| [Download PDF]
|Year : 2004 | Volume
| Issue : 23 | Page : 21--28
Low frequency noise and stress : Bronchitis and cortisol in children exposed chronically to traffic noise and exhaust fumes
H Ising1, H Lange-Asschenfeldt2, HJ Moriske3, J Born4, M Eilts5,
1 Berlin Centre for Public Health, Berlin, Germany
2 Federal Ministry of Health and Social Security, Berlin Office, Germany
3 Federal Environmental Agency, Berlin, Germany
4 Department of Neuroendocrinology, University of Lubeck, Germany
5 Practising ENT specialist in the district of Osterode/Harz, Germany
Rheinstr. 69, D-14612 Falkensee
A correlation of respiratory diseases to traffic related air pollution and noise was observed in an interview study. Since in that study the exposure was subjectively assessed, in the present field study nitrogen dioxide as indicator for vehicle exhausts and the mean night-time noise level were measured outside the children's windows in representative locations. Based on these measurements each child was placed in one of the following categories: low, medium or high traffic immission (ambient emissions). The physician contacts due to bronchitis of 68 children were assessed retrospectively from the files of the participating paediatricians. Saliva samples were collected from all children and the cortisol concentration was estimated. Children under high noise exposure (L night,8h= 54-70dB(A)) had in comparison to all other children significantly increased morning saliva cortisol concentrations, indicating an activation of the hypothalamus-pituitary-adrenal (HPA) axis. Analysing a subgroup of children without high noise exposure showed, that children with frequent physician contacts due to bronchitis did not have increased morning saliva cortisol. However, multiple regression analysis with stepwise exclusion of variables showed that bronchitis was correlated more closely to morning salvia cortisol than to traffic immissions. On the other hand, the rate of physician contacts due to bronchitis increased in a dose dependent manner and significantly with increasing traffic immissions.
From these results it can be concluded that high exposure to traffic noise, especially at nighttime, activates the HPA axis and this leads in the long term to an aggravation of bronchitis in children. This seems to be more important than the effect of exhaust fumes on bronchitis symptoms. The results of the present study should be subjected to further investigation using specially designed studies.
|How to cite this article:|
Ising H, Lange-Asschenfeldt H, Moriske H, Born J, Eilts M. Low frequency noise and stress : Bronchitis and cortisol in children exposed chronically to traffic noise and exhaust fumes.Noise Health 2004;6:21-28
|How to cite this URL:|
Ising H, Lange-Asschenfeldt H, Moriske H, Born J, Eilts M. Low frequency noise and stress : Bronchitis and cortisol in children exposed chronically to traffic noise and exhaust fumes. Noise Health [serial online] 2004 [cited 2023 Jan 29 ];6:21-28
Available from: https://www.noiseandhealth.org/text.asp?2004/6/23/21/31666
In a pilot field study, stress hormone excretion of children was measured in the first and the second half of the night. The results indicated an impairment of the normal rhythm of nocturnal cortisol excretion in children living in streets with high traffic load as compared to less exposed children. This endocrine change was correlated to restless sleep, difficulties in returning to sleep after awakening during the night, as well as to bronchial asthma and allergy prevalence (Ising H. and Ising M. 2002). Moreover, traffic-related immission correlated with signs of increased bronchitis. Our results suggest that lorry noise at night with mean indoor maximal levels of L max > 63 dB(C) could be detrimental to health even at indoor levels in dB(A) which up till now have not been rated as interfering with healthy sleep (Interdisziplinarer Arbeitskreis fur Larmwirkungsfragen beim Umweltbundesamt 1982). As previously reported (Ising H. and Ising M. 2002) the frequency spectrum of the lorry noise in the children's bedroom had its maximum below 100 Hz.
The present field study extends a previous study (Ising et al. 2003) using blind interviews in a greater sample of parents to investigate the combined effects of chronic exposure to traffic related air pollution and noise on the risk of skin and respiratory diseases in children. That study included the parents from all children between 512 years, who had consulted one of two participating paediatricians in the area of interest here. The paediatricians' diagnoses of 401 children were analysed together with their parents' answers regarding the density of road traffic on their street and several confounding factors. Multiple regression analyses resulted in relative risks of asthma, chronic bronchitis and neurodermitis, which increased significantly with increasing traffic load (Ising et al. 2003). Since in that interview study the exposure was subjectively assessed, it was decided here to measure the traffic related immissions in representative locations in order to determine immission categories objectively. Therefore, a field study was planned including a 20%subgroup of our previous study. It should be answered whether results of the interview study can be validated on the basis of objective measures of noise and air pollution.
While previous studies have provided considerable support for the notion that chronic exposure to traffic exhaust fumes aggravates bronchitis in children, the role of noise especially during night-time, is less clear in this context. The effect of night-time noise was a focus of the present study. It was hypothesized that noise presumably by elevating stress-related excretion of cortisol from the pituitary-adrenal system, contributes to the development of bronchitis. In previous study, traffic noise was revealed to be associated with significant increases in cortisol during the early night, a period which is normally characterized by complete suppression of any stress-related cortisol release (Ising H. and Ising M. 2002).
Most of the highly exposed children, who participated in that and in the here presented study, lived near the road B 243, especially used by heavy lorries. Even at night, on average every two minutes a heavy goods vehicle passed by the houses at distances as low as 2 m, causing also noticeable vibrations.
From the 401 children of the interview study four subgroups were selected with high and low traffic exposure and with and without diagnosed bronchitis respectively. [Table 1] indicates for each of the four groups the children selected for the field study from the initial greater sample of children whose parents were questioned in the previous interview study. The interview with the parents was taken by the nurses in order to document their estimation of the road traffic load in front of the children's room windows. All children had contact with their physician for at least two years. To quantify the physicians' diagnoses the frequencies of the patients' contacts with their physician were assessed from the physicians' files retrospectively for the last five years.
From all participating children, saliva samples were collected four times, in the evening, at 1 a.m., in the morning at waking up and half an hour later. Samples were kept frozen till analysis by standard radioimmunoassay.
Noise and nitrogen dioxide concentrations as indicator for traffic-related emissions were measured outside the bedroom window of the children. The A-weighted mean level L eq (night) was measured between 10 p.m. and 6 a.m. (Larson Davis dosimeter Type 700) in 25 representative locations. For nitrogen dioxide measurement, passive samplers (Palmes tubes) which had shown reliable results during prior outdoor measurements (Moriske et al., 1996 a;b) were exposed for 58-93 h, and the exposure time was documented. Nitrogen dioxide was measured in 41 representative locations as indicator for vehicle exhausts. Additionally, a protocol of air pollution measurements (nitrogen dioxide and sedimented dust) over one year was used to describe the long-term air quality (Deutscher Wetterdienst).
Noise and nitrogen dioxide measures were taken to classify each child in one of three categories: (i) 'Control area'children - L night 2 3* ), (ii)'High exposure' - L night > 53 dB(A) and NO 2 > 40 µg/m 3* ), (iii)'Medium exposure' - defining all other exposures in between.
* )Mean values up to four days exposure
To ensure correct classification in the control area, where no measurements were taken, only quiet streets with a minimum distance of 200 m from major through roads were considered. Children were accepted in the 'high exposure' group without measurement, if the window of their bedroom faced directly or parallel to a major through road and the distance to that road was less than 10 m.
Associations between exposure to road traffic emissions and health outcome were estimated by multiple regression analyses (Systat Version 6.0.1. 1996. SPSS INC.). For statistical adjustment, the potential confounding factors age, sex, smoking (environmental tobacco smoke (ETS) - at least one person at home is smoker), and the parents' highest educational level were considered as covariates in the multiple model.
According to the parents' statements only 5 out of the 68 children had any sleeping problems. 12 children felt disturbed by traffic noise. 63 children had been under treatment of the participating physicians for five years or longer. The total number of physician contacts due to bronchitis during the last five years amounted to 691.
Results of noise and nitrogen dioxide measurements for the three exposure classes are shown in [Table 2]. Results of the official measurement protocol (Deutscher Wetterdienst) for the year 2000 for Bad Lauterberg (Harz mountains), the town where most of the children lived, is given in [Table 3]. This protocol included one representative location for each of the three exposure classes defined here. The data in [Table 3] represent average values across 51-53 weeks. Important to note is, that the highly traffic exposed area of this protocol was located at a through road (Federal Road B 27), which had considerable less traffic than the other through road in this area (Federal Road B 243) where the majority of our 'High exposure' children lived. [Table 4] gives the numbers of children for the selected classes of noise and nitrogen dioxide, separately as well as in the combination, as used to defining the 3 classes of traffic-related emission exposure. Night time traffic noise and nitrogen dioxide were closely correlated (Spearman rank correlation coefficient R = 0.836, p 53 dB(A). The noise effect remained significant after controlling for the following covariates: age, sex, parent's highest level of school education, day of week of saliva sampling. The result of multiple regression analysis with stepwise exclusion of variables with error probabilities p>0.1 is shown in [Table 7].
To examine the effect of bronchitis on cortisol, the children with the highest noise exposure (L eq > 53 dB(A)) were excluded. This exclusion did not change the distribution of bronchitis contacts p.a. - the median and the upper hinge remained 1.2 and 2.0 respectively. However, in the reduced group (n = 34) there was no association between morning saliva cortisol and the annual rate of physician visits due to bronchitis (r = 0.101; n.s.) in contrast to a significant correlation in the whole group (n = 43, r = 0.321; p 0.1. As shown in [Table 8] the annual physician contacts due to bronchitis increase dose dependent and significantly with traffic immissions (covariates at the beginning of stepwise ANOVA: age, sex, smoking at home and parent's highest level of school education).
Elimination of children with missing saliva cortisol data led to a reduced correlation between physician contacts due to bronchitis and high traffic immission (p 3 ) was similar to other low polluted areas in Germany (Lahmann E. and Moriske H-J. 1995; and Moriske H-J. and Torowski E., 1998). At sampling points with medium, and especially at those with high exposure, traffic emissions correspondingly were enhanced to mean values of 28 and 63 µg/m 3 , respectively.
Comparison of our three days nitrogen dioxide measurements [Table 2] with the one year average values of the Deutscher Wetterdienst (DWD, [Table 3]) shows, that for the control area our short term estimate of nitrogen dioxide concentration (16µg/m 3 ) came close to the 95 th percentile value (18µg/m 3 ) of the DWD protocol. The range of nitrogen dioxide concentration in the control area extended from 12 - 20µg/m 3 . In contrast, in the medium exposure area, the 95th percentile of the DWD (19µg/m 3 ) was near the lower end of the variance of our nitrogen dioxide concentration (13 - 45µg/m 3 ). In the high exposure area the 95 th percentile of the DWD (36µg/m 3 ) fell below the minimum of our nitrogen dioxide estimate (42µg/m 3 ). Our sampling point was - like that used by the DWD protocol - located at road B 27, but was situated at a closed front of houses while that of the DWD was located at a "free space". Notably, on the other major road B243 traffic density is much higher, leading to short term nitrogen dioxide concentrations between 60 and 80 µg/m 3 . This would correspond to yearly mean values above 40 µg/m 3 , similar to areas in major cities (Moriske H-J. and Torowski E. 1998; Miicke H-G. et al., 2001).
The night-time mean noise level outside the bedroom windows of the children was very low in the control area: 29 - - 39 dB(A) and between high and very high in the high exposure area: 54 - - 70 dB(A). In Germany 16 % of the population are exposed to night time mean traffic noise levels > 55 dB(A) and 0.2% to levels > 70 dB(A) (Ising et al. 2001).
The region where the study was conducted is a mountain area known as a health resort because of its good air quality. Close to this area there is a major road (B 243), where air pollution and traffic noise exposure are extremely high. These conditions lead to the extreme exposure differences and clear effects of traffic emissions. The traffic on road B 243 increased drastically within the last 12 years after the reunification of Germany. Before 1990 this road ended at the border between the former East and West Germany, but now is the connection between two major industrial areas (Halle/Bitterfeld and Hannover/Salzgitter). Even at night, on average every two minutes a heavy goods vehicle passes by the houses (Ising H. and Ising M. 2002) at distances as low as 2 m. Most of the inhabitants own their houses, but their value has dropped drastically since the traffic load increased. Therefore only few people could sell their property and moved to a less exposed area. The majority of the people stay in spite of the traffic exposure and struggle for a bypass road. Remaining inhomogeneity of the social economic distribution in the different immission groups, were model adjusted using the standard of education reached by the parents.
Using blind interviews with subjective assessment of traffic exposure, in a previous study in an extended sample of the same population we had found that traffic load is associated with an increased risk of frequent bronchitis (Ising et al. 2003). This result is confirmed by the presented field study on the basis of objectively assessed traffic noise and exhaust fumes. Additionally, the occasional diagnosis of the physicians used in our former study was replaced by the retrospectively assessed contact rate during the last five years. Of special importance are the correlations between night-time noise exposure, saliva cortisol concentrations and bronchitis. The increase of saliva cortisol in the morning in children exposed to night-time noise levels of L eq = 54 - 70 dB((A) complements our previous results (Ising H. and Ising M. 2002) of an increase in cortisol excretion in the first half of the night in children exposed to mean indoor maximal levels of L max > 63 dB(C). Most of the highly exposed children, who participated in both studies, lived near the road B 243, especially used by heavy lorries. As previously reported (Ising H. and Ising M. 2002) the frequency spectrum of this lorry noise has its maximum below 100 Hz. In several houses vibrations caused by lorries were noticeable but not measured.
The human acoustical system is on permanent alert serving as the most important sensory warning system. Thus, environmental sound signals are perceived and processed also during sleep. In sound processing, specific sub-cortical arousing systems of the central nervous system (CNS) - which also influence the regulation of cortisol - are of great importance (Spreng, 2000). While asleep, sound signals such as the noise of a lorry passing by, can induce increases in cortisol release even at lower noise levels (Ising et al. 2001). In sleeping persons and in those seemingly habituated to night-time traffic noise, one of the effects of noise is obviously an increased activation of the pituitary-adrenal stress system (Ising and Brown 2000, Ising H. and Ising M. 2002). An effective suppression of cortisol release during nocturnal sleep and particularly during early sleep has been considered critical for the recovery of mental and bodily functions during sleep (Born and Fehm, 2000).
Divergent from our previous study (Ising H. and Ising M. 2002) where cortisol release during early sleep was found to be most sensitive to the effects of traffic-related noise, here average saliva cortisol concentrations in the morning showed a distinct association with traffic noise load. This discrepancy might stem from methodological differences in the two studies, since early night cortisol concentrations in saliva are at a minimum and close to the threshold of sensitivity of the radioimmunoassay used. Measurement of cortisol accumulating in urine over the 3-4 hour period of early sleep, as employed in the our previous study, may thus represent the more appropriate approach than saliva samples at single time points for assessing differences in cortisol release during this phase of the night. However, the effect of traffic noise here being limited to morning cortisol concentrations could also point to an alternative explanation. Increases in awakening cortisol concentrations have been consistently demonstrated to be particularly linked to subjectively perceived acute and chronic stress (e.g., Pruessner et al. 2003, Wuest et al. 2003). Thus, rather than immediate effects of night-time noise, the increase in morning cortisol concentrations in our children with high exposure to traffic noise could reflect a more general feeling of strain and reduced well-being, originating from living nearby a road with high traffic noise. This view may be supported by a recent study, indicating that experimental presentation of low frequency continuous noise during sleep did not enhance but reduce morning saliva cortisol concentration (Persson Waye et al 2003). However, there are important differences between the two types of noise. Persson Waye used a constant installation noise in contrast to the fluctuating noise of lorries, which additionally is a signal of danger.
Since in the subgroup without high noise exposure the correlation between saliva control and the bronchitis indicator was not significant, the increase in saliva cortisol in the morning cannot be primarily reduced to increased bronchitis. However, in the total group a significant correlation between saliva cortisol and noise categories was obvious, which in the stepwise analyses was overridden by the even more distinct correlation between saliva cortisol and physician contact rate. This pattern is consistent with the following 'pathogenic' model of bronchitis in our children:
High exposure to traffic noise, especially at night-time and with predominant low frequency spectra, is associated with enhanced activation of the hypothalamus-pituitary-adrenal (HPA) axis and this in turn leads in the long term to an aggravation of bronchitis in children.
This model substantially extends a model established in the literature, indicating that traffic exhaust fumes lead to an aggravation of bronchitis.
Since these conclusions are based on a correlational and rather limited set of empirical data, specially designed studies are required to further support and elaborate these models.
The authors wish to thank the parents, paediatricians and nurses, who co-operated in this study, as well as the Association for Water, Soil and Air Hygiene for financial support of this study.
|1||Born, J., Fehm, H. (2000). The neuroendocrine recovery function of sleep. Noise & Health 7:25-37|
|2||Interdisziplinarer Arbeitskreis fMr Larmwirkungsfragen beim Umweltbundesamt (1982), Beeintrachtigung des Schlafs durch Larm. Z. Larmbekampfung 29,13-16|
|3||Ising H. Braun C. (2000) Acute and chronic endocrine effects of noise: Review of the research conducted at the Institute for Water Air and Soil Hygiene. Noise & Health 7: 7-24|
|4||Ising H. Kruppa B. Babisch W. Gottlob D. Guski R. Maschke C. Spreng M. (2001). Kapitel VII-1 Larm, In: Wichmann, Schlipkoter, Fiilgraff (eds.) Handbuch der Umweltmedizin. Erg. Lfg.7/01 Ecomed, Landsberg.|
|5||Ising H. Ising M. (2002). Chronic cortisol increases in the first half of the night caused by road traffic noise. Noise & Health 4;16, 13-21|
|6||Ising H. Lange-Asschenfeldt H. Lieber G.-F. Weinhold H.|
|7||Eilts M. (2003). Respiratory and dermatological diseases in children with long-term exposure to road traffic immissions. Noise & Health; 5;19, 41-50|
|8||Lahmann E., Moriske H.-J. (1995).Trend and peaks of ambient air pollution with sulphur dioxide and nitrogen dioxide in Germany. Proceedings, 10th World Clean Air Congress, Espoo (Finland) 28.5.-2.6.95. Eds.: P. Anttila, J. Kameri und M. Tolvanen. Vol. 2, No. 234|
|9||Moriske H.-J., Schondube M., Menk G., Seifert B. (1996a). Erfassung von NO 2 -Konzentrationen in der Auβenluft mittels Passivsammlern nach Palmes. 1. Mitteilung: Laborversuche und Qualitatssicherung. Gefahrstoffe-Reinhaltung der Luft 56, 129-132|
|10||Moriske H.-J., Schondube M., Ebert G., Menk G., Seifert B., Abraham H.-J. (1996b). Erfassung von NO 2 Konzentrationen in der Auβenluft mittels Passivsammlern nach Palmes. 2. Mitteilung: Feldversuche. GefahrstoffeReinhaltung der Luft 56, 161-164|
|11||Moriske H.-J., Turowski E. (Eds.) (1998). Handbuch fir Bioklima und Lufthygiene. ecomed- Verlagsgesellschaft, Landsberg|
|12||Miicke H.-G. Koch T. Ranft U. (2001). Kapitel VIII-1.3.1 Kraftfahrzeugverkehr -Belastungssituation. In: Wichmann, Schlipkoter, Fiilgraf (eds.) Handbuch der Umweltmedizin. Erg. Lfg.3/01 Ecomed, Landsberg|
|13||Persson Waye K. Clow A. Edwards S. Hucklebridge F. Rylander R. (2003), Effects of nighttime low frequency noise on the cortisol response to awakening and subjective sleep quality Life Science, 72 (8) 863-875.|
|14||Pruessner M, Hellhammer DH, Pruessner JC, Lupien SJ (2003) Self-reported depressive symptoms and stress levels in healthy young men: associations with the cortisol response to awakening. Psychosom Med; 65, 92-9|
|15||Spreng, M. (2000). Central nervous system activation by noise. Noise & Health; 7, 49-57.|
|16||Wiist S. Wolf J. Hellhammer D.H. Federenko I. Schommer N. Kirschbaum C. (2000) The Cortisol awakening response - normal values and confounds. Noise & Health; 7,79-88.|