J. Frederiks

The importance of high-frequency paced breathing in spectral baroreflex sensitivity assessment

Objective Computation of the low-frequency (LF) blood pressure variability (BPV) to heart rate variability (HRV) transfer-index is a common method to assess baroreflex sensitivity (BRS), tacitly assuming that all LF-HRV is caused by baroreflex feedback of LF-BPV. However, respiration may also cause HRV by mechanisms not involving the baroreflex. Application of narrow-band (controlled) high-frequency breathing would keep such non-baroreflex-mediated HRV best out of the LF band. Spontaneous breathing, because of its broad-band character, might cause extra, non-baroreflex-mediated, HRV in the LF band, while paced LF breathing would even concentrate most non-baroreflex-mediated HRV in the LF band. Our study addresses the likely resulting BRS overestimation. Design We recorded HRV and BPV in 20 healthy young subjects in the sitting position. We varied the sympathovagal balance by gradual leg-lowering from horizontal till 60°. At each angle the subjects performed controlled 0.10 Hz, spontaneous, and controlled 0.25 Hz respiration. Results Resting BRS values were 15.5(7.2), 13.1(3.7), and 11.6(6.2) ms/mmHg, respectively. Both the 15/min and the free breathing values differed significantly, P < 0.01 and P = 0.04, from the 6/min breathing value. With lowered legs, the BRS values were 8.2(3.4), 8.3(2.9), and 8.3(3.4) ms/mmHg, respectively. Conclusion Controlled 6/min breathing caused significant BRS overestimation under resting conditions. For the group, spontaneous respiration yielded acceptable BRS values, but individual BRS values deviated sometimes considerably. Conversely, with gravitational load, the respiratory pattern had only minor impact on BRS. Our results demonstrate that the risk of an overestimated BRS value is realistic as long as respiration is not controlled and of high-frequency.

Noninvasive baroreflex sensitivity assessment in geriatric patients: feasibility and role of the coherence criterion

In baroreflex sensitivity (BRS) assessment by the blood pressure to heart rate transfer function modulus in the 0.05-0.15 Hz band, low coherence spectral components are normally rejected. This criterion has, however, no sound theoretical basis. We studied the impact of this criterion in low BRS low coherence subjects. Eleven geriatric patients with cardiac and/or pulmonary disease participated. Five measurement sessions were held. In only 8 measurements (in 5 subjects) there were coherence-components >0.7. Using all frequency components, BRS was 7.9/spl plusmn/5.1 ms/mmHg and coherence was 0.43/spl plusmn/0.14. Using only frequency components with coherence >0.7, BRS was 11.1/spl plusmn/11.3 ms/mmHg (NS) and coherence was 0.75/spl plusmn/0.06 (P<0.01). Our study demonstrates that the coherence criterion often precludes BRS assessment in patients. Our study suggests that application of the criterion introduces the risk of BRS bias (overestimation).

Correlated neurocardiologic and fitness changes in athletes interrupting training.

PURPOSE We studied nine male Dutch top marathon skaters during a 1-month interruption of their training schedules after their last contest in the winter to investigate a possible decline in baroreflex sensitivity. METHODS Before and after this period, a maximal exercise test was done, and at days 0, 4, 7, 14, and 28 neurocardiologic measurement sessions--heart rate and noninvasive baroreflex sensitivity, recumbent and tilt--were performed. RESULTS Interruption of training resulted in a significant and relevant decrease in the maximal oxygen uptake (from 65.7 +/- 5.8 to 61.6 +/- 4.7 mL O2 x kg(-1) x min(-1); P = 0.03), most likely associated with decreased competitive possibilities. Resting heart rate modestly increased (from 54.6 +/- 7.2 to 58.8 +/- 7.5 bpm), however, not significantly. Heart rate during 60 degrees tilt increased considerably (from 70.1 +/- 6.1 to 80.1 +/- 9.1 bpm; P = 0.01), possibly due to a decrease in blood volume and an increase in cardiopulmonary baroreflex gain. Arterial baroreflex sensitivity decreased significantly in the recumbent (from 13.3 +/- 5.4 to 9.8 +/- 3.8 ms x mm Hg(-1), P = 0.04), but not in the 60 degrees tilt position (from 6.7 +/- 2.0 to 6.0 +/- 2.5 ms x mm Hg(-1)). The relative decrease in baroreflex sensitivity and maximal oxygen uptake correlated significantly (r = 0.71, P = 0.02). CONCLUSIONS In summary, our data show that correlated detrimental changes in fitness and baroreflex sensitivity are measurable in these athletes after a month of interruption of training.

Non-baroreflex mediated heart rate variability causes overestimation of baroreflex sensitivity

Spectral assessment of baroreflex sensitivity (BRS) is hampered by non-baroreflex mediated heart rate variability (HRV), which adds to the baroreflex mediated HRV. Relating HRV (baroreflex output) to blood pressure variability (BPV, baroreflex input) may therefore yield BRS values that are too large. The authors propose to overcome this problem by computing BRS in the Mayer waves (0.05-0.15 Hz), while keeping the non-baroreflex mediated HRV in the high frequency band by 15/min (0.25 Hz) metronome breathing.

Similar heart rates, different QT-intervals

The authors hypothesized that the QT-interval depends not only on rate, but also on autonomic influences. To study this, the authors compared QT-intervals within 14 healthy subjects (7 male/7 female, ages 26.0/spl plusmn/3.1 years), at equal heart rates during handgrip and during head-up tilt. QT- and QTc-intervals during handgrip were significantly larger than during tilt (QT: 403/spl plusmn/24 vs. 394/spl plusmn/25 ms; QTc: 436/spl plusmn/16 vs. 426/spl plusmn/15 ms). This finding demonstrates that rate alone is not sufficient to interpret the QT-interval.

Short-term stability of baroreflex sensitivity

We studied the within-session stability of the noninvasive baroreflex sensitivity (BRS) in 23 young healthy male subjects, by analyzing continuous 8-minute recordings, and comparing the results of the first and second 4-minute periods. BRS recordings were made under 15/min (0.25 Hz) metronome respiration. BRS was computed by integrating the spectral systolic blood pressure to inter-beat-interval (IBI) transfer function in the 0.05-0.10 Hz band. For the group, BRS and mean IBI in the first and the second 4-minute period (mean+SD: 8.8/spl plusmn/6.0 vs. 8.2/spl plusmn/6. 5 ms/mmHg, and 927/spl plusmn/151 vs. 938/spl plusmn/153 ms, respectively) did not differ significantly. However, Bland-Altman plots (differences against averages) revealed rather large individual BRS differences between the first and second half of the recording period: the absolute values of the differences were 34.6/spl plusmn/30% of the average (range 0.8-111.2%, median 22.7%). BRS differences were not related to changes in mean IBI. We conclude that for reliable BRS assessment a recording period of 4 minutes is too short. A continuous 8-minute recording with metronome respiration is preferable and feasible.

Dissociation of heart rate and ECG morphology

The authors studied thirteen healthy subjects in which identical heart rates (/spl Delta/HR<1%) could be obtained under different autonomous conditions: by increasing the angle of the legs with the horizontal plane, with tilt angles ranging from 0 till 60/spl deg/ and back, or by performing a handgrip maneuver. During all measurements the thorax was kept at a 700 angle. Heart rate increased from 65.2a/spl plusmn/9.0 (control) to 72.1/spl plusmn/8. 7 (tilt) and 72.1/spl plusmn/8.8 (handgrip) bpm. A number of vector cardiographic parameters differed significantly (P<0.05) between tilt and handgrip, e.g. QRS azimuth (-33.5/spl plusmn/15.0 vs. -22.4/spl plusmn/22.5/spl deg/), QRS duration (103/spl plusmn/10 vs. 107/spl plusmn/13 ms), maximal T vector (646/spl plusmn/200 vs. 703/spl plusmn/184 /spl mu/V), T azimuth (45.3/spl plusmn/14.5 vs. 38.6/spl plusmn/13.6/spl deg/) and the heart rate corrected QT interval (418/spl plusmn/15 vs. 435/spl plusmn/21 ms). This study demonstrates that tilt and handgrip dissociate heart rate and ventricular depolarization and repolarization.