EBM Conference

Carla Nester, M.D.

October 25, 2001

 

CAT: Level of International Normalized Ratio that is adequate to prevent recurrent thrombosis in the Lupus Anticoagulant.

 

Clinical Bottom Line:  Treatment to produce an INR of >3 is significantly more effective in preventing the recurrence of thrombosis in patients with the lupus anticoagulant.

 

Our Patient:  35yo AAG with ESRD (FSGS) s/p KTx who has the lupus anticoagulant and anti-cardiolipin antibody and a hx of recurrent venous thrombosis in the right internal jugular innominant vein.

 

Citation:  The Management of Thrombosis in the Antiphospholipid-Antibody Syndrome. NEJM. 1995;332:993-997.

 

Clinical Question:  Is there an optimal INR at which to maintain a patient with the lupus anticoagulant to prevent recurrent thrombosis?

 

Search Terms:  "lupus anticoagulant" and "thrombosis" in MEDLINE 1966 to present.

 

Design:  Retrospective evaluation of one hundred forty-seven patients with the antiphospholipid-antibody syndrome and a history of thrombosis.  A three-page questionnaire was used to collect data on each patient with respect to risk factors for thrombosis, antithrombotic treatments received and immunosuppressive therapy prescribed. Table 2. Reflects this data. Antithrombotic therapy was subdivided into the following categories:  none, 75mg of aspirin daily, warfarin to obtain an INR of <3.0 (low intensity) and warfarin to obtain an INR of >3 (high intensity). Objective accepted radiologic evidence was required to diagnose thrombosis.  Lupus anticoagulants, anticardiolipin antibodies and prothrombin time tests with INRs were performed using standard available assays.  The relative risks of thrombosis per year were calculated relative to the non-treatment group. Kaplan-Meir calculations were used to define a five year probability of a thrombotic event.

 

The Patients:  One hundred eighty three patients with antiphospholipid antibody syndrome seen in referral to St Thomas's Hospital Lupus Clinic between 1983-1993.  Inclusion criteria required that each patient have positive tests for lupus anticoagulant, anticardiolipin antibodies, or both and a history of thrombosis.  Participants were allocated into three different study groups.  Group 1 included those patients with antiphospholipid antibody syndrome who also met 4 or more criteria for SLE (66).  Group 2 were those patients who had antiphospholipid antibody syndrome and met one to three criteria for SLE (19) and finally Group 3 were those patients who had antiphospholipid antibody syndrome and no evidence of SLE (62).

 

Exclusion Criteria:  Thirty six of the referred 183 were excluded from the study for the following reasons: follow-up of less that one year (3); loss to follow-up (8); recurrent fetal loss with no hx of thrombosis (18); thrombocytopenia with no hx of thrombosis (5); and lack of objective evidence of antiphospholipid antibodies or thrombosis (2).  Table 1. gives the demographics of the study participants.

 

Results:

Treatment                                                       # of Patients           RRR              ARR                 NNT      P Value

None

      84

 

 

 

 

Aspirin

      70

38%

 25%

 4.0

0.013

Warfarin, <3 INR

      67

15%

 10%

 10

0.270

                <3 INR with Aspirin

      14

 

 

 

0.531

Warfarin, >3 INR

      64

92%

 60%

 1.5

<0.001

                >3 INR with Aspirin

      17

 

 

 

<0.001

 

 

 

 

 

 

Comments:

I.  Are the results in the study valid?

  (Was there a representative and well-defined sample of patients at a similar point in the course of the

    disease?)

1.       This was a retrospective cohort study of referred patients therefore the study is likely to suffer from referral filter bias (as well as memory bias) making it difficult to ascertain whether the study group is representative of the general population of patients with the lupus anticoagulant and recurrent thrombosis.

2.       Patients had received different treatments at different times.

3.       Demographics in each treatment group were not clearly defined.

 

  (Was the follow-up sufficiently long and complete?)

4.       Eight patients were lost to follow-up and 3 were only followed for one year.  One hundred and one patients had recurrent thromboses (69% vs. 76.5%).

5.       Length of follow-up may have been somewhat short in some individuals/treatment groups.

 

  (Were objective and unbiased outcome criteria used?)

6.       No discussion was made of the blinded nature of the radiologic procedures used for confirmation of thromboses - though accepted techniques were used.

 

  (Was there adjustment for important prognostic factors?)

7.       Treatment groups were not allocated for risk of thrombotic events. (Though they do report that the only "risk" for thrombosis that decreased the interval to clotting was treatment.)

 

II.  What are the results?

  (How large is the likelihood of the outcome events(s) in a specified period of time?)

 

1.         Study provides relative risk data for the various treatment groups.

2.         95% Confidence Intervals are provided which seem appropriate - however the patient-years of follow-up are highly variable between groups lending to wider intervals in some treatment groups.

 

I.                   Will the Results Help Me in Caring for Our patients?

  (Were the study patients similar to my own?)

1.       Fairly good demographics provided. Likely to be generalizeable to our patient (except the ESRD variable). (1% were primary, 26% male)

 

  (Will the results lead directly to selecting or avoiding therapy?)

2.       The benefits of an increased INR will need to be weighed against the risk of serious bleeding.  In this study the RR of bleeding in the >3 INR group was 7.1% (1.7% serious), per patient year.

 

  (Are the results useful for reassuring or counseling patients?)

3.   Yes