Fredrick L. Kiechle, MD, PhD
Q. We are developing stroke center status at our four-hospital system. Core measures defined by clinical national groups are part of the credentialing and monitoring processes for this status. I am finding apparent conflicts with required laboratory tests and what I think are best lab practices. One example is the requirement that pa-tients post stroke be discharged on statins. This requires that lipid testing be part of mandatory in-house lab tests. I read in Henry’s book and it is my understanding that such tests in an acute setting are not reliable and that it is better to wait several months after the acute episode to establish these baselines.
Also, the neurosurgeons are insisting on running thrombosis panels on these patients before discharge (proteins S and C, factor V Leiden, etc.). The panels do not seem to offer any potential to alter immediate post-stroke medications and are better obtained in a more stable clinical environment, except perhaps the genetic tests.
In my view, these expensive practices are not in the patient’s best interest. Please advise.
Four people were asked to provide advice, and their comments follow. The first answer comes from Kerri S. Remmel, MD, PhD, chief of vascular neurology, University of Louisville (Ky.) School of Medicine, and A. Victoria McKane, MD, pathologist, Minneapolis.
A. Strokes can etiologically be ischemic or hemorrhagic. Ischemic strokes, wherein a blood clot obstructs a vessel in the brain, account for approximately 80 percent of all strokes, according to the National Institutes of Health’s National Institute of Neurological Disorders and Stroke. The other 20 percent are hemorrhagic, wherein a vessel ruptures and bleeds into the brain parenchyma. The majority of strokes are arterial in nature, with approximately one to two percent attributable to intracranial venous thrombosis. Thrombophilia, defined as a tendency—acquired or hereditaryto develop thrombosis, generally refers to the development of venous throm-bosis. Because of this, discrimination should be exercised in ordering thrombosis panels for patients with stroke.
There are three settings in which inherited thrombophilia has been associated with cerebral ischemia: 1) right-to-left cardiac shunts with paradoxical emboli, 2) cerebral vein thrombosis, and 3) children with stroke. In these settings, thrombophilia testing may be indicat-ed, as long as testing is performed at the appropriate time and the appropriate tests are ordered. Thrombophilia testing is also in order for patients with previous history of thrombosis, especially at unusual sites, as well as patients with stroke who are less than 50 years old. Cerebral venous thrombosis may be underdiagnosed. Patients with ischemic stroke who do not follow an arterial territory pattern may have a cerebral venous thrombosis and should be evaluated with magnetic resonance venography in addition to magnetic resonance angiography. Additionally, patients with stroke or transient ischemic attacks discovered to have a patent foramen ovale (PFO), by transesophageal or transthoracic echocardiogram, are further studied by lower extremity Doppler or pelvic and lower extremity venography to rule out a deep vein thrombosis (DVT) associated with paradoxical emboli. If a DVT is found, a thrombophilia panel is ordered. Many DVTs are silent, and diligence in looking for them in patients with PFO is warranted.
The battery of recommended thrombophilia tests includes free protein S antigen and protein C activity and antithrombin activity, factor V Leiden and prothrombin G20210A mutations, and factor VIII activity, as well as homocysteine measurement and evaluation for antiphospholipid antibodies, the lupus anticoagulant. There is controversy about whether evaluating for lipoprotein (a) and lipoprotein-associated phospholipase A2 (Lp-PLA2) is also indicated.
As most strokes are arterial in nature, there is a close association of stroke risk with the degree of underlying atherosclerotic vascular disease. In those stroke patients with significant atherosclerotic vascular disease, thrombophilia testing generally is not warranted. Those individuals without obvious atherosclerotic defect are more likely than those thus afflicted to benefit from investigation of a hereditary or acquired thrombophilic state. Thrombophilic assays associated with arterial thrombotic risk should be considered in particular, such as antiphospholipid anti-bodies including lupus anti-coagulant and plasma homocysteine.
In summary, in the stroke patient without evidence of venous thrombosis, tests for those problems that have a known association with arterial thrombosis are warranted. This typically constitutes risk factors for atherosclerotic vascular disease. Testing for thrombophilic risk factors may be indicated in those patients without evidence of venous thrombosis but who develop an arterial clot at a young age or without evidence of significant arterial disease.
The timing of a comprehensive thrombophilia laboratory evaluation with respect to an acute thrombotic event is challenging, given the situational acuity and initiation of and ongoing anticoagulant therapy. Immediately after a thrombotic episode, the titers of antiphospholipid anti-bodies present may transiently decrease. While antiphospholipid titers such as anticardiolipin and anti-β2GPI antibodies may be measured while the patient is anticoagulated, evaluating for presence of a lupus anticoagulant once anticoagulant therapy is initiated must be done cautiously and is best performed after the course of therapy has been completed.
Homocysteine levels may differ in the acute setting of stroke compared with the recovery phase, appearing decreased in the acute phase and increased later. Whether this represents a temporary depression of homocy-steine or a subsequent elevation after the event is unclear. Levels of this analyte are not influenced by the patient’s state of anticoagulation.
Acute thrombosis may transiently decrease protein C, protein S, and antithrombin, and it is recommended that these tests be repeated be-tween three and six months after the thrombotic event and not while the patient is anticoagulated. Heparin therapy may decrease antithrombin beyond that seen from the thrombosis alone, though it is unusual for this to fall below 50 to 60 percent. Vitamin K antagonists (warfarin) decrease the vitamin K-dependent proteins C and S. The latter are best assessed while the patient is not anticoagulated, but some have advocated using the total protein S antigen to factor VII antigen ratio and protein C antigen to factor VII antigen ratio, respectively, to presumptively iden-tify deficiencies of either of these respective proteins (with confirmatory testing performed minimally two weeks after anticoagulant therapy is terminated), though there is little published literature to substantiate this approach. Any anticoagulant therapy such as direct thrombin inhibitors or heparin therapies may falsely increase levels of proteins C and S when measured by clottable methods. Heparin and warfarin can cause false-positives in lupus anticoagulant testing. Ideally, these tests should all be performed at least two weeks after anticoagulation therapy is discontinued.
Genetic testing for factor V Leiden, prothrombin G20210A, and perhaps MTHFR C677T should not be affected by a patient’s medical acuity or his or her state of anticoagulation. Notwithstanding, the activated protein C ratio, a screening test for factor V Leiden, is affected by acute thrombosis (decreased) and heparin (decreased).
Despite these limitations, thrombophilia testing is appropriate in selected (not all) patients, as detailed earlier, while they are in the hospital. Blood for these tests should ideally be obtained before starting anticoagulant therapy. One of the authors (KSR) does not hesitate to contact hematology for a consultation during the acute event, if the results of the laboratory evaluation are complicated. Abnormal plasma-based results are repeated in three months, and a patient with a repeated abnormal result is also referred to a hematologist.
Similarly, lipid testing on a fasting blood sample should be performed acutely and not delayed. A LDL-C result greater than 70 mg/dL is the in-dication to begin statin therapy; the drugs themselves are anti-inflammatory and will decrease inflammatory markers studied by the laboratory. This approach is recommended in stroke and transient ischemic attack protocols. Similar to followup for thrombophilia testing, one of the authors (KSR) repeats a lipid panel and liver enzymes (AST and ALT) in three months.
1. Adams HP, del Zoppon G, Albers MJ. Guidelines for the early management of adults with ische-mic stroke. Stroke. 2007;38:1655-1711.
2. Bushnell CD, Goldstein LB. Diagnostic testing for coagulopathies in patients with ischemic stroke. Stroke. 2000;31:3067-3078.
3. Daif A, Awada A, Al-Rajeh S, et al. Cerebral venous thrombosis in adults. Stroke. 1995;26:1993-1995.
4. Feinbloom D, Bauer KA. Assessment of hemostatic risk factors in predicting arterial thrombotic events. Arterioscler Thromb Vasc Biol. 2005;25:2043-2053.
5. Gosselin RC, King JH, Janatpur KA, et al. Effects of pentasaccharide (Fonda-parinux) and direct thrombin inhibitors on coagulation testing. Arch Pathol Lab Med. 2004;128: 1142-1145.
6. Hart RG. What’s new in stroke? The top 10 studies of 2006-2008. Pol Arch Med Wewn. 2008;118(11):650-657.
7. Haywood S, Liesner R, Pindora S, et al. Thrombophilia and first arterial ischaemic stroke: a systematic review. Arch Dis Child. 2005;90:402-405.
8. Kottke-Marchant K, Duncan A. Anti-throm-bin deficiency: issues in laboratory diagno-sis. Arch Pathol Lab Med. 2002;126: 1326-1336.
9. Nagaraja D, Christopher R, Tripathi M. Plasma antithrombin III deficiency in ischaemic stroke in the young. Neurol India. 1999;47:155-156.
10. O’Brien AE, Tate GM, Shiach C. Evaluation of protein C and protein S levels during oral anticoagulation therapy. Clin Lab Haematol. 1998;20(4):245-252.
11. Sacco RL, Adams R, Albers G, et al. Guide-lines for prevention of stroke in patients with ischemic stroke or transient ischemic attack. Stroke. 2006;37:577-617.
12. Tripodi A, Mannucci PM. Laboratory investigation of thrombophilia. Clin Chem. 2001;47(9):1597-1606.
13. Van Cott EM, Laposata M, Prins MH. Laboratory evaluation of hypercoagulability with venous or arterial thrombosis. Arch Pathol Lab Med. 2002;126:1281-1295.
Daniel Laskowitz, MD, MHS, associate professor of medicine (neurology), anesthesiology, and neurobiology, and director of Neurovascular Laboratories, Duke University Medical Center, added the following:
A. My impression is that performing thrombosis panels need not be part of any routine workup for stroke. It is probably neither cost-effective nor would it have therapeutic implications for the majority of patients. In fact, inherited thrombophilias (such as protein C or protein S or antithrombin III deficiency, factor V Leiden or prothrombin G20210A mutations) rarely contribute to adult stroke, though they may play a somewhat larger role in children.
Thus, my view is that hypercoagulable state should be tested for only in select situations (for example, prior thromboembolic events without precipitating factors, family history of thrombosis, thrombosis during pregnancy, <45 years old, multiple spontaneous abortions, select patients with no other risk factors). More formal guidelines have been published (Bushnell CD, Goldstein LB. Stroke. 2000;31:3067–3078; Seligsohn U, Lubetsky A. N Engl J Med. 2001;344:1222–1231).
Regarding cholesterol testing, and this is a point that’s often confused, hypercholesterolemia and hyperlipidemia are not as well established as risk factors for first or recurrent stroke in contrast with what is seen in cardiac disease, though observational studies have demonstrated a weak association between cholesterol level and stroke. However, patients with ischemic stroke or transient ischemic attack with elevated cholesterol, comorbid coronary artery disease, or evidence of an atherosclerotic origin should be managed according to National Cholesterol Education Program III guidelines, which include lifestyle modification, dietary guidelines, and medication recommendations. Moreover, there are data that indicate statins may benefit patients with ischemic stroke or TIA, even in the absence of hypercholesterolemia. Acute illness may affect cholesterol levels.
Larry B. Goldstein, MD, director of the Duke Stroke Center, added this:
A. Testing for coagulopathy is reserved for specific patients and need not be done routinely.
The SPARCL (Stroke Prevention by Aggressive Reduction in Cholesterol Levels) trial found that patients with stroke or TIA, and LDL-C between 100 and 190 mg/dL and no known coronary heart disease (an independent indication for treatment), have a 16 percent reduction in the risk of recurrent stroke with statin treatment. This is reflected in current American Heart Association/American Stroke Association guidelines.
Some lipid parameters may be low in the acute setting, but they should not be falsely elevated; therefore, it is appropriate to test in the acute setting and treat accordingly.
Dr. Kiechle is medical director of clinical pathology, Memorial Healthcare, Hollywood, Fla.