Etiologic workup in cases of cryptogenic stroke: protocol for a systematic review and comparison of international clinical practice guidelines

Globally, stroke is the second most common cause of death [1] and the third most common cause of disability [2]. In the UK alone, the direct and indirect costs of stroke are around £26bn per year [3]. Those who survive an initial stroke are at high risk of further stroke [4], and a quarter of all stroke survivors will have a further stroke within 5 years [5]. These subsequent strokes are associated with worse outcomes and an even higher mortality than the initial stroke [6]. It is therefore essential that the aetiology of a stroke can be identified, not only to guide treatment for specific conditions, but also to initiate appropriate and timely secondary prevention to reduce the risk of recurrence [7].

Although most ischaemic strokes are caused by atherosclerosis or arteriosclerosis of the blood vessels within the brain, or embolism originating from the heart or the major extracranial blood vessels, in at least 30% of cases, the exact aetiology of an ischaemic stroke is not identifiable by existing investigative protocols [8]; this is classed as a cryptogenic stroke or stroke of unknown origin. Classifying an ischaemic stroke as ‘cryptogenic’ is therefore based on the exclusion of other well-established causes of stroke [9], and with more than 200 possible causes of a stroke, arriving at this classification can pose a formidable challenge to clinicians [10]. Standard diagnostic workup of patients tends to include brain imaging, vascular imaging and cardiac evaluations [8]. If these investigations fail to identify the aetiology of the stroke, further investigations tend to depend on local service availability and on the preference of the treating clinician. Therefore, a ‘diagnosis’ of cryptogenic stroke is often allocated loosely in the absence of a thorough workup, rather than the cause truly being unknown. This has important implications especially in light of the lack of available options for secondary prevention for strokes of unknown source. For example, two recent randomised trials of non-vitamin K antagonist oral anticoagulants did not show a reduction in recurrent stroke compared to aspirin [11, 12], and one of these showed possible harm with an excess of bleeding [11]. This is in contrast to the well-established benefits for stroke prevention with oral anticoagulation in patients where a cardiogenic cause such as atrial fibrillation is found [13]. Patients whose stroke is labelled as cryptogenic are unable to take steps to reduce their risk of a subsequent stroke.

Diagnostic workup in cases of seemingly cryptogenic stroke is highly complex, and from existing guidelines, it is difficult to discern specific recommendations for diagnostic pathways. Additionally, up to now, only one diagnostic algorithm has been proposed to classify cryptogenic stroke [10]. This algorithm has identified three levels of investigation in an hierarchical fashion : (1) standard level, which may be considered the absolute minimum, (2) advanced level, and (3) highly specialised level where an even more advanced set of investigations may be performed. These recommendations are in line with the American Heart Association–American Stroke Association and American Academy of Neurology guidelines [14‐17].

As such, while some guidelines and published diagnostic algorithms have been proposed to facilitate the identification of the aetiology of ischaemic stroke, there does not yet exist international consensus on the optimum level and timing of investigations required following a stroke. Therefore, we aim to review international clinical practice guidelines (CPGs) to compare recommendations and pathways for the diagnostic workup necessary before arriving at the diagnosis of cryptogenic stroke, to understand where there are agreement, disagreement and conflicting recommendations. The review findings will inform prioritisation of future research in this emerging field of inquiry.

The review has been designed based on input from experts in stroke and systematic review methodology and in accordance with the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) guidance. A PRISMA-P reporting checklist is provided in Additional file 1. The review is registered on PROSPERO, number: CRD42019127822.

This review forms part of a wider programme of work exploring the optimisation of assessment and investigation of the aetiology of ischaemic stroke and the diagnosis of cryptogenic stroke. To our knowledge, this will be the first systematic review to compare the content of recommendations in relevant international CPGs. The review findings will allow for a better understanding of similarities and differences in international recommendations for identifying the aetiology of a stroke, the optimum level of investigations required following a stroke and the quality of evidence underpinning practice recommendations. Additionally, the review findings will have practical implications for the detection and management of patients with cryptogenic stroke. This review will follow a robust systematic method to identify all relevant publications. It will however only include English language publications which are a limitation.