Persistent or permanent atrial fibrillation is associated with severe cardioembolic stroke in patients with non-valvular atrial fibrillation

The Hirosaki Stroke and Rehabilitation Center (HSRC) has both a stroke care unit for acute therapy and a stroke rehabilitation unit for further rehabilitation therapy. Therefore, all patients with acute ischemic stroke admitted to HSRC receive consistent therapies in the acute phase, and subsequently in the chronic phase during hospitalization.

Over the 4-year period from April 2011 to March 2015, a total of 846 consecutive patients with CE stroke were admitted to the HSRC for acute therapy and further rehabilitation within 60 days after the onset of CE stroke. Among them, 358 patients with NVAF who were admitted to the HSRC within 48 h of the onset of stroke and had a modified Rankin Scale (mRS score) of 0 or 1 (i.e., without any limitation in physical activities) before the onset of stroke were included in the present study (Fig. 1). The clinical characteristics, stroke severity on admission, and functional outcome at discharge were compared between patients with PAF (n = 127) and PerAF (n = 231).

According to the guidelines of the European Society of Cardiology [9], AF is classified into PAF that returns to sinus rhythm spontaneously within 7 days (usually within 48 h), persistent AF that lasts longer than 7 days but can be returned to sinus rhythm by treatment, long-standing persistent AF that lasts longer than 1 year but can be considered to return to sinus rhythm, and permanent AF that is accepted by the patient and physician, and no further attempts to restore/maintain sinus rhythm. In the present study, the type of AF was determined either as a PAF or as a PerAF that included persistent AF, long-standing persistent AF, and permanent AF. Patients with AF that returned to sinus rhythm spontaneously within 2 weeks in the acute phase of stroke treatment were also classified as a PAF.

The subjects were registered in the Hirosaki Stroke Registry (UMIN Clinical Trials Registry: UMIN000016880).

All patients underwent brain computed tomography on admission. When intracerebral hemorrhage was not detected, we further performed brain magnetic resonance imaging including transversal diffusion-weighted imaging, T2-weighted imaging, fluid-attenuated inversion recovery, and magnetic resonance angiography (Signa Excite HD 1.5 T; GE Medical System, Waukesha, WI). Carotid ultrasonography, chest X-ray, 12-lead and 24 h Holter electrocardiogram (ECG), and standard blood tests were performed on all patients. If necessary, an ECG monitor was installed for 2 weeks during the acute treatment phase of stroke and transesophageal echocardiography was performed. CE stroke was diagnosed according to the Trial of Org10172 in Acute Stroke Treatment Classification [10].

Thrombolysis therapy with intravenous recombinant tissue plasminogen activator (rt-PA) was performed according to the Japanese Guideline [11]. Treatment with OACs before onset was also assessed. Stroke severity was assessed based on National Institutes of Health Stroke Scale (NIHSS) score on admission. Severe stroke was defined as an NIHSS score ≥ 8 [12, 13]. The mRS score at discharge was evaluated as a measure of functional outcome. The CHADS₂ and CHA₂DS₂-VASc scores, for risk stratification of thromboembolism before stroke onset, and the HAS-BLED score, for bleeding risk stratification before stroke onset, were also determined in each patient, as described previously [9, 13‐15]. The risk factors were as follows: congestive heart failure (left ventricular ejection fraction < 40%, New York Heart Association class II or higher heart failure symptoms within 6 months before stroke onset), hypertension (treatment with antihypertensive medication or documented systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg), diabetes mellitus (treatment with insulin or antidiabetic medication, or at least two determinations of diabetic type on separate days as evaluated by an oral glucose tolerance test, fasting blood glucose ≥126 mg/dL, casual blood glucose ≥200 mg/dL, or HbA1c ≥6.5%), vascular disease (coronary artery disease, ankle brachial index ≤0.9, or aortic plaque), and dyslipidemia (treatment with lipid-lowering medication, low-density lipoprotein cholesterol ≥140 mg/dL, high-density lipoprotein cholesterol < 40 mg/dL, or triglycerides ≥150 mg/dL).

Data were expressed as the median (25th–75th percentiles) or n (%). A Mann–Whitney U test, chi-squared test, or Fisher’s exact test was used to compare differences between two groups, as appropriate. The effects of AF type on the NIHSS score on admission and the mRS score at discharge were assessed by multivariate logistic regression analysis as an odds ratio (OR) adjusted for age, sex, body mass index (BMI), congestive heart failure, hypertension, diabetes mellitus, prior cerebral infarction or transient ischemic attack, vascular disease, serum creatinine (Cre) level, and OAC treatment. All statistical analyses were performed using the JMP Pro 11 software (SAS, Cary, NC). Significance was set at P <  0.05.

A comparison of the clinical characteristics of the patients with PAF and PerAF in this study is shown in Table 1. The median age, BMI, and proportion of males were similar between the two groups. The CHADS₂ score, CHA₂DS₂-VASc score, and HAS-BLED score were significantly higher in the PerAF than in the PAF group. Congestive heart failure, diabetes mellitus, and prior cerebral infarction or transient ischemic attack were more frequent in the PerAF than in the PAF group. The rates of hypertension, vascular disease, antiplatelet use, smoking, and dyslipidemia did not differ between the two groups. Cre and Cre clearance (CCr) were similar between the two groups. The proportion of patients undergoing OAC therapy before stroke onset was higher in the PerAF group than in the PAF group, even though still underuse of OAC in both groups. There were no differences between the two groups regarding the proportion of patients treated with rt-PA thrombolysis.

Stroke severity on admission was assessed using the NIHSS. The NIHSS score tended to be higher in the PerAF group than in the PAF group, although not statistically significant (12 [5–20] vs. 9 [4–18]; P = 0.12) (Table 2). Functional outcome at discharge, as assessed by mRS score, was unfavorable in the PerAF group compared with the PAF group (3 [1–5] vs. 2 [1–4]; P = 0.01 (Fig. 2). Consistently, mortality was significantly higher in the PerAF group than in the PAF group (13% vs. 4%; P = 0.005) (Fig. 2 and Table 2).

A multivariate logistic regression analysis showed that PerAF was a significant determinant of severe stroke (NIHSS score ≥ 8) on admission (OR to PAF = 1.80; 95% confidence interval [CI], 1.08–2.98; P = 0.02) (Table 3). PerAF was also a significant determinant of having an mRS score ≥ 3 at discharge (OR to PAF = 2.07; 95% CI, 1.24–3.46; P = 0.006) (Table 3).

Internal carotid artery occlusion, which is indicative of severe cerebral embolism, was examined by MR angiography on admission in patients without OAC therapy before stroke onset (PAF, n = 105 and PerAF, n = 143). The proportion of internal carotid artery occlusion was higher in the PerAF group than in the PAF group (35/143 (24%) vs. 8/105 (8%); P = 0.0005) (Fig. 3).