Impact of atrial fibrillation on the cognitive decline in Alzheimer’s disease

Patient clinical characteristics

There were 14 (8.2%) patients in the AF group and 156 (91.8%) in the SR group. The clinical characteristics of all the patients are shown in Table 1. The percentage of AD patients in the AF group was significantly higher than that in the SR group (p = 0.0382: 92.9% and 65.4%, respectively). Patients in the AF group had significantly lower MMSE-J scores (average [SD]) than those in the SR group (19.4 [3.1] vs 22.0 [4.4]: p = 0.0347, respectively). The average [SD] age at the first visit was not significantly different between the AF and SR groups (81.6 [4.1] and 79.3 [7.6] years old, respectively; p = 0.2575). The frequencies of hypertension and dyslipidemia were not significantly different between the AF and SR groups, but the frequency of diabetes mellitus was significantly higher in the AF group than in the SR group (50.0% and 23.7%, respectively; p = 0.0315). The average [SD] HbA1c was not significantly different between the AF and SR groups (6.4 [0.78]% and 6.1 [0.87]%, respectively; p = 0.2211). The prevalence of dysthyroidism (hyper- or hypothyroidism assessed by laboratory data) did not differ between the two groups. The cardiothoracic ratio (CTR) measured on the chest X-ray image was significantly higher for the AF group than for the SR group (mean [SD]%, 57.6 [5.9] and 50.0 [5.6], respectively; p < 0.0001). There were no significant differences between the systolic and diastolic blood pressures of the two groups. Renal function and hemoglobin concentration were not different between the two groups. The inflammation-related data, i.e., WBC, CRP, and homocysteine, were not significantly different between the two groups. Antithrombotic therapy was administered to 88.2% of the patients with AF. However, the doses of three patients were not optimal; that is, three patients taking warfarin showed a prothrombin time international normalized ratio (PT-INR) of < 1.6.

Table 1 Patient characteristicsAffected white matter lesions by AF

There was no significant difference in cerebral volume between the AF and SR groups (average [SD] mL: 1014 [74.9] and 1029 [97.0], respectively; p = 0.5741) (Fig. 2A). The volume of PVWML was significantly greater for the AF group than for the SR group (mean [SD] mL: 6.85 [3.78] vs 4.37 [3.21], respectively; p = 0.0070) (Fig. 2B). The volume of DWML was relatively greater in the AF group than in the SR group (mean [SD] mL: 11.34 [8.25] vs 7.26 [8.00], respectively; p = 0.0699) (Fig. 2C). Even though patients with AD and aMCI were separately analyzed, these differences showed the same results (Fig. S1). The existence of any old ischemic lesions in the cerebrum was observed in 15.4%, 8.8%, and 9.3% of AD patients in the AF group, AD patients in the SR group, and aMCI patients in the SR group, respectively. These percentages were not significantly different (Additional file 2: Table S1). The numbers of CMBs in the subcortical and the deep white matter regions showed no significant difference between the AF and SR groups (average [SD]; 1.2 [4.3] vs 2.5 [7.9], p = 0.4463, and 0.2 [0.4] vs 0.3 [1.5], p = 0.5201, respectively). The trend was preserved even if patients were separated into AD and aMCI (Additional file 3: Table S2). According to the MRA findings, no patient showed significant stenosis in the main arteries of the brain (data not shown).

Fig. 2figure 2

Scatter graphs of cerebral volume (A), PVWML volume (B), and DWML volume (C). There was no difference in the distributions of cerebral volume in the AF and SR groups. The PVWML volume in the AF group was significantly increased (§p = 0.0070). Regarding the distribution of DWML volume, the AF group had a greater volume than the SR group, although the difference was not significant (p = 0.0699). The Y-axis represents the volume in milliliters. The gray circles and bars on the right side of the scatter plots indicate the averages and standard deviations, respectively. PVWML, periventricular white matter lesion; DWML, deep white matter lesion

Multivariate analysis

Confounding factors which presented the significant difference between the AF and SR groups by the bivariate analysis were adopted for the multivariate analysis, i.e., MMSE-J score, diabetes mellitus, CTR, and PVWML. CTR and PVWML still showed a significant correlation with AF (correlation coefficient [CC] = 0.231, p = 0.003, and CC = 0.365, p < 0.001, respectively). However, the MMSE-J score did not present a significant correlation with AF (CC = − 0.122, p = 0.120). Diabetes mellitus also showed no significant correlation (CC = 0.147, p = 0.062). Moreover, patients with AD were separately analyzed, significant correlation with AF was also observed in CTR and PVWML (CC = 0.348, p < 0.001, and CC = 0.214, p = 0.022, respectively). MMSE-J score and diabetes mellitus did not show a significant correlation with AF (CC = − 0.085, p = 0.367, and CC = 0.122, p = 0.194, respectively). When the involvement of each factor in the lowering of the MMSE-J score was assessed by Spearman’s rank correlation coefficient test, only the presence of AF showed a significant contribution (Table 2). While the stepwise regression analysis was adopted with minimal BIC rule, AF also showed a significant correlation with the lowering of the MMSE-J score (Additional file 4: Table S3).

Table 2 Spearman’s rank correlation coefficient test for the lowering of MMSE-J scoreRetained cerebral blood flow in AF

rCBF was calculated using the SEE analysis of SPECT data in a specific area of the brain, which is expected to be related to AD pathology (Table 3). Both the AF and SR groups showed decreased rCBF in these areas. However, there were no significant differences in the data between the AF and SR groups. Instead, the rCBF reduction was relatively remarkable in the AD-related areas for the SR group than for the AF group. These trends became more remarkable, when patients with AD and aMCI were separately analyzed (Additional file 5: Table S4).

Table 3 Stereotactic extraction estimation analysis for SPECT

留言 (0)

沒有登入
gif