosity syndrome score was obtained by summing the points. Subjects with scores of 2 or greater were labeled: ‘presence of hyperviscosity syndrome’ [7].

2.3 Measurement of blood viscosity

WBV and PV were measured using a scanning capillary tube viscometer, Hemovister (Ubiosis, Seongnam, Korea) according to the manufacturer’s recommendations. The analytical performance of a scanning capillary tube viscometer is stable and comparable with Brookfield rotational viscometer. [1, 3]. WBV was measured with EDTA-treated whole blood. PV was measured with plasma following centrifugation at 2,600 g for 10 minutes. To improve the sensitivity of detecting movement of the plasma in the device, we added black dyes (McCormick, Highland, USA) to plasma. At concentrations of less than 2%, the impact of the dye on viscosity will be negligible [12]. In this study, 6 μl (0.2%) of dye was used per 3 ml of blood. All samples were preheated and maintained at 37°C throughout viscosity measurement. Both WBV and PV were measured under high shear rate (HSR) of 300 s^-1 and low shear rate (LSR) of 5 s^-1. Tissue oxygen delivery index (TODI) was calculated (ie, hematocrit divided by WBV at LSR). Reproducibility was validated by measuring viscosity of control material for 20 days; the coefficient of variation was 6.4%. Reference ranges of viscosities were determined by calculating 5.0 and 95.0 percentiles of healthy controls (WBV at HSR, 3.65– 6.10; WBV at LSR, 10.75– 18.88; PV at HSR, 1.64– 2.0; PV at LSR, 2.29– 3.07 cP). Based on reference ranges, PCD patients were divided into 2 groups; increased viscosity and normal viscosity groups. The sensitivities and specificities of WBV and PV were calculated across the presence of hyperviscosity syndrome, that is, hyperviscosity syndrome score≥2 or <2. $Sensitivity = \frac{No . of patients with increased viscosity}{No . of patients with score ⩾ 2}$ (1) $Specificity = \frac{No . of patients with normal viscosity}{No . of patients with score < 2}$ (2)

2.4 Other laboratory assays

Complete blood count (CBC) was performed using SE2100 (Sysmex, Japan). Total protein, albumin, total cholesterol and high-sensitivity C-reactive protein (hs-CRP) were measured using Toshiba 200FR (Toshiba, Tokyo, Japan).

2.5 Statistical analysis

We used IBM SPSS Statistics version 21 (IBM Corporation, Armonk, NY, USA) for all statistical analyses. Patients were classified as having PCD, MPN, or lymphoma. Patients with MPN were further classified as P. vera, ET, primary myelofibrosis (PMF), and CML. Subgroup analyses for blood viscosity and laboratory profiles were performed with one-way ANOVA (Analysis of variance) with Bonferroni correction and t-test. Chi-square and Fisher’s exact tests were performed for categorical data. A probability value (P) of less than 0.05 was considered significant. Results were presented as “mean ± standard deviation” in the tables.

3 Results

3.1 Hematologic parameters, WBV, and PV

There were no significant differences among the characteristics (ie, age and gender) in patients with hematologic diseases and healthy controls (Table 1). When compared to healthy controls, WBC counts were significantly higher in patients with MPN and RBC counts were significantly lower in patients with PCD and lymphoma. Total protein levels were significantly higher in patients with PCD. Albumin levels were significantly lower in patients with PCD and lymphoma. Total cholesterol levels were significantly lower in patients with PCD and MPN. Hs-CRP levels were elevated in patients with lymphoma.

Table 1
Laboratory results of patients with hematologic diseases and healthy controls

Parameters PCD (n = 26) MPN (n = 25) Lymphoma (n = 24) Healthy controls (n = 104)

Gender (Male, %) 19 (73.1%) 19 (76.0%) 16 (33.3%) 74 (71.2%)

Age (years) 63.6 ± 9.2 54.9 ± 16.5 52.8 ± 15.6 55.0 ± 14.5

WBC (×10⁹/L) 6.10 ± 3.96 37.89 ± 82.03^* 12.48 ± 14.89 5.39 ± 1.57

RBC (×10¹²/L) 3.32 ± 0.80^* 4.90 ± 1.53 4.17 ± 0.67^* 4.74 ± 0.44

Hb (g/dL) 10.49 ± 2.15^* 13.98 ± 3.82 12.55 ± 2.03^* 14.88 ± 1.41

Platelet (×10⁹/L) 178.2 ± 90.7 630.0 ± 339.5^* 221.6 ± 97.0 214.2 ± 48.2

Total protein (g/dL) 7.71 ± 1.86^* 7.36 ± 0.56 7.19 ± 0.53 7.08 ± 0.43

Albumin (g/dL) 3.53 ± 0.59^* 4.20 ± 0.40 4.02 ± 0.45^* 4.33 ± 0.27

Total cholesterol (mg/dL) 141.4 ± 51.2^* 155.7 ± 36.3^* 177.7 ± 50.8 186.4 ± 32.1

hs-CRP (mg/dL) 0.90 ± 2.25 0.36 ± 0.79 2.21 ± 4.04^* 0.12 ± 0.25

ESR (mm/hr) 44.32 ± 32.82 16.52 ± 18.22 34.38 ± 29.74 –

WBV at HSR (cP) 4.75 ± 2.39 5.59 ± 1.91 4.43 ± 0.91 4.99 ± 0.82

WBV at LSR (cP) 13.22 ± 6.24 16.63 ± 5.23 13.43 ± 3.23 15.03 ± 2.63

PV at HSR (cP) 2.57 ± 1.81^* 1.86 ± 0.16 1.95 ± 0.24 1.84 ± 0.11

PV at LSR (cP) 5.46 ± 6.04^* 2.96 ± 0.45 3.08 ± 0.54 2.71 ± 0.24

TODI 2.83 ± 1.06 2.70 ± 0.46 2.93 ± 0.44 3.00 ± 0.37

Parameters	PCD (n = 26)	MPN (n = 25)	Lymphoma (n = 24)	Healthy controls (n = 104)
Gender (Male, %)	19 (73.1%)	19 (76.0%)	16 (33.3%)	74 (71.2%)
Age (years)	63.6 ± 9.2	54.9 ± 16.5	52.8 ± 15.6	55.0 ± 14.5
WBC (×10⁹/L)	6.10 ± 3.96	37.89 ± 82.03^*	12.48 ± 14.89	5.39 ± 1.57
RBC (×10¹²/L)	3.32 ± 0.80^*	4.90 ± 1.53	4.17 ± 0.67^*	4.74 ± 0.44
Hb (g/dL)	10.49 ± 2.15^*	13.98 ± 3.82	12.55 ± 2.03^*	14.88 ± 1.41
Platelet (×10⁹/L)	178.2 ± 90.7	630.0 ± 339.5^*	221.6 ± 97.0	214.2 ± 48.2
Total protein (g/dL)	7.71 ± 1.86^*	7.36 ± 0.56	7.19 ± 0.53	7.08 ± 0.43
Albumin (g/dL)	3.53 ± 0.59^*	4.20 ± 0.40	4.02 ± 0.45^*	4.33 ± 0.27
Total cholesterol (mg/dL)	141.4 ± 51.2^*	155.7 ± 36.3^*	177.7 ± 50.8	186.4 ± 32.1
hs-CRP (mg/dL)	0.90 ± 2.25	0.36 ± 0.79	2.21 ± 4.04^*	0.12 ± 0.25
ESR (mm/hr)	44.32 ± 32.82	16.52 ± 18.22	34.38 ± 29.74	–
WBV at HSR (cP)	4.75 ± 2.39	5.59 ± 1.91	4.43 ± 0.91	4.99 ± 0.82
WBV at LSR (cP)	13.22 ± 6.24	16.63 ± 5.23	13.43 ± 3.23	15.03 ± 2.63
PV at HSR (cP)	2.57 ± 1.81^*	1.86 ± 0.16	1.95 ± 0.24	1.84 ± 0.11
PV at LSR (cP)	5.46 ± 6.04^*	2.96 ± 0.45	3.08 ± 0.54	2.71 ± 0.24
TODI	2.83 ± 1.06	2.70 ± 0.46	2.93 ± 0.44	3.00 ± 0.37

^*P < 0.05 versus healthy controls. Abbreviations: PCD, plasma cell dyscrasia; MPN, myeloproliferative neoplasm; hs-CRP, high sensitivity C-reactive protein; ESR, erythrocyte sedimentation rate; WBV, whole blood viscosity; HSR, high shear rate; LSR, low shear rate; PV, plasma viscosity; TODI, tissue oxygen delivery index.

There were no significant differences in WBVs at HSR and LSR in patients with hematologic diseases and healthy controls. Interestingly, PVs at HSR and LSR were significantly higher in PCD patients than in healthy controls. TODI were not different across groups.

3.2 Blood viscosity measurements for the detection of hyperviscosity syndrome in patients PCD

After the exclusion of 2 patients with PCD because of incomplete surveys (final n = 24), mean blood viscosity levels were quantified using the hyperviscosity syndrome score (Table 2). When comparing patients with score 0, 1, 2-3, and 4, there were no statistically significant differences in viscosities and TODI (P values of WBV at HSR and LSR, PV at HSR and LSR, and TODI were 0.202, 0.434, 0.091, 0.216, and 0.665, respectively.). Although not reaching statistical significance, individuals with a hyperviscosity syndrome score of 4 showed higher levels of WBV at HSR/LSR, PV with HSR/LSR, compared with those with scores of 0– 3. Numerically, the average TODI of individuals with a score of 4 was lower than those with score 0– 3; again however, statistical significance was not reached.

Table 2
Differences of blood viscosities relative to hyperviscosity syndrome scores in patients with plasma cell dyscrasia

Score WBV at HSR (cP) WBV at LSR (cP) PV at HSR (cP) PV at LSR (cP) TODI

0 (n = 8) 4.50 ± 1.15 12.95 ± 4.31 2.13 ± 0.52 5.23 ± 5.05 2.84 ± 1.18

1 (n = 9) 4.14 ± 1.21 11.73 ± 4.85 2.01 ± 0.29 3.16 ± 0.74 3.08 ± 1.03

2-3 (n = 5) 5.20 ± 3.25 14.76 ± 7.73 3.24 ± 2.45 7.02 ± 7.67 2.54 ± 0.92

4 (n = 2) 3.4– 12.8^* 8.7– 30.6^* 1.8– 9.0^* 2.8– 24.1^* 0.88– 3.38^*

Score	WBV at HSR (cP)	WBV at LSR (cP)	PV at HSR (cP)	PV at LSR (cP)	TODI
0 (n = 8)	4.50 ± 1.15	12.95 ± 4.31	2.13 ± 0.52	5.23 ± 5.05	2.84 ± 1.18
1 (n = 9)	4.14 ± 1.21	11.73 ± 4.85	2.01 ± 0.29	3.16 ± 0.74	3.08 ± 1.03
2-3 (n = 5)	5.20 ± 3.25	14.76 ± 7.73	3.24 ± 2.45	7.02 ± 7.67	2.54 ± 0.92
4 (n = 2)	3.4– 12.8^*	8.7– 30.6^*	1.8– 9.0^*	2.8– 24.1^*	0.88– 3.38^*

^*“Minimum-maximum” values were presented. Abbreviations: WBV, whole blood viscosity; HSR, high shear rate; LSR, low shear rate; PV, plasma viscosity; TODI, tissue oxygen delivery index.

The PVs at HSR and LSR were significantly increased in patients with hyperviscosity syndrome (n = 7), compared with healthy controls (P = 0.009 and 0.002, respectively) (Fig. 1). In patients without hyperviscosity syndrome (n = 17), PV at LSR was significantly increased compared with healthy controls (P = 0.035), but PV at HSR was not significantly increased (P = 0.146). The sensitivity and specificity of WBV and PV for detection of hyperviscosity syndrome were 28.6% and 94.1%, and 71.4% and 66.7%, respectively (Table 3).

Fig.1

Blood viscosity differences in patients with plasma cell dyscrasia (n = 24) with and without hyperviscosity syndrome, and healthy controls. Presence of hyperviscosity syndrome was arbitrarily defined as a hyperviscosity syndrome score or 2 or more points. The average levels of PVs at HSR and LSR in patients with hyperviscosity syndrome (n = 7) (3.86 and 8.86 cP) were significantly higher than those in healthy controls (1.84 and 2.71 cP) (P = 0.009 and 0.002). The average levels of PVs at LSR in patients without hyperviscosity syndrome (n = 17) (4.26 cP) were significantly higher than those in healthy controls (P = 0.035), but average levels of PVs at HSR (2.07 cP) were not significantly higher than those in healthy controls (P = 0.146). Abbreviations: HSR, high shear rate; LSR, low shear rate, WBV, whole blood viscosity; PV, plasma viscosity. ^*P < 0.05 versus healthy controls.

Table 3

Diagnostic performance of blood viscosity values for detecting hyperviscosity syndrome in plasma cell dyscrasia patients

Viscosity		Hyperviscosity syndrome (Score≥2 or <2)		Sensitivity (%)	Specificity (%)	Diagnostic accuracy (%)
		Presence (n = 7)	Absence (n = 17)
WBV at	Increased^a (n = 3)	2 (8.3%)	1 (4.2%)	28.6	94.1	75.0
HSR	Normal(n = 21)	5 (20.8%)	16 (66.7%)
WBV at	Increased^a (n = 3)	2 (8.3%)	1 (4.2%)	28.6	94.1	75.0
LSR	Normal(n = 21)	5 (20.8%)	16 (66.7%)
PV at HSR^c	Increased^a (n = 10)	5 (22.7%)	5 (22.7%)	71.4	66.7	68.2
Normal(n = 12)	2 (9.1%)	10 (45.5%)
PV at LSR^c	Increased^a (n = 10)	5 (22.7%)	5 (22.7%)	71.4	66.7	68.2
Normal(n = 12)	2 (9.1%)	10 (45.5%)

^a‘Increased’ indicates that the viscosity levels were >the cutoff value (Reference ranges: WBV at HSR, 3.65– 6.10; WBV at LSR, 10.75– 18.88; PV at HSR, 1.64– 2.0; PV at LSR, 2.29– 3.07 cP). ^bData on PV were missing in 2 patients. Abbreviations: WBV, whole blood viscosity; HSR, high shear rate; LSR, low shear rate; PV, plasma viscosity.

3.3 WBV and PV in subgroups of patients with myeloproliferative neoplasms

Among patients with MPN, those with P. vera showed significantly higher WBV at HSR and LSR (8.14 and 23.44 cP; P < 0.001 at both HSR and LSR) and higher RBC counts (7.19×10¹²/L; P < 0.001) than healthy controls (Table 4). Patients with ET showed no difference of blood viscosity compared with healthy controls despite significantly higher platelet counts (919.6×10⁹/L; P < 0.001). Patients with PMF showed significantly lower WBV at LSR (11.60 cP; P = 0.044) and RBC counts (3.79×10¹²/L; P = 0.001) than healthy controls. Patients with CML showed significantly higher PV at LSR (3.20 cP; P = 0.002) and WBC counts (107.58×10⁹/L; P < 0.001), and lower TODI (2.56; P = 0.039) than healthy controls.

Table 4
Differences of blood viscosities and blood cell counts among subgroups of patients with myeloproliferative neoplasm

Diagnosis P. vera (n = 5) ET (n = 7) PMF (n = 6) CML (n = 7) Healthy controls (n = 104)

WBV at HSR (cP) 8.14 ± 2.08^* 5.40 ± 0.61 4.07 ± 0.91 5.27 ± 1.67 4.99 ± 0.82

WBV at LSR (cP) 23.44 ± 3.71^* 16.74 ± 1.94 11.60 ± 2.72^* 15.97 ± 5.08 15.03 ± 2.63

PV at HSR (cP) 1.80 ± 0.14 1.80 ± 0.16 1.90 ± 0.14 1.91 ± 0.17 1.85 ± 0.11

PV at LSR (cP) 3.15 ± 0.35 2.70 ± 0.33 2.90 ± 0.36 3.20 ± 0.55^* 2.71 ± 0.24

TODI 2.65 ± 0.21 2.65 ± 0.15 2.95 ± 0.28 2.56 ± 0.80^* 3.00 ± 0.37

WBC (×10⁹/L) 8.53 ± 1.80 8.23 ± 2.77 15.65 ± 11.29 107.58 ± 137.45^* 5.39 ± 1.57

RBC (×10¹²/L) 7.19 ± 1.54^* 4.88 ± 0.51 3.79 ± 0.71^* 4.23 ± 0.99 4.74 ± 0.44

Platelet (×10⁹/L) 350.0 ± 241.0 919.6 ± 186.7^* 456.3 ± 277.0^* 689.3 ± 355.0^* 214.2 ± 48.2

Diagnosis	P. vera (n = 5)	ET (n = 7)	PMF (n = 6)	CML (n = 7)	Healthy controls (n = 104)
WBV at HSR (cP)	8.14 ± 2.08^*	5.40 ± 0.61	4.07 ± 0.91	5.27 ± 1.67	4.99 ± 0.82
WBV at LSR (cP)	23.44 ± 3.71^*	16.74 ± 1.94	11.60 ± 2.72^*	15.97 ± 5.08	15.03 ± 2.63
PV at HSR (cP)	1.80 ± 0.14	1.80 ± 0.16	1.90 ± 0.14	1.91 ± 0.17	1.85 ± 0.11
PV at LSR (cP)	3.15 ± 0.35	2.70 ± 0.33	2.90 ± 0.36	3.20 ± 0.55^*	2.71 ± 0.24
TODI	2.65 ± 0.21	2.65 ± 0.15	2.95 ± 0.28	2.56 ± 0.80^*	3.00 ± 0.37
WBC (×10⁹/L)	8.53 ± 1.80	8.23 ± 2.77	15.65 ± 11.29	107.58 ± 137.45^*	5.39 ± 1.57
RBC (×10¹²/L)	7.19 ± 1.54^*	4.88 ± 0.51	3.79 ± 0.71^*	4.23 ± 0.99	4.74 ± 0.44
Platelet (×10⁹/L)	350.0 ± 241.0	919.6 ± 186.7^*	456.3 ± 277.0^*	689.3 ± 355.0^*	214.2 ± 48.2

^*P < 0.05 versus healthy controls. Abbreviations: P. vera, polycythemia vera; ET, essential thrombocythemia; PMF, primary myelofibrosis; CML, chronic myelogenous leukemia; WBV, whole blood viscosity; HSR, high shear rate; LSR, low shear rate; PV, plasma viscosity; TODI, tissue oxygen delivery index.

4 Discussion

This study demonstrated that PV was elevated in patients with PCD when compared with healthy controls and that the sensitivity and specificity of PV to detect hyperviscosity syndrome were 71.4% and 66.7%. Compared to healthy controls, WBV was elevated in patients with P. vera. Among RBC, WBC, and platelet, WBV was correlated with RBC.

Unlike WBV, PV is not dependent on circulating blood cell counts, but can depend on plasma macromolecular components [11]. Plasma protein levels are proportional to PV. In this study we observed that individuals with PCD had higher PV, but similar WBV levels than healthy controls. Since M-protein levels are elevated but circulating blood cell counts are usually normal in PCD, it is likely that only plasma viscosity—which is dependent on M-protein levels—was elevated. Our results are consistent with a previous study showing that PV was elevated in individuals with PCD [4].

Here, we also show that WBV strongly correlated with both RBC counts and total protein levels. Patients with PCD usually had low RBC counts despite high total protein levels. Therefore, the WBV in PCD is likely to be normal because low RBC counts might compensate for the increase in WBV normally caused by high total protein levels.

According to the National Comprehensive Cancer Network guidelines for multiple myeloma [13], measurement of PV is recommended to evaluate hyperviscosity symptoms. However, there are few laboratories in which blood viscosity measurements are routinely performed in clinical practice. In the present study, individuals with PCD experiencing hyperviscosity symptoms showed markedly elevated PV compared to healthy controls, indicating that PV may be a diagnostic marker of hyperviscosity syndrome (see Table 2 and Fig. 1).

Considering that erythrocytosis might contribute to prothrombotic conditions [9], it is likely that high levels of RBC in P. vera may increase WBV. In this study, individuals with P. vera showed increased levels of WBV compared to healthy controls, suggesting that erythrocytosis can significantly increase WBV. To our knowledge, the present study is the first report demonstrating increased levels of WBV in patients with P. vera, although there have been previous reports about high viscosity levels in infants with physiological polycythemia [14, 15]. Erythrocytosis may induce more complex 3 dimensional clumps, which can increase blood viscosity and vascular resistance, resulting in decreased blood flow and blood sludging [5, 16]. In addition to the impaired rheology, RBC from P. vera patients showed a capacity for increased adhesion to endothelium which can easily recruit platelets, promoting thrombosis [9]. Although erythrocytosis in individuals with P. vera is a major cause of vascular disturbances [17], there have been few reports on indicators for thrombosis prediction or phlebotomy performance. In our view, viscosity can be used as a potential indicator for thrombosis through future clinical studies.

Meanwhile, thrombocytosis in ET did not increase WBV, suggesting that thrombotic risk in ET may come from causes other than high viscosity (eg, platelet activation) [17, 18]. PV did not increase in individuals with ET because plasma did not contain circulating blood cells. Hyperleukocytosis can be implicated in hyperviscosity, but an accurate assessment using a laboratory test is not currently possible. In our study, patients with CML and hyperleukocytosis did not show increases in WBV. Our results are consistent with a previous report showing hypo-viscosity in patients with hyperleukocytosis [8]. Although an increase in leukocytes can biologically activate endothelial cells and other blood cells [18], they are not considered to increase blood viscosity.

This study has some limitations. First, we could not assess the association of viscosity levels with thrombosis due to a small sample size. However, we did demonstrate a significant association between viscosity and symptoms of hyperviscosity in individuals with PCD. Second, our study was performed as a cross-sectional study, meaning that we were unable to investigate the long-term effect of viscosity on clinical outcomes of patients. Future prospective studies may help characterize the causal relationship of viscosity with thrombosis. Third, since there was no specific guideline about the level of hyperviscosity syndrome, we arbitrarily categorized the level according to the number of hyperviscosity symptoms in patients.

In summary, PCD patients showed increased PV compared to healthy controls, especially in those with hyperviscosity syndrome. Patients with PCD and hyperviscosity syndrome showed more significant increases in PV than those without hyperviscosity syndrome. We suggest that PV should be considered as an indicator of hyperviscosity syndrome in patients with PCD. Among patients with MPN, those with P. vera exhibited increased WBV and RBC counts, but patients with ET and high platelet counts and CML patients with high WBC counts showed no increase in WBV. Our study revealed hemorheologic disturbances in PCD and MPN, suggesting that the hemorheologic disturbances besides blood hypercoagulation can contribute to blood clot growth. It is expected that the hemorheologic condition measured by a viscometer may be a potential therapeutic target for antithrombotic management.

Footnotes

Acknowledgment

This work (2016R1A2B4015571) was supported by Mid-career Researcher Program through NRF grant funded by the Korea government (MSIP).

References

Kim

, Cho

, Lee

, Park

, Moon

, Hur

, et al. Analytical performance evaluation of the scanning capillary tube viscometer for measurement of whole blood viscosity. Clinical Biochemistry. 2013;46(1-2):139–42.

Graf

, Barras

. Rheological properties of human blood plasma–a comparison of measurements with three different viscometers. Experientia. 1979;35(2):224–5.

Kim

, Cho

, Hogenauer

, Kensey

. A method of isolating surface tension and yield stress effects in a U-shaped scanning capillary-tube viscometer using a Casson model. J Nonnewton Fluid Mech. 2002;103(2-3):205–19.

Uggla

, Nilsson

. Whole blood viscosity in plasma cell dyscrasias. Clinical Biochemistry. 2015;48(3):122–4.

Vaya

, Suescun

. Hemorheological parameters as independent predictors of venous thromboembolism. Clinical Hemorheology and Microcirculation. 2013;53(1-2):131–41.

Kwaan

. Hyperviscosity in plasma cell dyscrasias. Clinical Hemorheology and Microcirculation. 2013;55(1):75–83.

Mehta

, Singhal

. Hyperviscosity syndrome in plasma cell dyscrasias. Seminars in Thrombosis and Hemostasis. 2003;29(5):467–71.

Nwose

, Richards

. Whole blood viscosity issues VII: The correlation with leucocytosis and implication on leukapheresis. North American Journal of Medical Sciences. 2010;2(12):576–9.

El Nemer

, De Grandis

, Brusson

. Abnormal adhesion of red blood cells in polycythemia vera: A prothrombotic effect? Thrombosis Research. 2014;133(Suppl 2):S107–11.

10.

Thurston

. Viscoelasticity of human blood. Biophysical Journal. 1972;12(9):1205–17.

11.

Kesmarky

, Kenyeres

, Rabai

, Toth

. Plasma viscosity: A forgotten variable. Clinical Hemorheology and Microcirculation. 2008;39(1-4):243–6.

12.

Kim

, Cho

. The effect of dye concentration on the viscosity ofwater in a scanning capillary-tube viscometer. J Nonnewton Fluid Mech. 200311163–68.

13.

NCCN practice guidelines for multiple myeloma. National Comprehensive Cancer Network. Oncology. 1998;12(11A):317–51.

14.

Ramamurthy

, Berlanga

. Postnatal alteration in hematocrit and viscosity in normal and polycythemic infants. The Journal of Pediatrics. 1987;110(6):929–34.

15.

Sarkar

, Rosenkrantz

. Neonatal polycythemia and hyperviscosity. Seminars in Fetal & Neonatal Medicine. 2008;13(4):248–55.

16.

Harding

, Knisely

. Settling of sludge in human patients; a contribution to be biophysics of disease. Angiology. 1958;9(6):317–41.

17.

Falanga

, Marchetti

. Thrombosis in myeloproliferative neoplasms. Seminars in Thrombosis and Hemostasis. 2014;40(3):348–58.

18.

Casini

, Fontana

, Lecompte

. Thrombotic complications of myeloproliferative neoplasms: Risk assessment and risk-guided management. Journal of Thrombosis and Haemostasis: JTH. 2013;11(7):1215–27.