AF affects one-third of patients after cryptogenic stroke, TIA
Almost 30 percent of patients who experience a cryptogenic stroke or transient ischaemic attack (TIA) go on to develop atrial fibrillation (AF), most of them asymptomatic, according to results of the NOR-FIB* study presented at EHRA 2022.
“In this study, we found that an implantable cardiac monitor (ICM) was effective for diagnosing underlying AF, which was identified in 29 percent of patients with a stroke or TIA of indeterminate cause,” presented Dr Barbara Ratajczak-Tretel from the Østfold Hospital Trust, Sarpsborg, and Institute of Clinical Medicine, University of Oslo, Norway.
“More than 90 percent of stroke patients found to have AF had no symptoms of the heart rhythm disorder,” she continued.
The multicentre (18 hospitals in Norway, Sweden, and Denmark), prospective, observational study involved 259 patients with cryptogenic stroke or TIA without a history of AF. All patients had an ICM implanted within 14 days following symptom onset (median 9 days post-stroke). Data transmitted by the device was assessed weekly by physicians (two neurologists and two cardiologists) at a core lab who recommended anticoagulant therapy upon detection of AF.
Patients were followed up for 12 months. During this time, paroxysmal AF, defined as detected atrial arrhythmia episodes lasting ≥2 minutes, occurred in 28.6 percent (n=74) of patients. Of these, 93 percent were asymptomatic. [EHRA 2022, abstract N° 40566]
Compared with those without AF, patients with AF were older (mean age 72.6 vs 62.2 years; p<0.001), had greater stroke severity (as per National Institute Stroke Scale Score) upon admission (median 2 vs 1; p=0.002), and had a higher CHA₂DS₂VASc score prior to stroke (median 3 vs 2; p<0.001). More patients with than without AF had hypertension or dyslipidaemia.
About 87 percent of AF cases (n=64) were detected soon after the index stroke, ie, within the first 2 months of monitoring (mean 47.7 days post-ICM insertion). Ninety-two percent of patients (n=68) experienced recurrent AF episodes.
All patients who developed AF were recommended to receive oral anticoagulation (OAC) treatment. At 12 months, 97.3 percent of patients (n=72) were on OACs.
Eleven patients experienced recurrent stroke during the follow-up period, two and nine in the AF and non-AF groups, respectively (2.7 percent vs 4.9 percent). The two strokes in the AF group occurred before AF was detected or anticoagulation initiated.
“Since most of the patients with detected AF were switched** to OAC, the 12-month risk of recurrent stroke in this group was low,” remarked Ratajczak-Tretel.
Three patients (1.2 percent) experienced device complications, with one incident each of infection at implantation site warranting device removal, skin tear warranting new positioning of device, and subcutaneous haematoma.
“The best therapy to prevent another stroke depends on the underlying cause,” said Ratajczak-Tretel. “Those with AF should receive OACs but a definitive diagnosis is needed before these drugs can be prescribed. AF can be transient and asymptomatic making it difficult to detect.”
“Prolonged cardiac rhythm monitoring to rule out occult AF in cryptogenic stroke [has been] recommended since 2014 and continuous electrocardiogram (ECG) monitoring [for] ≥72 hours whenever possible since 2020,” noted Ratajczak-Tretel.
“[However,] ICMs, most appropriate for this purpose, are still not implemented as a preferred tool in any stroke guidelines,” she continued.
“[In this study,] for many patients, AF would have gone undiagnosed and untreated without the continuous monitoring, putting them at risk of another stroke.”
“[T]he use of ICM was manageable for neurologists and stroke physicians [and] ICM use seems feasible to be implemented in stroke unit evaluation of cryptogenic ischaemic events while waiting for updated guidelines,” she pointed out.