已HRV和RSA進行禪坐組與呼吸控制組的心肺交互作用探討

Abstract

本研究主要是探討禪坐練習者和呼吸控制組在不同呼吸率下的心肺交互作用的情形，藉由量化呼吸性竇性心律不整(Respiratory sinus arrhythmia)和心率變異分析(Heart Rate Variability, HRV)來探討交感-副交感活躍的程度和心肺氣體交換的效率。SDNN(心跳間隔標準差)為評估HRV特性的時域分析方法，但是需要有長時間的RR資料才能獲致可靠的分析結果；其解決之道是採取頻域分析方法、以心率訊號的頻譜資訊來估測HRV，則僅需要短時間的心電圖或RR資料即可獲得可靠結果。在頻域分析方法中，HRV的量化分析是藉由低頻高頻功率比(LF/HF)來估測。在低呼吸率時，傳統的HRV分析(tHRV)方法會造成『低頻高頻功率比』的估測值過高(尤其當呼吸率低於每分鐘8次時)。造成估測過高是因為呼吸影響的高頻頻帶（反應副交感神經活動）與反應交感神經活動的低頻頻帶互相重疊。因此傳統頻域分析方法無法從HR頻譜中正確區分交感神經和副交感神經影響的區間。故本研究分別使用兩種方法改進上述的缺點:(1)適應性頻域範圍(Adaptive frequency range, AFR) 和 (2) LMS (Least Mean Square)適應性濾波 (LMS adaptive filtering, AFLMS)。本論文也評估RSA和SDNN以為比較，為了分析以上所述三種指標，我們同步記錄心電圖和呼吸訊號，對象包括兩個群組，其中實驗組為10位具有禪坐經驗者，而控制組為14位健康受測者，且與實驗組同年齡層、但無禪坐經驗。初步結果在高呼吸率(呼吸每分鐘18~26次) 控制組的SDNN高於實驗組，而在低呼吸率(呼吸每分鐘6~14次)時兩個群組幾乎相同。然而在HRV的低高頻功率比中實驗組比控制組在低呼吸率具有較好的交感副交感的平衡(LFLMS/HFLMS ratio~1)。

This thesis reports the results of our study on cardiorespiratory interactions for the control group under pre-designed breathing control paradigm and the experimental group under Chan-meditation natural respiration. Quantification of RSA (respiratory sinus arrhythmia) and HRV (heart rate variability) was employed in the evaluation of regulation of sympathetic-parasympathetic activity and efficiency of pulmonary gas exchange. SDNN (standard deviation of normal-to-normal heart-beating intervals) provides a time-domain approach for evaluating the HRV activity, yet, requires a long-term record of the NN intervals for reliable estimate. An alternative approach based on power spectrum of the NN intervals is feasible for short-time analysis. In the frequency-domain scheme, HRV behavior is quantified by the ratio of low-frequency power to high-frequency power (denoted by LF/HF) of HRV spectrum. In traditional HRV analysis (tHRV), the ratio LF/HF is very likely to be overestimated in the case of low respiratory rate (particularly, slower than 8 breaths/min). Such overestimate is resulted from the overlap of respiratory spectrum with low-frequency band reflecting the sympathetic activity. Accordingly, tHRV cannot accurately distinguish sympathetic-driven from parasympathetic-driven frequency band. In this study, two methods are developed to deal with the issue: adaptive frequency range (AFR) method and least-mean-square adaptive filtering methods (AFLMS). For comparison, RSA and SDNN were also evaluated. To analyze all the above three indexes, ECG (electrocardiograph) and respiratory signals were recorded from two groups, experimental and control group including respectively ten Chan-meditation practitioners and fourteen normal, healthy subjects in the same age range, yet, without any meditation experience. Preliminary results show that, in the high RR (respiration-rate) range (18~26 breaths/min), SDNN of control group is larger than that of experimental group; while in the slow RR range (6~14 breaths/min) SDNN’s of both groups are approximately the same. Moreover, LFLMS/HFLMS ratio characterizing the HRV reveals a better sympathetic-parasympathetic balance (ratio~1) in the experimental than in the control group for the slow RR range.