Is the Stroke Caused by a Hypercoagulable State?

Two recent cases have stimulated my interest in the appropriate work up for the possibility of a hypercoagulable state causing a stroke.

The first case involved a 50-year-old gentleman who had suffered a right parietal infarct. His work up revealed evidence for neurosyphillis with a positive VDRL at 1:64 in both the serum and CSF, as well as a mild CSF pleocytosis with an increase in protein and reduced glucose concentration. His labs also revealed a positive anticardiolipin antibody at >150 of both IgG and IgM classes. His lupus anticoagulant was negative. Several discussions ensued: was this a stroke due to the neurosyphillis? Could he have a hypercoagulable condition related to the anticardiolipin antibodies?

Another patient presented with bilateral renal infarcts and subsequently had a large left hemisphere stroke. The questions that arose from these cases were:

1. When is it appropriate to consider a hypercoagulable etiology of stroke?

2. What are the appropriate lab tests to request?

In my CPOE orders, I had bundled together those tests that are associated with hypercoagulability: protein S, protein C, antithrombin III, factor V leiden, homocysteine level, and antiphospholipid antibody. Further, it was my impression that factor V leiden was extremely rare.

I’m sure we all remember Virchow’s triad regarding the factors predisposing to venous thrombosis- endothelial wall factors, blood flow (stasis), and intrinsic blood factors (hypercoagulability). These factors do not have the same predictive value for causing arterial thrombosis, however. Therefore, the tests that would indicate the cause of a venous thrombosis do not necessarily indicate a propensity for stroke. In fact, when screening for the cause of a cryptogenic stroke (ie, no cardiac or vascular cause determined after appropriate testing), testing for the inherited protein deficiencies (protein S, protein C, and antithrombin III) is of extremely low yield. Furthermore, their presence in the setting of stroke is likely coincidental, and treatment with anticoagulants has not shown any superiority to the use of antiplatelet agents.

As it turns out, the factor V leiden (arginine for glutamine substitution at residue 506) is the most common cause of congenital venous thrombosis. In the presence of this mutation, there is resistance to activated protein C, which enables continuation of the thrombotic cascade. The second most common cause of an inherited thrombophilia is an abnormal prothrombin mutation called prothrombin 20210A— one that was completely off of my radar before investigating this topic.

Of the acquired hypercoagulable states, the presence of antiphospholipid antibodies (aPL) is possibly relevant to causation. These antibodies, directed against phospholipids-dependent clotting tests such as the aPTT, require a cofactor-beta glycoprotein I that bind to cardiolipin. These aPLs include the so-called lupus anticoagulant, which has to be considered way up on the medical misnomer list. These antibodies are present in patients without lupus and cause clinical thrombosis and are not anticoagulants.

However, common tests for clotting—aPTT or the Russel Venom Viper Test, dRVVT—are prolonged and do not correct with 1:1 dilution (a correction would imply lack of serum clotting factors). The anti-phosphlipid syndrome (APS) consists of both the presence of the APL and clinical evidence of thrombosis. This could be recurrent venous thrombosis, an unusual location of thrombosis, fetal waste, or multiple miscarriages. Since the titer of the aPL can vary, especially after an acute thrombotic event such as a stroke, it is necessary to repeat titers perhaps eight weeks after the presenting event to definitively detect the presence or absence of these antibodies. Red flags to consider the presence of these antibodies are the above clinical features and lab abnormalities of a biologically false positive syphilis test, thrombocytopenia, and a prolonged aPTT. The specific lab tests to order are:

· Lupus anticoagulant

· Anticardiolipin antibody- ELISA

· Anti beta 2 glycoprotein I antibody- ELISA

As a result of my review, I have concluded that I was way too liberal in my criteria for triggering the ordering of the complete “hypercoagulable workup” battery. At this time, I would limit ordering tests for an abnormality of blood clotting as the cause of a stroke to:

· Young patients (<45 yo) with cryptogenic strokes, especially those with:

o Lupus or symptoms suggestive of lupus

o Previous episodes of thrombosis

o Symptoms of the APL syndrome - livedo reticularis, previous miscarriages , migraines

Given the low yield and dubious causative relationship between the inherited thrombophilias, I am eliminating the requests for protein S, protein C, factor V leiden, and antithrombin from my battery of tests for stroke and hypercoagulability. I was blissfully ignorant regarding the prothrombin 20210A condition, so I will continue to not order tests for that entity. Whether to test for hyperhomocysteinemia depends on the data supporting the ability of vitamin therapy to reduce the homocysteine level and reduce recurrent stroke risk. I have seen conflicting reports regarding the efficacy of lowering these levels so will likely continue to request this result. Finally, given the above selection criteria, I will continue to order the tests for the presence of aPL.

If you were expecting a definitive answer or explanation for the two clinical cases, you will be sorely disappointed. In fact, I welcome any suggestions or input you have on those patients and this topic in general.

References:

http://www.gbhn.ca/ebc/documents/HYPERCOAGULABLESCREENINFOSHEET.pdf

http://www.springerlink.com/content/978-1-934115-01-5/#section=26842&page=2&locus=16

http://www.cnsspectrums.com/aspx/articledetail.aspx?articleid=346