Elsevier

Sleep Medicine

Volume 5, Issue 4, July 2004, Pages 339-343
Sleep Medicine

Original article
Excessive daytime sleepiness in obstructive sleep apnea: prevalence, severity, and predictors

https://doi.org/10.1016/j.sleep.2004.01.021Get rights and content

Abstract

Objectives: To assess prevalence, severity, and predictive factors of excessive daytime sleepiness (EDS) in obstructive sleep apnea (OSA) in an Asian population.

Methods: A retrospective, cross-sectional study of data from patients diagnosed with OSA over a period of three years and having had overnight polysomnography (PSG) followed by daytime multiple sleep latency test (MSLT). Respiratory disturbance index (RDI) was used for diagnosis and assessment of severity. OSA was classified as mild (RDI 5–20), moderate (RDI 20–40), and severe (RDI>40). EDS was objectively assessed using MSLT. According to MSLT, patients were categorized into two groups; EDS (mean sleep latency:MSL<10) and no EDS (MSL>10). PSG, MSLT and demographic data were subjected to univariate and multivariate analyses to ascertain predictive factors of EDS.

Results: There were 195 patients comprising 89.4% males and 10.6% females. The severity of OSA was mild in 35.9%, moderate in 27.2%, and severe in 36.9%. EDS was demonstrated in 87.2%. Sleep onset REM periods were detected in the MSLT of 28.2% patients. Univariate analysis demonstrated age, RDI, sleep efficiency, total arousals, arousals with apnea, arousal index, number of desaturations, and severity of snoring as significant predictors of EDS. However, stepwise logistic regression analysis identified only sleep efficiency, total arousals, and severity of snoring as significant predictive factors.

Conclusions: OSA causes EDS in the majority of patients. Severe snoring, higher sleep efficiency and increased total arousals in polysomnography seem to predict EDS.

Introduction

Excessive daytime sleepiness (EDS) is a well recognized consequence of obstructive sleep apnea (OSA), increasingly considered an important health problem leading to accidents, psychosocial morbidity and poor quality of life [1], [2], [3], [4]. However, the subjective evaluation of EDS is complicated by the fact that patients may complain of fatigue, tiredness, and lack of energy rather than sleepiness itself [5]. Therefore, objective assessment of prevalence, severity, and predictors of EDS in OSA would be useful in understanding the magnitude of the problem and identifying high risk groups.

Several studies have looked into the predictive factors of EDS in OSA patients from predominantly Caucasian populations [6], [7], [8], [9], [10]. A study from Singapore has indicated prevalence of snoring and OSA syndrome to be around 77 and 15%, respectively, in that country [11]. In this context, we were interested in determining prevalence, severity, and predictors of EDS in the local OSA population of Singapore.

Section snippets

Sample and acquisition of data

A total of 195 consecutive patients diagnosed with OSA at the Sleep Disorders Unit of Singapore General Hospital between January 2000 and December 2002, having gone through both overnight polysomnography (PSG) and multiple sleep latency test (MSLT) were included in the study. All patients were seen by a single clinician (KP), who routinely performed MSLT and PSG in every case. The vast majority of cases were referred because of snoring, and a few for the complaint of daytime sleepiness.

Demographic characteristics

The study population consisted of 89.4% males and 10.6% females, with a mean age distribution of 45.5±11.1 years, median of 45 years and range of 20–74 years.

Polysomnographic characteristics

Mild OSA was determined in 35.9% of cases, with 27.2 and 36.9% falling into the moderate and severe OSA groups, respectively. Descriptive statistics for both the EDS and no-EDS groups are presented in Table 1. The majority of snorers (86.8%) belonged to the EDS groups (Table 2).

MSLT characteristics

EDS proved to be highly prevalent (87.2%) among OSA patients,

Discussion

The number of studies published on possible predictors of EDS in OSA underscores the clinical importance of this subject [6], [7], [8], [9], [10]. However, the studies are not absolutely homogeneous. There are some differences evident in the study design, statistical methodology, and tools used to evaluate EDS among different studies. Yet they provide a useful insight into this phenomenon.

Various tools such as Epworth Sleepiness Scale (ESS) and MSLT have been used to evaluate and quantify

References (22)

  • K Puvanendran et al.

    From snoring to sleep apnea in a Singapore population

    Sleep Res Online

    (1999)
  • Cited by (136)

    • Elevated body mass index increases the risk of cardiovascular events in hypertensive patients accompanied with obstructive sleep apnea: A cohort study

      2022, Obesity Research and Clinical Practice
      Citation Excerpt :

      Fourth, it is reported that excessive daytime sleepiness (EDS) is the predictor of cardiovascular events [56], whereas we failed to assess EDS in the present study, possibly bringing unknown bias to the current observation. However, EDS, as a major symptom, is reported to be 15–87% in OSA and increases with OSA severity [57–59]. We have adjusted for OSA severity, which may have, in part, attenuated the potential bias from unmeasured EDS to the results.

    View all citing articles on Scopus

    This data was presented and published in abstract form at the 7th World Congress on Sleep Apnea, Helsinki, Finland in July 2003.

    View full text