Introduction: Nonstenotic carotid plaque is found in some patients with otherwise cryptogenic stroke (CS) but also in normal individuals and patients with stroke of known cause (KS). Clinical management would be aided by estimates of how often nonstenotic carotid plaques in patients with CS are causal versus incidental. Methods: Through systematic search, we identified all cohort studies applying identical carotid artery imaging modalities to both CS and KS patients. We performed a random-effects meta-analysis across these studies to determine the best estimate of the differential frequency of nonstenotic carotid plaque in CS and KS patients. We then applied Bayes’ theorem (method of Alsheikh-Ali et al, Stroke 2009) to determine the probability that nonstenotic carotid plaque is causal or incidental. Results: The systematic search identified 6 cohort studies enrolling 1551 patients (761 CS, 790 KS). Methods of vessel imaging were carotid duplex ultrasound, CTA, and MRA. Overall, nonstenotic carotid stenosis ipsilateral to ischemic stroke was found in 33.5% (255/761) CS patients vs 19.2% (152/790) KS patients. In formal meta-analysis, the summary risk ratio for nonstenotic carotid plaque in CS vs KS patients was 1.54 (95%CI 1.15-2.05) (Figure). With application of Bayes theorem, the corresponding probabilities indicated that, when present, nonstenotic carotid plaque in CS patients is causal in 52.7% and incidental in 47.3%. Given the prevalence of nonstenotic carotid plaque, these findings indicate that nonstenotic carotid plaque is the cause of 17.7% of all anterior circulation ischemic strokes in the cryptogenic stroke population. Conclusions: In patients with cryptogenic stroke, when nonstenotic carotid plaque is found, it is likely causal in about half of patients, who may benefit from intensified anti-atherosclerotic therapy, and likely incidental in about half of patients. Nonstenotic carotid plaque is a common cause of cryptogenic stroke in the anterior circulation, accounting for about 1 of every 6 cases.