Objective To study the association between exposure to transportation noise and blood pressure (BP) reduction during nighttime sleep.
Methods 24-h ambulatory BP measurements at 15-min intervals were carried out on 149 persons living near four major European airports. Noise indicators included total and source-specific equivalent indoor noise, total number of noise events, annoyance scores for aircraft and road traffic nighttime noise. Long-term noise exposure was also determined. Multivariate linear regression analysis was applied.
Results The pooled estimates show that the only noise indicator associated consistently with a decrease in BP dipping is road traffic noise. The effect shows that a 5 dB increase in measured road traffic noise during the study night is associated with 0.8% (−1.55, −0.05) less dipping in diastolic BP. Noise from aircraft was not associated with a decrease in dipping, except for a non-significant decrease noted in Athens, where the aircraft noise was higher. Noise from indoor sources did not affect BP dipping.
Conclusions Road traffic noise exposure may be associated with a decrease in dipping. Noise from aircraft was not found to affect dipping in a consistent way across centres and indoor noise was not associated with dipping.
- Ambulatory blood pressure monitoring
- blood pressure
- blood pressure dipping
- environmental noise
Statistics from Altmetric.com
- Ambulatory blood pressure monitoring
- blood pressure
- blood pressure dipping
- environmental noise
Transportation noise, defined as undesirable sound mainly from road and air traffic, is an environmental stressor that has been associated with cardiovascular endpoints such as ischaemic heart disease and hypertension.1 2 Excessive central nervous system activation by environmental stimuli such as noise, leading to hyperactivity of the sympathetic autonomic nervous system (ANS) and the hypothalamic–pituitary–adrenal axis is the mechanism proposed to explain the effects of noise on the cardiovascular system.3 4 One of the most important aspects of environmental noise exposure with regard to health effects is sleep disturbance.5
During sleep, a state of reduced sympathetic and preponderant parasympathetic ANS tone, blood pressure (BP) shows a physiological decline (dipping),6 measured usually as a percentage BP decline with reference to daytime values. Dipping shows a normal distribution in the general population7 and its absence (‘non-dipping') is consistently found in situations with impairment of the ANS, either central8 9 or peripheral.10 11 Non-dipping, together with enhanced BP variability, have also been reported as persisting consequences of major stressful events such as earthquakes.12 Moreover, there is evidence that non-dipping may be associated with the quality of sleep and that sleep disturbance may contribute to its causation due to pathological conditions of the individual13 or to extrinsic factors.14 Finally, non-dipping has been proposed as an independent risk factor for cardiovascular morbidity,15 16 although it is not always reproducible on an individual basis when a second ambulatory blood pressure monitoring (ABPM) is performed.17
Within the wider framework of the European HYENA project,18 in the present study we investigate the a-priori hypotheses that exposure to noise, either long term or acute during the specific study night, can affect BP dipping, assessed by ABPM, in individuals living in the vicinity of four major European airports.
The hypotheses for the present study were stated a priori and included in the initial protocol as a separate working package of the HYENA study. The sample was selected from the main sample of the HYENA project (age range 45–70 years) and consisted of individuals living around four European airports with night flights: Athens (Greece), Malpensa (Italy), Arlanda (Sweden) and London Heathrow (UK). Details of the sampling procedures of the HYENA project are reported elsewhere.18 The desired sample size for the present study was based on power calculations for finding a difference in BP dipping between participants chronically exposed to aircraft noise and non-exposed subjects. Based on the data reported by Staessen et al7 and according to our power calculations, a total of 100 subjects chronically exposed to aircraft noise and 100 non-exposed subjects would give us a power of 80% to detect an increase of 4.8 units in the night/day BP systolic ratio and 5.8 units in the night/day BP diastolic ratio.
The following exclusion criteria were applied in the selection of subjects: (1) antihypertensive medication; (2) a diagnosis of diabetes mellitus; (3) a diagnosis of obstructive sleep apnoea syndrome; (4) a diagnosis of secondary hypertension; (5) working on a night shift; (6) using sleeping pills and sedatives; (7) a diagnosis of (or self-reported) hearing impairment; (8) regular use of ear plugs; or (9) a diagnosis of atrial fibrillation (confirmation of absence of this condition was also made from the ABPM recordings). Criteria 1–6 were applied as they affect nighttime BP, criteria 7 and 8 as they modify noise exposure and criterion 9 as it hinders ABPM.
Approval for the study was granted by each centre's ethical committee and all participants gave written informed consent.
Measurements and data management
Long-term noise exposure assessment to transportation noise (aircraft and road traffic separately) for each participant according to his/her residence is based on the HYENA project's A-weighted equivalent noise level for a whole 24 h period averaged over a year (LAeq 24 h), precisely the year 2002 for the HYENA project, and is described elsewhere.18
Noise exposure assessment during the study night was performed by continuous noise measurement using a type I CESVA SC310 (CESVA instruments, SL, Barcelona, Spain)19 noise metre (integration time 125 ms LAF) as well as noise recording with an mp3 recorder connected to the noise metre's high-quality microphone during the study night in each participant's bedroom. Each participant was followed for one night. From the 125 ms values we calculated the A-weighted equivalent noise level for every second (LAeq 1 s) and subsequently for the whole nighttime sleeping period (defined by a sleep log filled in by the participants) as:where t is the sleeping time in seconds.
Using playback and visualisation of sound recordings on a computer, the source of each event was identified and synchronised with the sound measurements with a program written for this purpose. An event was defined as present if its indoor A-weighted maximum level (LAmax) exceeded 35 dB. Noise events were classified into four categories according to source: indoor source, aircraft, road traffic and other outdoor source. Other outdoor events were very rare and were thus excluded from the analysis. We calculated for each noise event the LAmax and the sound exposure level (SEL), which represents the event's sound energy. From the SEL we calculated the equivalent sound level of noise events averaged over the whole sleeping period for each participant.
Non-invasive 24-h ABPM, with heart rate measurements, was performed at 15 min intervals with the validated Mobil-O-Graph device (I.E.M. GmbH, Stolberg, Germany),20 including the night of sound measurements. The 15-min frequency has been implemented before21 and was chosen as optimal for frequent measurements without excessive sleep disturbance. Each participant completed a questionnaire shortly after waking indicating sleeping hours and providing subjective information on his/her quality of sleep, and on disturbance from the ABPM device. ‘Waking’ and ‘sleeping’ BP values were defined, in keeping with current practice, from the questionnaire data22 and BP dipping from day to night was calculated as mean BP during awake time minus mean BP during nighttime and divided by mean BP during awake time. BP values during any daytime sleep (‘siesta’) were excluded because their inclusion as awake values is known to underestimate nocturnal BP dipping.23 The three instruments (noise metre, noise recorder and ABPM device) and the participants' alarm clock were synchronised at 1 min precision.
Specially trained staff installed the noise equipment, placed the ABPM device on the participants and gave them written instructions, during a home visit at least 3 h before normal sleeping time. Each participant was instructed not to engage in unusually heavy activity during the measurement period and filled in a sleep log indicating actual sleep times.
We also used, alternatively to the objective noise levels, the annoyance scores provided by the participants by means of a questionnaire, administered to the whole HYENA sample by interview. As a result of the limited availability of instruments, the study was performed at different time periods in each centre: in Athens from April 2004 to December 2004; in Milan from December 2004 to July 2005; in Stockholm from September 2005 to December 2005 and finally in London from December 2005 to February 2006. Subjects were examined on different days.
Multiple linear regression models were applied for the data from each centre separately in order to assess long-term or nighttime noise effects on BP dipping (systolic and diastolic). The noise exposure based on the annual estimate (LAeq 24 h and Lnight) for aircraft and road traffic were used as long-term noise exposure variables. Other noise exposure variables used alternatively were the A-weighted equivalent noise level measured indoors during the study night (LAeq night) and the equivalent sound levels of source-specific noise events (aircraft, road traffic and indoor source) averaged over the whole sleeping period for each participant. The number of source-specific events (aircraft, road traffic and indoor source) with LAmax exceeding 35 dB during the study night and the annoyance scores from chronic exposure to air and road traffic during nighttime were also used as exposure variables. Moreover, self-reported disturbance from the ABPM device and quality of sleep, as assessed by questionnaire, were used alternatively as independent variables as they have been reported to affect nocturnal BP.24 To account for potential confounding effects we applied multiple linear regression models with BP dipping as outcome and sex, age (years), current smoking habits (smoker/non-smoker), body mass index (as weight in kg divided by height squared in m2) and working status on the day of the measurements (working/not working) as covariates, because these factors have been reported to affect BP dipping.25–27 After obtaining the four centre-specific effect estimates using multiple linear regression models, we then combined the centre-specific results using either fixed or random effects meta-analysis depending on the absence or presence of heterogeneity.28 All reported p values are based on two-sided hypotheses and the significance level used was 5%. Data analysis was performed using Intercooled Stata 8.2.
The initial sample size target of 50 subjects per centre was not reached in most centres, as the field work was particularly demanding. Furthermore, 12 subjects were excluded due to technical problems with the sound monitoring or ABPM equipment. Exclusion criteria relative to ABPM included duration of examination less than 21 h, fewer than 70 valid readings, reading success lower than 80%, and no-reading interval(s) longer than 2 h. The total sample consisted of 149 subjects, of which 62 were in the non-exposed noise category and 87 were in the exposed noise category (table 1). When taking into account noise long-term noise exposure variables and noise annoyance scores the sample consisted of 149 subjects, while when using indicators obtained only from the measured indoor noise during the specific study night (total noise equivalent of the specific night and source-specific noise) nine subjects were excluded from the analysis due to noise equipment failure during that specific study night. Table 1 shows the descriptive characteristics of the samples. Systolic and diastolic BP dipping both displayed a normal distribution (data not shown) with values similar to those previously reported from large databases.6
Table 2 and figure 1 show descriptive data on the noise and annoyance indicators used. The number of aircraft events during the nighttime sleep was higher in Athens compared with the other centres. The presence of subjects with no or few night aircraft events was explained by the sampling procedure through which a number of subjects were selected from the main subsample of low exposure to aircraft noise and by the fact that an event was defined by noise measured indoors with LAmax greater than 35 dB. The median indoor noise equivalent measured during the study nighttime sleep (LAeq night) was higher in Athens.
Compared with the three other centres, the median aircraft noise level (based on SEL10) measured indoors during the study night was higher in Athens. However, the median road traffic noise levels (based on SEL10) were similar in the four samples.
Table 3 shows the pooled effect estimates of the noise exposure indicators, of long-term annoyance from aircraft and road traffic, of sleep quality during the study night and of disturbance from the device on BP dipping, after adjusting for potential confounding effects. The pooled estimates from all four centres show that the only noise indicator associated consistently with a decrease in BP dipping is higher road traffic noise during the study night. The effect is statistically significant only on diastolic dipping and shows that a 5 dB increase in measured road traffic noise during the study night is associated with 0.8% less dipping in diastolic BP. The effect estimates of LAeq 24 h and Lnight, which express the effect of modelled long-term aircraft and road traffic noise exposure, respectively, were consistently identical between the four cities (data not shown). Other variables expressing noise exposure or long-term annoyance from aircraft or road traffic noise were not associated with BP dipping in a consistent way across the four centres (figures 2–4). However, in Athens, where measured aircraft noise was much higher, there is a consistent decrease in dipping in all models with aircraft, road traffic, indoor source and measured indoor noise indicators (figures 2–4).
Finally, better reported quality of sleep was associated with an increase in dipping, which was nearly significant (p=0.09), while annoyance from the ABPM itself was associated with decreased dipping only in the Athens sample. The corresponding results for systolic BP dipping were similar, with the exception of significant heterogeneity in the effects of the annual estimate of the aircraft noise exposure, adjusted for road traffic noise. The effect of the reported disturbance from the ABPM device showed significant heterogeneity between centres.
We tested the hypothesis that exposure to noise, either short-term, that is during the study night, or long-term, can affect the physiological BP decline during sleep. In the present paper a decrease in BP dipping (especially diastolic) is associated only with road traffic noise during the study night, even though the attributed noise to road traffic events was low. No overall effect was observed of aircraft or indoor noise in the four centres. However, the measured nighttime noise from aircraft, which was done indoors, in the subjects' bedrooms, was considerably and significantly higher in the Athens sample. If the results of the aircraft-related noise effects and annoyance are observed, it is seen that consistent estimates of a decrease in dipping are obtained for Athens, which are not statistically significant due to the low statistical power in one sample. This may imply that the aircraft noise indoors in the other samples was generally below a threshold for the studied effect. Noise, and more specifically recorded transportation noise, is known to be able to cause BP elevations during sleep in laboratory conditions.29 30 In our previously published paper on the same individuals31 we showed that higher BP measurements are associated with noise events from aircraft, road traffic, or indoor sources.
Therefore, although noise events were associated with BP elevations,31 these were not of sufficient magnitude or number in each individual in all centres so as to affect the day–night BP ratio. A larger sample or a larger number of noise events might have been able to detect this. Another explanation might be that individuals with more noise, BP elevations and possibly arousals during the study night also had higher BP during the morning after (most 24 h BP monitorings started in the afternoon before the study night) and thus their day–night BP ratio was not affected. Higher awake BP has been reported in individuals with arousals and sleep fragmentation.32
Modelled long-term noise exposure was not found to be associated with BP dipping during our study night. A possible explanation is that environmental noise at the exposure levels of our study subjects may not activate the sympathetic ANS on a long-term basis, at least not to the extent needed to produce non-dipping, or non-dipping may not be a sensitive indicator of ANS disturbance. It should be noted that non-dipping is found more consistently in individuals with diseases associated with severe, clinically manifested or paraclinically diagnosed peripheral or central ANS impairment such as diabetes mellitus, familial amyloid polyneuropathy, stroke and Shy–Drager syndrome, and therefore the exclusion of individuals with diabetes mellitus and diagnosed hypertension, who might respond more to environmental stressors, may have led to the underestimation of the effect.8–11 Furthermore, it has been reported that non-dipping is not always reproducible when a second ABPM is performed on the same individual,17 supporting the view that it may not be an indicator of chronic or permanent dysfunction of the organism. It would have been preferable to include more study nights per subject, which was impossible considering the burden imposed on the subjects and the limited availability of the equipment.
Alternatively, the specific estimation of long-term noise may be subject to misclassification error that lowers the power to detect an effect. This may have resulted in the discrepancy in the results of measured and estimated noise. Furthermore, it should be noted that measured noise indoors takes into account the home's noise insulation, which may be especially important for nighttime exposure.
Finally, although BP measurements on noise-exposed or unexposed subjects were all performed with ABPM, the effect of cuff inflation on sleep and on the nocturnal BP of subjects sensitive to the disturbance from the device24 may have masked the chronic (and presumably small) effects of noise we were looking for. Our results of the effects of greater self-reported disturbance from the device on a decrease in BP dipping showed significant heterogeneity between centres and that of self-reported sleep quality during the study night did not reach the nominal statistical significance level, although it was nearly significant and considerably low showing that better sleep and less disturbance from the device were associated with greater BP dipping.
What is already known on this subject
Noise is an environmental stressor that has been associated with hypertension, mainly in laboratory studies. In particular, there is evidence that transportation noise can cause blood pressure elevations during sleep in laboratory conditions.
What this study adds
The present study was performed in order to investigate the association between exposure to transportation noise and BP reduction during nighttime sleep. To our knowledge, this study is the first based on noise measurements in a population living under their normal conditions at home. Our results provide some evidence that road traffic noise may contribute to less BP dipping during sleep and they also indicate that there may be an effect of aircraft noise in locations with nighttime flights. Preventive measures should thus be considered to reduce transportation noise during nighttime.
The authors would like to thank all the participants for their willingness to contribute. They also thank the members of the HYENA study team: Joy Read, Yvonne Tan, Yousouf Soogun, Gabriele Wölke, Jessica Kwekkeboom, Birgitta Ohlander, Eva Thunberg, Elli Davou, Yannis Zahos, Venetia Velonaki, Ageliki Athanasopoulou, Alessandro Borgini, Maria Chiara Antoniotti, Salvatore Pisani, Giorgio Barbaglia, Matteo Giampaolo, Mauro Mussin, Cristina Degli Stefani, Tiziana Vanoli, Federica Mathis, Laura Cianfrocca and Clara Tovo for assistance during various parts of the study.
Funding The HYENA study was funded by the European Commission DG Research (contract no QLK4-CT-2002-02501) in the fifth framework programme, quality of life and management of living resources.
Competing interersts None.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of each centre's ethical committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.