- Email: [email protected]
Autonomic nervous function in patients with Meniere’s disease evaluated by power spectral analysis of heart rate variability
Auris Nasus Larynx 26 (1999) 419 – 426 www.elsevier.com/locate/anl
Autonomic nervous function in patients with Meniere’s disease evaluated by power spectral analysis of heart rate variability Masahiko Yamada *, Keisuke Mizuta, Yatsuji Ito, Michiya Furuta, Shigeo Sawai, Hideo Miyata Department of Otolaryngology, Gifu Uni6ersity School of Medicine, 40 Tsukasa-machi, Gifu shi, Gifu-ken, 500 -8705, Japan Received 28 August 1998; received in revised form 11 December 1998; accepted 22 January 1999
Abstract Objecti6e: A hypothesis has been advanced that the autonomic nervous dysfunction (AND) relates to the development of vertigo in Meniere’s disease (MD). We also studied the causal relationship between AND and vertigo in MD. Methods: We evaluated autonomic nervous function in 17 patients with MD (five men and 12 women ranging in age from 16 to 70 years) by classifying them by their stages of attack and interval of vertigo and with power spectral analysis (PSA) of heart rate variability. Fourteen healthy volunteers were also tested as controls. Results: At the interval stage, parasympathetic nervous hypofunction and significant depression of sympathetic response due to postural changes from the supine to the standing position were observed in many of those patients. At the attack stage, sympathetic nervous hypofunction was observed in some of the patients. Conclusion: These findings lead us to the conclusion that AND relates to vertigo in MD as a predisposing factor. However, the question of whether AND relates as a trigger or as a consequence of vertigo in MD has not been adequately solved in this study. We will make further studies on circadian variation of autonomic nervous function. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Meniere’s disease; Autonomic nervous dysfunction; Power spectral analysis
1. Introduction A hypothesis has been advanced that the autonomic nervous dysfunction (AND) relates to the development of vertigo in Meniere’s disease (MD) * Corresponding author. Present address: Yoro Chuo General Hospital, 986 Oshikoshi, Yoro-cho, Yoro-gun, Gifu-ken, 503-1394, Japan. Tel.: +81 584 321161; fax: +81 584 322856.
[1–5]. Although studies have been made using various methods [1–14], it is still unclear that whether AND plays a role as a predisposing factor, a trigger or as a consequence of vertigo. The aim of this study, therefore, is to determine the causal relationship between AND and vertigo in MD by the use of PSA, which has been shown to be a reliable non-invasive technique for the quantitative analysis of the two components of
0385-8146/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 8 5 - 8 1 4 6 ( 9 9 ) 0 0 0 2 1 - 8
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
420
the autonomic nervous system. In this study, we classified the patients into the interval stage and the attack stage of vertigo, considering that the autonomic nervous function in MD may vary in those two stages.
2. Patients and methods Seventeen patients (five men and 12 women) with MD were subjected in this study. They varied in age from 16 to 70 years (48.4 years in mean). Fourteen healthy volunteers (nine men and five women) ranging in age from 26 to 54 years (34.7 years in mean) with no history of vertigo were served as controls. Clinical data and results of equilibrium test of 17 patients with MD were presented in Table 1. For the patients in the middle of vertigo attack stage, the same antivertiginous and circulatory controller were given as symptomatic treatment.
2.1. Power spectral analysis of heart rate 6ariability After resting for 20 min, each participant received Schellong test, which is very useful for
diagnosing orthostatic dysregulation. The procedure is that a participant is put in the supine position for 10 min rest, and put in the standing position for 10 min burden, and then blood pressure, pulse and pulse pressure are recorded. During this test, electrocardiograph was simultaneously conducted, and the findings were input to a data recorder. RR intervals, which is an interval between an R wave and the subsequent R wave in ECG wave form and is considered to vary in patients with autonomic nervous dysfunction, were determined for 2 min after having the participants resting for 7 min in the supine position, and same procedure was repeated on them by modifying the supine position to the standing position of burden. The frequency of RR interval was analyzed with personal computer by the use of Fast Fourier transformation. The period for the analysis was 128 s, and the sampling point was set to 1024. Based on the obtained power spectrum, integrated power values were calculated on the respective components in low frequency bands (LF: 0.04–0.15 Hz), high frequency bands (HF: 0.15–0.40 Hz) and total frequency bands (T:B 0.40 Hz). The analysis was carried out using the rate of HF against T (HF/T) as an index of the parasympathetic nervous function, and the
Table 1 Clinical data and results of equilibrium test of patients with Meniere’s diseasea Patient no.
Sex
Age
Caloric test (CP%)
OD criteria
Schellong test
Vertigo stage
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
F F F M M F M F F F F F F F F M M
38 69 67 56 51 50 27 67 35 51 27 55 70 61 36 16 47
L 11 L 30 L 22 R 90 R4 L 13 L 12 R 90 L 16 R 16 L 49 R 16 R6 R 70 L 12 L 100 L 12
+ − − + − + − − + + + − + − + − +
+ − − + − + − − + − + − − − + − −
Int Int Int Int Int Att Att Att Att Att Att Int&Att Int&Att Int&Att Int&Att Int&Att Int&Att
a
CP: canal paresis; Int/Att: interval/attack; M/F: male/female; OD: orthostatic dysregulation; R/L: right/left.
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
421
Fig. 1. The mean HF/T rate and the LF/HF rate after resting in supine position during the interval stage. Although there was no significant difference between those results, the individual HF/T rates of eight out of the 11 patients were lower than the mean − SE of the controls.
Table 2 HF/T rate and LF/HF rate of patients with MD in both of the stages Patient no.
1 2 3 4 5 6 7 8 9 10 11 12
Vertigo stage
Int Int Int Int Int Att Att Att Att Att Att Int Att Int Att Int Att Int Att Int Att Int Att
13 14 15 16 17
a
Supine positiona
Standing position
HF/T
LF/HF
HF/T
LF/HF
0.07¡ 0.24¡ 0.12¡ 0.16¡ 0.16¡ 0.24 0.05 0.19 0.65 0.54 0.34 0.27 0.15 0.46 0.57 0.30 0.59 0.17¡ 0.30 0.14¡ 0.34 0.22¡ 0.11
2.88 0.07 1.85 1.01 3.06 0.34¡ 2.89 1.30 0.19¡ 1.13 1.08 1.36 0.92¡ 0.13 0.09¡ 0.46 0.18¡ 0.91 0.26¡ 1.28 0.94¡ 0.50 0.80¡
0.14 0.15 0.09 0.13 0.18 0.43 0.06 0.16 0.22 0.25 0.19 0.84 Not can 0.39 0.24 0.20 0.57 0.25 0.32 0.05 0.17 0.13 0.18
1.23 0.30 1.07 1.02 1.71 0.18 6.31 1.73 1.68 0.45 2.57 1.12 Not can 0.21 0.39 0.75 0.17 1.14 0.93 6.90 2.03 0.38 0.47
¡: Downward arrow shows the lower rate than the mean−SE of the controls.
422
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
rate of LF against HF (LF/HF) as an index of the sympathetic nervous function [15 – 21].
3. Results
3.1. E6aluation on the inter6al stage 2.2. In6estigation methods The autonomic nervous function of patients with MD are considered to vary during repeating attacks of vertigo. We, therefore, divided the patients subjected to this study into two groups, those in the attack stage (within 2 weeks after the last attack of vertigo), and in the interval stage (more than 2 weeks after the last attack of vertigo). The testing was able to be made on 12 cases in the attack stage, and on 11 cases in the interval stage. All the patients in the attack stage underwent the test within 2 days after the vertigo attack except case 17. Six patients underwent the tests in both of the stages. The analysis was made on the items: (i) data obtained after resting in the supine position; and (ii) those obtained after burdening in the standing position. Among the data of the six patients, on whom the testing was made at both interval and attack stages, those obtained after resting in the supine position were analyzed. Each result was compared with those of the controls. Mann–Whitney U-test was used to compare the HF/T rate and LF/HF rate in the two groups. Wilcoxon rank-sum test was used to analyze postural changes in the two groups. A P value less than 0.05 was considered as significant.
3.1.1. The e6aluation after resting in the supine position (Fig. 1) The mean HF/T rate of patients with MD was 0.219 0.03 (mean9 standard error (SE)), and that of the controls was 0.30 9 0.05. Although there was no significant difference between those results, the individual rates of eight out of the 11 patients were lower than the mean − SE of the controls (Table 2). The mean LF/HF rate of the patients was 1.2390.31, and that of the controls was l.21 90.26. Again, there was no significant difference between those results. 3.1.2. The effects of burden in the standing position (Fig. 2) There was no marked difference between the mean HF/T rate of the patients after resting in the supine position (0.2190.03) and that after burdening in the standing position (0.239 0.07). On the other hand, the mean HF/T rate of the controls after the rest (0.309 0.05) showed a tendency of dropping after the burden (0.169 0.03). There was no marked difference between the mean LF/HF rate of the patients after the rest (1.239 0.31) and that after the burden (1.44 9
Fig. 2. The effects of burden in the standing position during the interval stage. There was no marked difference between the mean LF/HF rate of the patients after the rest and that after burdening in the standing position. In the case of the controls, however, the difference was significant (*PB 0.05). And the difference between the patients with MD and the controls was significant (*PB0.05).
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
423
Fig. 3. The mean HF/T rate and the LF/HF rate after resting in supine position during the attack stage. Although there was no significant difference between those results, the individual LF/HF rates of eight out of the 12 patients were lower than the mean −SE of the controls.
0.56). In the case of the controls, however, the difference between after the rest (1.219 0.26) and after the burden (2.859 0.82) was significant (P B 0.05). And the difference between the patients with MD and the controls was significant (P B 0.05).
patients (0.34 9 0.06). Although the mean LF/HF rate of the patients (0.849 0.22) was not significantly different from that of the controls (l.219 0.26), the individual rates of eight out of the 12 patients were lower than the mean − SE of the controls (Table 2).
3.2. E6aluation on the attack stage 3.2.1. The e6aluation after resting in the supine position (Fig. 3) There was no marked difference in the mean HF/T rates between the controls (0.3090.05) and
3.2.2. The effects of burden in the standing position (Fig. 4) Similar to that of the controls, the mean HF/T rate of the patients after the rest (0.34 9 0.04) was showed a tendency of dropping after the burden
Fig. 4. The effects of burden in the standing position during the interval stage. There was no marked difference between the patients with MD and the controls.
424
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
(0.25 90.04). Although the mean LF/HF rate of the patients became grater after the burden (1.44 90.56) than that after the rest (1.23 9 0.56), the difference was not significant, and those results were not markedly different from those of the controls. One patient who was not able to undergo the burden was excluded from the evaluation.
3.3. E6aluation of the six patients on whom the tests were performed at both stages (Fig. 5) The HF/T rates of four patients were higher, on the other hand, those of two patients were lower during the attack stage than those during the interval stage. Regarding the LF/HF rates, five patients presented lower rates during attack stage than during interval stage, and one patients presented higher. The patient who presented higher LF/HF rate during attack stage was the only one on whom the test was able to be performed only at the seventh day after the attack.
4. Discussion
4.1. The autonomic ner6ous function in patients with MD It has been considered that AND may play a role in the development of endolymphatic hy-
drops; one of the pathological associates with Meniere’s disease [1–5]. Although many researches have been made using various methods [1–14], the correlation between them has not still been clearly revealed. This is because the autonomic nervous function is easily affected by various stimulative factors from outside, since their natures are to maintain the homeostasis of a living body. Furthermore, AND with personal characteristics and those affected by vestibular and autonomic interactions on the development of vertigo are intermingled, and that makes it difficult to make out whether the AND relates to MD as trigger or as a consequence of vertigo [7].
4.2. E6aluation by the use of PSA In undertaking the present study, we considered that the autonomic nervous function together with the cardio-circulative aspect should be reasonable to research [7]. Among various methods, we chose PSA which has been receiving much attention recently [22–26]. The technique of PSA is non-invasive, and quantitative determination on the activities of both sympathetic and parasympathetic nervous systems can be done with it. The obtained components of low frequency mainly reflect the function of sympathetic nerves, but at the same time, it is considered to be partially affected by the function of the parasympathetic nerves. The components of high frequency mainly
Fig. 5. The evaluation of the six patients on whom the tests were performed at both stages. The LF/HF rates of five patients were lower, on the other hand, those of one patients were higher during the attack stage than those during the interval stage.
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
reflect the function of parasympathetic nerves. In this study, LF/HF and HF/T were used as indices of sympathetic and parasympathetic nervous function, respectively [22 – 26]. In assuming that the function of the sympathetic nervous system would be evaluated more clearly when they were tested not only in the resting condition but also in burdening condition of the participants, we conducted the tests under both conditions [13].
4.2.1. AND during the inter6al stage In comparison with healthy controls, eight of the 11 patients (73%) presented lower HF/T rate with which parasympathetic nervous hypofunction was indicated. On the other hand, six of the 11 patients (55%) presented lower LF/HF rate than those of the controls. When parasympathetic nervous hypofunction exists, relative acceleration of sympathetic nervous function is very likely to occur. The patients with MD, therefore, were considered to have sympathetic nervous hypofunction as well. Regarding the effects of the burden, the controls showed significantly elevated LF/HF rate. The patients, however, showed no acceleration in sympathetic nervous function. Most probably, they had weakened activation of sympathetic nervous function. Based on these findings, it was considered that the patients with MD during the interval stage had weakened parasympathetic nervous function, and inhibited sympathetic response against the postural change from the supine to the standing position. These phenomena agree with those reported by Kawasaki [6], and suggest that they can be part of predisposing factors of MD. Igarashi et al. [4] have pointed out sympathetic nervous dysfunction during the interval stage, and Ogino et al. [14] reported that not only parasympathetic nervous hypofunction but also sympathetic nervous dysfunction were observed during the interval stage of MD. Seki [13] studied sympathetic nervous function using pulse wave velocity and power spectral analysis. They reported that the same patient could present different results with the difference in methods used, and combinations of several methods, therefore, should be used. Even though sympathetic nervous dysfunction was observed in the patients with MD in this study, further researches will be needed.
425
4.2.2. AND during the attack stage At the attack stage, the HF/T rates of the patients after the rest were not significantly different from those of the healthy controls. The LF/HF rates of eight of the 12 patients (67%) were lower than those of the controls, which suggested that sympathetic nervous function was depressed. As acceleration of sympathetic response due to the postural change was not observed in the patients with MD, the burden seemed to have small effects on them even during the attack stage. In addition, the results were similar to the patients who underwent the tests during both attack and interval stages, as five of the six patients (83%) presented low LF/HF rates. Although one patient showed a high LF/HF rate, it was attributed to the inappropriate classification of the patient because the patient was the only one who received the test just 7 days after the attack of vertigo. These results are different from those reported by Matsunaga [7], that the acceleration of sympathetic function was observed in patients with MD during the attack stage. We consider that this difference is caused by the difference in the classification of attack stage. That is, all the patients in this study were in a state well enough to undergo the burden, but not in a so-to-speak acute attack stage of vertigo in which they hardly move their bodies. Even in this study, acceleration of sympathetic nervous function was found in one patient with frequent attacks of vertigo. However, the acceleration of sympathetic function during the attack stage may well be considered as a result of the vestibular and autonomic interactions. In this study, patients in the middle of vertigo attack stage presented a tendency to sympathetic nervous hypofunction rather than hyperfunction which was reported by other works [6,7,14], although no statistical significance was proven by the present study. This suggests that sympathetic nervous dysfunction can be not only a consequence but also one of the triggers of vertigo in MD. The conclusion that AND with parasympathetic nervous hypofunction during the interval stage relates to vertigo in MD as a predisposing factor is not different from those of other reports [5–7,13,14]. We consider, however, that further studies are needed to clarify whether AND during
426
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
vertigo attacks is a consequence or a trigger in MD. In order to solve this problem, a continuous analysis should be made on the temporal course of MD in each patient throughout the periods of attack and interval instead of classifying patients by those two stages. In addition, as it has been reported that sympathetic dysfunction after the attacks presents quicker recovery than parasympathetic dysfunction [13], we consider that the continuous analyses on the temporal course have to be made in same timing, and as long as possible. We, therefore, are planning to have patients with MD installed with 24-h Holter electrocardiography, and make a study by analyzing the circadian variation of the autonomic nervous function.
References [1] Tokita T, Aoki S, Miyata H. The loading equilibrium examination for etiologic diagnosis of Meniere’s disease. Acta Otolaryngol (Stockh) 1971;72:107–16. [2] Uemura T, Itoh M, Kikuchi N. Autonomic dysfunction on the affected side in Meniere’s disease. Acta Otolaryngol (Stockh) 1980;89:109–17. [3] Inoue H, Uemura T. Sluggishness pupillary light contraction in patients with Meniere’s disease. Acta Otolaryngol (Stockh) 1988;105:582–6. [4] Igarashi S, Seki S, Sasaki H, Nagaba A, Nakano Y. Autonomic nerve function in patients with Meniere’s disease. An analysis using pulse wave velocity, RR interval and thermography. Equilibrium Res 1994;53:417–21. [5] Seki S, Igarashi S, Sasaki H, Nakano Y. Autonomic nerve function in patients with Meniere’s disease. An analysis using thermography. Equilibrium Res 1993;52:433 – 7. [6] Kawasaki Y. Autonomic nerve function of vertiginous patients. Assessment by spectral analysis of heart rate variability. J Otol Jpn 1993;96:444–56. [7] Matsunaga T. Autonomic nerve system function test in patients with vertigo. Pract Otol (Kyoto) 1987;80:865– 78. [8] Burton AC. The range and variability of the human fingers and the vasomotor regulation of body temperature. Am J Physiol 1939;127:437–53. [9] Kitney R, Rompelman O. Thermal entailment patterns in heart rate variability. J Physiol 1977;270:41–2. [10] Pappas DG, Crawford W, Coghlan HC. Dizziness and dysautonomic syndrome. Otolaryngol Head Neck Surg 1986;94:186 – 94.
[11] Pappas DG, Crawford W. Dizziness due to dysautonomia: response tospecific therapy. Southern Med J 1988;81:531 – 2. [12] Nozawa I, Imamura S, Hisamatsu K, Murakami Y. The relationship between orthostatic dysregulation and orthostatic test in dizzy patients. Eur Arch Otorhinolaryngol 1996;253:268 – 72. [13] Seki S. Autonomic nerve function in patients with vertigo. Evaluation for static function, variation and dynamic change using power spectral analysis of RR interval. J Otol Jpn 1997;100:457 – 66. [14] Ogino H. Autonomic nerve function in patients with Meniere’s disease. Pract Otol (Kyoto) 1991;41:80 – 93. [15] Pomeranz B, Macaulay RJB, Caudil MA, Caudill MA, Kutz I, Adam D, et al. Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol 1985;249:151 – 3. [16] Pagani M, Lonbardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a maker of sympatovagal interaction in man and conscious dog. Circ Res 1986;59:178 – 93. [17] Hayano J, Yamada M, Ohte N, Hashimoto T, Fujinami T, Takata K. Severity of coronary atheroscrelosis correlate with the respiratory component heart rate variability. Am Heart J 1991;121:1070 – 9. [18] Sayers BMcA. Analysis of heart rate variability. Ergonomics 1973;16:17 – 32. [19] Hyndman HW, Kitney RI, Sayers BMcA. Spontaneous rhythms in physiological control system. Nature 1971;233:339 – 41. [20] Taniguchi I, Kageyama S, Aihara K, Isogai Y, Katoh F, Fukuhara T. Assessment of power spectral analysis of heart rate functions. Effects of postural change and respiration. Auton Nerv Syst 1994;31:79 – 84 (in Japanese). [21] Berger RD, Saul JP, Cohen RJ. Transfer function analysis of autonomic regulation. I. Canine atrial rate response. Am J Physiol 1989;256:142 – 52. [22] Akselrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectral analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science 1981;213:220 – 3. [23] Wheeler T, Watkins PJ. Cardiac denervation in diabetes. Br Med J 1973;4:584 – 6. [24] Karemaker JM. Heart rate variability: why do spectral analysis? Heart 1997;77:99 – 101. [25] Vardas PE, Kochiadakis E, Manios EG, Kanoupakis EM, Zouridakis EG, Chlouverakis GI. Spectral analysis of heart rate variability before and during episodes of nocturnal ischaemia in patients with extensive coronary artery disease. Eur Heart J 1996;17:388 – 93. [26] Cavalcanti S, Severi S, Chiari L, Avanzolini G, Enzmann G, Bianco F, et al. Autonomic nervous function during haemodialysis assessed by spectral analysis of heart rate variability. Clin Sci 1997;92:351 – 9.
Autonomic nervous function in patients with Meniere’s disease evaluated by power spectral analysis of heart rate variability Masahiko Yamada *, Keisuke Mizuta, Yatsuji Ito, Michiya Furuta, Shigeo Sawai, Hideo Miyata Department of Otolaryngology, Gifu Uni6ersity School of Medicine, 40 Tsukasa-machi, Gifu shi, Gifu-ken, 500 -8705, Japan Received 28 August 1998; received in revised form 11 December 1998; accepted 22 January 1999
Abstract Objecti6e: A hypothesis has been advanced that the autonomic nervous dysfunction (AND) relates to the development of vertigo in Meniere’s disease (MD). We also studied the causal relationship between AND and vertigo in MD. Methods: We evaluated autonomic nervous function in 17 patients with MD (five men and 12 women ranging in age from 16 to 70 years) by classifying them by their stages of attack and interval of vertigo and with power spectral analysis (PSA) of heart rate variability. Fourteen healthy volunteers were also tested as controls. Results: At the interval stage, parasympathetic nervous hypofunction and significant depression of sympathetic response due to postural changes from the supine to the standing position were observed in many of those patients. At the attack stage, sympathetic nervous hypofunction was observed in some of the patients. Conclusion: These findings lead us to the conclusion that AND relates to vertigo in MD as a predisposing factor. However, the question of whether AND relates as a trigger or as a consequence of vertigo in MD has not been adequately solved in this study. We will make further studies on circadian variation of autonomic nervous function. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Meniere’s disease; Autonomic nervous dysfunction; Power spectral analysis
1. Introduction A hypothesis has been advanced that the autonomic nervous dysfunction (AND) relates to the development of vertigo in Meniere’s disease (MD) * Corresponding author. Present address: Yoro Chuo General Hospital, 986 Oshikoshi, Yoro-cho, Yoro-gun, Gifu-ken, 503-1394, Japan. Tel.: +81 584 321161; fax: +81 584 322856.
[1–5]. Although studies have been made using various methods [1–14], it is still unclear that whether AND plays a role as a predisposing factor, a trigger or as a consequence of vertigo. The aim of this study, therefore, is to determine the causal relationship between AND and vertigo in MD by the use of PSA, which has been shown to be a reliable non-invasive technique for the quantitative analysis of the two components of
0385-8146/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 8 5 - 8 1 4 6 ( 9 9 ) 0 0 0 2 1 - 8
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
420
the autonomic nervous system. In this study, we classified the patients into the interval stage and the attack stage of vertigo, considering that the autonomic nervous function in MD may vary in those two stages.
2. Patients and methods Seventeen patients (five men and 12 women) with MD were subjected in this study. They varied in age from 16 to 70 years (48.4 years in mean). Fourteen healthy volunteers (nine men and five women) ranging in age from 26 to 54 years (34.7 years in mean) with no history of vertigo were served as controls. Clinical data and results of equilibrium test of 17 patients with MD were presented in Table 1. For the patients in the middle of vertigo attack stage, the same antivertiginous and circulatory controller were given as symptomatic treatment.
2.1. Power spectral analysis of heart rate 6ariability After resting for 20 min, each participant received Schellong test, which is very useful for
diagnosing orthostatic dysregulation. The procedure is that a participant is put in the supine position for 10 min rest, and put in the standing position for 10 min burden, and then blood pressure, pulse and pulse pressure are recorded. During this test, electrocardiograph was simultaneously conducted, and the findings were input to a data recorder. RR intervals, which is an interval between an R wave and the subsequent R wave in ECG wave form and is considered to vary in patients with autonomic nervous dysfunction, were determined for 2 min after having the participants resting for 7 min in the supine position, and same procedure was repeated on them by modifying the supine position to the standing position of burden. The frequency of RR interval was analyzed with personal computer by the use of Fast Fourier transformation. The period for the analysis was 128 s, and the sampling point was set to 1024. Based on the obtained power spectrum, integrated power values were calculated on the respective components in low frequency bands (LF: 0.04–0.15 Hz), high frequency bands (HF: 0.15–0.40 Hz) and total frequency bands (T:B 0.40 Hz). The analysis was carried out using the rate of HF against T (HF/T) as an index of the parasympathetic nervous function, and the
Table 1 Clinical data and results of equilibrium test of patients with Meniere’s diseasea Patient no.
Sex
Age
Caloric test (CP%)
OD criteria
Schellong test
Vertigo stage
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
F F F M M F M F F F F F F F F M M
38 69 67 56 51 50 27 67 35 51 27 55 70 61 36 16 47
L 11 L 30 L 22 R 90 R4 L 13 L 12 R 90 L 16 R 16 L 49 R 16 R6 R 70 L 12 L 100 L 12
+ − − + − + − − + + + − + − + − +
+ − − + − + − − + − + − − − + − −
Int Int Int Int Int Att Att Att Att Att Att Int&Att Int&Att Int&Att Int&Att Int&Att Int&Att
a
CP: canal paresis; Int/Att: interval/attack; M/F: male/female; OD: orthostatic dysregulation; R/L: right/left.
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
421
Fig. 1. The mean HF/T rate and the LF/HF rate after resting in supine position during the interval stage. Although there was no significant difference between those results, the individual HF/T rates of eight out of the 11 patients were lower than the mean − SE of the controls.
Table 2 HF/T rate and LF/HF rate of patients with MD in both of the stages Patient no.
1 2 3 4 5 6 7 8 9 10 11 12
Vertigo stage
Int Int Int Int Int Att Att Att Att Att Att Int Att Int Att Int Att Int Att Int Att Int Att
13 14 15 16 17
a
Supine positiona
Standing position
HF/T
LF/HF
HF/T
LF/HF
0.07¡ 0.24¡ 0.12¡ 0.16¡ 0.16¡ 0.24 0.05 0.19 0.65 0.54 0.34 0.27 0.15 0.46 0.57 0.30 0.59 0.17¡ 0.30 0.14¡ 0.34 0.22¡ 0.11
2.88 0.07 1.85 1.01 3.06 0.34¡ 2.89 1.30 0.19¡ 1.13 1.08 1.36 0.92¡ 0.13 0.09¡ 0.46 0.18¡ 0.91 0.26¡ 1.28 0.94¡ 0.50 0.80¡
0.14 0.15 0.09 0.13 0.18 0.43 0.06 0.16 0.22 0.25 0.19 0.84 Not can 0.39 0.24 0.20 0.57 0.25 0.32 0.05 0.17 0.13 0.18
1.23 0.30 1.07 1.02 1.71 0.18 6.31 1.73 1.68 0.45 2.57 1.12 Not can 0.21 0.39 0.75 0.17 1.14 0.93 6.90 2.03 0.38 0.47
¡: Downward arrow shows the lower rate than the mean−SE of the controls.
422
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
rate of LF against HF (LF/HF) as an index of the sympathetic nervous function [15 – 21].
3. Results
3.1. E6aluation on the inter6al stage 2.2. In6estigation methods The autonomic nervous function of patients with MD are considered to vary during repeating attacks of vertigo. We, therefore, divided the patients subjected to this study into two groups, those in the attack stage (within 2 weeks after the last attack of vertigo), and in the interval stage (more than 2 weeks after the last attack of vertigo). The testing was able to be made on 12 cases in the attack stage, and on 11 cases in the interval stage. All the patients in the attack stage underwent the test within 2 days after the vertigo attack except case 17. Six patients underwent the tests in both of the stages. The analysis was made on the items: (i) data obtained after resting in the supine position; and (ii) those obtained after burdening in the standing position. Among the data of the six patients, on whom the testing was made at both interval and attack stages, those obtained after resting in the supine position were analyzed. Each result was compared with those of the controls. Mann–Whitney U-test was used to compare the HF/T rate and LF/HF rate in the two groups. Wilcoxon rank-sum test was used to analyze postural changes in the two groups. A P value less than 0.05 was considered as significant.
3.1.1. The e6aluation after resting in the supine position (Fig. 1) The mean HF/T rate of patients with MD was 0.219 0.03 (mean9 standard error (SE)), and that of the controls was 0.30 9 0.05. Although there was no significant difference between those results, the individual rates of eight out of the 11 patients were lower than the mean − SE of the controls (Table 2). The mean LF/HF rate of the patients was 1.2390.31, and that of the controls was l.21 90.26. Again, there was no significant difference between those results. 3.1.2. The effects of burden in the standing position (Fig. 2) There was no marked difference between the mean HF/T rate of the patients after resting in the supine position (0.2190.03) and that after burdening in the standing position (0.239 0.07). On the other hand, the mean HF/T rate of the controls after the rest (0.309 0.05) showed a tendency of dropping after the burden (0.169 0.03). There was no marked difference between the mean LF/HF rate of the patients after the rest (1.239 0.31) and that after the burden (1.44 9
Fig. 2. The effects of burden in the standing position during the interval stage. There was no marked difference between the mean LF/HF rate of the patients after the rest and that after burdening in the standing position. In the case of the controls, however, the difference was significant (*PB 0.05). And the difference between the patients with MD and the controls was significant (*PB0.05).
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
423
Fig. 3. The mean HF/T rate and the LF/HF rate after resting in supine position during the attack stage. Although there was no significant difference between those results, the individual LF/HF rates of eight out of the 12 patients were lower than the mean −SE of the controls.
0.56). In the case of the controls, however, the difference between after the rest (1.219 0.26) and after the burden (2.859 0.82) was significant (P B 0.05). And the difference between the patients with MD and the controls was significant (P B 0.05).
patients (0.34 9 0.06). Although the mean LF/HF rate of the patients (0.849 0.22) was not significantly different from that of the controls (l.219 0.26), the individual rates of eight out of the 12 patients were lower than the mean − SE of the controls (Table 2).
3.2. E6aluation on the attack stage 3.2.1. The e6aluation after resting in the supine position (Fig. 3) There was no marked difference in the mean HF/T rates between the controls (0.3090.05) and
3.2.2. The effects of burden in the standing position (Fig. 4) Similar to that of the controls, the mean HF/T rate of the patients after the rest (0.34 9 0.04) was showed a tendency of dropping after the burden
Fig. 4. The effects of burden in the standing position during the interval stage. There was no marked difference between the patients with MD and the controls.
424
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
(0.25 90.04). Although the mean LF/HF rate of the patients became grater after the burden (1.44 90.56) than that after the rest (1.23 9 0.56), the difference was not significant, and those results were not markedly different from those of the controls. One patient who was not able to undergo the burden was excluded from the evaluation.
3.3. E6aluation of the six patients on whom the tests were performed at both stages (Fig. 5) The HF/T rates of four patients were higher, on the other hand, those of two patients were lower during the attack stage than those during the interval stage. Regarding the LF/HF rates, five patients presented lower rates during attack stage than during interval stage, and one patients presented higher. The patient who presented higher LF/HF rate during attack stage was the only one on whom the test was able to be performed only at the seventh day after the attack.
4. Discussion
4.1. The autonomic ner6ous function in patients with MD It has been considered that AND may play a role in the development of endolymphatic hy-
drops; one of the pathological associates with Meniere’s disease [1–5]. Although many researches have been made using various methods [1–14], the correlation between them has not still been clearly revealed. This is because the autonomic nervous function is easily affected by various stimulative factors from outside, since their natures are to maintain the homeostasis of a living body. Furthermore, AND with personal characteristics and those affected by vestibular and autonomic interactions on the development of vertigo are intermingled, and that makes it difficult to make out whether the AND relates to MD as trigger or as a consequence of vertigo [7].
4.2. E6aluation by the use of PSA In undertaking the present study, we considered that the autonomic nervous function together with the cardio-circulative aspect should be reasonable to research [7]. Among various methods, we chose PSA which has been receiving much attention recently [22–26]. The technique of PSA is non-invasive, and quantitative determination on the activities of both sympathetic and parasympathetic nervous systems can be done with it. The obtained components of low frequency mainly reflect the function of sympathetic nerves, but at the same time, it is considered to be partially affected by the function of the parasympathetic nerves. The components of high frequency mainly
Fig. 5. The evaluation of the six patients on whom the tests were performed at both stages. The LF/HF rates of five patients were lower, on the other hand, those of one patients were higher during the attack stage than those during the interval stage.
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
reflect the function of parasympathetic nerves. In this study, LF/HF and HF/T were used as indices of sympathetic and parasympathetic nervous function, respectively [22 – 26]. In assuming that the function of the sympathetic nervous system would be evaluated more clearly when they were tested not only in the resting condition but also in burdening condition of the participants, we conducted the tests under both conditions [13].
4.2.1. AND during the inter6al stage In comparison with healthy controls, eight of the 11 patients (73%) presented lower HF/T rate with which parasympathetic nervous hypofunction was indicated. On the other hand, six of the 11 patients (55%) presented lower LF/HF rate than those of the controls. When parasympathetic nervous hypofunction exists, relative acceleration of sympathetic nervous function is very likely to occur. The patients with MD, therefore, were considered to have sympathetic nervous hypofunction as well. Regarding the effects of the burden, the controls showed significantly elevated LF/HF rate. The patients, however, showed no acceleration in sympathetic nervous function. Most probably, they had weakened activation of sympathetic nervous function. Based on these findings, it was considered that the patients with MD during the interval stage had weakened parasympathetic nervous function, and inhibited sympathetic response against the postural change from the supine to the standing position. These phenomena agree with those reported by Kawasaki [6], and suggest that they can be part of predisposing factors of MD. Igarashi et al. [4] have pointed out sympathetic nervous dysfunction during the interval stage, and Ogino et al. [14] reported that not only parasympathetic nervous hypofunction but also sympathetic nervous dysfunction were observed during the interval stage of MD. Seki [13] studied sympathetic nervous function using pulse wave velocity and power spectral analysis. They reported that the same patient could present different results with the difference in methods used, and combinations of several methods, therefore, should be used. Even though sympathetic nervous dysfunction was observed in the patients with MD in this study, further researches will be needed.
425
4.2.2. AND during the attack stage At the attack stage, the HF/T rates of the patients after the rest were not significantly different from those of the healthy controls. The LF/HF rates of eight of the 12 patients (67%) were lower than those of the controls, which suggested that sympathetic nervous function was depressed. As acceleration of sympathetic response due to the postural change was not observed in the patients with MD, the burden seemed to have small effects on them even during the attack stage. In addition, the results were similar to the patients who underwent the tests during both attack and interval stages, as five of the six patients (83%) presented low LF/HF rates. Although one patient showed a high LF/HF rate, it was attributed to the inappropriate classification of the patient because the patient was the only one who received the test just 7 days after the attack of vertigo. These results are different from those reported by Matsunaga [7], that the acceleration of sympathetic function was observed in patients with MD during the attack stage. We consider that this difference is caused by the difference in the classification of attack stage. That is, all the patients in this study were in a state well enough to undergo the burden, but not in a so-to-speak acute attack stage of vertigo in which they hardly move their bodies. Even in this study, acceleration of sympathetic nervous function was found in one patient with frequent attacks of vertigo. However, the acceleration of sympathetic function during the attack stage may well be considered as a result of the vestibular and autonomic interactions. In this study, patients in the middle of vertigo attack stage presented a tendency to sympathetic nervous hypofunction rather than hyperfunction which was reported by other works [6,7,14], although no statistical significance was proven by the present study. This suggests that sympathetic nervous dysfunction can be not only a consequence but also one of the triggers of vertigo in MD. The conclusion that AND with parasympathetic nervous hypofunction during the interval stage relates to vertigo in MD as a predisposing factor is not different from those of other reports [5–7,13,14]. We consider, however, that further studies are needed to clarify whether AND during
426
M. Yamada et al. / Auris Nasus Larynx 26 (1999) 419–426
vertigo attacks is a consequence or a trigger in MD. In order to solve this problem, a continuous analysis should be made on the temporal course of MD in each patient throughout the periods of attack and interval instead of classifying patients by those two stages. In addition, as it has been reported that sympathetic dysfunction after the attacks presents quicker recovery than parasympathetic dysfunction [13], we consider that the continuous analyses on the temporal course have to be made in same timing, and as long as possible. We, therefore, are planning to have patients with MD installed with 24-h Holter electrocardiography, and make a study by analyzing the circadian variation of the autonomic nervous function.
References [1] Tokita T, Aoki S, Miyata H. The loading equilibrium examination for etiologic diagnosis of Meniere’s disease. Acta Otolaryngol (Stockh) 1971;72:107–16. [2] Uemura T, Itoh M, Kikuchi N. Autonomic dysfunction on the affected side in Meniere’s disease. Acta Otolaryngol (Stockh) 1980;89:109–17. [3] Inoue H, Uemura T. Sluggishness pupillary light contraction in patients with Meniere’s disease. Acta Otolaryngol (Stockh) 1988;105:582–6. [4] Igarashi S, Seki S, Sasaki H, Nagaba A, Nakano Y. Autonomic nerve function in patients with Meniere’s disease. An analysis using pulse wave velocity, RR interval and thermography. Equilibrium Res 1994;53:417–21. [5] Seki S, Igarashi S, Sasaki H, Nakano Y. Autonomic nerve function in patients with Meniere’s disease. An analysis using thermography. Equilibrium Res 1993;52:433 – 7. [6] Kawasaki Y. Autonomic nerve function of vertiginous patients. Assessment by spectral analysis of heart rate variability. J Otol Jpn 1993;96:444–56. [7] Matsunaga T. Autonomic nerve system function test in patients with vertigo. Pract Otol (Kyoto) 1987;80:865– 78. [8] Burton AC. The range and variability of the human fingers and the vasomotor regulation of body temperature. Am J Physiol 1939;127:437–53. [9] Kitney R, Rompelman O. Thermal entailment patterns in heart rate variability. J Physiol 1977;270:41–2. [10] Pappas DG, Crawford W, Coghlan HC. Dizziness and dysautonomic syndrome. Otolaryngol Head Neck Surg 1986;94:186 – 94.
[11] Pappas DG, Crawford W. Dizziness due to dysautonomia: response tospecific therapy. Southern Med J 1988;81:531 – 2. [12] Nozawa I, Imamura S, Hisamatsu K, Murakami Y. The relationship between orthostatic dysregulation and orthostatic test in dizzy patients. Eur Arch Otorhinolaryngol 1996;253:268 – 72. [13] Seki S. Autonomic nerve function in patients with vertigo. Evaluation for static function, variation and dynamic change using power spectral analysis of RR interval. J Otol Jpn 1997;100:457 – 66. [14] Ogino H. Autonomic nerve function in patients with Meniere’s disease. Pract Otol (Kyoto) 1991;41:80 – 93. [15] Pomeranz B, Macaulay RJB, Caudil MA, Caudill MA, Kutz I, Adam D, et al. Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol 1985;249:151 – 3. [16] Pagani M, Lonbardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a maker of sympatovagal interaction in man and conscious dog. Circ Res 1986;59:178 – 93. [17] Hayano J, Yamada M, Ohte N, Hashimoto T, Fujinami T, Takata K. Severity of coronary atheroscrelosis correlate with the respiratory component heart rate variability. Am Heart J 1991;121:1070 – 9. [18] Sayers BMcA. Analysis of heart rate variability. Ergonomics 1973;16:17 – 32. [19] Hyndman HW, Kitney RI, Sayers BMcA. Spontaneous rhythms in physiological control system. Nature 1971;233:339 – 41. [20] Taniguchi I, Kageyama S, Aihara K, Isogai Y, Katoh F, Fukuhara T. Assessment of power spectral analysis of heart rate functions. Effects of postural change and respiration. Auton Nerv Syst 1994;31:79 – 84 (in Japanese). [21] Berger RD, Saul JP, Cohen RJ. Transfer function analysis of autonomic regulation. I. Canine atrial rate response. Am J Physiol 1989;256:142 – 52. [22] Akselrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectral analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science 1981;213:220 – 3. [23] Wheeler T, Watkins PJ. Cardiac denervation in diabetes. Br Med J 1973;4:584 – 6. [24] Karemaker JM. Heart rate variability: why do spectral analysis? Heart 1997;77:99 – 101. [25] Vardas PE, Kochiadakis E, Manios EG, Kanoupakis EM, Zouridakis EG, Chlouverakis GI. Spectral analysis of heart rate variability before and during episodes of nocturnal ischaemia in patients with extensive coronary artery disease. Eur Heart J 1996;17:388 – 93. [26] Cavalcanti S, Severi S, Chiari L, Avanzolini G, Enzmann G, Bianco F, et al. Autonomic nervous function during haemodialysis assessed by spectral analysis of heart rate variability. Clin Sci 1997;92:351 – 9.