Hemorheologic Abnormalities Associated with HIV Infection: Altered Erythrocyte Aggregation and Deformability

Abstract

purpose. To investigate possible alterations of erythrocyte aggregation and deformability, which are factors that can influence blood flow, in human immunodeficiency virus (HIV)–infected individuals and to determine whether these factors are related to the severity of immunodeficiency.

methods. Laboratory evaluations were performed on 46 HIV-infected individuals and 44 HIV-negative control subjects. Current and nadir (lowest previous) CD4⁺ T-lymphocyte counts were identified for each subject. Erythrocyte aggregation was measured using a fully automatic erythrocyte aggregometer. Factors related to erythrocyte aggregation were also determined: erythrocyte sedimentation rate (ESR), zeta sedimentation ratio (ZSR), and plasma fibrinogen levels. Erythrocyte deformability was observed at various fluid shear stress levels, with a laser diffraction ektacytometer. Correlations were sought between each of these measures and current or nadir CD4⁺ T-lymphocyte counts, and each measure was compared between three subgroups based on current and nadir CD4⁺ T-lymphocyte counts (severely immunosuppressed, immune reconstituted, never severely immunosuppressed).

results. The following parameters were significantly different between HIV-infected subjects and controls: increased erythrocyte aggregation, at stasis (P < 0.001) and low shear stress (P < 0.001), increased ESR (P < 0.001), increased ZSR (P < 0.028), increased serum fibrinogen (P = 0.015), and decreased erythrocyte deformability (P < 0.001). Only erythrocyte aggregation at stasis correlated significantly with current CD4⁺ T-lymphocyte count (r = − 0.344, P = 0.022). None of the parameters was significantly different between HIV-infected subgroups.

conclusions. Increased aggregation and decreased deformability of erythrocytes are associated with HIV-infection regardless of the severity of immunodeficiency. HIV-infected individuals may be at risk for progressive retinal microvascular damage from persistent hemorheologic abnormalities, despite immune reconstitution associated with potent antiretroviral drug therapies.

Ultrastructural abnormalities of the retinal microvasculature are considered to be a universal finding among people with human immunodeficiency virus (HIV) infection. ^{1 2} These changes include pericyte degeneration and loss, thickening of the basal lamina, and narrowing and occlusion of vascular lumina. ¹ In addition, alterations in blood flow are common, as manifested by sludging of flow in conjunctival vessels, ^{3 4} by decreased retinal perifoveal capillary blood flow on scanning laser ophthalmoscopic fluorescein angiography, ⁵ and by decreased velocity of leukocytes in macular vessels. ⁶ Although anatomic abnormalities of the microvasculature may contribute to these hemorheologic alterations, hematologic factors are also believed to influence the dynamics of flow. Among HIV-infected individuals, these factors include increased plasma fibrinogen, which is a determinant of erythrocyte aggregation, ³ and increased leukocyte rigidity. ^{7 8} Hemorheolgic abnormalities can damage the microvasculature further through diminished oxygen exchange and hypoxia. 

In severely immunosuppressed individuals, the combination of microvasculopathy and hemorheologic abnormalities results in the clinical signs of HIV retinopathy (cotton-wool spots, retinal hemorrhages), which resembles the clinical appearance of diabetic retinopathy. ³ It has been hypothesized that these changes may contribute to visual disturbances, such as reduced contrast sensitivity and altered color vision, among HIV-infected individuals without opportunistic infections of the retina. ^{9 10}  

Increased leukocyte rigidity and decreased velocity of leukocytes through macular vessels appear to persist, even with immune reconstitution attributable to potent antiretroviral therapies. ^{6 8} Because erythrocyte aggregation is the primary factor responsible for increased blood viscosity and is a major factor in blood flow resistance at low flow rates, ¹¹ we sought to determine whether erythrocyte aggregation also is independent of the severity of immunodeficiency, as reflected in CD4⁺ T-lymphocyte counts. Furthermore, we were interested in determining whether decreased erythrocyte deformability, which is known to affect tissue perfusion adversely, ¹² was associated with HIV infection. Understanding these hemorheologic changes may provide insight into the pathogenesis of HIV-related microvascular disease and its clinical sequelae. 

Methods

Adult HIV-infected subjects were recruited from the Jules Stein Eye Institute and the Center for Clinical AIDS Research and Education (CARE) Clinic, UCLA Medical Center, and from Community Eye Medical, a private ophthalmology practice in Pasadena, California, without regard to specific age, gender, race and ethnicity, or the presence or absence of clinically apparent HIV-related ocular disease. Adult HIV-negative control subjects were recruited from the Jules Stein Eye Institute and from the Department of Physiology and Biophysics, USC Keck School of Medicine (Los Angeles, CA), without regard to specific age, gender, or race and ethnicity. Studies on HIV-infected subjects and HIV-negative control subjects were conducted concurrently. All subjects were recruited from June 2002 through March 2003. Excluded were individuals with diabetes mellitus, hypertension, any form of vasculitis, rheumatologic disease, tobacco use, or blood transfusion within the past 3 months. None of the HIV-infected subjects in this study were participants in our previous studies of hemorheologic abnormalities. The study adhered to the tenets of the Declaration of Helsinki and was approved by the UCLA and USC Institutional Review Boards. All participants provided written informed consent. 

The following clinical and laboratory data were obtained from medical records for each HIV-infected subject: use of antiretroviral medications, CD4⁺ T-lymphocyte count within the 3 months before the study (current CD4⁺ T-lymphocyte count), and lowest previous (nadir) CD4⁺ T-lymphocyte count. 

Blood specimens (20 mL) were obtained from all subjects by venipuncture with a 19-gauge needle and vacuum tubes containing sodium citrate (10 U/mL) or EDTA (1.5 mg/mL). Erythrocyte mean cell volume (MCV) was measured with an automated analyzer (ABX Micros, Irvine, CA) that was routinely calibrated with standards supplied by the manufacturer. Hematocrit was determined using the microhematocrit method (12,000g, 4 minutes). Fibrinogen concentration was determined on frozen citrated plasma Specialty Laboratories (Santa Monica, CA). 

Measurements of Erythrocyte Aggregation

The hematocrit of whole blood drawn into tubes containing EDTA was adjusted to 40% ± 1% by addition or removal of autologous plasma, then used to determine the following parameters: erythrocyte aggregation, Westergren erythrocyte sedimentation rate (ESR), and the zeta sedimentation ratio (ZSR). Erythrocyte aggregation was quantified using a photometric rheoscope (Myrenne Aggregometer; Myrenne GmbH, Rötgen, Germany) at stasis (M) and at a low shear rate (3/s; M1), as previously described. ¹³ ZSR, a modified sedimentation test, was performed using a specialized low-speed inclined blood centrifuge (Zetafuge; Coulter Electronics, Hialeah, FL), in which slightly increased gravitational forces (7–8g) promote sedimentation of whole blood. ZSR has a truncated range of 0.4 to 1.0. A ZSR of 0.4 (corresponding to the 40% hematocrit of the tested erythrocyte suspension) indicates a lack of aggregation, whereas a ZSR of 1.0 represents aggregation of such intensity that the low-g hematocrit equals the true hematocrit. ZSR was determined using previously published protocols. ¹⁴ For each of the parameters, larger values reflect increased erythrocyte aggregation. All data represent averages of duplicate measurements. 

Measurements of Erythrocyte Deformability

Erythrocyte deformability was quantitated at various shear stress (i.e., deforming force) levels using a laser-diffraction ektacytometer system (Laser-Assisted Optical Rotational Cell Analyzer [LORCA]; RR Mechatronics, Hoorn, The Netherlands). ¹⁵ This device consists of a laser diode, a concentric cylinder shearing section containing a low-hematocrit suspension of erythrocytes in an isotonic, viscous polymer solution, a stepper motor to rotate the outer cylinder, and a video camera connected to a microcomputer. The microcomputer controls the speed of the motor and analyzes the diffraction that results when the laser passes through the erythrocyte suspension being sheared. The shape of the laser diffraction pattern is analyzed to obtain an elongation index (EI), determined as: EI = (L − W)/(L + W), where L is the length of the image and W is its width. The EI was determined at room temperature over a range of shear stress levels (1.58, 2.81, 5, 8.89, and 49.71 Pa). An increased EI represents greater cell deformation. All data represent averages of duplicate measurements. 

Subgroup Analyses

We compared erythrocyte aggregation, related factors, and erythrocyte deformability between three subgroups of HIV-infected subjects, based on current and nadir CD4⁺ T-lymphocytes values, as defined in our previous studies of hemorheologic abnormalities associated with HIV disease. ^{6 8} Immunosuppressed subjects were those with a current CD4⁺ T-lymphocyte count <50 cells/μL. Immune-reconstituted subjects were those with a nadir CD4⁺ T-lymphocyte count <50 cells/μL and a current CD4⁺ T-lymphocyte count >200 cells/μL. Never severely immunosuppressed subjects were those whose CD4⁺ T-lymphocyte count was never below 200 cells/μL. 

Results were also compared among the following three subgroups based on antiretroviral drug use: individuals not using antiretroviral drugs, individuals using zidovudine or stavudine (either alone or in combination, with or without other antiretroviral drugs), and individuals using antiretroviral drugs other than zidovudine or stavudine. 

Statistical Analysis

Statistical analyses were performed using statistical software SAS version 9.1 (SAS Institute, Cary, NC). P < 0.05 was considered to be statistically significant. All comparisons between HIV-infected subjects and HIV-negative control subjects were conducted using Student’s t-test, allowing for unequal variances. For HIV-infected individuals, Pearson correlation and/or linear regression analyses were used to evaluate relationships between hemorheologic parameters and the following factors: current and nadir CD4⁺ T-lymphocyte counts, HIV RNA blood level, and MCV. Differences among subgroups of HIV-infected subjects were assessed using analysis of variance (ANOVA), and pair-wise comparisons between any two subgroups were corrected with the Bonferroni method. 

Results

Studies were performed on 46 HIV-infected individuals and 44 HIV-negative control subjects. Among the HIV-infected individuals, six were considered severely immunosuppressed, 10 were considered immune reconstituted, and 20 were considered never severely immunosuppressed. The remaining 10 subjects did not meet our inclusion criteria for these three subgroups and were not considered in subgroup analyses dealing with immune status. Potent antiretroviral drugs were being used by 40 HIV-infected subjects, 18 of whom were using zidovudine, stavudine, or both. The remaining six subjects were not using antiretroviral drugs. 

Age, laboratory values, and antiretroviral drug use are summarized for all subjects in Table 1 . Although the mean age of the controls was significantly lower than the mean age of the HIV-infected subjects, the absolute difference in age was small. In previous studies, we have found that age differences of this magnitude have no meaningful effect on fibrinogen levels and erythrocyte aggregation. ¹³  

Table 2summarizes laboratory values related to erythrocyte aggregation and deformability. All values related to erythrocyte aggregation were significantly elevated in the HIV-infected group. Erythrocyte deformability was significantly reduced in HIV-infected individuals at all shear stress levels (Table 2 , Fig. 1 ). 

Among HIV-infected individuals, erythrocyte aggregation, as determined by the Myrenne Aggregometer, correlated inversely with current CD4⁺ T-lymphocyte counts both at stasis (M; r = − 0.344, P = 0.022) and at a low shear rate (M1; r = − 0.233, P = 0.128), indicating that erythrocyte aggregation was greater for individuals with lower CD4⁺ T-lymphocytes counts. Such a relationship was not seen for the other factors related to erythrocyte aggregation (ZSR, r = − 0.177, P = 0.251; ESR, r = − 0.248, P = 0.105,) or for plasma fibrinogen levels (r = − 0.231, P = 0.131). No significant Pearson correlations were seen between erythrocyte aggregation and nadir CD4⁺ T-lymphocyte count (P ≥ 0.226) or HIV blood level (P ≥ 0.376). 

There was no association between erythrocyte deformability and current CD4⁺ T-lymphocyte count (all P > 0.171) or HIV blood level (all P > 0.643). Erythrocyte deformability in HIV-infected individuals was inversely correlated with nadir CD4⁺ T-lymphocyte counts at a moderate shear stress (8.89 Pa; r = − 0.399, P = 0.053), but not at the other shear stress levels (all P > 0.105). Nadir CD4⁺ T-lymphocyte counts were available for only 24 subjects, raising the possibility of an effect of selection bias on results. 

Table 3shows results for HIV-infected subgroups based on immune status. There were no significant differences for any of the measures, when all subgroups were compared by ANOVA. Furthermore, pair-wise comparisons with Bonferroni correction also did not reveal any significant differences between subgroups. 

Because MCV was significantly different between HIV-infected and control subjects (Table 1) , we investigated the relationship between MCV and erythrocyte deformability. We excluded two HIV-infected subjects and one control subject because of low MCV values that were statistical outliers. For HIV-infected individuals, EI was inversely related to MCV at all shear stress levels (all P ≤ 0.0013). Similar relationships were not identified for HIV-negative control subjects (all P ≥ 0.061). 

Table 4shows that MCV was different between subjects grouped by antiretroviral drug use, when all subgroups were compared by the ANOVA technique. In pair-wise comparisons, HIV-infected individuals taking any antiretroviral drugs had higher MCV than those not taking antiretroviral drugs (P = 0.03). Those taking zidovudine or stavudine had higher MCV than those taking other antiretroviral drugs (P = 0.019; P = 0.068 after adjustment for multiple comparisons). Those taking antiretroviral drugs, but not zidovudine or stavudine, had higher MCV than those not taking antiretroviral drugs (P = 0.05; P = 0.433 after adjustment for multiple comparisons). 

With regard to erythrocyte deformability, antiretroviral drug use also had an effect on EI, with a pattern inverse to that seen with MCV. Antiretroviral drug treatment, in particular treatment with zidovudine and stavudine, was associated with decreased EI. With adjustment for multiple comparisons, subjects being treated with zidovudine or stavudine had significantly reduced EI when compared with subjects taking other antiretroviral drugs (P ≤ 0.0052 at all shear stress levels) and when compared with control subjects (P < 0.0001 at all shear stress levels). Subjects taking zidovudine or stavudine also had lower EI than HIV-infected subjects not taking antiretroviral drugs, but the difference was statistically significant only at shear stress levels of 5 and greater (P ≤ 0.046). 

We could not confirm statistically that HIV infection in the absence of antiretroviral drug treatment was associated with higher MCV (P = 0.949 after correction for multiple comparisons) or decreased EI (P ≥ 0.428 at all shear stress levels), but the sample size was small (n = 6). The sample size was also too small to determine whether there was an inverse relationship between MCV and EI for HIV-infected subjects who were not taking any antiretroviral drugs. 

The same pattern of results presented in Table 4for MCV and EI was seen when only subjects with CD4⁺ T-lymphocyte counts >200/μL were grouped on the basis of antiretroviral drug treatment (data not shown). 

Discussion

Altered blood flow is believed to contribute to microvascular disease, and thus, studying the rheologic behavior of formed elements in the blood may provide insight into the pathogenesis of the microvascular abnormalities associated with HIV disease. ^{1 2 3 6 7 8} Erythrocyte aggregation and deformability, hematocrit, and plasma viscosity all influence in vivo blood flow. ¹⁶ Fibrinogen is the plasma protein most responsible for cell–cell attraction that results in erythrocyte aggregation. ^{3 17} The effects of hemorheologic abnormalities are most significant in the microcirculation, primarily due to the similarity of erythrocyte diameter and aggregate size to vessel diameters. 

Our previous study of erythrocytes and microvascular disease in HIV-infected individuals ³ was undertaken before the availability of potent antiretroviral agents and did not take into account the subjects’ level of immunodeficiency, as reflected by CD4⁺ T-lymphocyte count. Subsequent studies have shown that leukocyte rigidity, another determinant of blood flow, as well as leukocyte velocity through macular vessels in vivo, are abnormal in subjects who are immune reconstituted. ^{6 8} We therefore chose to investigate the relation between erythrocyte aggregation and CD4⁺ T-lymphocyte counts in HIV-infected individuals and to extend our studies to include erythrocyte deformability. Our findings suggest that factors other than the level of immune function may be responsible for the increased erythrocyte aggregation and decreased erythrocyte deformability seen in HIV-infected individuals, and that these abnormalities do not improve with immune reconstitution. 

When suspended in plasma, erythrocytes have the ability to form large, linear rouleaux aggregates that can disturb flow streamlines, because of their increased size when compared with an individual erythrocyte. ¹⁸ The effect of erythrocyte aggregation is particularly apparent under conditions of low shear stress and low flow. Increased erythrocyte aggregation has been associated with inflammation and can be seen in conditions such as diabetes mellitus, ^{19 20 21} hypertension, ²² ischemic heart disease, ²³ ischemic stroke, ²⁴ sepsis, ²⁵ and ischemia–reperfusion injury. ²⁶ Pathologically increased erythrocyte aggregation may compromise capillary tissue perfusion and oxygen delivery, ¹⁶ resulting in ischemia. 

The most important plasma protein to promote erythrocyte aggregation and increase plasma viscosity is the acute-phase reactant fibrinogen. ^{11 17} The mean fibrinogen level for our HIV-infected subjects was within the normal range (364 mg/dL; normal, 200–400 mg/dL), but it was significantly greater than the mean fibrinogen level for our control subjects (P = 0.015). Fibrinogen levels have been associated with the severity of HIV disease, based on assessments of conjunctival microvascular changes, with sludging of blood flow through capillary vessels, and with the presence of cotton-wool spots. ³  

Erythrocyte aggregation can be assessed by several methods, including the assessment of light transmission through erythrocyte suspensions, as performed with the aggregometer (Myrenne GmbH), and the measurement of sedimentation rates. Higher sedimentation rates are associated with an elevated concentration of plasma acute-phase reactants and represent greater erythrocyte aggregate formation. The fact that results from each of the several tests used in this study were consistent with increased erythrocyte aggregation in HIV-infected individuals argues against experimental or procedural artifacts. 

We found that erythrocyte deformability was significantly lower in HIV-infected individuals across a wide range of shear stress levels. In the microvasculature, erythrocytes must deform to enter and traverse vessels with lumina narrower than the resting diameter of the cell. ²² Evidence suggests that the reduced erythrocyte deformability in sepsis, ^{27 28} cardiovascular disorders, ²⁹ and hypertension ²² may impair the passage of erythrocytes through such small vessels and thus compromise oxygen delivery. The major determinants of erythrocyte deformability are the geometric features of the cell (membrane surface area-to-cell volume ratio, cell shape), rheologic properties of the intracellular fluid (hemoglobin concentration, and hence viscosity), and the rheologic properties of the cell membrane (elastic and bending modulus, membrane viscosity). ¹²  

The cause of the altered erythrocyte properties in HIV-infected individuals remains unknown, but is probably multifactorial. Erythrocyte deformability was inversely related to the MCV in HIV-infected individuals, and MCV was significantly greater in HIV-infected individuals than in control subjects. Certain nucleoside reverse transcriptase inhibitors commonly used as antiretroviral drugs, including zidovudine and stavudine, have been associated with macrocytosis, defined as MCV ≥ 100 fL. ^{30 31} Our study confirmed this relationship, which was found to be independent of CD4⁺ T-lymphocyte count. The use of these two drugs was not the only cause of macrocytosis, however; MCV was also significantly elevated in those subjects taking other antiretroviral drugs when compared with HIV-negative controls. The effect of zidovudine and stavudine, and to a lesser extent other antiretroviral drugs, on MCV and EI was not an indirect reflection of disease severity (and thus a need for antiretroviral drug therapy), as the associations were also seen in patients with high CD4⁺ T-lymphocyte counts. We could not determine whether macrocytosis was solely an effect of antiretroviral drug therapy, but our results suggest a small additional effect of HIV infection alone. Although not statistically significant, EIs were higher (i.e., more deformable erythrocytes) in the control subjects at every shear stress level than in HIV-infected subjects not taking any antiretroviral drugs (Table 4) ; the lack statistical significance may reflect inadequate power. By our calculations, the power to confirm the observed differences statistically ranged from only 17% to 36%, depending on the shear stress level at which comparisons were made. The relationship between HIV infection, MCV, and EI should be explored further in a larger group of subjects before antiretroviral drug therapy is begun. 

Activated polymorphonuclear leukocytes (PMNs) have been reported to induce structural and functional changes in neighboring erythrocytes. ^{32 33 34} Increased erythrocyte aggregation has been demonstrated after incubation with activated PMNs, attributable to changes in erythrocyte surface properties. ³³ Activation causes increased PMN rigidity and secretory activity, with the production and release of chemotactic agents, oxygen free radicals, and proteolytic enzymes. ³⁵ As a result, there is increased lipid peroxidation of erythrocyte membranes ³⁴ and alterations in erythrocyte membrane cytoskeletons, with increased cross-linking between spectrin and hemoglobin, ³² making the cells less deformable. Experimental studies show these effects can be reduced both by antioxidant enzymes and inhibitors of proteolytic enzymes. ^{32 33} Enhanced PMN activation has been demonstrated in HIV-infected individuals. ³⁶ Goldenberg et al. ⁸ showed that the PMN rigidity in HIV-infected subjects was unrelated to CD4⁺ T-lymphocyte counts. Persistent activation of PMN may thus play a role in the increased aggregation and decreased deformability of erythrocytes in HIV-infected individuals. 

In summary, increased erythrocyte aggregation and decreased erythrocyte deformability are features of HIV disease. These changes appear to be unrelated to an individual’s level of immunodeficiency. Thus, they are not expected to improve with immune reconstitution. In fact, some antiretroviral agents may make these changes worse, by causing macrocytosis. Increased erythrocyte aggregation and decreased erythrocyte deformability can be expected to alter blood flow, eventually placing patients at risk for the sequelae of retinal microvascular disease. 

²  Contributed equally to the work and therefore should be considered equivalent authors.

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Ft. Lauderdale, Florida, May 2003.

 Supported in part by Research to Prevent Blindness (RPB), Inc. (GNH), the Skirball Foundation (GNH), National Heart Lung and Blood Institute Grants HL15722 and HL70595 (HJM), National Eye Institute Grant EY08057 (GNH), the David May II Endowed Professorship (GNH), the Emily Plumb Estate, and the Trust Gift for resources utilized in the Jules Stein Eye Institute Clinical Research Center. GNH is the recipient of an RPB Physician-Scientist Award.

 Submitted for publication February 7, 2006; revised April 8, 2006; accepted July 12, 2006.

 Disclosure: A. Kim, None; H. Dadgostar, None; G.N. Holland, None; R. Wenby, None; F. Yu, None; B.G. Terry, None; H.J. Meiselman, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Characteristics of HIV-Infected Individuals

	HIV-Negative Control Subjects	All HIV-Infected Subjects	P *	Subgroups Based on Immune Function
								Severely Immunosuppressed, †	Immune Reconstituted, ‡	Never Severely Immunosuppressed, §
Subjects (n)	44	46		6	10	20
Hematocrit (%; mean ± SD)	44 ± 2.5	42 ± 5.8	0.19	41 ± 4.6	42 ± 7.4	44 ± 5.7
MCV (fL; mean ± SD)	89 ± 6.1	98 ± 10.8	<0.0001	95 ± 5.5	101 ± 17.0	98 ± 9.6
Age (y; mean ± SD)	38 ± 10.9	45 ± 7.5	0.0007	43 ± 4.2	47 ± 7.6	44 ± 7.2
CD4+ T-lymphocyte counts (cells/μL)
Current, ∥
Median	NE	368	—	14.5	348	610.5
Range	NE	10–1239		10–41	212–556	219–1239
Nadir, ¶
Median	NE	25	—	10	5	NE
Range	NE	2–57		6–28	2–50	>200
Current HIV blood level (×103 copies/μL), ∥
Median	NA	0.798	—	163.0	0.339	0.358
Range	NA	ND-2100		ND-2100	ND-9.48	ND-232
Use of antiviral drugs, n (%)	NA	40 (87%)	—	6 (100%)	10 (100%)	15 (75%)

 NA, not applicable; NE, not examined; ND, nondetectable.

*  Student’s t-test comparisons between all HIV-infected subjects and control subjects.

†  Current CD4⁺ T-lymphocyte count <50 /μL.

‡  Current CD4⁺ T-lymphocyte count >200/μL and nadir CD4⁺ T-lymphocyte count <50/μL.

§  CD4⁺ T-lymphocyte always >200/μL.

∥  Current values were determined within 3 months before the determination of hemorheologic factors.

¶  Lowest previously measured CD4⁺ T-lymphocyte count at any point before determination of hemorheolic factors.

Table 2.

 

Measures of Erythrocyte Aggregation and Deformability

	All HIV-Infected Subjects	HIV-Negative Control Subjects	P *
Number of subjects	46	44
Fibrinogen (mg/dL; mean ± SD)	364 ± 151	299 ± 88	0.015
Erythrocyte sedimentation rate (mm/h; mean ± SD)	31 ± 28	10 ± 7.9	<0.001
Zeta sedimentation ratio (mean ± SD)	0.64 ± 0.11	0.60 ± 0.06	0.028
Aggregation, †
M (mean ± SD), ‡	26.2 ± 6.8	20.7 ± 5.7	<0.001
M1 (mean ± SD), §	43.4 ± 10.5	33.9 ± 7.8	<0.001
Deformability (EI at various shear stress levels [Pa; mean ± SD])
1.58	0.14 ± 0.04	0.17 ± 0.03	<0.001
2.81	0.23 ± 0.05	0.27 ± 0.03	<0.001
5.00	0.32 ± 0.04	0.36 ± 0.03	<0.001
8.89	0.40 ± 0.04	0.44 ± 0.03	<0.001
49.7	0.53 ± 0.04	0.56 ± 0.02	<0.001

 EI, elongation index.

*  Student’s t-test.

†  Determined using a fully automatic erythrocyte aggregometer (Myrenne Aggregometer; Myrenne GmbH).

‡  Value determined at stasis.

§  Value determined at 3/s shear rate.

Figure 1.

 

Elongation index versus shear stress for HIV-infected subjects and HIV-negative control subjects. The elongation index is a measure of cellular deformability. It was measured at five shear stress levels (1.58, 2.81, 5.00, 8.89, and 49.71 Pa). The two curves are significantly different (P < 0.001, ANOVA).

Table 3.

 

Comparison of Erythrocyte Aggregation and Deformability between Patients Grouped by Level of Immune Function

	Severely Immunosuppressed	Immune Reconstituted	Never Severely Immunosuppressed	P *
Subjects (n)	6	10	20
Fibrinogen (mg/dL; mean ± SD)	377 ± 233	370 ± 121	328 ± 127	0.66
Erythrocyte sedimentation rate (mm/h; mean ± SD)	35 ± 28	32 ± 30	25 ± 22	0.62
Zeta sedimentation ratio (mean ± SD)	0.62 ± 0.09	0.62 ± 0.15	0.65 ± 0.10	0.71
Aggregation, †
M (mean ± SD), ‡	29.6 ± 5.3	23.0 ± 6.4	24.9 ± 6.0	0.12
M1 (mean ± SD), §	46.7 ± 10.3	38.4 ± 10.4	42.9 ± 9.5	0.26
Deformability (EI at various shear stress levels [Pa; mean ± SD])
1.58	0.13 ± 0.02	0.14 ± 0.03	0.14 ± 0.05	0.93
2.81	0.23 ± 0.03	0.23 ± 0.03	0.23 ± 0.06	0.90
5	0.32 ± 0.02	0.33 ± 0.03	0.32 ± 0.06	0.81
8.89	0.40 ± 0.02	0.40 ± 0.04	0.39 ± 0.05	0.83
49.7	0.53 ± 0.02	0.52 ± 0.03	0.52 ± 0.04	0.76

 Levels of immune function are as defined in Table 1 .

*  ANOVA.

†  Determined using a fully automatic erythrocyte aggregometer (Myrenne Aggregometer; Myrenne GmbH).

‡  Value determined at stasis.

§  Value determined at 3/s shear rate.

Table 4.

 

Comparison of Erythrocyte Deformability between Patients Grouped by Antiretroviral Drug Use

	HIV-Negative Control Subjects	HIV-Infected Subjects					P *
				No Antiretrovirals	Antiretrovirals Excluding Zidovudine, Stavudine	Antiretrovirals Including Zidovudine, Stavudine
Subjects (n)	44	6	22	18
MCV (fL; mean ± SD)	89 ± 6.1	91 ± 6.5	97 ± 4.9	103 ± 14.7	<0.001
Deformability (EI at various shear stress levels [Pa; mean ± SD])
1.58	0.17 ± 0.03	0.15 ± 0.05	0.15 ± 0.02	0.11 ± 0.04	<0.001
2.81	0.27 ± 0.03	0.25 ± 0.06	0.24 ± 0.03	0.20 ± 0.05	<0.001
5.00	0.36 ± 0.03	0.34 ± 0.06	0.34 ± 0.02	0.29 ± 0.05	<0.001
8.89	0.44 ± 0.03	0.42 ± 0.04	0.41 ± 0.02	0.38 ± 0.05	<0.001
49.7	0.56 ± 0.02	0.55 ± 0.02	0.54 ± 0.02	0.50 ± 0.04	<0.001

 EI, elongation index.

*  ANOVA.