微血管內皮細胞擴張功能及微循環血流參數之量測及分析

Measurement and Analysis of Microvascular Endothelial Dilatation and Microcirculation Parameters

林建廷(Jian-Ting Lin)

指導教授 ： 張慧柔

共同指導教授 ： 王家鍾(Jia-Jung Wang)

義守大學/醫學院/生物醫學工程學系/碩士(2015年)

https://doi.org/10.6343/ISU.2015.00150

摘要

微循環功能障礙會引起不同的血管疾病，像是高血壓，糖尿病和外周血管等疾病，所以人體的微循環與生理系統的運作之間是有密切相關的。本計畫擬以非侵入性的雷射都普勒血流測量技術(LDF)，來擷取微血流振盪訊號及分析微血管內皮細胞的舒張功能及相關參數。
對33位正常受測者，進行臂動脈3分鐘阻塞的外在生理刺激試驗，以LDF來擷取他們的基態期及反應充血期之微血流振盪訊號，並使用自製呼吸訊號量取裝置連續量取人體呼吸變化以及Biopac量取心電圖訊號，又透過快速傅立葉和小波轉換進行頻譜分析。針對微血流振盪訊號中六個不同特徵頻帶的調節機制(0.005至2.0 Hz)：心原性的(0.6-2.0Hz)，呼吸性的(0.15-0.6Hz)，肌原性(0.05-0.15 Hz)，神經原性的(0.02-0.05Hz)，內皮細胞NO依賴型(0.0095-0.02Hz)及內皮細胞NO非依賴型(0.005-0.0095Hz)，加以分析及比較。
由33位受測者的時域分析結果顯示，在3分鐘的阻塞之後，發生最大微血流振盪振幅的平均時間為10.9±5.3秒，而微血管內皮細胞舒張功能指標的平均值為287±131 %。頻域分析結果顯示，以快速傅立葉轉換來分析阻塞前後微血流振盪訊號的特徵頻率變化，在反應充血期間相對於心原性的頻譜功率密度面積及平均功率密度，均顯著大於基態期(2.522.58 vs. 1.771.66, p<0.05; 1.801.84 vs. 1.261.19, p<0.05)，而在反應充血期相對於肌原性的頻譜功率密度面積及平均功率密度，也均顯著大於基態期(p<0.05; p<0.05)。以小波轉換來分析阻塞前後微血流振盪訊號的特徵頻率變化，在反應充血期間相對於全部頻段的平均頻譜功率密度，均顯著大於基態期(尤其是一氧化氮依賴型之代謝性內皮細胞：84.39±36.01 vs. 43.84±30.74；一氧化氮非依賴型之代謝性內皮細胞：176.95±83.36 vs. 55.06±39.97，兩者P<0.001)。再者，以阻塞前後特徵頻率的變化來評估微血管內皮細胞舒張功能的指標(包括FMDEB１：一氧化氮依賴型，FMDEB２：一氧化氮非依賴型，與FMDEB：合併一氧化氮依賴型及非依賴型)，顯示由平均頻譜功率密度所計得的FMDEB１、FMDEB２與FMDEB分別為162.81±180.59 %、360.13±378.23 %及268.41±287.53 %，而由頻譜功率密度百分比所推得的FMDEB１、FMDEB２與FMDEB分別為4.06±21.64 %、74.33±54.30 %及36.03±28.98 %。
從實驗數據歸納獲知，藉分析微血管振盪訊號特徵頻帶的變化，可定量出微血管內皮細胞的舒張功能，進而可作為微循環功能是否病變的參考指標。

關鍵字

微循環 ； 雷射杜卜勒 ； 內皮細胞舒張功能 ； 小波轉換

並列摘要

Microcirculation dysfunction can cause different diseases, such as hypertension, diabetes, peripheral vascular abnormality, and so forth. So, the human microcirculation is closely related to the operation of physiological systems. In the thesis, we measured the microvascular perfudion/flow signals with the laser Doppler flowmetry (LDF) and analyzed the endothelial dilatation and other hemodynamic parameters of the microcirculation.
Two 10-minute recordings of microvascular flow signals on the surface of the index finger were measured with the LDF before (the baseline) and after (the reactive hyperemia) a 3-minute occlusion of brachial arteries in 33 normal subjects. Both the respiratory and ECG signals were also recorded using the self-designed device and with the Biopact device, respectively. Six characteristic frequency intervals of the LDF signal that represent distinct mechanisms were analyzed by the fast Fourier transform and wavelet transform, including the cardiac (0.6–2 Hz), respiratory(0.145–0.6 Hz), myogenic (0.05–0.15 Hz), neurogenic (0.02–0.05 Hz) and endothelium-NO dependent (0.0095–0.02 Hz) and -NO independent (0.005–0.0095 Hz).
Time domain analysis showed that after the occlusion, the average time of occurrence of the maximum amplitude in the LDF signals was 10.9 ± 5.3 sec, and the average of flow-mediated endothelial dilatation was 287±131 %. With the fast Fourier transform, the spectral power density area and the average power density corresponding to the cardiogenic frequency band were significantly greater during the hyperemia than during the baseline (2.522.58 vs. 1.771.66, p<0.05; 1.801.84 vs. 1.261.19, p<0.05). During the hyperemia, the myogenic frequency band had significantly greater spectral power density area and average power density than those during the baseline (p<0.05; p<0.05). With the wavelet transform, significantly larger average power densities corresponding to the six characteristic frequency bands were found as compared with those during the baseline (all p<0.05), specially with the endothelium-NO dependent frequency band (84.39±36.01 vs. 43.84±30.74, p<0.001) and the endothelium-NO independent frequency band (176.95±83.36 vs. 55.06±39.97, p<0.001). Furthermore, based on the average power densities before and after the occlusion, we found that the FMDEB１(endothelium-NO dependent), the FMDEB２(endothelium-NO independent), and the FMDEB (endothelium-related) were 162.81±180.59 %, 360.13±378.23 % and 268.41±287.53 %, respectively. Also, based on the percentages of the spectral power densities, the FMDEB１, FMDEB２ and FMDEB were found to be 4.06±21.64 %, 74.33±54.30 % and 36.03±28.98 %.
In conclusion, we can quantify the indexes associated with the microvascular endothelial dilatation based on the change in the characteristic frequency components of the microvascular blood flow signals, suggesting that those indexes may be useful for diagnosis of microcirculation dysfunction.

並列關鍵字

Microcirculation ； Laser Doppler ； Endothelial dilatation ； Wavelet transform

參考文獻

[15]陳天送(成功大學醫學工程研究所)，「應用近紅外光技術於糖尿病患者足部血管病變之研究」，96 年度國科會專題研究計畫成果報告(2007/08/01~2010/07/31)。

[18]許昕(臺灣科技大學醫學工程研究所)，「微循環血流變異度之研究與評估指標開發」，國科會專題研究計畫成果報告(2010/08/01~2012/07/31)。

[5]Shin SY, Na JO, Lim HE, Choi CU, Choi JI, Kim SH, Kim EJ, Park SW, Rha SW, Park CG, Seo HS, Oh DJ, Kim YH. Improved endothelial function in patients with atrial fibrillation through maintenance of sinus rhythm by successful catheter ablation. J CardiovascElectrophysiol. 22(4): 376-382, 2011.

[6]Selvaraj N, Jaryal AK, Santhosh J, Anand S, Deepak KK. Monitoring of reactive hyperemia usingphotoplethysmographic pulse amplitude and transit time. J ClinMonitComput. 23(5): 315-322, 2009.

[7]McEniery CM, Wallace S, Mackenzie IS, McDonnell B, Yasmin, Newby DE, Cockcroft JR, WilkinsonIB. Endothelial function is associated with pulse pressure, pulse wave velocity, and augmentation index inhealthy humans. Hypertension. 48(4): 602-608, 2006.