What Proportion of Stroke Is Not Explained by Classic Risk Factors?

Catalina C. Ionita, MD; Andrew R. Xavier, MD; Jawad F. Kirmani, MD; Subasini Dash, MD; Afshin A. Divani, PhD; Adnan I. Qureshi, MD


Prev Cardiol. 2005;8(1):41-46. 

In This Article

Cryptogenic Stroke

Cryptogenic stroke (i.e., stroke with no identifiable cause) is more common among young patients defined as less than 45 years old. The classic risk factors for stroke are usually absent and yet cerebral angiography and transesophageal echocardiography do not reveal large-artery occlusive disease or an unequivocal cardioembolic source. Echocardiographic studies in patients with cryptogenic stroke, however, show an increased incidence of PFO, raising the possibility of a cardioembolic pathogenesis in a subgroup of patients. Prevalence of PFO in the general population is reported between 17%-35%. In patients with cryptogenic stroke, PFO was found in up to 45% of cases. Early reports of the rate of recurrent stroke PFO-related were in the range of 1.7%-4.7%/yr and suggest a correlation with the size of the PFO. Recently, the Patent foramen ovale In Cryptogenic Stroke Study (PICSS)[44] reported a much higher rate of recurrent ischemic events in cryptogenic stroke subtype with no significant difference between the patients with and without PFO (a 2-year event rate of 20.4% vs. 16.6%, respectively). As opposed to the original belief, the recurrence rate has not been influenced by the size of PFO or the coexistence of an atrial septal aneurysm.[44] Ischemic stroke in patients with PFO is likely multifactorial. A higher incidence of deep venous thrombosis in patients with PFO, as well reports of trapped thrombus in PFO, make paradoxical embolism one of the possible stroke mechanisms. Moreover, recurrent strokes have been reported even after percutaneous or surgical closure of PFO, suggesting that the cause of the initial or recurrent stroke may be local thrombus formation.[46] Potential atrial vulnerability to arrhythmia related to the presence of PFO can also trigger local thrombus formation. Therefore, a subgroup of cryptogenic strokes may be truly cardioembolic. A recent meta-analysis concluded that PFO alone does not represent an increased risk for subsequent stroke or death. Patients with cryptogenic stroke and PFO do not have a higher risk for recurrent stroke than patients with cryptogenic stroke without PFO. The combined presence of a PFO and atrial septal aneurysm may confer an increased risk of subsequent stroke in patients <55 years of age. The best strategy for preventing recurrent ischemic events remains undetermined.[47] Although a prothrombotic state has been proposed as a possible culprit for cryptogenic stroke, hemostatic abnormalities and genetic abnormalities or variations for coagulation factors are rarely found.[46] Other risk factors and the adjusted RRs associated with cryptogenic stroke are: 1) hypertension (2.31); 2) migraine headache (2.09); and 3) recent heavy drinking within the preceding week (4.68) or within the preceding 24 hours (3.84).[49]

In conclusion, most strokes can be explained by the classic risk factors, however, in a third of cases, even a thorough investigation fails to reveal the risk factor that triggered the vascular event. Unidentified environmental or genetic factors might play a role. Besides the interventions directed against modifiable risk factors as measures of primary and secondary stroke prevention, identification of other risk factors yet unnamed should be the subject of future investigations.