1
Online Resource for
Can short-term heart rate variability be used to monitor fentanyl-midazolam induced changes in ANS preceding respiratory depression?
Anne-Louise Smith Harry Owen Karen J. Reynolds[1]
This supplementary document holds:
a) Tables of confidence interval data that are displayed in figures in the main document:
Table 3 Description of HRV indices
Table 4 Post-fentanyl 95% confidence intervals
Table 5Peri-CD 95% confidence intervals
Table 6Peri-UAWO 95% confidence intervals
b) Thumbnails of the individual subject responses to fentanyl for the traditional indices:
Figure 8 Individual responses to fentanyl for time domain indices
a)mean HR, b) SDNNmc, c) RMSSDmc and d) pNN20
Figure 9 Individual responses for Lomb Scargle spectral indices
a) LombLF, b) LombHFnu, c) LombLF/HF and d) LombTotal
c) References for the HRV indices in Table 3
Table 3 Description of HRV indices
Index / Units / Reference / Description1: SDNN / ms / Task Force [1] / Standard deviation of NN intervals
2: SDNNmc
3: RMSSD / ms / Task Force [1] / RMS of successive differences
4: RMSSDmc
5: pNN20 ^ / % / Task Force [1] / Percent of NN intervals >50 ms
6: Lomb LF / ms2 / Moody [2] / Low frequency power (0.04-0.15 Hz)
7: LombHF / ms2 / Moody [2] / High frequency power (0.15-0.4 Hz)
8: Lomb HFnu / Moody [2] / HF normalised units (over LF+HF)
9: Lomb LF/HF / Moody [2] / Ratio of low to high frequency
10: LombTotal / ms2 / Moody [2] / Total power,LF+HF (0.04 to 0.4 Hz)
11: accel / Guzik[3] from Piskorski[4] / Asymmetry – accelerations
12: acv0x / Nikolopoulos[5] / Lag of auto covariance first zero crossing
13: assym(R/L) / Kovatchev[6] / SAA, sample asymmetry analysis
14: CVdRR / Tateno [7] / Coefficient of variation of dRR
15: gradRR / Marciano [8] / Gradient RR-intervals (average), DR
16: kurtRRz / Olesen[9] Griffin [10] / Kurtosis, peakiness of RR histogram, normalised (z)
17: mean.r(L1-6) / Sosnowski[11] / Mean of correlation coefficient for return maps with lags 1-6, estimate of RSA
18: meanRR / ms / Goldberger [12] / Mean RR-interval, sympathovagal balance
19: norm.dRR / Tateno [7] / Difference of distribution to normal measured by Kolmogorov-Smirnov or Lilliefors test statistic
20: normRR / Tateno [7]
21: PolVar20 ^ / % / Voss [13] Wessel [14] / Probability of low variability, <20 ms difference for 6 beats in succession
22: pQa / % / Raetz[15] / Proportion of sequences by quadrant:
a: decreased RR-interval then increase,
23: pQb / % / b: increase followed by increase,
24: pQc / % / c: decrease followed by decrease,
25: r(RR) / Otzenberger[16] / Interbeat autocorrelation coefficient, nonlinear estimate of RSA
26: SDNN/RMSSD / Balocchi[17] / Ratio similar to LF/HF and SD2/SD1, sympathovagal balance
27: skew.dRR / Tateno [7] / Asymmetry RR-interval differences
28: skewRRz / Griffin [10] / Asymmetry of RR histogram (negative has left tail), normalised (z)
29: sign(dRR) ^ / Ashkenazy [18] / Sign of differences
30: TACI(10) / Arif[19] / Threshold based acceleration index with 10
31: TACI(20) / or 20 ms thresholds
Table 4Post-fentanyl 95% confidence intervals for median of indices for baseline and 3 of 10 consecutive 30-beat intervals(F1 to F10)aftersedative administration (results from every third interval shown) and intervals reaching significancefor difference between oxygen mask baseline and post-fentanyl windows(a to j)
Index / Units / O2 maskBaselinea / F3(c) / F6 (f) / F9 (i) / p<0.006 / p<0.04
1: SDNN / ms / 33.1–47.5 / 13.1–49.1 / 12.4–50.4 / 24.8–71.4 / d
2: SDNNmc / 3.98–4.98 / 2.02–5.01 / 1.66–5.24 / 2.94–6.09
3: RMSSD / ms / 22.6–39.2 / 8.53–54.56 / 9.21–29.30 / 13.0–47.0 / d
4: RMSSDmc / 2.80–4.28 / 1.20–4.80 / 1.25–3.26 / 1.56–5.36 / d,f
5: pNN20 ^ / % / 34.5–55.2 / 1.72–51.72 / 1.72–44.83 / 10.3–55.2 / d
6: Lomb LF / ms2 / 20.3–33.2 / 14.8–31.5 / 18.9–37.3 / 19.5–32.5
7: LombHF / ms2 / 12.1–26.1 / 12.6–26.2 / 11.4–23.8 / 9.17–24.17
8: Lomb HFnu / 26.1–52.3 / 31.7–55.9 / 25.6–56.7 / 19.9–56.1
9: Lomb LF/HF / 0.91–2.83 / 0.83–2.07 / 0.76–2.91 / 0.81–4.02
10: LombTotal / ms2 / 49.6–52.0 / 39.5–51.6 / 32.5–52.7 / 35.9–51.1 / e, g, j / b,e,g, h,j
11: accel / 0.48–0.55 / 0.52–0.63 / 0.41–0.62 / 0.48–0.65
12: acv0x / 3.00–3.00 / 0.00–3.00 / 2.00–5.00 / 0.00–4.00
13: assym(R/L) / 0.80–1.07 / 0.45–3.08 / 0.71–2.73 / 0.20–1.92
14: CVdRR / 124–128 / 124–136 / 123–139 / 125–141
15: gradRR / -0.55–1.17 / -0.40–1.11 / -1.07–3.72 / -1.03–2.54
16: kurtRRz / 2.19–2.55 / 1.95–2.89 / 2.17–2.56 / 2.06–3.19 / d
17: mean.r(L1-6) / -0.03–0.05 / -0.02–0.35 / -0.00–0.35 / 0.10–0.34 / e, h, i / d,e, g,h,i,j
18: meanRR / ms / 757–947 / 622–1042 / 648–1035 / 672–1017
19: norm.dRR / 0.11–0.13 / 0.10–0.14 / 0.10–0.16 / 0.10–0.19
20: normRR / 0.11–0.13 / 0.11–0.19 / 0.10–0.15 / 0.12–0.19 / h / h
21: PolVar20 ^ / % / 0.00–2.08 / 0.00–87.50 / 0.00–95.83 / 0.00–22.92 / a, b, c, d, e, f, g, h,j
22: pQa / % / 15.1–18.5 / 17.9–27.0 / 18.1–25.5 / 15.1–25.0 / c, e,f
23: pQb / % / 25.5–30.8 / 18.5–36.3 / 23.1–37.0 / 23.2–41.9 / h
24: pQc / % / 29.6–35.7 / 18.5–32.0 / 22.3–32.1 / 21.4–33.9 / d
25: r(RR) / 0.63–0.74 / 0.43–0.79 / 0.56–0.84 / 0.53–0.91
26: SDNN/RMSSD / 1.17–1.43 / 0.96–1.60 / 1.10–1.78 / 1.07–2.31
27: skew.dRR / 0.68–0.89 / 0.48–1.06 / 0.64–1.43 / 0.78–1.43
28: skewRRz / -0.17–0.03 / -0.23–0.60 / -0.50–0.53 / -0.71–0.20 / g / g
29: sign(dRR) ^ / -2.00–1.00 / -2.50–3.00 / -0.50–3.00 / -3.00–4.00
30: TACI(10) / 0.20–0.33 / 0.13–0.51 / 0.00–0.60 / 0.20–0.39 / d, e / d,e
31: TACI(20) / 0.15–0.28 / 0.00–1.00 / 0.00–0.50 / 0.00–0.40 / d
Note: Correction for multiple comparisons (greyed) removed all but five significant results for four indices
a baseline pooled 7x30 beat windows
^ corrected percentile method of bootstrap, otherwise bias corrected and accelerated percentile method
Consecutive 30 beat windows post-fentanyl a: F1, b: F2, c: F3, d: F4, e: F5, f: F6, g: F7, h: F8, i: F9 and j: F10.
Table 5Peri-CD 95% confidence intervals of indices for baseline and 3 of 10 consecutive 30-beat intervals(CD-7 to CD+2) aligned with 8th period (h) at CDevent (results from every third interval shown) and intervals reaching significance for difference between oxygen mask baseline and CD windows (a-j)
Index / Baselinea / CD-4 (d) / CD-2 (f) / CD (h) / p<0.006 / p<0.041: SDNN / 34.8–49.7 / 14.5–61.7 / 13.3–68.6 / 19.5–45.0
2: SDNNmc / 4.00–5.08 / 2.10–5.86 / 1.88–6.29 / 2.14–4.94
3: RMSSD / 22.6–39.3 / 11.0–45.4 / 8.71–55.35 / 9.41–36.79
4: RMSSDmc / 2.74–4.22 / 1.62–4.42 / 1.22–5.07 / 1.33–3.71
5: pNN20 ^ / 32.8–53.4 / 5.17–62.07 / 0.00–65.52 / 1.72–48.28 / i
6: Lomb LF / 21.0–32.0 / 15.4–36.2 / 14.4–32.0 / 24.1–32.9
7: LombHF / 12.2–23.1 / 11.5–29.4 / 9.68–34.02 / 7.80–23.63 / i
8: Lomb HFnu / 26.0–50.1 / 26.0–62.2 / 23.5–64.9 / 20.9–48.3
9: Lomb LF/HF / 1.00–2.84 / 0.61–3.12 / 0.54–3.25 / 1.08–3.78
10: LombTotal / 48.8–51.4 / 41.9–52.3 / 39.4–47.6 / 37.6–49.3 / f, g, j / c,f,g, h, i,j
11: accel / 0.48–0.55 / 0.39–0.58 / 0.47–0.57 / 0.53–0.70 / h
12: acv0x / 2.00–3.00 / 1.00–4.00 / 2.00–6.00 / 1.50–5.50
13: assym(R/L) / 0.73–1.07 / 0.57–0.95 / 0.62–1.67 / 1.11–4.02 / h / a,h
14: CVdRR / 125–129 / 117–131 / 118–126 / 119–147 / g, j
15: gradRR / -0.45–0.38 / -0.07–2.85 / -3.84–1.55 / 0.34–2.38 / d, h
16: kurtRRz / 2.18–2.56 / 1.91–2.81 / 2.38–3.01 / 2.08–4.49 / f
17: mean.r(L1-6) / -0.01–0.06 / 0.01–0.25 / 0.06–0.34 / 0.01–0.38 / f, g, j / f,g,i,j
18: meanRR / 757–947 / 670–1018 / 676–958 / 687–1022
19: norm.dRR / 0.11–0.13 / 0.10–0.15 / 0.09–0.12 / 0.10–0.20
20: normRR / 0.11–0.14 / 0.10–0.16 / 0.10–0.14 / 0.10–0.21 / j / c,j
21: PolVar20 ^ / 0.00–4.17 / 0.00–83.33 / 0.00–100.00 / 0.00–83.33 / c, d, e, f, g, h,i, j
22: pQa / 15.1–18.5 / 18.2–22.6 / 16.3–25.0 / 16.3–29.6 / b, j / b, d,j
23: pQb / 25.0–30.9 / 25.4–39.3 / 25.9–35.7 / 14.8–32.1
24: pQc / 29.6–35.2 / 17.9–34.5 / 17.9–32.1 / 21.4–32.7 / i, j
25: r(RR) / 0.61–0.74 / 0.51–0.80 / 0.55–0.80 / 0.47–0.87 / i
26: SDNN/RMSSD / 1.16–1.42 / 1.09–1.63 / 1.07–1.63 / 1.04–2.04 / i
27: skew.dRR / 0.68–0.92 / 0.27–1.51 / 0.17–1.08 / 0.52–1.76 / j
28: skewRRz / -0.21–0.08 / -0.38–0.27 / -0.22–0.50 / -0.21–1.11
29: sign(dRR) ^ / -2.50–0.00 / -0.50–7.00 / -1.00–5.00 / -3.50–1.50 / d, f
30: TACI(10) / 0.21–0.33 / 0.20–0.56 / 0.27–0.57 / 0.46–0.59 / g, h / g,h
31: TACI(20) / 0.18–0.29 / 0.00–0.60 / 0.00–0.44 / 0.00–1.00 / g
Note: Correction for multiple comparisons removed (greyed) all but seven significant results for three indices
a baseline pooled 7x30 beat windows
^ corrected percentile method of bootstrap, otherwise bias corrected and accelerated percentile method
Difference between baseline and peri-CD windows: a, CD-7; b, CD-6; c, CD-5; d, CD-4; e, CD-3; f, CD-2; g, CD-1; h, CD; i, CD+1; j, CD+2.
Table 6Peri-UAWO 95% confidence intervals of indices for baseline and 10 consecutive 30-beat intervals (UO-7 to UO+2, corresponding with a-j) aligned with 8th period (h) at UO (results from every third interval shown)
Index / Baselinea / UO-4 (d) / UO-2 (f) / UO (h) / p<0.006 / p<0.041: SDNN / 34.8–49.7 / 18.5–61.9 / 17.4–70.3 / 15.8–49.0
2: SDNNmc / 4.00–5.11 / 2.59–7.38 / 2.20–7.21 / 2.05–5.58 / g
3: RMSSD / 22.6–39.6 / 11.0–45.5 / 10.7–51.1 / 12.4–37.8
4: RMSSDmc / 2.72–4.20 / 1.68–4.84 / 1.40–5.00 / 1.60–3.62
5: pNN20 ^ / 32.8–55.2 / 5.17–62.07 / 5.17–72.41 / 5.17–37.93 / h
6: Lomb LF / 21.0–31.4 / 18.6–34.2 / 13.4–43.7 / 14.7–37.6
7: LombHF / 12.3–24.0 / 10.2–32.4 / 6.44–24.60 / 8.36–27.46 / i
8: Lomb HFnu / 26.0–49.6 / 22.8–63.6 / 12.8–55.0 / 15.7–61.2
9: Lomb LF/HF / 0.95–2.84 / 0.57–3.61 / 0.71–6.96 / 0.56–4.50
10: LombTotal / 48.8–51.4 / 36.7–50.5 / 39.0–51.3 / 35.1–50.3 / i / g, h,i,j
11: accel / 0.48–0.55 / 0.33–0.58 / 0.49–0.64 / 0.41–0.65 / c / c
12: acv0x / 2.00–3.00 / 0.00–4.00 / 0.00–4.00 / 1.00–4.00
13: assym(R/L) / 0.73–1.07 / 0.68–3.37 / 0.52–2.90 / 0.32–6.31
14: CVdRR / 125–129 / 120–130 / 120–130 / 123–147 / i
15: gradRR / -0.45–0.38 / -3.08–1.63 / 0.28–3.63 / -1.30–3.28 / f
16: kurtRRz / 2.18–2.56 / 1.89–2.38 / 2.11–2.97 / 2.24–3.80
17: mean.r(L1-6) / -0.01–0.06 / -0.00–0.36 / 0.01–0.32 / 0.04–0.27 / j / g,h,j
18: meanRR / 757–947 / 671–1018 / 714–1021 / 731–1033
19: norm.dRR / 0.11–0.13 / 0.09–0.16 / 0.09–0.15 / 0.10–0.20
20: normRR / 0.11–0.14 / 0.10–0.17 / 0.10–0.18 / 0.13–0.21 / h
21: PolVar20 ^ / 0.00–2.08 / 0.00–64.58 / 0.00–66.67 / 0.00–60.42 / b, c, d, e, f, g, h, i,j
22: pQa / 15.4–18.5 / 14.8–21.4 / 17.9–25.0 / 18.5–29.6 / h / f, h, i,j
23: pQb / 25.0–30.9 / 23.5–42.3 / 25.0–38.2 / 14.8–39.3 / b
24: pQc / 29.6–34.6 / 21.4–35.7 / 21.7–29.1 / 16.8–32.1 / f / f
25: r(RR) / 0.61–0.74 / 0.58–0.82 / 0.64–0.84 / 0.43–0.84 / i
26: SDNN/RMSSD / 1.15–1.42 / 1.11–1.87 / 1.22–1.79 / 1.01–1.74 / i
27: skew.dRR / 0.68–0.92 / 0.71–1.44 / 0.34–1.32 / 0.47–1.78
28: skewRRz / -0.21–0.08 / -0.33–0.37 / -0.42–0.41 / -0.66–0.94
29: sign(dRR) ^ / -2.00–1.00 / -3.50–6.00 / 0.00–7.00 / -3.00–7.00 / f
30: TACI(10) / 0.21–0.33 / 0.00–0.50 / 0.36–0.56 / 0.25–0.50 / f, g / f, g
31: TACI(20) / 0.18–0.29 / 0.00–1.00 / 0.33–0.67 / 0.00–1.00 / a,f,j / a,f, j
Note: Correction for multiple comparisons removed (greyed) all but five results for four indices. Abbreviation: UO, upper airway obstruction
a baseline pooled 7x30 beat windows
p<0.006 (or ‘ p<0.04) for difference between baseline and peri-UAWO periods: a, UO-7; b, UO-6; c, UO-5; d, UO-4; e, UO-3; f, UO-2; g, UO-1; h, UO; i, UO+1; j, UO+2.
b)
c)
d)
Figure 8 Individual subject responses to fentanyl-midazolam for time domain indicesfrom 460 beats before to 460 beats after sedative administration: a)mean HR, b) SDNNmc, c) RMSSDmc and d) pNN20.
Vertical lines: left side of each plot- mask on (dot, mid); sedative (mid-line); right side of plot - initial CD (dash); UAWO ETCO2=0 or RRspiro=0 (dot), clinical UAWO (dash-dot). Titles: subject number, UAWO event (F, E, or EF), observed clinical event.
Abbreviations: E, ETCO2=0; F, RRspiro = 0.
b)
c)
d)
Figure 9 Individual responses to fentanyl for Lomb Scargle spectral indicesa) LombLF, b) LombHFnu, c) LombLF/HF and d) LombTotal (explanation as for Figure 8)
References
1. Task Force of European Society of Cardiology (1996) Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Eur Heart J 17 (3):354-381
2. Moody GB (1993) Spectral analysis of heart rate without resampling. Paper presented at the Computers in Cardiology 1993. Proc., London, UK, Sep. 5-8,
3. Guzik P, Piskorski J, Krauze T, Wykretowicz A, Wysocki H (2006) Heart rate asymmetry by Poincaré plots of RR intervals. Biomed Tech (Berl) 51 (4):272-275
4. Piskorski J, Guzik P (2007) Geometry of the Poincaré plot of RR intervals and its asymmetry in healthy adults. Physiol Meas 28 (3):287-300
5. Nikolopoulos S, Alexandridi A, Nikolakeas S, Manis G (2003) Experimental analysis of heart rate variability of long-recording electrocardiograms in normal subjects and patients with coronary artery disease and normal left ventricular function. J Biomed Inform 36 (3):202-217
6. Kovatchev BP, Farhy LS, Cao H, Griffin MP, Lake DE, Moorman JR (2003) Sample asymmetry analysis of heart rate characteristics with application to neonatal sepsis and systemic inflammatory response syndrome. Pediatr Res 54 (6):892-898
7. Tateno K, Glass L (2001) Automatic detection of atrial fibrillation using the coefficient of variation and density histograms of RR and deltaRR intervals. Med Biol Eng Comput 39 (6):664-671
8. Marciano F, Migaux ML, Acanfora D, Furgi G, Rengo F (1994) Quantification of Poincaré maps for the evaluation of heart rate variability. Paper presented at the Computers in Cardiology 1994, Bethesda, MD, Sep 25-28
9. Olesen RM, Thomsen PE, Saermark K, Glikson M, Feldman S, Lewkowicz M, Levitan J (2005) Statistical analysis of the DIAMOND MI study by the multipole method. Physiol Meas 26 (5):591-598
10. Griffin MP, Moorman JR (2001) Toward the early diagnosis of neonatal sepsis and sepsis-like illness using novel heart rate analysis. Pediatrics 107 (1):97-104
11. Sosnowski M, Petelenz T, Leski J (1994) Return maps: a non-linear method for evaluation of respiratory sinus arrhythmia. Paper presented at the Computers in Cardiology 1994, Bethesda, MD, Sep 25-28
12. Goldberger JJ (1999) Sympathovagal balance: how should we measure it? Am J Physiol 276 (4 Pt 2):H1273-1280
13. Voss A, Kurths J, Kleiner HJ, Witt A, Wessel N, Saparin P, Osterziel KJ, Schurath R, Dietz R (1996) The application of methods of non-linear dynamics for the improved and predictive recognition of patients threatened by sudden cardiac death. Cardiovasc Res 31 (3):419-433
14. Wessel N, Malberg H, Bauernschmitt R, Schirdewan A, Kurths J (2006) Nonlinear additive autoregressive model-based analysis of short-term heart rate variability. Med Biol Eng Comput 44 (4):321-330. doi:10.1007/s11517-006-0038-0
15. Raetz SL, Richard CA, Garfinkel A, Harper RM (1991) Dynamic characteristics of cardiac R-R intervals during sleep and waking states. Sleep 14 (6):526-533
16. Otzenberger H, Gronfier C, Simon C, Charloux A, Ehrhart J, Piquard F, Brandenberger G (1998) Dynamic heart rate variability: a tool for exploring sympathovagal balance continuously during sleep in men. Am J Physiol 275 (3 Pt 2):H946-950
17. Balocchi R, Cantini F, Varanini M, Raimondi G, Legramante JM, Macerata A (2006) Revisiting the potential of time-domain indexes in short-term HRV analysis. Biomed Tech (Berl) 51 (4):190-193
18. Ashkenazy Y, Ivanov PC, Havlin S, Peng CK, Goldberger AL, Stanley HE (2001) Magnitude and sign correlations in heartbeat fluctuations. Phys Rev Lett 86 (9):1900-1903
19. Arif M, Aziz W (2005) Application of threshold-based acceleration change index (TACI) in heart rate variability analysis. Physiol Meas 26 (5):653-665
[1]Anne-Louise Smith
The Medical Device Research Institute, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia
Biomedical Engineering Dept, Flinders Medical Centre, Adelaide, Australia
Ph: +61 (0)8 8222 5534
Fax: +61 (0)8 8222 5944
Email:
Harry Owen
School of Medicine, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia
Karen J. Reynolds
The Medical Device Research Institute, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia