Purpose Cerebrovascular reactivity (CVR) is commonly assessed via transcranial Doppler ultrasonography (TCD) through the temporal bone window. However, inadequate temporal acoustic window in some individuals necessitates alternative approaches. This study aimed to evaluate the feasibility of using the submandibular window to measure breath holding index (BHI) and hyperventilation index (HVI) from the distal internal carotid artery (ICA) and to compare them with measurements of middle cerebral artery (MCA) obtained via the temporal window. Methods Fifty healthy volunteers (25 females, 25 males; age range: 25–64 years) underwent CVR evaluation using TCD. BHI and HVI were calculated both from the MCA via the temporal window and from the distal ICA via the submandibular window. Each test was repeated twice and averaged. Spearman’s correlation analysis was used to assess relationships between variables. Intra-session reproducibility (first vs. second measurement) and inter-vessel agreement (MCA vs. distal ICA) of cerebrovascular parameters were evaluated using the Intraclass Correlation Coefficient (ICC) and Bland-Altman analysis. Results The mean BHI was significantly higher in the MCA (0.70 ± 0.38) than the ICA (0.55 ± 0.36) (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(p = 0.017\)</EquationSource> </InlineEquation>), while HVI values were comparable (MCA: 0.56 ± 0.15; ICA: 0.51 ± 0.15; <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(p = 0.248\)</EquationSource> </InlineEquation>). A weak-to-moderate, yet significant, positive correlation was found between MCA and ICA values for both BHI (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(r_s = 0.366\)</EquationSource> </InlineEquation>, <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(p = 0.009\)</EquationSource> </InlineEquation>) and HVI (<InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(r_s = 0.314\)</EquationSource> </InlineEquation>, <InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(p = 0.026\)</EquationSource> </InlineEquation>). Both MCA and distal ICA measurements showed high reproducibility for standard TCD parameters. The reproducibility of BHI was moderate for MCA (ICC <InlineEquation ID="IEq7"> <EquationSource Format="TEX">\(= 0.694\)</EquationSource> </InlineEquation>) and weak for ICA (ICC <InlineEquation ID="IEq8"> <EquationSource Format="TEX">\(= 0.453\)</EquationSource> </InlineEquation>), while reproducibility of HVI showed the opposite pattern; weak for MCA (ICC <InlineEquation ID="IEq9"> <EquationSource Format="TEX">\(= 0.436\)</EquationSource> </InlineEquation>) and moderate for ICA (ICC <InlineEquation ID="IEq10"> <EquationSource Format="TEX">\(= 0.679\)</EquationSource> </InlineEquation>). Bland–Altman analyses demonstrated a small mean bias for BHI (0.15 ± 0.44) and HVI (0.05 ± 0.18), with narrower limits of agreement for HVI indicating greater measurement stability. Agreement analysis revealed low-to-moderate inter-site reproducibility, with the highest concordance observed for PI (ICC <InlineEquation ID="IEq11"> <EquationSource Format="TEX">\(= 0.739\)</EquationSource> </InlineEquation>; <InlineEquation ID="IEq12"> <EquationSource Format="TEX">\(p < 0.001\)</EquationSource> </InlineEquation>) and RI (ICC <InlineEquation ID="IEq13"> <EquationSource Format="TEX">\(= 0.646\)</EquationSource> </InlineEquation>; <InlineEquation ID="IEq14"> <EquationSource Format="TEX">\(p < 0.001\)</EquationSource> </InlineEquation>), while other parameters, including BHI (ICC <InlineEquation ID="IEq15"> <EquationSource Format="TEX">\(= 0.439\)</EquationSource> </InlineEquation>; 95% CI: 0.012-−0.682; <InlineEquation ID="IEq16"> <EquationSource Format="TEX">\(p = 0.023\)</EquationSource> </InlineEquation>) and HVI (ICC <InlineEquation ID="IEq17"> <EquationSource Format="TEX">\(= 0.454\)</EquationSource> </InlineEquation>; 95% CI: 0.038-−0.690; <InlineEquation ID="IEq18"> <EquationSource Format="TEX">\(p = 0.018\)</EquationSource> </InlineEquation>), showed weaker agreement. Conclusion Although the correlation between MCA and distal ICA measurements was statistically significant, the strength of association was weak and may have limited clinical relevance. Submandibular ICA measurements may serve as supportive or alternative tools, especially in cases where the temporal window is insufficient, but should not replace conventional approaches.

Assessment of cerebrovascular reactivity with the TCD through the submandibular window

期刊论文
出版模式：

开放获取
发表日期：

2026年4月15日

查看全文

Burak Özaltun,
Mine Hayriye Sorgun,
Canan Togay Işıkay

摘要

Purpose

Cerebrovascular reactivity (CVR) is commonly assessed via transcranial Doppler ultrasonography (TCD) through the temporal bone window. However, inadequate temporal acoustic window in some individuals necessitates alternative approaches. This study aimed to evaluate the feasibility of using the submandibular window to measure breath holding index (BHI) and hyperventilation index (HVI) from the distal internal carotid artery (ICA) and to compare them with measurements of middle cerebral artery (MCA) obtained via the temporal window.

Methods

Fifty healthy volunteers (25 females, 25 males; age range: 25–64 years) underwent CVR evaluation using TCD. BHI and HVI were calculated both from the MCA via the temporal window and from the distal ICA via the submandibular window. Each test was repeated twice and averaged. Spearman’s correlation analysis was used to assess relationships between variables. Intra-session reproducibility (first vs. second measurement) and inter-vessel agreement (MCA vs. distal ICA) of cerebrovascular parameters were evaluated using the Intraclass Correlation Coefficient (ICC) and Bland-Altman analysis.

Results

The mean BHI was significantly higher in the MCA (0.70 ± 0.38) than the ICA (0.55 ± 0.36) ( \(p = 0.017\) ), while HVI values were comparable (MCA: 0.56 ± 0.15; ICA: 0.51 ± 0.15; \(p = 0.248\) ). A weak-to-moderate, yet significant, positive correlation was found between MCA and ICA values for both BHI ( \(r_s = 0.366\) , \(p = 0.009\) ) and HVI ( \(r_s = 0.314\) , \(p = 0.026\) ). Both MCA and distal ICA measurements showed high reproducibility for standard TCD parameters. The reproducibility of BHI was moderate for MCA (ICC \(= 0.694\) ) and weak for ICA (ICC \(= 0.453\) ), while reproducibility of HVI showed the opposite pattern; weak for MCA (ICC \(= 0.436\) ) and moderate for ICA (ICC \(= 0.679\) ). Bland–Altman analyses demonstrated a small mean bias for BHI (0.15 ± 0.44) and HVI (0.05 ± 0.18), with narrower limits of agreement for HVI indicating greater measurement stability. Agreement analysis revealed low-to-moderate inter-site reproducibility, with the highest concordance observed for PI (ICC \(= 0.739\) ; \(p < 0.001\) ) and RI (ICC \(= 0.646\) ; \(p < 0.001\) ), while other parameters, including BHI (ICC \(= 0.439\) ; 95% CI: 0.012-−0.682; \(p = 0.023\) ) and HVI (ICC \(= 0.454\) ; 95% CI: 0.038-−0.690; \(p = 0.018\) ), showed weaker agreement.

Conclusion

Although the correlation between MCA and distal ICA measurements was statistically significant, the strength of association was weak and may have limited clinical relevance. Submandibular ICA measurements may serve as supportive or alternative tools, especially in cases where the temporal window is insufficient, but should not replace conventional approaches.