Background Patients with treated obstructive sleep apnoea have a greater chance of treatment success compared to those untreated, when undergoing ablation for atrial fibrillation (AF) (Patel et al 2010). The STOP-BANG questionnaire has previously been validated as a screening tool for detecting sleep apnoea in surgical patients (Chung et al 2008), but has not been validated for use in AF patients. We present the results from a study investigating its predictive value in AF patients.

Methods Patients with paroxysmal AF undergoing ablation were approached. Those with previously diagnosed sleep apnoea were excluded. Participants completed the STOP-BANG questionnaire and underwent overnight oximetry. Of 228 patients approached, 101 participated. Sleep apnoea was defined as a 4% desaturation index of >5 per hour, and participants were reviewed by an experienced sleep physician to determine if treatment with continuous positive airway pressure (CPAP) was necessary.

Results Of the 101 patients screened, 36 had sleep apnoea, and 13 were offered treatment with CPAP. The STOP-BANG questionnaire (with a cut off of 3 or more out of 8 questions answered positively considered ‘high risk’) had a sensitivity of 97.2% and a specificity of 43.1% for detecting sleep apnoea in this group. The STOP questions alone (with a cut off of 2 or more out of 4 considered ‘high risk’) had a sensitivity of 75.0% and a specificity of 63.1%, however the sensitivity was 100% and specificity 56.8% for predicting the need for CPAP.

Conclusion There is a high prevalence of undiagnosed sleep apnoea in this patient group. The STOP-BANG questionnaire has a high sensitivity for detecting sleep apnoea in AF patients. The STOP questions alone have a high sensitivity for detecting sleep apnoea requiring treatment.

References

1. Patel D, Mohanty P, Di Biase L, et al. Circ Arrhythm Electrophysiol. 2010 Oct;3(5):445-51

2. Chung F, Yegneswaran B, Liao P, et al. Anesthesiology. 2008 May;108(5):812-21