670Risks of unfavorable clinical outcomes in patients with first diagnosed stroke-associated atrial fibrillation

Atrial fibrillation (AF) in patients who have survived a cardioembolic stroke is associated with high risks of death and secondary thromboembolic complications. In some cases AF is diagnosed at admission to a hospital with the stroke, and this observation requires comprehensive research. Aim of this study was to investigate the prognosis in patients with AF that was diagnosed in the acute phase of cardioembolic stroke (at admission). Methods: From October 1, 2013 to March 31, 2015 1291 consecutive patients with AF and cardioembolic stroke were screened. 661 patients (63.1% females, average age 68.0 [62.0; 74.0] years) met inclusion (cardioembolic stroke ≤ 90 days; non-valvular AF; absence of hemodynamic stenosis in the carotid arteries) and exclusion (thyroid diseases; cancer; severe neurological deficiency) criteria. Results: A high level of co-morbidity was observed among enrolled patients: hypertension (100%), IHD (61.4%), DM (52.3%), CKD (32.1%), and neurological deficiency: NIHSS scale index 9,0 [7,0; 11.0] points; the Bartell index 35.0 [30.0, 45.0] points. In 154 (23.3%) patients AF was first diagnosed (fdAF) at the onset of cardioembolic stroke. Among these patients the HAS–BLED score, age, and hypertension and IHD rates were comparable with the rest of the group; however, the percentage of patients with CKD (48,7% vs. 32,1%; p=0,008), DM (53,9% vs. 46,1%; p=0,011) or MI (5,8% vs. 7,3%; p=0,046) was significantly higher. Neurological deficiency was more severe in the rest of the studied group: NIHSS scale index (7,01 vs 9,2 points; p=0,013) and the Bartell index (32,8 vs. 35,12 points; p=0,009). According to Kaplan-Meier analysis the highest death rate during one year was observed in the group of patients with fdAF, where the number of deaths was almost two-fold higher (46 [29.9%] patients, of which 24 were women) than in the whole group of patients with AF (p < 0.001). The age of patients with fdAF at time of death was significantly (p < 0.001) greater than that of those who survived (71.72 ± 6.65 years vs. 65.80 ± 6.14 years). Findings from the multivariate Cox regression analysis demonstrate that throughout the whole group of patients with AF and a history of cardioembolic stroke mortality hazards were as follows: NIHSS score for stroke severity >5 (95% CI: 1.34 (1.09 – 1.64); p = 0.006); fdAF (95% CI: 3.99 (1.29 – 12.35); p = 0.016); treatment with ASA (95% CI: 3.84 (1.18 – 12.48); p = 0.025); history of CKD (95% CI: 3.41 (2.14 – 5.44); p < 0.001) and age greater than 70 years (95% CI: 0.92 (0.88 – 0.96); p<0.001). In patients with fdAF mortality hazards are associated with stroke severity and the history of IHD, CHF or CKD. Conclusions: Our results demonstrate a high rate of fdAF and its poor prognosis in patients with cardioembolic stroke, despite at less severe neurological deficiency.