Background The association between atrial fibrillation (AF) and the prognosis of acute ischaemic stroke (AIS) remains controversial; whether the recombinant tissue plasminogen activator dose influences this association remains poorly understood.

Methods Patients who had an AIS were enrolled from eight stroke centres in China. According to the recombinant tissue plasminogen activator dose, patients treated with intravenous recombinant tissue plasminogen activator within 4.5 hours after symptom onset were divided into a low-dose group (recombinant tissue plasminogen activator <0.85 mg/kg) and a standard-dose group (recombinant tissue plasminogen activator ≥0.85 mg/kg). Patients who had an AIS in the low-dose group and the standard dose group were divided into whether or not they had AF. The main outcomes were major disability (modified Rankin scale (mRS) score 3–5), mortality and vascular events occurring within 3 months.

Results The study included 630 patients who received recombinant tissue plasminogen activator after AIS, including 391 males and 239 females, with a mean age of 65.8 years. Of these patients, 305 (48.4%) received low-dose recombinant tissue plasminogen activator and 325 (51.6%) received standard dose recombinant tissue plasminogen activator. The recombinant tissue plasminogen activator dose significantly influenced the association between AF and death or major disability (p-interaction=0.036). After multivariate adjustment, AF was associated with an increased risk of death or major disability (OR 2.90, 95% CI 1.47 to 5.72, p=0.002), major disability (OR 1.93, 95% CI 1.04 to 3.59, p=0.038) and vascular events (HR 5.01, 95% CI 2.25 to 11.14, p<0.001) within 3 months in patients with standard-dose recombinant tissue plasminogen activator. No significant association was found between AF and any clinical outcome in patients with low-dose recombinant tissue plasminogen activator (all p>0.05). With AF, the mRS score distribution showed a significantly worse shift in patients with standard-dose recombinant tissue plasminogen activator (p=0.016) than in those with low-dose recombinant tissue plasminogen activator (p=0.874).

Conclusions AF may be a strong predictor of poor prognosis in patients who had an AIS receiving standard-dose recombinant tissue plasminogen activator, suggesting that low-dose recombinant tissue plasminogen activator should be administered to patients who had a stroke with AF to improve their prognosis.