To challenge the notion that embolic stroke of unknown source (ESUS) is required to justify implantation of an implantable loop recorder (ILR) in cryptogenic stroke patients.

Detection of paroxysmal atrial fibrillation (AF) in cryptogenic stroke is difficult, but essential as this diagnosis changes management. Crystal AF demonstrated that ILRs are efficacious in detecting paroxysmal AF when compared to conventional follow-up. Herein we describe our experience with ILR in cryptogenic stroke.

Using cardiology’s ILR registry list, we used standardized case report forms to perform a chart abstraction of consecutive cryptogenic stroke patients (7/2017-7/2019). Cases with incomplete data were excluded. Demographic, laboratory, and clinic data were compared in patients found to have AF on ILR and those without evidence of AF on ILR. Clinical prediction scores (e.g., LADS, CHA2DS2VASc) were calculated. Groups were compared using Chi-square (or Fisher’s Exact) and Mann Whitney U tests.

97 stroke patients with ILRs met inclusion criteria.  AF was detected in 12% of cases (9% at 6 months, 10% at 12 months). The median time from ILR placement to AF detection was 102 days (9–543); median time from AF detection to the start of anticoagulation was 7 days (0–19). The AF group did not significantly differ from the no AF group with the exception of age (AF 75 vs. no AF 65 years). Among those found to have AF, 100% (7/7) had a LADS score>3 vs. 70% with no AF (44/63). If evidence of ESUS were required for ILR placement, 50% of our AF cases would not have been detected. 

Our experience to date with ILRs suggests that we cannot rely on ESUS or clinical scoring systems alone in determining who may benefit from ILR placement.