On Spectral Analysis of Heart Rate Variability during Very Slow Yogic Breathing Emil Jovanov Electrical and Computer Engineering Dept. University of Alabama in Huntsville http://www.ece.uah.edu/~jovanov email: jovanov@ece.uah.edu Introduction Slow breathing provides valuable insights into mechanisms of cardio respiratory synchronization Normal breathing (> 9 b/min) HF (.5-.4Hz) Slower breathing LF (.4-.5Hz) Very slow breathing (>25 sec/breath) VLF (.3-.4Hz) Reported resonant characteristic Vaschillo, Song & Lehrer Available physiological margin? Very slow yogic breathing ~ breath/min EMBS 25, Shanghai, China 2
Methods Eight measurements of Nadi Shodhana Pranayama on a single practitioner Healthy practitioner with year experience Measurements Respiration, HRV, and SpO2 Before, during, and after exercise HRV WHRM system ( ms precision) Respiration (NIMS Respitrace 2 system) Ribcage, Abdominal, and total Volume, Flow Blood oxygen saturation (Ares / Advanced Brain Monitoring) External synchronization very hard EMBS 25, Shanghai, China 3 Nadi Shodhana Pranayama Yogic breathing - three phases abdominal, chest, clavicular) Breathing phases. Inhalation 2. Internal retention 3. Exhalation 4. External retention 2 3 4 Alternating nostrils Duration ratio :2:2: (:2:2: sec) EMBS 25, Shanghai, China 4
Lung volume during exercise 4 Lung volume during exercise 3.5 3 2.5 2.5.5 2 3 4 5 6 7 EMBS 25, Shanghai, China 5 Lung volumes detail 2 Volume, V-Abdomen, RRint during three breath cycles.5.5 -.5 - -.5-2 62 64 66 68 7 72 74 76 78 8 Time [s] EMBS 25, Shanghai, China 6
Ventilation during exercise Ventilation during exercise 3 2.5 Ventilation 3.64 times lower during exercise 2.5.5 4 5 6 7 8 9 2 3 Time [s] EMBS 25, Shanghai, China 7 Blood oxygen saturation Breath phase (symbolic lung volume).8.6.4.2 2 3 4 5 6 7 8 9 SpO2 [%] 95 9 85 2 3 4 5 6 7 8 9 EMBS 25, Shanghai, China 8
RRint during exercise 2 After 2 9 8 7 9 6 25 3 35 4 45 5 55 6 Before 2 8 7 LF/HF=46 6 5 5 2 25 3 35 During 9 8 7 6 4 5 6 7 8 9 EMBS 25, Shanghai, China 9 Typical changes during one cycle HR [bpm] 5 Inhalation One cycle of slow breathing ( breath/min) External retention Internal retention Exhalation 95 9 85 8 75 7 65 6 55 5 II III I 94 95 96 97 98 99 EMBS 25, Shanghai, Time [s] China
Ectopic beats Probably due to overstimulation of vagus during rapid abdominal inhalation Atrial extra-systoles Number of ectopic events in the given set of experiments 5 E (3), E3(), 4.5 and E5(4) 4 Average relative 3.5 position of 6 events 3 74 s (63s) 2.5 Two events at 2 9 and 49 seconds.5 Typical example.5 453 453.5 454 454.5 455 455.5 456 456.5 457 EMBS 25, Shanghai, China Spectral Components HR [bpm] One cycle of slow breathing ( breath/min) 5 Inhalation External retention Internal retention Exhalation 95 9 85 8 75 7 65 6 55 II III I 5 94 95 96 97 98 99 Time [s] T 3 3 sec (.32Hz) T 22 sec (.45Hz) T 2 6 sec (.6Hz) EMBS 25, Shanghai, China 2
HRV spectrum VLF LF HF % 8% 6% 4% 2% % Before NS After EMBS 25, Shanghai, China 3 Typical spectral components Parameter Before exercise During exercise After exercise Mean RRint (SE) 959.29 (.26) 94.3 * (.8) 998.98 (4.45) Total HRV power [ms 2 ] (SE),625.62 (,386.7) 9,36.96 (836.64),93.74 (,75.58) VLF power [ms 2 ] (%) 2,329.4 (2.7) 7,472.99 * (8.96 * ) 734.28 * (7.7 ) LF power [ms 2 ] (%) 8,494.94 (7.54),352.76 * (5.8 * ) 9,823. (88.65 ) HF power [ms 2 ] (%) 8.54 (7.75) 2.2 * (2.24 ) 374.36 * (3.64 ) LF/HF (SE) 4.33 (3.73) 9.5 (.8) 5.93 (6.35) LF max. frequency [Hz] (SE).9875 (.494).7525 (.956).725 (.37) HF max. frequency [Hz] (SE).7425 (.679).6 (.42).85625 (.438) The results of raw HRV data processing; ectopic heart beats are not filtered out. Significant differences when compared against the state before exercise: p <.5, p <., * p <. EMBS 25, Shanghai, China 4
Results Rapid abdominal inhalation stimulates vagus and increases heart rate Arrhythmia ectopic beats Relative power of VLF band is significantly smaller after exercise M=2.7% M=7.7%, t(7)=-4.2, p<.5. HF power significantly reduced during exercise, and remains small after exercise Decrease of breathing frequency after exercise 5.5 4.3 b/min Increase of HF/LF 4.33 5.93 EMBS 25, Shanghai, China 5 Conclusions Model development ANS characterization in the absence of RSA (extended quiet periods) Serious problems with instrumentation Similar state before and after Conditioning after years of practice Resonant characteristics of the RSA Resonant frequency not fixed, influenced by state Available physiological margin increased by systematic practice Future experiments more participants - experienced practitioners/beginners Blood pressure measurement and correlation EMBS 25, Shanghai, China 6