4 Efficacy

Guidance

This section describes efficacy outcomes from the published literature that the committee considered as part of the evidence about this procedure. For more detailed information on the evidence, see NICE's interventional procedure overview.

In a systematic review and meta-analysis of 200 patients, there was a statistically significant decrease in the apnoea–hypopnoea index (AHI; a normal AHI is less than 5 events per hour). At 3-, 6-, and 12‑month follow-up the mean differences from baseline were -23.94 (95% confidence interval [CI] -31.45 to -16.43, 34 patients), -25.60 (95% CI -31.18 to -20.01, 60 patients) and -17.51 (95% CI -20.69 to -14.34, 170 patients) respectively (p<0.001 for all time points).

4.1 In a randomised controlled therapy-withdrawal trial of 46 'responders' from a prospective case series of 126 patients (23 therapy-maintenance responders compared with 23 therapy-withdrawal responders), there was a statistically significant increase in the mean AHI from 7.6 at 1‑year follow-up (before randomisation into the trial) to 25.8 at 1 week after randomisation, in the group in which the device was turned off for 1 week (p<0.001). There was no statistical difference in mean AHI within the therapy-maintenance group, who continued to use the device (7.2 compared with 8.9). At 18‑month follow-up, the mean AHI scores were 9.6 in the therapy-maintenance group and 10.7 in the group who had the device turned off for 1 week (p<0.05 for the differences compared with baseline within groups). There was a statistically significant difference between the therapy-withdrawal group and the therapy-maintenance group for change in mean AHI, from assessment at 1 year to assessment at the end of the therapy-withdrawal study (p<0.001).

4.2 In the systematic review and meta-analysis of 200 patients, there was a statistically significant decrease in the oxygen desaturation index (defined as the number of times per hour of sleep that the blood oxygen level drops by 4 or more percentage points from baseline). At 3-, 6-, and 12‑month follow-up the mean differences from baseline were -10.04 (CI -16.31 to -3.78, 34 patients), -11.68 (95% CI -17.16 to -6.19, 60 patients) and -13.73 (95% CI -16.87 to -10.58, 170 patients) respectively (p<0.01 at 3 months and p<0.001 at 6 and 12 months).

4.3 In the systematic review and meta-analysis of 200 patients, there was a statistically significant decrease in the Epworth sleepiness scale (scores range from 0 to 24 with higher scores indicating more daytime sleepiness). At 3-, 6-, and 12‑month follow-up the mean differences from baseline were -4.17 (CI -6.45 to -1.90, 34 patients), -3.82 (95% CI -5.37 to -2.27, 60 patients) and -4.42 (95% CI -5.39 to -3.44, 170 patients) respectively (p<0.001 for all time points).

4.4 In a follow-up study of 95 patients from the prospective case series of 126 patients, there was a statistically significant increase in the mean functional outcomes of sleep questionnaire score (FOSQ, ranging from 5 to 20 with higher scores indicating better subjective sleep quality) from 14.6±3.0 at baseline to 17.5±2.9 at 4‑year follow-up (p<0.05).

4.5 In the follow-up study of 95 patients from the prospective case series of 126 patients, the rates of bed-partner reported 'no snoring' or 'soft snoring' were 17% (18/108) at baseline and 85% at 4‑year follow-up.

4.6 In a prospective case series of 46 patients, there was a statistically significant improvement in the mean sleep apnoea quality of life index from 4.3±1.0 at baseline to 4.7±1.2 at 6‑month follow-up (p=0.019).

4.7 The specialist advisers listed the key efficacy outcomes as: reduction in severity of obstructive sleep apnoea, improved sleep and reduced daytime sleepiness.