Free thyroxine, brain frailty and clock drawing test performance in patients with acute minor stroke or transient ischaemic attack

Abstract

Objective: Thyroid dysfunction is associated with an elevated risk of cognitive decline, but the mechanism underlying this relationship is elusive. In this study, we investigate the relationships between free thyroxine (FT4), brain frailty and clock drawing test (CDT) performance in patients with acute minor stroke or transient ischaemic attack (TIA). Design, Patients and Measurements: A total of 204 consecutive patients admitted to our hospital within 72 h after the onset of acute minor stroke or TIA were prospectively enroled and categorized in terms of quartiles of FT4 between March 2018 and August 2019. Brain frailty on magnetic resonance imaging was rated according to previously published criteria. Cognitive performance was assessed with the CDT.Results: Generalized linear analysis revealed that FT4 was independently associated with higher brain frailty score after adjusting potential confounders (β, 0.03; 95% confidence interval [CI], 0.00-0.06; p = 0.0205), which is consistent with the result of FT4 (quartile) as a categorical variable (β, 0.34; 95% CI, 0.01-0.68; p = 0.0059; p trend = 0.0807). A nonlinear relationship was detected between FT4 and brain frailty score, which had an inflection point of 1.19. FT4 was also associated with poor CDT performance (odds ratio, 1.15; 95% CI, 1.04-1.26; p = 0.0051). And mediation analysis found that brain frailty partially mediated the positive relationship between FT4 and poor CDT performance (indirect effect = 0.0024; 95% CI, 0.0003-0.01, p = 0.04).Conclusions: Our findings suggested that a higher FT4 level was associated with a higher brain frailty score and poorer CDT performance, and brain frailty might play an important effect on the association between FT4 and cognitive decline.

Eosinophils, Stroke-Associated Pneumonia, and Outcome After Mechanical Thrombectomy for Acute Ischemic Stroke

“…Two-sided values of P < 0.05 were considered statistically significant. The total procedure of statistical analysis was divided into four steps (Guo et al, 2021 ; Yu et al, 2021 ).…”

“…For multivariate analysis, we first included age and women (model 1) and then included variables in the final models if they were significantly associated with SAP ( P < 0.10) or changed the estimates of eosinophils on SAP by more than 10% (model 2; age, hyperlipidemia, baseline National Institutes of Health Stroke Scale (NIHSS), ASPECTS, occluded artery, dysphagia, stroke etiology, and collateral score). Tables II–V in Appendix A show the associations of each confounder with the outcomes of interest (Guo et al, 2021 ; Yu et al, 2021 ). (2) We used generalized additive models to identify the non-linear relationships because eosinophils were a continuous variable.…”

“…If a non-linear relationship was observed, a two-piecewise linear regression model was used to calculate the threshold effect of the eosinophils on SAP in terms of the smoothing plot. When the ratio between eosinophils and SAP appeared obvious in a smoothed curve, the recursive method automatically calculates the inflection point, where the maximum model likelihood will be used (Guo et al, 2021 ; Yu et al, 2021 ). (3) We conducted subgroup analyses to assess the robustness of the association between low eosinophils and SAP using stratified regression models.…”

“…(3) We conducted subgroup analyses to assess the robustness of the association between low eosinophils and SAP using stratified regression models. The modifications and interactions between eosinophils and subgroup variables on the SAP were tested by likelihood ratio tests (Guo et al, 2021 ; Yu et al, 2021 ).…”

Oxygen saturation before and after mechanical thrombectomy and functional outcome in patients with acute ischemic stroke

“…Subsequently, variables were included in model 2 if they were associated with poor outcome ( P < 0.10) or changed the estimated effect of oxygen saturation on poor outcome by more than 10%. The association of each confounder with the outcomes of interest is shown in Supplementary Tables II–VI in Supplementary Appendix A ( 16 , 17 ).…”