(Circulation. 1997;96:4380-4384.)
© 1997 American Heart Association, Inc.
Articles |
From the Division of Rheumatology, Department of Medicine (M.H.E.), the Department of Surgery (J.H.B.), and the Department of Physiology (G.A.), University of Louisville (Ky); and Morehouse School of Medicine (S.P., X.-W.L., E.N.H.), Atlanta, Ga.
Correspondence to Michael H. Edwards, MD, Medicine/Rheumatology, University of Louisville, Louisville, KY 40292.
| Abstract |
|---|
|
|
|---|
Methods and Results Three groups of nine mice were injected with purified immunoglobulin G (IgG) from a patient with the antiphospholipid syndrome (IgG-APS) and then fed with hydroxychloroquine at various doses (100, 6, and 3 mg/kg body wt). Three control groups of mice were also studied, including mice injected with IgG-APS and then fed with placebo, as well as two other groups injected with IgG from normal human serum and fed either hydroxychloroquine or placebo. A standardized thrombogenic injury was subsequently induced in the femoral vein of each mouse and the area (size) of thrombus measured as well as the total period of time that thrombus was present. Mice treated with hydroxychloroquine and IgG-APS showed significantly smaller thrombi that persisted for a shorter period of time compared with animals treated with IgG-APS and placebo.
Conclusions Hydroxychloroquine significantly diminished both thrombus size and total time of thrombus formation in mice previously injected with IgG-APS.
Key Words: thrombus antibodies anticoagulants drugs hydroxychloroquine
| Introduction |
|---|
|
|
|---|
Currently, warfarin is generally the long-term anticoagulant of choice for patients with known APS.3 4 If effective, hydroxychloroquine could be a much safer alternative to warfarin as prophylaxis against thrombosis in patients with this syndrome. However, convincing evidence that hydroxychloroquine can be effective (and the mechanism by which it might work) remains unclear.
We have recently devised a mouse model of APS in which human polyclonal and monoclonal anticardiolipin antibodies have been demonstrated to increase both the size of an induced thrombus and the duration of time that the clot persists.12 13 14
This study was performed to determine whether administration of hydroxychloroquine to mice preimmunized with immunoglobulin G (IgG) from a patient with APS (IgG-APS) might result in reduction of thrombus formation. Further experiments were performed to determine whether the effects of hydroxychloroquine are dose dependent.
| Methods |
|---|
|
|
|---|
Hydroxychloroquine
Hydroxychloroquine sulfate (powder form) was kindly provided by
Dr Charles Nicol (Sanofi-Winthrop Pharmaceuticals; New York, NY).
Hydroxychloroquine suspensions were prepared in 1% tragacanth gum
suspension (Sigma Chemical Co) at the desired concentration. Three
concentrations of hydroxychloroquine were used in this study: 100, 6,
and 3 mg/kg body wt.
Isolation of Immunoglobulin G From an APS Patient
IgG was isolated from the serum of a patient with the
antiphospholipid syndrome (IgG-APS) with the use of DEAE ion exchange
chromatography as described elsewhere.15
Purity was determined by SDS-PAGE (single band at 150 kD), and the
absence of contamination with ß2GP1 was ruled out by
immunoblot with a rabbit anti-human ß2GP1
antiserum. Protein concentration was determined by the method of Lowry.
After adjustment of the protein concentration with a sterile saline
solution, these preparations were filter-sterilized before injections.
Similarly, IgG from normal healthy controls was obtained (IgG-NHS).
Determination of Hydroxychloroquine Levels in Mouse Blood
To ensure that hydroxychloroquine was being absorbed through the
gastrointestinal tract and that significant blood levels could be
achieved, 0.5 to 1.0 mL of whole blood was obtained from additional
groups of mice
2 hours after being fed with a variable dose of
the drug. All whole blood samples were sent to the Pharmacokinetics
Drug Analysis Laboratory at North Dakota State University for
determination of hydroxychloroquine concentrations as described
elsewhere.16
Experimental Design
To achieve high serum levels of anticardiolipin antibodies in
each mouse, intraperitoneal injections of IgG-APS
or equivalent amounts of IgG-NHS were administered at times 0 and 48
hours, as described elsewhere.12
Kinetic experiments were initially performed to determine an
appropriate time that oral hydroxychloroquine should be given. Because
levels of hydroxychloroquine were found to reach a plateau within 1 to
2 hours after feeding (Figure
), surgery,
as described below, was performed 2 hours after oral administration of
the drug. This was
72 hours after the first
intraperitoneal injection of IgG.
|
Four groups of mice were studied, each group consisting of nine animals. Group A was injected intraperitoneally with IgG-APS and fed with hydroxychloroquine (100, 6, or 3 mg/kg body wt); group B was injected intraperitoneally with IgG-APS and fed with placebo (1% tragacanth gum suspension); group C was injected intraperitoneally with IgG-NHS and fed with hydroxychloroquine, and group D was injected with IgG-NHS and fed with placebo.
Determination of Anticardiolipin Antibody Levels by ELISA
Human anticardiolipin antibody levels in immunized mice were
determined with an ELISA assay as described
elsewhere.12 13 17 Alkaline phosphatase anti-human IgG
sera were used as secondary antibodies in the ELISA system. The color
reaction was stopped when a positive control (of
100 GPL units)
reached 1.0 OD units (20 to 30 minutes). The levels of anticardiolipin
antibodies were determined in GPL units by extrapolation of a curve
constructed with calibrators (Louisville APL Diagnostics,
Inc).
Surgical Procedure and Measurement of the Dynamics of In Vivo
Thrombus Formation
To determine the dynamics of thrombus formation, we used a
modification of the surgical procedure described
previously.12 13 14 18 This procedure enables continuous and
quantitative measurements of a standardized, focally induced,
nonocclusive mural thrombus in a surgically exposed mouse femoral
vein.
Using sodium pentobarbital (60 mg/kg IP) as anesthetic, a longitudinal incision was made in the right groin of the CD-1 mice. A standardized thrombogenic injury was produced in the vein with a standard pinch injury, as described elsewhere.12 13 18 A transilluminator was used to visualize the vein, and a stereoscopic operating microscope, equipped with a closed-circuit video system and a recorder, was used to visualize and observe the induced thrombus.
Measurements were made of the total period of time (minutes) that thrombus was present. Thrombus area was also measured (µm2). Three to five thrombi were induced in each animal, and mean thrombus area and time were computed for each group. The person (X.-W.L.) performing the surgery and measurements was blinded as to what treatment had been given to each animal.
Statistical Analysis
One-way ANOVA was used to compare the means of thrombus sizes
and times of thrombus duration for the four groups. Tukey's honest
significant difference was used for post hoc analysis of the
means. Data were checked for normality with the Kolmogorov-Smirnov
test. The null hypothesis of normality was not rejected.
| Results |
|---|
|
|
|---|
The animals in groups A and B, injected with IgG-APS, had high levels
of anticardiolipin antibodies immediately before the surgical
experiments. Mean levels of anticardiolipin antibodies in mice from
both groups was >100 GPL units. Animals in groups C and D, injected
with IgG-NHS, did not have detectable levels of anticardiolipin
antibodies. Animals treated with hydroxychloroquine (groups A and C)
had high levels of the drug in their blood at 2 hours after ingestion
(Fig 1
), whereas animals from groups B and D, both given placebo,
showed levels <10 ng/mL (data not shown).
Animals treated with IgG-APS and placebo produced significantly larger
thrombi that also persisted longer than animals treated with IgG-NHS
and placebo, indicating that the IgG-APS enhanced thrombus formation,
as shown previously12 (Table 1
). Thrombus size and the time that
thrombus persisted in animals treated with IgG-APS and
hydroxychloroquine (group A) were significantly reduced compared with
animals treated with IgG-APS and placebo (group B). There was no
significant difference in thrombus size and thrombus duration between
IgG-NHS mice treated with hydroxychloroquine versus placebo (groups C
and D). In addition, values obtained with
IgG-APShydroxychloroquinetreated animals (group A) were not
significantly different from IgG-NHShydroxychloroquinetreated
animals (group C) (Table 1
). These results indicate that a high dose of
hydroxychloroquine prevented the enhanced thrombus formation associated
with IgG anticardiolipin antibodies.
|
Effect of Different Doses of Hydroxychloroquine on the Dynamics of
Thrombus Formation
In a second series of experiments, animals were injected with
IgG-APS or IgG-NHS as before, but treated with lower doses of
hydroxychloroquine (6 and 3 mg/kg). At 2 hours after administration of
the drug, whole blood levels of hydroxychloroquine in mice treated with
6 mg/kg were
230 ng/mL, whereas in mice treated with 3 mg/kg, levels
were
50 ng/mL. As shown in Table 2
,
hydroxychloroquine given at doses of 6 mg/kg to IgG-APSinjected mice
was equally effective in reducing thrombus size and the time that
thrombus persisted compared with mice treated with 100 mg/kg
hydroxychloroquine as described above. However, with the 3 mg/kg dose
of hydroxychloroquine, thrombus sizes were significantly larger
compared with groups treated with 6 and 100 mg/kg of the drug, though
still smaller than the group fed placebo. Total time of thrombus
duration was similar in all three treated groups. Thus while the
kinetics of thrombus formation was not significantly affected by the
dose of hydroxychloroquine, thrombus size may be altered in a
dose-dependent fashion.
|
| Discussion |
|---|
|
|
|---|
Clinically, there is evidence that hydroxychloroquine may have a role to play in the prevention of thromboembolic disease, though this has remained controversial. The incidence of deep venous thrombosis was found to be 5% in 107 general surgical patients who received hydroxychloroquine in the perioperative period compared with an incidence of 16% in a similar control group.5 In the 1970s, Sir John Charnley and colleagues became enthusiastic supporters of the use of hydroxychloroquine as prophylaxis against thromboembolism in patients undergoing hip replacement. In their series, a significant reduction in the incidence of postoperative pulmonary embolism was observed in patients receiving anywhere from 600 to 1600 mg of hydroxychloroquine daily, beginning the day before surgery and continuing until the time of hospital discharge.6 21 Unfortunately, these large case series were not rigorously tested in a double-blind, prospective trial. One such double-blind, randomized trial involving hydroxychloroquine versus placebo being given to patients after elective hip surgery failed to reveal any significant difference in the incidence of thromboembolism between the two groups.22
There are several reports that indicate that antimalarial drugs such as chloroquine and hydroxychloroquine have a possible antithrombotic effect. Platelet aggregation induced by ADP, collagen, and ristocetin can be inhibited by chloroquine, and the inhibition is dose dependent.23 24 More recently, chloroquine was found to inhibit the release of arachidonic acid from the membrane phospholipids of thrombin-stimulated platelets.25 This appears to be mediated through inhibition of activated phospholipase A2 by chloroquine. Quinacrine, a related antimalarial compound, has also been identified as an inhibitor of phospholipase A2 and similarly blocks thrombin-induced secretion and release of arachidonic acid from platelet phospholipids.26 27 This inhibition can be overcome by the addition of lysophosphatidic acid, which is normally formed as a result of the action of phospholipase A2 on phosphatidic acid.27
In addition to their effect on platelets, antimalarial drugs also appear to inhibit intravascular aggregation of erythrocytes. This "desludging" effect has been well described in the treatment of malaria.7 However, it has also been noted to occur in the vessels of the bulbar conjunctiva in patients with other vascular disease processes as well as in the retinal veins of rheumatoid arthritis patients.8 28 In spite of this, it remains controversial as to whether hydroxychloroquine and chloroquine truly alter the rheological properties of red blood cells. Indeed, whole blood and plasma viscosities, as well as erythrocyte sedimentation rates, were found to be the same in patients with rheumatoid arthritis irrespective of treatment with hydroxychloroquine, sodium aurothiomalate, or D-penicillamine.29 On the other hand, a more recent clinical study found a significant reduction in plasma viscosity as well as whole blood viscosity in postoperative patients treated with hydroxychloroquine.30
Whether the administration of hydroxychloroquine for a longer period of time could have resulted in a stronger antithrombotic effect remains unknown. Hydroxychloroquine has a calculated half-life of >45 days16 ; thus the drug would need to be administered for several months before any steady state could be expected. Metabolites of hydroxychloroquine, including desethylhydroxychloroquine, may also play a role in inhibiting thrombus formation, though any therapeutic benefit of such metabolites is also not clear.31
To our knowledge, this study is the first of its kind to demonstrate a direct effect of hydroxychloroquine on induced thrombi in blood containing high titers of anticardiolipin antibodies. The exact mechanism by which hydroxychloroquine reverses thrombogenic properties of anticardiolipin antibodies is unclear. However, one possible mechanism is that hydroxychloroquine may affect the enhanced platelet aggregation and activation of platelets produced by antiphospholipid antibodies, as shown previously.32 33 Other antiplatelet drugs such as aspirin, ticlopidine, and persantine may have a similar effect, though these have not been studied in this model.
The potential role of hydroxychloroquine in the management of patients with APS remains unclear. It is conceivable that it may be of benefit in patients who are unable to tolerate high levels of anticoagulation with warfarin due to hemorrhagic side effects, or in those who continue to experience thrombotic events despite warfarin therapy. Hydroxychloroquine may also be useful in patients found to have significant titers of anticardiolipin antibodies or a positive lupus anticoagulant test, who have not had any previous thromboembolic events.
| Acknowledgments |
|---|
Received April 21, 1997; revision received August 11, 1997; accepted September 1, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G Ruiz-Irastorza and M. Khamashta Hydroxychloroquine: the cornerstone of lupus therapy Lupus, April 1, 2008; 17(4): 271 - 273. [PDF] |
||||
![]() |
X. Feng, H. Li, A. A. Rumbin, X. Wang, A. La Cava, K. Brechtelsbauer, L. W. Castellani, J. L. Witztum, A. J. Lusis, and B. P. Tsao ApoE-/-Fas-/- C57BL/6 mice: a novel murine model simultaneously exhibits lupus nephritis, atherosclerosis, and osteopenia J. Lipid Res., April 1, 2007; 48(4): 794 - 805. [Abstract] [Full Text] [PDF] |
||||
![]() |
G Ruiz-Irastorza, M-V Egurbide, J-I Pijoan, M Garmendia, I Villar, A Martinez-Berriotxoa, J-G Erdozain, and C Aguirre Effect of antimalarials on thrombosis and survival in patients with systemic lupus erythematosus Lupus, September 1, 2006; 15(9): 577 - 583. [Abstract] [PDF] |
||||
![]() |
M Petri Lupus in Baltimore: evidence-based 'clinical pearls' from the Hopkins Lupus Cohort Lupus, December 1, 2005; 14(12): 970 - 973. [Abstract] [PDF] |
||||
![]() |
F. C. Hall and N. Dalbeth Disease modification and cardiovascular risk reduction: two sides of the same coin? Rheumatology, December 1, 2005; 44(12): 1473 - 1482. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. T. Ho, C. W. Ahn, G. S. Alarcon, B. A. Baethge, F. K. Tan, J. Roseman, H. M. Bastian, B. J. Fessler, G. McGwin Jr, L. M. Vila, et al. Systemic lupus erythematosus in a multiethnic cohort (LUMINA): XXVIII. Factors predictive of thrombotic events Rheumatology, October 1, 2005; 44(10): 1303 - 1307. [Abstract] [Full Text] [PDF] |
||||
![]() |
E Toubi, A Kessel, I Rosner, M Rozenbaum, M Lorber, D Paran, E Sabo, and T D Golan Quinacrine added to ongoing therapeutic regimens attenuates anticardiolipin antibody production in SLE Lupus, April 1, 2003; 12(4): 297 - 301. [Abstract] [PDF] |
||||
![]() |
D. Erkan, Y. Yazici, M. G. Peterson, L. Sammaritano, and M. D. Lockshin A cross-sectional study of clinical thrombotic risk factors and preventive treatments in antiphospholipid syndrome Rheumatology, August 1, 2002; 41(8): 924 - 929. [Abstract] [Full Text] [PDF] |
||||
![]() |
J-C Piette, A Sbai, and C Frances Warning: thalidomide-related thrombotic risk potentially concerns patients with lupus Lupus, February 1, 2002; 11(2): 67 - 70. [PDF] |
||||
![]() |
D D'Cruz Antimalarial therapy: a panacea for mild lupus? Lupus, March 1, 2001; 10(3): 148 - 151. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |