Correspondence
Thrombosis in the Antiphospholipid-Antibody Syndrome
N Engl J Med 1995; 333:665-667September 7, 1995
- Article
To the Editor:
The finding by Khamashta et al. (April 13 issue)1 supports those of two recent studies,2,3 which demonstrated that patients with antiphospholipid antibodies are in an ongoing prothrombotic state. In this condition there are high plasma values of prothrombin fragment F1+2, a peptide derived from the conversion of prothrombin to thrombin and considered a marker of thrombin generation in vivo.
Even if Khamashta et al. show that the risk of bleeding with warfarin therapy is low, an important issue is whether all patients with antiphospholipid-antibody syndrome and a history of thrombosis should be treated with oral anticoagulants for long periods or whether there is a low-risk category for which this treatment is unnecessary. We base the latter suggestion on our finding that not all patients with antiphospholipid-antibody syndrome are in an ongoing prothrombotic state. We showed that levels of fragment F1+2 are high only in patients with lupus anticoagulant and anticardiolipin antibodies or lupus anticoagulant alone, whereas it is normal in patients positive for anticardiolipin antibodies only.3 This suggests that a positive test for the lupus anticoagulant indicates a high thrombotic risk, whereas the presence of anticardiolipin antibody alone may be indicative of a low risk.
Knowledge of the rate of recurrent thromboses in patients positive for lupus anticoagulant or anticardiolipin antibodies may provide information on whether these two antibodies are markers for different risks of recurrent thrombosis and whether the time between the initial thrombosis and the first recurrence is related to positivity for one of these antibodies.
3 ReferencesFrancesco Violi, M.D.
Domenico Ferro, M.D.
Guido Valesini, M.D.
Università La Sapienza, 00185 Rome, Italy1
Khamashta MA ,Cuadrado MJ ,Mujic F ,Taub NA ,Hunt BJ ,Hughes GRV . The management of thrombosis in the antiphospholipid-antibody syndrome. N Engl J Med 1995;332:993-997
Full Text | Web of Science | Medline2
Ginsberg JS ,Demers C ,Brill-Edwards P , et al. Increased thrombin generation and activity in patients with systemic lupus erythematosus and anticardiolipin antibodies: evidence for a prothrombotic state. Blood 1993;81:2958-2963
Web of Science | Medline3
Ferro D ,Quintarelli C ,Valesini G ,Violi F . Lupus anticoagulant and increased thrombin generation in patients with systemic lupus erythematosus. Blood 1994;83:304-304
Web of Science | Medline
To the Editor:
Khamashta and coworkers reported that anticoagulation producing an international normalized ratio (INR) of 2 to 3 may be insufficient to prevent new thromboses and recommended keeping the INR at or above 3. Many patients with the antiphospholipid-antibody syndrome have a lupus anticoagulant causing prolongation of the prothrombin time by up to four seconds, as compared with a control value. Yet, when warfarin therapy is monitored, a normal prothrombin time is used as a control to calculate the patient's prothrombin-time ratio, which is raised to the power of the international sensitivity index to calculate the INR. Thus, both the anticoagulant effect of warfarin and the lupus anticoagulant contribute to the value obtained for the patient's INR.
To correct for this we determine a target INR for patients whose base-line prothrombin time exceeds the normal prothrombin time by relating the INR to the result of a different type of test of prothrombin time, the prothrombin–proconvertin test.1 This test, for which a reagent is commercially available (Simplastin A, Organon Teknika), has a recommended therapeutic range of 10 to 30 percent.2 Because the patient's plasma is diluted 1:10 in the prothrombin–proconvertin test, the effect of a lupus anticoagulant is minimized. We have found3 that a prothrombin–proconvertin result ranging from 13 to 30 percent corresponds to an INR of between 3 and 2 but that in patients with a lupus anticoagulant an INR higher than 3 may be required to yield prothrombin–proconvertin values within the therapeutic range. Thus, we agree with Khamashta et al. that the INR should be kept at or above 3 in many patients with antiphospholipid antibodies. However, we believe the reason for this is not that these patients require a higher than usual intensity of anticoagulation but that the INR frequently underestimates the anticoagulant effect of warfarin when plasma contains a lupus anticoagulant.
We recommend that warfarin therapy in patients with antiphospholipid antibodies whose base-line prothrombin time exceeds the laboratory's control prothrombin time by more than one second be monitored initially with both the prothrombin–proconvertin test and the INR. One can then select an INR range corresponding to results on the prothrombin–proconvertin test between 15 and 20 percent, which is the intensity of anticoagulation we usually aim for in patients with antiphospholipid antibodies. Periodically, thereafter, we check the INR against the prothrombin–proconvertin test because plasma lupus-anticoagulant activity may vary from month to month and year to year.
3 ReferencesSamuel I. Rapaport, M.D.
Dzung The Le, M.D., Ph.D.
University of California, San Diego, Medical Center, San Diego, CA 921031
Owren PA ,Aas K . The control of dicumarol therapy and the quantitative determination of prothrombin and proconvertin. Scand J Clin Lab Invest 1951;3:201-208
CrossRef | Web of Science | Medline2
Owren PA. Long-term dicumarol treatment in cardiovascular disease — technique and results. In: Proceedings of the First International Conference on Thrombosis and Embolism, Basel, Switzerland, 1954. Basel, Switzerland: Schwabe, 1955:1085-94.
3
Le DT ,Weibert RT ,Sevilla BK ,Donnelly KJ ,Rapaport SI . The international normalized ratio (INR) for monitoring warfarin therapy: reliability and relation to other monitoring methods. Ann Intern Med 1994;120:552-558
Web of Science | Medline
To the Editor:
We found the study by Khamashta et al. interesting and provocative, but its retrospective and nonrandomized design does not justify the unqualified conclusion that patients with thrombosis associated with antiphospholipid antibodies should receive long-term anticoagulation in which an INR of >3 is maintained.
More data from their study are required before even a qualified recommendation of high-intensity coagulation for their patients can be accepted. Since patients were treated according to the clinical judgment of their physicians, knowing what criteria were used to decide which patients were treated with aspirin or warfarin, and which were not treated is critical.
Since the conclusion of Khamashta et al. was based on an analysis of high- and low-intensity anticoagulation, the reason for the INR cutoff of 3 should be stated. Was the goal of the treating physicians to obtain an INR of more than 3 or less than 3? For most conditions the recommended therapeutic INR range is 2.0 to 3.0.1,2 Using a cutoff of 2.0 thus seems more reasonable unless there were compelling reasons to use a higher cutoff.
We are also concerned about the implications of recommending long-term high-intensity anticoagulation. Although the risk of hemorrhage reported in the study was relatively low, the study population was young. As the population ages, however, the cumulative risk of hemorrhage is likely to increase, particularly if the INRs are maintained above 3.3,4 In fact, all the bleeding complications in the study occurred in patients with INRs above 3. The intensity of anticoagulation has been found to correlate strongly and consistently with the risk of hemorrhage.2-4
For the reasons outlined above, we do not believe the study data support an unqualified recommendation of high-intensity long-term warfarin therapy in patients with thrombosis associated with antiphospholipid antibodies. If the authors believe their data, it is not clear why they think that controlled, randomized, prospective studies of the problem are urgently needed. Given the potential risks of such treatment we believe their recommendations may be dangerous and should be tempered until more compelling evidence in favor of such therapy is forthcoming.
4 ReferencesAndrew Slivka, M.D.
Elizabeth Walz, M.D.
Ohio State University Medical Center, Columbus, OH 432101
Litin SC ,Gastineau DA . Current concepts in anticoagulant therapy. Mayo Clin Proc 1995;70:266-272
CrossRef | Web of Science | Medline2
Landefeld CS ,Beyth RJ . Anticoagulant-related bleeding: clinical epidemiology, prediction, and prevention. Am J Med 1993;95:315-328
CrossRef | Web of Science | Medline3
Fihn SD ,McDonell M ,Martin D , et al. Risk factors for complications of chronic anticoagulation: a multicenter study. Ann Intern Med 1993;118:511-520
Web of Science | Medline4
Landefeld CS ,Goldman L . Major bleeding in outpatients treated with warfarin: incidence and prediction by factors known at the start of outpatient therapy. Am J Med 1989;87:144-152
CrossRef | Web of Science | Medline
To the Editor:
The article by Khamashta et al. on the management of thrombosis in patients with the antiphospholipid-antibody syndrome validated the long-term use of warfarin, which has in many cases been the common practice. There have been some reports of the antiphospholipid-antibody syndrome developing after the initiation of treatment with certain drugs, including oral contraceptives.1,2
We have a patient who had a stroke and a positive test for anticardiolipin antibodies (>20 IgG phospholipid units) two months after starting oral contraceptives. The oral contraceptives were discontinued, and the anticardiolipin-antibody level became negative and remained so three years later.
This case brings up two important points: Is there an association between oral contraceptives and the antiphospholipid-antibody syndrome? If there is an association and it is transient, perhaps this population does not need life-long high-intensity warfarin therapy.
2 ReferencesAnne J. Cole, M.D.
Ochsner Medical Foundation, New Orleans, LA 701211
Asherson RA ,Harris EN ,Hughes GR ,Farquharson RG . Complications of oral contraceptives and antiphospholipid antibodies. Arthritis Rheum 1988;31:575-576
CrossRef | Web of Science | Medline2
Asherson RA ,Khamashta MA ,Ordi-Ros J , et al. The “primary“ antiphospholipid syndrome: major clinical and serological features. Medicine (Baltimore) 1989;68:366-374
Web of Science | Medline
To the Editor:
In their retrospective study, Khamashta et al. concur with previous recommendations1,2 of long-term, high-dose anticoagulation (i.e., to maintain an INR of >3) for patients with phospholipid antibodies who have had clinical thromboses. Their paper leads Lockshin to conclude in his editorial (April 13 issue)3 that “the standard of treatment in 1995 . . . [is] life-long warfarin therapy to achieve an international normalized ratio of 3.0 or higher” for patients with antiphospholipid-antibody syndrome and a major thrombosis. We beg to differ.
We have studied 33 patients with the antiphospholipid-antibody syndrome (23 fulfilling criteria for systemic lupus erythematosus and 10 with primary antiphospholipid-antibody syndrome or a lupus-like syndrome), who have all had at least one thrombotic event and who have been followed by one of us for a minimum of three years.
Using the Kaplan–Meier survival model to compare the data for three treatment groups — oral anticoagulants (INR, <3), low-dose aspirin alone, and no treatment — we found no significant difference in the rate of rethrombosis between the treatment groups: after 80 months of follow-up the probability that thrombosis would not recur was 0.6 for the untreated patients and those given low-dose aspirin and 0.74 for patients given oral anticoagulants. There were 36 episodes of recurrent thrombosis during a mean follow-up of 100 patient-months (range, 36 to 288). Fifty-one percent of the patients never had a recurrence of thrombosis despite having had no anticoagulant therapy for many years. All had persistently positive tests for antiphospholipid antibodies.
Even with an INR of less than 3, 4 of 23 patients (17 percent) had serious bleeding during anticoagulation. Since we began treatment with high-dose oral anticoagulants (INR, >3) cerebral bleeding and uncontrollable bleeding from the bladder have occurred in two of seven patients.
We believe that it is dangerous to set the standard of care for patients with antiphospholipid-antibody syndrome until we have a better understanding of the factors and the triggers that predispose some of these patients to recurrent thrombosis and the appropriate prospective studies are done. Our concern is that these recommendations are not necessary for all patients and may result in serious iatrogenic disease.
3 ReferencesSamer Z. Nasr, M.D.
Ann L. Parke, M.D.
University of Connecticut Health Center, Farmington, CT 06030-13101
Rosove MH ,Brewer PM . Antiphospholipid thrombosis: clinical course after the first thrombotic event in 70 patients. Ann Intern Med 1992;117:303-308
Web of Science | Medline2
Derksen RH ,de Groot PG ,Kater L ,Nieuwenhuis HK . Patients with antiphospholipid antibodies and venous thrombosis should receive long term anticoagulant treatment. Ann Rheum Dis 1993;52:689-692
CrossRef | Web of Science | Medline3
Lockshin MD . Answers to the antiphospholipid-antibody syndrome? N Engl J Med 1995;332:1025-1027
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Violi et al. suggest from their previous studies that patients with the antiphospholipid-antibody syndrome and a previous thrombosis who have increased levels of anticardiolipin antibodies, but not a positive test for the lupus anticoagulant, have the lowest thrombotic risk. Our study showed no evidence of a decreased risk of recurrent thrombosis in such patients, regardless of whether an adjustment was made for other covariates.
Rapaport and Le state that patients with lupus anticoagulant may have a prolonged prothrombin time. However, this phenomenon is uncommon unless the thromboplastin used in performing the test is recombinant tissue factor, which is what Rapaport and Le used. We used only conventional thromboplastins. Rapaport and Le suggest a method of allowing for the effect of the lupus anticoagulant in INRs using recombinant tissue factor, but we think it is simpler and cheaper to use a conventional thromboplastin.
Slivka and Walz criticize our study because it was retrospective and the treatment was prescribed on the basis of the physician's judgment — i.e., it was not randomized. We did not collect data on the physicians' own criteria for prescribing treatment because we thought that such subjective self-assessments would not be comparable. We believe that it is extremely important to discover whether prospective, randomized trials will confirm our experience in the care of these patients.1 Slivka and Walz also ask why the cutoff point for the INR was 3. By convention in the United Kingdom, patients receive anticoagulant therapy with the aim of maintaining an INR of 2 to 3 or 3 to 4. In no case was the INR below 2. We agree with Slivka and Walz that the risk of hemorrhage is higher with an INR of more than 3 than with an INR of less than 3, but we believe the benefits of anticoagulation in this group of patients outweigh the risks.
We could not examine the immediate effect of oral contraceptives on the risk of thrombosis: the effect on the risk of initial thromboses could not be analyzed because only patients with a history of previous thrombosis were enrolled in the study. We did find that women taking oral contraceptives immediately before their first thrombosis were proportionately more likely to have a venous rather than an arterial thrombosis (P = 0.016). The effect on the risk of recurrent thrombosis could not be assessed because most women were advised to stop taking oral contraceptives after their first thrombotic event. With respect to the two points raised by Cole, we believe that oral contraceptives can heighten the thrombotic tendency present in some patients who are positive for antiphospholipid antibodies.2,3 The second question as to whether this subgroup of patients should be treated differently cannot be answered at present.
The size of the study described by Nasr and Parke (only 33 patients) may explain why they found no evidence of an association between treatment and risk of recurrent thrombosis. Our study found a much lower proportion of untreated patients (i.e., receiving neither aspirin nor warfarin) who did not have a recurrent thrombosis after 80 months: 0.24 (95 percent confidence interval, 0.15 to 0.33).
3 ReferencesMunther A. Khamashta, M.D.
Rayne Institute, London SE1 7EH, United KingdomNick A. Taub, M.Sc.
United Medical and Dental Schools of Guy's and St. Thomas's Hospitals, London SE1 9RT, United KingdomBeverley J. Hunt, M.D.
St. Thomas's Hospital, London SE1 7EH, United Kingdom1
Lockshin MD . Answers to the antiphospholipid-antibody syndrome? N Engl J Med 1995;332:1025-1027
Full Text | Web of Science | Medline2
Vianna JL ,Khamashta MA ,Ordi-Ros J , et al. Comparison of the primary and secondary antiphospholipid syndrome: a European Multicenter Study of 114 patients. Am J Med 1994;96:3-9
CrossRef | Web of Science | Medline3
Rivier G ,Herranz MT ,Khamashta MA ,Hughes GR . Thrombosis and antiphospholipid syndrome: a preliminary assessment of three antithrombotic treatments. Lupus 1994;3:85-90
CrossRef | Web of Science | Medline
Author/Editor Response
Given the ambiguities of retrospective studies, such as that by Khamashta et al., data from other clinics are important. Nasr and Parke report that 14 percent of their patients (36 events in 33 patients seen for 100 patient-months) had a recurrent thrombosis each year; their treatment-complication rate was apparently higher than that reported by Khamashta et al. Nasr and Parke's recurrence rate does not differ much from that of Khamashta's low-dose warfarin group (23 percent). A definitive answer regarding the best treatment must await the results of a prospective trial currently under way in which entry criteria, compliance, and other factors are carefully controlled. Until the trial is completed, responsible physicians will have to make their treatment choices on the basis of their own experiences and available peer-reviewed published data, of which the paper by Khamashta et al. is currently the best.
Michael D. Lockshin, M.D.
National Institute of Arthritis and Musculoskeletal and Skin Diseases, Bethesda, MD 20892- Citing Articles (5)
Citing Articles
1
Hayan Moualla, David Garcia. (2011) Vitamin K Antagonists – Current Concepts and Challenges. Thrombosis Research 128:3, 210-215
CrossRef2
Jack Hirsh, Valentin Fuster, Jack Ansell, Jonathan L Halperin. (2003) American Heart Association/American College of Cardiology Foundation guide to warfarin therapy. Journal of the American College of Cardiology 41:9, 1633-1652
CrossRef3
Armando Tripodi, Veena Chantarangkul, Marigrazia Clerici, Barbara Negri, Monica Galli, Pier Mannuccio Mannucci. (2001) Laboratory control of oral anticoagulant treatment by the INR system in patients with the antiphospholipid syndrome and lupus anticoagulant. Results of a collaborative study involving nine commercial thromboplastins. British Journal of Haematology 115:3, 672-678
CrossRef4
P. Stratta, C. Canavese, S. Ferrero, A. Grill, M. Salomone, P. C. Schinco, E. Fusaro, B. Montaruli, S. Santi, G. Piccoli. (1999) Catastrophic Antiphospholipid Syndromes in Systemic Lupus Erythematosus. Renal Failure 21:1, 49-61
CrossRef5
Masahide Yamazaki, Hidesaku Asakura, Masanori Saito, Hiroshi Jokaji, Chika Uotani, Ichiro Kumabashiri, Eriko Morishita, Keiji Aoshima, Tamotsu Matsuda, Douglas A. Triplett. (1998) Prothrombin Fragment 1+2 Measures Treatment Effect in Patients with Antiphospholipid Syndrome. Thrombosis Research 91:3, 121-128
CrossRef
