Excessive Daytime Sleepiness as a Risk Factor for Obstructive Sleep Apnoea among Public Transport Drivers: A Cross-Sectional Study

Abstract

Context:

Excessive daytime sleepiness (EDS) due to obstructive sleep apnoea (OSA) is reported to be a major contributor to many road traffic accidents. Lack of awareness and diagnosis of OSA among public transport workers remains a threat to the society.

Aims:

The primary aim of this study was to assess the risk of OSA among transport drivers of south Kerala using modified Berlin questionnaire. The secondary objective included craniofacial assessment of the high-risk patients identified through the questionnaire using lateral cephalogram.

Settings and Design:

A cross-sectional study was conducted among 180 transport drivers of south Kerala.

Methods and Material:

Modified Berlin questionnaire and limited physical examination [body mass index (kg/m²), neck circumference (cm), waist circumference (cm), hip circumference and waist to hip ratio, blood pressure (mm Hg)] were recorded. The screened subjects were categorized as high-risk snorers and low-risk snorers based on modified Berlin questionnaire. Craniofacial morphological variations of high-risk group were assessed by lateral cephalograms.

Statistical Analysis Used:

The descriptive statistics were represented as mean and standard deviation and percentage. Inter-group comparison was performed with independent sample t test.

Results:

The study demonstrated 64.4% of subjects were non-snorers and 35.6% were snorers. Furthermore, among the snorers, 46.9% were identified as high-risk snorers, whereas the remaining 53.1% represented low-risk snorers.

Conclusions:

The study revealed the concealed risk of OSA among transport drivers could be screened through the questionnaires and demographics assessment. The application of the proposed screening protocol would triage and enhance safety of OSA affected transport drivers.

INTRODUCTION

Obstructive sleep apnoea (OSA) is the repeated reduction or cessation of breathing caused by the narrowing of the upper airways during sleep.[1] Symptoms of OSA includes snoring, excessive daytime sleepiness (EDS), gasping at night, cognitive impairment, heartburn, morning headaches, insomnia, erectile dysfunction, nocturia, etc.[2] Several parameters such as high body mass index (BMI), increased waist circumference, increased neck circumference, and increased waist-to-hip ratio (WHR) leading to obesity are considered as risk factors for OSA.[3] Altered craniofacial morphology such as inferiorly positioned hyoid bone, posteriorly positioned mandible, enlarged tongue, soft palate, etc. are also identified as contributing factors in the pathogenesis of OSA.[4]

A medical history and use of screening questionnaires for sleep disorders can provide right patient selection for further investigations. The Berlin Questionnaire (BQ) is accepted as a reliable tool to determine high-risk patients for OSA.[5,6] The Epworth Sleepiness Scale (ESS) is a commonly used tool for the evaluation of EDS.[7]

Driving is a complex task that involves various motor, cognitive, and perceptual processes and requires constant interaction with the road and the environment. Falling asleep at the wheel is a significant cause of road traffic accident (RTA) involving public transport drivers.[8] The neurocognitive effects of OSA are extremely important in persons involved in daytime activities such as driving that require extreme concentration, simultaneous visuo-perceptual computations, prompt response time, and decision taking. Thus, it is important to screen the public transport drivers for OSA and related excessive day time sleepiness to reduce the increasing number of RTAs.[9]

The association between RTA and OSA has been studied as early as 1988, by Findley et al.[10] and he found that drivers with OSA had a seven times greater rate of RTA than healthy counterparts. In Indian setting, Parameswaran et al.[11] reported the association between the drivers and the OSA. The present study was conducted to evaluate the risk of OSA among transport drivers of south Kerala using modified BQ. Additionally, lateral cephalograms of the high-risk group was evaluated for the assessment of craniofacial characteristics and airway.

SUBJECTS AND METHODS

This community-based cross-sectional questionnaire study was conducted among 180 public transport drivers of south Kerala. Institutional ethical committee clearance was obtained for the study and all the participants were given consent forms to provide their willingness for the study. The study population included male public transport drivers of age between 18 and 56 yrs.

All participants were given the modified version of BQ translated into Malayalam language which included questions about demographics, sleep symptoms like snoring and ESS. Fidelity of questionnaire translated in Malayalam was confirmed by back translation into English.[12] The BQ contains three sets of questions about the symptoms and severity of OSA. The responses to first and second sets of questions were recorded positive if the responses indicated recurrent symptoms (≥3-4 times/week). Whereas a history of high blood pressure or BMI >30 kg/m²; alone or combination gave positive response in the third set of questions. Based on the score obtained from BQ the participants are categorized as snorers and non-snorers and the snorers were again classified as high-risk group and low-risk group as previously described by Netzer et al.[6]

ESS is a self-administered questionnaire with eight queries. Respondents were requested to rate on a 4-point scale (0-3), for every question depending on their probabilities of dozing. A cumulative score of 11 or more indicated an excessive day-time drowsiness.[7]

Limited physical examination was performed in all the participants, which included measurement of neck circumference, waist circumference, hip circumference, blood pressure, and calculation of BMI. Neck circumference is measured at the level of cricothyroid membrane.[13] The BMI was calculated as weight (kg)/height (m²), waist circumference, hip circumference, and WHR were measured according to the WHO protocol 1995.[14] Blood pressure is measured with the help of a digital sphygmomanometer. Subjects were considered hypertensive if they fulfil the Joint National Committee 7 (JNC7) criteria for hypertension.[15]

For evaluation of craniofacial morphologic characterization and airway assessment, digital lateral cephalograms were recorded in a natural head position of the high-risk snorers and analyzed with cephalometric analysis program (Dolphin Imaging Cephalometric and Tracing Software, Chatsworth, CA, USA).

Sample size and statistical analysis

Based on the prevalence rate of 35.7% OSA, reported by Garbarino et al.[16] the minimum sample size was calculated as 173 (95% confidence level, 20% allowable error). Statistical analysis was performed using IBM SPSS 20 (SPSS Inc, Chicago, USA). Descriptive statistics were represented as mean ± SD and percentage. Inter-group comparison was performed using independent sample t-test.

RESULTS

Based on the responses to the questionnaire, subjects were identified as non-snorers (n = 116) and snorers (n = 64) [Table 1]. The descriptive comparison of demographics and the risk factors among snorers and non-snorers are described in [Table 2].

Table 1.

Frequency distribution of snoring status

Variable	Frequency (n)	Percentage (%)
Non-snorers	116	64.4
Snorers	64	35.6
Total	180	100

Table 2.

Comparison of descriptive statistics among snorers and non-snorers

Variable	n	Mean	SD	P
Weight (kg)
Snorer	64	75.30	13.05	<0.001*
Non-snorer	116	66.41	10.90
Height (m)
Snorer	64	1.81	0.15	0.683
Non-snorer	116	1.82	0.19
BMI
Snorer	64	26.02	4.52	0.003*
Non - snorer	116	24.06	3.33
Neck circumference
Snorer	64	40.94	2.46	<0.001*
Non - snorer	116	39.16	2.30
Waist circumference
Snorer	64	100.18	7.90	0.011*
Non-snorer	116	97.15	7.35
Hip circumference
Snorer	64	100.81	8.23	0.009*
Non-snorer	116	97.93	6.31
SBP
Snorer	64	136.86	22.25	0.025*
Non-snorer	116	129.74	15.17
DBP
Snorer	64	87.63	13.56	0.304
Non-snorer	116	85.57	11.23
WH ratio
Snorer	64	0.99	0.03	0.631
Non-snorer	116	0.99	0.03

Independent sample t-test is used for comparison of means, *P<0.05 (statistically significant)

The group of snorers (n = 64) were further; categorized as high-risk snorers (n = 30) and low risk snorers (n = 34) [Table 3]. The mean height, systolic BP, and diastolic BP among high-risk snorers were significantly higher than the low-risk snorers with as shown in [Table 4].

Table 3.

Risk assessment among snorers

Risk	Frequency (n)	Percentage (%)
High	30	46.9
Low	34	53.1
Total	64	100

Table 4.

Comparison between high- and low-risk snorers

Parameters	Risk	n	Mean	SD	P
Weight (kg)	High	30	76.20	11.27	0.607
	Low	34	74.50	14.56
Height (m)	High	30	1.86	0.16	0.010*
	Low	34	1.76	0.13
BMI	High	30	26.67	3.49	0.286
	Low	34	25.45	5.26
Neck circumference	High	30	41.17	2.49	0.490
	Low	34	40.74	2.46
Waist	High	30	100.26	7.06	0.941
	Low	34	100.11	8.68
Hip	High	30	100.20	6.27	0.580
	Low	34	101.35	9.69
SBP	High	30	145.77	23.57	0.002*
	Low	34	129.00	17.93
DBP	High	30	95.63	13.53	<0.001*
	Low	34	80.56	8.94
W/H ratio	High	30	1.00	0.036	0.157
	Low	34	0.98	0.029

Independent sample t-test is used for comparison of means, *P<0.05 (statistically significant)

Craniofacial morphologic characteristics among high-risk snorers (n = 30) were compared with the normal norms as depicted in [Table 5]. There was significant difference in the mean superior airway space (11.07 ± 2.52), mean middle airway space (9.04 ± 1.88), mean inferior airway space (10.78 ± 2.70), soft palate length (36.23 ± 3.85), and the hyoid bone position than the normal (P ≤ 0.05).

Table 5.

Craniofacial characteristics among high-risk snorers

Parameters	High-risk group (n=30)	Normal values	P
SNA (°)	85.18±4.84	82±2	0.001*
SNB (°)	79.85±5.14	80±2	0.874
ANB (°)	5.32±2.80	2±2	<0.001*
GoGn (mm)	75.80±6.79	86±4	<0.001*
GoGn-SN°	29.42±7.46	32±0	0.069
Pog- NB (mm)	1.41±1.30	2±2	0.020*
SPAS (mm)	11.07±2.52	17.4±2.5	<0.001*
MAS (mm)	9.04±1.88	10.9±2.8	<0.001*
IAS (mm)	10.78±2.70	19.7±2.6	<0.001*
PNS- P (mm)	36.23±3.85	38.3±1.9	0.007*
H-Mp (mm)	17.16±5.97	19.75±6.57	0.024*

Independent sample t-test is used for comparison of means, *P<0.05 (statistically significant)

Based on ESS, 3.9% of the study population presented with EDS [Graph 1].

Graph 1.

Epworth Sleepiness Scale

DISCUSSION

The relationship of EDS and increasing number of RTA has been reported previously by Leechawengwongs et al.[17] and Canani et al.[18] A positive association between the EDS and increasing risk of accidents and its association with other risk factors of Sleep Disordered Breathing (SDB) were reported among the public transport drivers of Trivandrum city by Parameswaran et al.[11] Since EDS increases the risk of road traffic accidents and appears as an early symptom of OSA; a protocol needs to be developed to screen public transport drivers for EDS and risk factors associated with OSA.

Based on the study reported by Wahida et al.[19] in Malaysian population, 14.6% of drivers were categorized as having high risk of OSA; while 85.4% having low risk of OSA using BQ. Based on ESS of transport drivers in our study, only 3.9% of the respondents had ESS >10 which is lesser than that found by Sadeghniiat and Labbafinejad et al.[20] in an Iranian study (9.1%) and Canani et al.[18] in Brazilian study (26%). This difference may be attributed to the hesitance of the participants to disclose their sleep status. Yet another, major disadvantage of ESS is subjective variation based on perception of sleepiness by an individual.[18]

Previous studies have reported association of BMI and increased motor vehicle accidents.[21,22] Stoohs et al.[21] found an increased risk of accidents when the BMI ≥32 kg/m², whereas Horstmann et al.[22] found that the drivers with mean BMI of 35.1 kg/m² had higher risk of accidents than the drivers with a mean BMI of 30.9 kg/m². Yamamoto et al.[23] found that the drivers having a mean BMI of 32.4 kg/m² had a chance of road crash than the drivers with mean BMI of 28 kg/m².

Ahbab et al.[24] and Kawaguchi et al.[25] denoted that increased neck circumference was a risk factor and the most important predictor of OSA. Parameswaran et al.[11] noted that the drivers who met with RTA had a positive correlation with higher BMI, high Epworth sleepiness score, snoring, and SDB.

Increased waist and hip circumference and waist-hip ratio is an indicator of obesity; thus, a predisposing factor for OSA.[23] Seidell et al.[26] and Gasa et al.[27] reported the association between increased waist circumference, hip circumference, and waist-hip ratio to the development of OSA similar to the results of this study. Anthropometric measurements like neck circumference, waist circumference, and hip circumference are easily obtainable measurements with the use of only a tape measure, which could be a useful measure in clinical practice to assess obesity and related risk of OSA.

Among the high-risk and low-risk snorers, statistically significant difference was found in the height, systolic and diastolic blood pressure in this study. Stooh et al.[21] also reported a comparable association between hypertension and OSA among drivers.

Previous studies have reported craniofacial characteristics like position of maxilla, mandible, mandibular body length, position of hyoid bone and the evaluation of upper, middle, and inferior airway space to have correlation between craniofacial morphology and OSA.[28,29,30] Lateral cephalograms in patients with high-risk OSA when compared to the cephalometric standards reported by McNamara et al.[31] showed considerable narrowing in the upper, middle, and inferior airway space. Lavanya et al.[32] and Szymańska et al.[33] also reported airway constriction in high-risk OSA patients.

The position of maxilla was found to be slightly increased compared to the standard values previously reported by Steiner et al.[34] The mandibular position was normal in accordance with Steiner cephalometric analysis. The maxillomandibular relationship is increased in the high-risk group in this study with mean ANB angle of 5.32 ± 2.80°. These findings were in accordance with the results of Baik et al.[29] and Laxmi et al.[30]

The mean mandibular body length (GoGn) was found to be 75 ± 6.79 mm in this group which is significantly reduced than the normal values. The results obtained expressed a similar pattern to the findings of Baik et al.[29] and Gungor et al.[35] This could be a plausible reason for the constricted airway. The hyoid bone was positioned inferiorly in the high-risk OSA group with a mean value of 17.6 ± 5.97. The inferior positioning of hyoid bone and its relation to the narrow airway was also reported previously by Gungor et al.[35]

CONCLUSION AND RECOMMENDATION

Recent increase in number of road traffic accidents demands screening of transport drivers for sleep deprivation and associated risk of OSA. Further studies are planned in larger populations of transport drivers to screen the unreported cases of OSA for awareness and timely management.

The anthropometric measurements along with the screening questionnaires could be included in the screening protocol for driving license applicants in India to assess the potential risk of OSA. Along with the medical examination, a screening for OSA should also be recommended to aid in identification and timely management. An early screening, awareness, and treatment is expected to improve the occupational quality of life in transport drivers and thereby, reduce the incidence of road traffic accidents.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.