9 October 2011

Dear Editor,

We read with interest the article by Marshall et al. in the August issue of this journal entitled ‘Snoring is not associated with adverse effects on blood pressure, arterial structure or function in 8-year-old children: The Childhood Asthma Prevention Study (CAPS)’.1 This community-based study found no association between parentally reported snoring and adverse measures of metabolic markers or vascular structure/function when the study participants were 8 years.

Several points about this study should be highlighted. Snoring is the hallmark of sleep-disordered breathing (SDB) which ranges in severity from primary snoring (snoring with no gas exchange abnormalities or evidence of sleep disruption) to obstructive sleep apnoea (OSA) which manifests with repeated breathing pauses which result in reduced oxygenation and sleep disruption. The gold standard for determining the severity of SDB is polysomnography (PSG). The distinction between parental report of snoring, which is notoriously inaccurate at determining the existence or severity of SDB,2 and SDB, which has been objectively diagnosed on PSG, must be emphasized. As this study was not designed to investigate sleep, a formal PSG was not performed.

The definition of snoring also deserves comment. The authors report that at this age, 41% of caregivers reported that their child snored. This is much higher than previous reports, probably owing to the fact that any snoring was included in the analysis, as well as snoring frequency as an ordinal variable, rather than limiting analysis to habitual snoring (more than 3 days a week)3 as in most publications. This analysis will have weakened the likelihood of finding an association with snoring, as occasional and infrequent snorers are unlikely to have cardiovascular effects especially of the major nature studied in this paper. Roughly 10% of habitual snorers would be expected to have OSA, meaning that with 76 habitual snorers in this study, eight might have had OSA. Thus, the study is substantially underpowered to detect the given outcome. The authors also imply that snoring in children only manifests as SDB later in adulthood. This is simply incorrect, as children exhibit the broad range of SDB severities with 2–4% exhibiting OSA.4 In addition, the authors have not reported on the association between asthma and snoring, history of allergic rhinitis, previous or current treatment for these conditions (such as adenotonsillectomy or nasal steroids), or the existence of objectively diagnosed SDB in study participants, all of which may have impacted on study results.

The association between severe OSA and cardiovascular risk factors such as hypertension was raised by the authors. However, contrary to their conclusions, several recent papers suggest that adverse cardiometabolic effects linked to OSA are not limited to the severe cases only, with reports by ourselves from clinical populations and others from similar community-based studies that primary snoring, confirmed by PSG, is associated with elevated blood pressure in children of a similar age to those in this study.5–9

We agree with the suggestion by the authors that parental report of snoring alone does not provide sufficient evidence to initiate treatment. However, it should be an indication for the physician to investigate for SDB, which if untreated, is associated with adverse neuro-behavioural outcomes as well as effects on the cardiovascular system.5–9 It is important that readers are not falsely reassured by this study ‘that parentally reported primary snoring is probably below a detectable threshold of harm to children.’ Until longitudinal data on cardiovascular risk factors and measures of arterial structure and function are assessed in objectively diagnosed SDB, stratified according to severity and body mass index, the question regarding early detection and prevention of cardiovascular risk factors in children with SDB, and whether they are precursors of adult disease cannot be answered.


  1. Top of page
  2. References