Articles
The NoSAS score for screening of sleep-disordered breathing: a derivation and validation study

https://doi.org/10.1016/S2213-2600(16)30075-3Get rights and content

Summary

Background

Diagnosis of sleep-disordered breathing requires overnight recordings, such as polygraphy or polysomnography. Considering the cost and low availability of these procedures, preselection of patients at high risk is recommended. We aimed to develop a screening tool allowing identification of individuals at risk of sleep-disordered breathing.

Methods

We used the participants from the population-based HypnoLaus cohort in Lausanne, Switzerland, who had a clinical assessment and polysomnography at home, to build a clinical score (the NoSAS score) using multiple factor analysis and logistic regression to identify people likely to have clinically significant sleep-disordered breathing. The NoSAS score was externally validated in an independent sleep cohort (EPISONO). We compared its performance to existing screening scores (STOP-Bang and Berlin scores).

Findings

We used the 2121 participants from the HypnoLaus cohort who were assessed between Sept 1, 2009, and June 30, 2013. The NoSAS score, which ranges from 0 to 17, allocates 4 points for having a neck circumference of more than 40 cm, 3 points for having a body-mass index of 25 kg/m2 to less than 30 kg/m2 or 5 points for having a body-mass index of 30 kg/m2 or more, 2 points for snoring, 4 points for being older than 55 years of age, and 2 points for being male. Using a threshold of 8 points or more, the NoSAS score identified individuals at risk of clinically significant sleep-disordered breathing, with an area under the curve (AUC) of 0·74 (95% CI 0·72–0·76). It showed an even higher performance in the EPISONO cohort, with an AUC of 0·81 (0·77–0·85). The NoSAS score performed significantly better than did the STOP-Bang (AUC 0·67 [95% CI 0·65–0·69]; p<0·0001) and Berlin (0·63 [0·61–0·66]; p<0·0001) scores.

Interpretation

The NoSAS score is a simple, efficient, and easy to implement score enabling identification of individuals at risk of sleep-disordered breathing. Because of its high discrimination power, the NoSAS score can help clinicians to decide which patients to further investigate with a nocturnal recording.

Funding

Faculty of Biology and Medicine of the University of Lausanne, Lausanne University Hospital, Swiss National Science Foundation, Leenaards Foundation, GlaxoSmithKline, and Vaud Pulmonary League.

Introduction

Sleep-disordered breathing is a highly prevalent disease, characterised by repetitive collapse of the upper airway during sleep. Moderate-to-severe sleep-disordered breathing has been shown to affect half of men and almost a quarter of women in the general middle-to-old-age population (40–85 years old).1 The disorder affects quality of life and different aspects of health domains;2, 3 is associated with an increased incidence of cardiovascular events,4, 5 hypertension,6 diabetes,7 metabolic syndrome,8 and car accidents;9 and represents an important public health issue. Diagnosis requires overnight recordings with portable limited-channel recorders (respiratory polygraphy) or full polysomnography (PSG), either at home or in a sleep laboratory. These investigations are time-consuming and expensive and cannot be used as routine screening.

Different clinical scores, such as the Berlin10 or STOP-Bang11 scores, have therefore been previously proposed as screening tools. The Berlin Questionnaire10 includes information about snoring, daytime sleepiness or fatigue, obesity, and hypertension. It was created with use of a general clinical practice sample of 744 individuals, of whom 100 (13%) had a polygraphic recording at home to confirm the diagnosis. The STOP-Bang score11 combines information from a self-administered questionnaire about complaints of snoring, tiredness, observed apnoea, and high blood pressure with clinical characteristics such as body-mass index (BMI), age, neck circumference, and sex. It was created based on a large group of patients assessed before a surgical procedure (2477 patients), of whom 211 (9%) had a full PSG to confirm the diagnosis.

These instruments are used in current clinical practice, although they were developed with use of less sensitive recording technology (the Berlin score was developed with thermistors) or older respiratory event scoring criteria (STOP-Bang) than those used nowadays, which therefore do not match current standards. Considering that these technical differences have been shown to have a substantial impact on sleep-disordered breathing diagnosis and perceived prevalence,1 we aimed to develop a new screening tool for sleep-disordered breathing using current recording standards as a reference and external validation to ensure its reliability in different populations.

Research in context

Evidence before this study

Sleep-disordered breathing is a highly prevalent condition associated with neurocognitive deficits and increased cardiovascular risk. Diagnostic procedures, such as respiratory polygraphy or polysomnography, are time-consuming and expensive, so they cannot be used as routine screening for sleep-disordered breathing. We searched PubMed with the terms “sleep apnea”, “sleep apnoea”, and “screening score” up to Nov 30, 2015 (in English). We found that the most commonly used scores for sleep apnoea screening in clinical settings were the Berlin and the STOP-Bang scores. However, clinicians often complain about the complexity of the Berlin score and the high positive rate and lack of discrimination of the STOP BANG score. We thus aimed to develop a new screening score for sleep apnoea based on HypnoLaus, a large population-based sleep cohort.

Added value of this study

We propose a new simple clinical score (the NoSAS score) to screen for sleep-disordered breathing on the basis of clinical and polysomnographic data from a large population-based cohort. Using variables easily available in primary care practice, the NoSAS score allows identification of individuals at risk and ruling out of clinically significant sleep-disordered breathing, with an NPV of 90% and 98% in two ethnically different population-based cohorts.

Implications of all the available evidence

Compared with existing screening scores, the NoSAS score helps clinicians decide with a higher accuracy than at present which individuals should be referred for further testing, thereby reducing the number of missed sleep-disordered breathing diagnoses and unnecessary nocturnal recordings.

Section snippets

Study design and participants

The HypnoLaus Sleep Cohort study1 included a random subset of the population-based CoLaus/PsyCoLaus cohort12, 13 who had a full PSG at home in Lausanne, Switzerland. Sleep-related complaints and habits were investigated with use of the Pittsburgh Sleep Quality Index14 and Epworth Sleepiness Scale,15 and the probability of sleep-disordered breathing was investigated using the Berlin Questionnaire10 and STOP-Bang11 score. Clinical and demographic data were also collected (appendix). The ethics

Results

Of the 3043 participants contacted from the CoLaus/PsyCoLaus study, 2168 (71%) agreed to have a PSG at home in the HypnoLaus study, assessed between Sept 1, 2009, and June 30, 2013. Among the 2168 participants who had a full PSG at home, 2121 (98%) were included in the HypnoLaus study and 47 (2%) were excluded (41 because of a total sleep time of <4 h and six because of technical problems). Compared with the whole CoLaus/PsyCoLaus cohort, individuals who had PSG in the HypnoLaus study were

Discussion

Using a large population-based cohort, we propose a new simple and easy-to-interpret clinical score to screen for clinically significant sleep-disordered breathing. Our results show that—using variables easily available in primary care practice—the NoSAS score allows clinically significant sleep-disordered breathing to be reliably ruled out, with an NPV of 90% and 98% in two ethnically different population-based cohorts. Compared with existing screening scores, the NoSAS score allows a

References (27)

  • PE Peppard et al.

    Prospective study of the association between sleep-disordered breathing and hypertension

    N Engl J Med

    (2000)
  • T Kendzerska et al.

    Obstructive sleep apnea and incident diabetes. A historical cohort study

    Am J Respir Crit Care Med

    (2014)
  • SR Coughlin et al.

    Obstructive sleep apnoea is independently associated with an increased prevalence of metabolic syndrome

    Eur Heart J

    (2004)
  • Cited by (202)

    View all citing articles on Scopus

    Contributed equally

    View full text