Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea.

Cochrane Database of Systematic Reviews

Effects of opioid, hypnotic and sedating medications on sleepdisordered breathing in adults with obstructive sleep apnoea (Review) Mason M, Cates CJ, Smith I

Mason M, Cates CJ, Smith I. Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea. Cochrane Database of Systematic Reviews 2015, Issue 7. Art. No.: CD011090. DOI: 10.1002/14651858.CD011090.pub2.

www.cochranelibrary.com

Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Remifentanil vs placebo, Outcome 1 AHI (events/h). . . . . . Analysis 1.2. Comparison 1 Remifentanil vs placebo, Outcome 2 Obstructive apnoea (events/h). Analysis 1.3. Comparison 1 Remifentanil vs placebo, Outcome 3 Central apnoea (events/h). . . Analysis 1.4. Comparison 1 Remifentanil vs placebo, Outcome 4 Arousal Index. . . . . . . Analysis 1.5. Comparison 1 Remifentanil vs placebo, Outcome 5 Adverse events. . . . . . . Analysis 1.6. Comparison 1 Remifentanil vs placebo, Outcome 6 Minimum SpO2 (%). . . . Analysis 2.1. Comparison 2 Eszopiclone vs placebo, Outcome 1 AHI (events/h). . . . . . . Analysis 2.2. Comparison 2 Eszopiclone vs placebo, Outcome 2 Apnoea index (events/h). . . . Analysis 2.3. Comparison 2 Eszopiclone vs placebo, Outcome 3 Hypopnoea index (events/h). . Analysis 2.4. Comparison 2 Eszopiclone vs placebo, Outcome 4 Arousal index. . . . . . . Analysis 2.5. Comparison 2 Eszopiclone vs placebo, Outcome 5 Adverse events. . . . . . . Analysis 3.1. Comparison 3 Nitrazepam vs placebo, Outcome 1 Apnoea index (events/h). . . . Analysis 3.2. Comparison 3 Nitrazepam vs placebo, Outcome 2 Minimum SpO2 (%). . . . . Analysis 4.1. Comparison 4 Temazepam vs placebo, Outcome 1 AHI (events/h). . . . . . . Analysis 4.2. Comparison 4 Temazepam vs placebo, Outcome 2 Minimum SpO2 (%). . . . . Analysis 4.3. Comparison 4 Temazepam vs placebo, Outcome 3 Arousal Index. . . . . . . Analysis 4.4. Comparison 4 Temazepam vs placebo, Outcome 4 RDI (events/h). . . . . . . Analysis 4.5. Comparison 4 Temazepam vs placebo, Outcome 5 Adverse events. . . . . . . Analysis 5.1. Comparison 5 Triazolam vs placebo, Outcome 1 Minimum SpO2 (%). . . . . Analysis 5.2. Comparison 5 Triazolam vs placebo, Outcome 2 Adverse events. . . . . . . . Analysis 6.1. Comparison 6 Sodium oxybate vs placebo, Outcome 1 AHI (events/h). . . . . Analysis 6.2. Comparison 6 Sodium oxybate vs placebo, Outcome 2 Mean SpO2 (%). . . . . Analysis 6.3. Comparison 6 Sodium oxybate vs placebo, Outcome 3 Minimum SpO2 (%). . . Analysis 6.4. Comparison 6 Sodium oxybate vs placebo, Outcome 4 Arousal index. . . . . . Analysis 6.5. Comparison 6 Sodium oxybate vs placebo, Outcome 5 ESS. . . . . . . . . Analysis 6.6. Comparison 6 Sodium oxybate vs placebo, Outcome 6 Adverse events. . . . . . Analysis 7.1. Comparison 7 Ramelteon vs placebo, Outcome 1 AHI (events/h). . . . . . . Analysis 7.2. Comparison 7 Ramelteon vs placebo, Outcome 2 Mean SpO2 (%). . . . . . . Analysis 7.3. Comparison 7 Ramelteon vs placebo, Outcome 3 ESS. . . . . . . . . . . ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 2 4 5 7 7 10 11 14 17 19 19 20 23 40 42 43 43 44 44 45 45 46 46 47 47 48 48 49 49 50 50 51 51 52 52 53 53 54 54 55 55 56 56 56 58 64 65 65 65


i

INDEX TERMS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65


ii

[Intervention Review]

Effects of opioid, hypnotic and sedating medications on sleepdisordered breathing in adults with obstructive sleep apnoea Martina Mason1 , Christopher J Cates2 , Ian Smith1 1 Respiratory Support and Sleep Centre, Papworth Hospital, Cambridge, UK. 2 Population Health Research Institute, St George’s, University of London, London, UK

Contact address: Martina Mason, Respiratory [email protected].

Support and Sleep Centre, Papworth Hospital, Cambridge,

UK.

Editorial group: Cochrane Airways Group. Publication status and date: New, published in Issue 7, 2015. Review content assessed as up-to-date: 4 March 2015. Citation: Mason M, Cates CJ, Smith I. Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea. Cochrane Database of Systematic Reviews 2015, Issue 7. Art. No.: CD011090. DOI: 10.1002/14651858.CD011090.pub2. Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT Background Obstructive sleep apnoea (OSA) is a common sleep disorder characterised by partial or complete upper airway occlusion during sleep, leading to intermittent cessation (apnoea) or reduction (hypopnoea) of airflow and dips in arterial oxygen saturation during sleep. Many patients with recognised and unrecognised OSA receive hypnotics, sedatives and opiates/opioids to treat conditions including pain, anxiety and difficulty sleeping. Concerns have been expressed that administration of these drugs to people with co-existing OSA may worsen OSA. Objectives To investigate whether administration of sedative and hypnotic drugs exacerbates the severity of OSA (as measured by the apnoeahypopnoea index (AHI) or the 4% oxygen desaturation index (ODI)) in people with known OSA. Search methods We searched the Cochrane Airways Group Specialised Register (CAGR) of trials. The search was current as of March 2015. Selection criteria Randomised, placebo-controlled trials including adult participants with confirmed OSA, where participants were randomly assigned to use opiates or opioids, sedatives, hypnotics or placebo. We included participants already using continuous positive airway pressure (CPAP) or a mandibular advancement device. Data collection and analysis We used standard methodological procedures as recommended by The Cochrane Collaboration. Main results Fourteen studies examining the effects of 10 drugs and including a total of 293 participants contributed to this review. Trials were small, with only two trials, which used sodium oxybate, recruiting more than 40 participants, and all but three trials were of only one to three nights in duration. Most participants had mild to moderate OSA with a mean AHI of 11 to 25 events/h, and only two trials Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1

recruited patients with severe OSA. Two trials investigating the effects of ramelteon, a treatment option for insomnia, recruited adults over 60 years of age with OSA and concomitant insomnia. The drugs studied in this review included remifentanil (infusion) 0.75 mcg/kg/h, eszopiclone 3 mg, zolpidem 10 and 20 mg, brotizolam 0.25 mg, flurazepam 30 mg, nitrazepam 10 mg to 15 mg, temazepam 10 mg, triazolam 0.25 mg, ramelteon 8 mg and 16 mg and sodium oxybate 4.5 g and 9 g. We were unable to pool most of the data, with the exception of data for eszopiclone and ramelteon. None of the drugs in this review produced a significant increase in AHI or ODI. Two trials have shown a beneficial effect on OSA. One study showed that a single administration of eszopiclone 3 mg significantly decreased AHI compared with placebo (24 ± 4 vs 31 ± 5; P value < 0.05), and a second study of sodium oxybate 4.5 g showed a significant decrease in AHI compared with placebo (mean difference (MD) -7.41, 95% confidence interval (CI) -14.17 to -0.65; N = 48). Only four trials reported outcome data on ODI. No significant increase, in comparison with placebo, was shown with eszopiclone (21 (22 to 37) vs 28.0 (15 to 36); P value = NS), zolpidem (0.81 ± 0.29 vs 1.46 ± 0.53; P value = NS), flurazepam (18.6 ± 19 vs 19.6 ± 15.9; P value = NS) and temazepam (6.53 ± 9.4 vs 6.56 ± 8.3; P value = 0.98). A significant decrease in minimum nocturnal peripheral capillary oxygen saturation (SpO2 ) was observed with zolpidem 20 mg (76.8 vs 85.2; P value = 0.002), flurazepam 30 mg (81.7 vs 85.2; P value = 0.002), remifentanil infusion (MD -7.00, 95% CI -11.95 to 2.05) and triazolam 0.25 mg in both rapid eye movement (REM) and non-REM (NREM) sleep (MD -14.00, 95% CI -21.84 to 6.16; MD -10.20, 95% CI -16.08 to -4.32, respectively. One study investigated the effect of an opiate (remifentanil) on patients with moderate OSA. Remifentanil infusion did not significantly change AHI (MD 10.00, 95% CI -9.83 to 29.83); however it did significantly decrease the number of obstructive apnoeas (MD -9.00, 95% CI -17.40 to -0.60) and significantly increased the number of central apnoeas (MD 16.00, 95% CI -2.21 to 34.21). Similarly, although without significant effect on obstructive apnoeas, central apnoeas were increased in the sodium oxybate 9 g treatment group (MD 7.3 (18); P value = 0.005) in a cross-over trial. Drugs studied in this review were generally well tolerated, apart from adverse events reported in 19 study participants prescribed remifentanil (n = 1), eszopiclone (n = 6), sodium oxybate (n = 9) or ramelteon (n = 3). Authors’ conclusions The findings of this review show that currently no evidence suggests that the pharmacological compounds assessed have a deleterious effect on the severity of OSA as measured by change in AHI or ODI. Significant clinical and statistical decreases in minimum overnight SpO2 were observed with remifentanil, zolpidem 20 mg and triazolam 0.25 mg. Eszopiclone 3 mg and sodium oxybate 4.5 g showed a beneficial effect on the severity of OSA with a reduction in AHI and may merit further assessment as a potential therapeutic option for a subgroup of patients with OSA. Only one trial assessed the effect of an opioid (remifentanil); some studies included CPAP treatment, whilst in a significant number of participants, previous treatment with CPAP was not stated and thus a residual treatment effect of CPAP could not be excluded. Most studies were small and of short duration, with indiscernible methodological quality.Caution is therefore required when such agents are prescribed for patients with OSA, especially outside the severity of the OSA cohorts and the corresponding dose of compounds given in the particular studies. Larger, longer trials involving patients across a broader spectrum of OSA severity are needed to clarify these results.

PLAIN LANGUAGE SUMMARY Effects of opioid, hypnotic and sedating medications on obstructive sleep apnoea (OSA) in adults with known OSA Background Obstructive sleep apnoea (OSA) is a common sleep disorder characterised by intermittent apnoeas (pauses in breathing) leading to dips in oxygen levels in the blood during sleep. Many people with known or unknown (undiagnosed) OSA receive hypnotics, sedatives and opiate/opioid drugs to treat other conditions including pain, anxiety and difficulty sleeping. Opiates/opioids are commonly prescribed to treat pain after major surgery. These drugs might make sleep apnoea worse - increasing the frequency and duration of apnoeas. Review question We set out to look at evidence derived from randomised placebo-controlled trials to identify the risks of these drugs for patients with known OSA. Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

2

What we did We searched and reviewed all randomised placebo-controlled trials involving adult patients with known OSA who received these drugs. Most of the trials were short at just one or two nights’ duration, involved only a small number of patients and used poorly described methodology. Therefore it was difficult to find out the exact effects of these drugs on patients with OSA. Results We found 14 studies examining the effects of 10 drugs in trials including 293 people. Opiates and opioids (treatment of acute and chronic pain) Remifentanil infusion Remifentanil infusion did not significantly change severity of OSA as measured by numbers and duration of pauses in breathing during sleep, but it significantly lowered minimum oxygen levels during the night when compared with placebo. Sedatives and hypnotics ’Z drugs’ (a treatment for insomnia, inability to fall asleep or inability to maintain sleep) Eszopiclone and zolpidem Eszopiclone did not worsen OSA as measured by the numbers and duration of pauses in breathing during sleep. In one study, it reduced the severity of OSA, and it might be beneficial in treating this condition, although further studies are needed to assess this effect. Zolpidem did not significantly worsen OSA as measured by the numbers and duration of pauses in breathing during sleep, but in one trial, it significantly lowered minimum oxygen levels during the night when compared with placebo. Benzodiazepines (short-term relief of severe anxiety, behavioural disturbance or agitation and panic disorders) Brotizolam, flurazepam, nitrazepam, temazepam and triazolam None of the drugs examined had significantly worsened OSA as measured by the numbers and duration of pauses in breathing during sleep. In small, single-night studies, zolpidem 20 mg, flurazepam 20 mg and triazolam 0.25 mg showed a tendency to increase numbers and duration of pauses during sleep, which was not statistically significant, but these drugs significantly lowered minimum oxygen levels during the night when compared with placebo. Sodium oxybate (a treatment for narcolepsy, a condition causing excessive sleepiness during the day) Sodium oxybate was compared with placebo in two trials. Results showed that sodium oxybate did not worsen OSA as measured by the numbers and duration of pauses in breathing during sleep. In one study, sodium oxybate 4.5 g reduced the severity of OSA and might have proved beneficial in treating this condition, although further studies are needed to assess this effect. Melatonin and melatonin-related drugs Ramelteon (a treatment drug for insomnia) Ramelteon was assessed in older adults, over 60 years of age, with OSA and insomnia and was found not to worsen OSA. Results of this review show that the drugs studied do not worsen the severity of OSA as measured by the numbers and duration of pauses in breathing during sleep, but significant clinical and statistical decreases in minimum oxygen levels during the night were observed with remifentanil, zolpidem and triazolam; therefore prescribing these drugs for patients with OSA still warrants caution. Sodium oxybate 4.5 g and eszopiclone reduced the severity of OSA and might be beneficial in treating this condition; nonetheless further studies are needed to support these results. Bottom line The long-term effect and potential side effects of sedative drugs in people with OSA need to be assessed in larger, longer and methodologically robust studies.


3

S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]

Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea Patient or population: participants with sleep apnoea prescribed sedatives, hypnotics (including benzodiazepines (BZDs), non-BZD GABAA receptor agonists (’Z drugs’), sodium oxybate (SXB), melatonin agonists) and opioids Settings: community Intervention: 10 drugs were included in the review: remifentanyl (infusion) 0.75 mcg/kg/h, eszopiclone 3 mg, zolpidem 10 and 20 mg, brotizolam 0.25 mg, flurazepam 30 mg, nitrazepam 10 mg to 15 mg, temazepam 10 mg, triazolam 0.25 mg, ramelteon 8 mg and 16 mg and sodium oxybate 4.5 g and 9 g Comparison: placebo Outcomes

Effects of interventions

Quality of the evidence (GRADE)

Apnoea- hypopnoea index (AHI) • Zolpidem 20 mg: no ⊕⊕

(events/h) significant increase in AHI Low1 (intervention 30.0 vs control 17. 0; P value = 0.12) • Flurazepam 30 mg (21.5 vs 17.0; P value = 0.12) • Triazolam 0.25 mg (84.1 ± 6.5 vs 77.9 ± 6; NS) • Eszopiclone 3 mg: significantly decreased AHI compared with placebo (24 ± 4 vs 31 ± 5; P value 60 events/h) as well as those with a high arousal threshold (-25 cm H2 O to -63 cm H2 O as measured by an epiglottic pressure catheter), and Rosenberg 2007 excluded patients with severe OSA, which was defined as AHI of 40 events or more/h. Two trials recruited participants with co-existing insomnia ( Camacho 1995; Gooneratne 2010). Interventions

A total of 10 different study drugs were compared with placebo. No studies with gabapentin, pregabalin or melatonin fulfilling the inclusion criteria were identified. Opiates and opioids • Remifentanil (opioid-based analgesic) (Bernards 2009). Sedatives and hypnotics • Eszopiclone: non-BZD GABAA receptor agonist (Eckert 2011; Rosenberg 2007). • Zolpidem: non-BZD GABAA receptor agonist (Berry 2006; Cirignotta 1988). • Flurazepam: benzodiazepine (Cirignotta 1992). • Brotizolam: benzodiazepine analogue, similar in effect to short-acting benzodiazepines, which are not approved for sale in the UK (Cirignotta 1992). • Temazepam: short-acting benzodiazepine with anxiolytic properties, for treatment of insomnia and used as pre-medication during the preoperative period (Camacho 1995; Wang 2011). • Nitrazepam: long-acting benzodiazepine, used for shortterm treatment of anxiety and insomnia (Höijer 1994). • Triazolam: benzodiazepine (Berry 1995). • Sodium oxybate: central nervous system depressant (George 2010; George 2011). Melatonin and melatonin-related drugs • Ramelteon: melatonin receptor agonist, indicated for treatment of insomnia (Gooneratne 2010; Kryger 2007). All but one drug were administered orally. Remifentanil was administered as intravenous infusion (Bernards 2009). The duration of studies varied: Seven single-night studies were conducted


12

in laboratory settings (Berry 1995; Berry 2006; Cirignotta 1988; Eckert 2011; George 2010; Kryger 2007; Wang 2011), three studies were run over two to three nights (Cirignotta 1992; Höijer 1994; Rosenberg 2007), one study was of two weeks’ duration (George 2011) and two were of one to two months’ duration (Camacho 1995; Gooneratne 2010).

Outcomes

All studies reported the effects of drug therapy on at least some measurements of sleep-disordered breathing (namely, AHI; AI; HI; ODI; duration of apnoea and arousals). In addition, two studies reported outcomes of symptoms on the ESS (George 2011; Gooneratne 2010). Five studies reported TST (George 2010; George 2011; Gooneratne 2010; Höijer 1994; Wang 2011).

Funding

Five studies were supported by drug companies (Cirignotta 1992; George 2011; Gooneratne 2010; Kryger 2007; Rosenberg 2007), four were supported by grants from research councils (Bernards 2009; Berry 2006; Camacho 1995; Höijer 1994) and one study was supported by both (Eckert 2011). Four studies did not clearly state funding (Berry 1995; Cirignotta 1988; George 2010; Wang 2011). Excluded studies See Characteristics of excluded studies.

Risk of bias in included studies We assessed risk of bias using the Cochrane risk of bias tool. Full details can be found in the Characteristics of included studies. For a summary of the risk of bias, see Figure 2.


13

Figure 2. Risk of bias summary: review authors’ judgements about each risk of bias item for each included study.


14

Allocation Only three studies reported adequate randomisation and allocation concealment procedures (Eckert 2011; Gooneratne 2010; Wang 2011); all remaining studies were described as randomised, but no further information was provided on randomisation or concealment of allocation, making adequate judgement impossible. Blinding Three studies described adequate blinding of participants and personnel (Eckert 2011; Gooneratne 2010; Wang 2011). In Bernards 2009, blinding of participants and outcome assessors but not personnel was adequate. Three studies reported adequate blinding of outcome assessors (Bernards 2009; Eckert 2011; Wang 2011). The remaining studies did not report enough details on blinding. Incomplete outcome data All randomly assigned participants completed the intervention in two studies, and the number of dropouts in seven studies did not cause particular concern (Bernards 2009; George 2010; George 2011; Gooneratne 2010; Höijer 1994; Rosenberg 2007; Wang 2011). In five studies, details were insufficient to allow a judgement (Berry 1995; Berry 2006; Camacho 1995; Cirignotta 1988; Cirignotta 1992)

One parallel trial involving 19 adults with moderate OSA (Bernards 2009) compared remifentanil infusion (0.075 mcg/ kg/h) versus placebo. Remifentanil infusion did not significantly change AHI (MD 10.00, 95% CI -9.83 to 29.83, Analysis 1.1); however it significantly decreased the number of obstructive apnoeas per hour (MD -9.00, 95% CI -17.40 to -0.60; Analysis 1.2) and significantly increased the number of central apnoeas (MD 16.00, 95% CI -2.21 to 34.21; Analysis 1.3). The increase in the average number of central apnoeas in the remifentanil group was not uniform across the group and reflected very large increases in a subset of four participants. The central apnoea index for the subset of four participants at baseline was 0.8 ± 0.9 apnoeas per hour, and during the intervention 43 ± 34 apnoeas per hour. In addition, remifentanil infusion significantly decreased minimum SpO2 (MD -7.00, 95% CI -11.95 to -2.05; Analysis 1.6) and increased the total arousal index . No significant adverse events were reported in the remifentanil group (Analysis 1.5), but one participant in this group was given supplemental oxygen as the result of a drop in SpO2 to less than 80% for longer than five minutes. Sedatives and hypnotics

’Z drugs’

Four cross-over studies comparing “Z drugs” with placebo recruited a total of 67 participants. Selective reporting None of the studies appear to have issues with incomplete reporting of outcomes, although we did not have access to study protocols.

Eszopiclone versus placebo

See: Summary of findings for the main comparison Studies reported diverse outcome measures; however AHI as the common outcome measure allowed direct comparison between studies. For an overview of the results, see Table 1.

Two trials with 39 participants (Eckert 2011; Rosenberg 2007) compared eszopiclone 3 mg versus placebo, and pooled data show no significant change in AHI (MD -3.40, 95% CI -10.36 to 3.56; participants = 39; Analysis 2.1), although in Eckert 2011, a single administration of eszopiclone 3 mg significantly decreased AHI compared with placebo (24 ± 4 vs 31 ± 5; P value < 0.05). In addition, researchers in Rosenberg 2007 found no significant differences in AI, HI and arousal indices (Analysis 2.2; Analysis 2.3; Analysis 2.4). No serious adverse events or adverse events led to discontinuation of treatment. Six participants treated with eszopiclone experienced an unpleasant taste, and two developed infection, which was deemed by investigators as not related to study drug.

Opiates and opioids

Zolpidem versus placebo

Other potential sources of bias One cross-over study (George 2010) included no washout period, leading to the potential for differential carry-over effects.

Effects of interventions

Remifentanil versus placebo

Two cross-over trials with 28 participants examined zolpidem versus placebo (Berry 2006; Cirignotta 1988). Short-term administration of zolpidem 10 mg to 16 participants with severe OSA on


15

CPAP for at least six months (Berry 2006) did not significantly change AHI (2.7 ± 0.47 vs 4.8 ±1.4) or minimum SpO2 (91.0 ± 0.7 vs 91.4 ± 0.6 ). A small decrease in the arousal index was observed in the zolpidem group (16.5 ± 2 vs 19 ± 1.6; P value ≤ 0.03). Compared with placebo, a single dose of zolpidem 20 mg to 12 participants (Cirignotta 1988) increased AHI, although not significantly (30.0 vs 17.0; P value = 0.12), and significantly lowered minimum SpO2 (76.8 vs 85.2; P value = 0.002). No adverse events were reported in these two cross-over trials. (Berry 2006; Cirignotta 1988).

A parallel study (Camacho 1995) and a cross-over trial (Wang 2011) with 35 participants compared temazepam versus placebo. Temazepam (15 mg to 30 mg) administered to 15 participants for eight weeks (Camacho 1995) did not significantly change RDI (MD -1.60, 95% CI -5.81 to 2.61). Compared with placebo, temazepam (10 mg) (Wang 2011) did not alter AHI (MD 1.18, 95% CI -1.76 to 4.12), apnoea index, minimum SpO2 nor arousal index, respectively (Analysis 4.1; Analysis 4.2; Analysis 4.2). Researchers reported no adverse events in these trials.

Benzodiazepines

Triazolam versus placebo

Five trials comparing benzodiazepines versus placebo recruited a total of 70 participants.

One cross-over study with 12 participants examined triazolam 0.25 mg versus placebo (Berry 1995). The AHI was slightly but not significantly higher on triazolam nights (84.1 ± 6.5 vs 77.9 ± 6.6 events/h). The minimum SpO2 was significantly lower in the triazolam group in both REM and NREM sleep (MD -14.00, 95% CI -21.84 to -6.16; Analysis 5.1; MD -10.20, 95% CI -16.08 to -4.32, respectively). This paper describes no adverse events.

Brotizolam versus placebo

One cross-over trial with 12 participants compared brotizolam 0.25 mg versus placebo (Cirignotta 1992). Compared with placebo, brotizolam 0.25 mg caused no significant change in RDI (15.9 ± 14.8 vs 18.5 ± 14.4), mean SpO2 (93.9 ± 1.6 vs 93.9 ± 1.4) or minimum SpO2 (88.5 ± 3.0 vs 88.5 ± 2.5). Investigators reported no adverse events.

Flurazepam versus placebo

Two cross-over trials with 24 participants compared flurazepam with placebo (Cirignotta 1988; Cirignotta 1992). In one crossover trial with 12 participants (Cirignotta 1988), flurazepam 30 mg in comparison with placebo led to a non-significant increase in AHI (21.5 vs 17.0; P value = 0.12) and a significant decrease in minimum SpO2 (81.7 vs 85.2; P value = 0.002). In contrast, in Cirignotta 1992, flurazepam 30 mg compared with placebo (n = 12) did not significantly alter RDI (19.6 ± 15.9 vs 18.5 ± 14.4), mean SpO2 (93.7 ± 1.6 vs 93.9 ± 1.4) or minimum SpO2 (88.5 ± 2.5 vs 88.5 ± 2.5). These trials (Cirignotta 1988; Cirignotta 1992) reported no adverse events.

Temazepam versus placebo

Sodium oxybate versus placebo

Two trials with 90 participants compared sodium oxybate versus placebo (George 2010; George 2011). In one cross-over trial involving 42 participants (George 2010), 9 g sodium oxybate did not significantly increase AHI compared with placebo (18.6 ± 11.1 vs 22.5 ± 10.3; P value = 0.06) nor mean SpO2 (94.8 ± 1.4 vs 94.6 ± 2.5; P value = 0.03). However, CSA was increased in the sodium oxybate treatment group (Mean 7.3 (SD 18)). A parallelgroup trial (George 2011) found that compared with placebo (n = 22), administration of 4.5 g sodium oxybate (n = 26) significantly decreased AHI (MD -7.41, 95% CI -14.17 to -0.65; Analysis 6.1). In addition, investigators reported no differences between treatment groups in mean SpO2 , minimum SpO2 or arousal index (Analysis 6.2; Analysis 6.3; Analysis 6.4). Researchers reported no differences between treatment and placebo groups on ESS (MD 0.10, 95% CI -4.90 to 5.10; Analysis 6.5). Adverse events, most commonly headache, were reported in nine of 26 participants and in six of 22 of those receiving sodium oxybate and placebo, respectively.

Nitrazepam versus placebo

One cross-over trial with 11 participants compared nitrazepam 5 mg and 10 mg versus placebo (Höijer 1994). Investigators found no significant change in apnoea index among participants receiving 5 mg and 10 mg, respectively (MD -0.64, 95% CI -3.98 to 2.70; Analysis 3.1; MD -1.00, 95% CI -4.34 to 2.34). In addition, they reported no change in minimum SpO2 between treatment groups receiving 5 mg or 10 mg nitrazepam (Analysis 3.2). No adverse events were reported in this trial.

Melatonin and melatonin-related drugs

Ramelteon versus placebo

Two trials with 47 participants compared ramelteon versus placebo (Gooneratne 2010; Kryger 2007). Pooled data showed no significant change in AHI (MD 0.26, 95% CI -2.05 to 2.57; participants = 47; Analysis 7.1). In a parallel-group trial, ramelteon 8 mg (n =


16

8) compared with placebo (n = 13) did not significantly change AHI (MD 0.50, 95% CI -11.60 to 12.60; Analysis 7.1) or ESS (MD 0.10, 95% CI -4.90 to 5.10; Analysis 7.3) in participants using CPAP (Gooneratne 2010). In one cross-over trial with 26 adults with mild to moderate OSA (Kryger 2007), ramelteon 16 mg compared with placebo did not significantly alter AHI (MD 0.25, 95% CI -2.10 to 2.60) or mean SpO2 (MD 0.50, 95% CI 0.00 to 1.00) among participants not using CPAP. In Gooneratne 2010, four adverse events were reported in the ramelteon treatment group (diarrhoea n = 1, skin ulcer n = 1, sinusitis n = 1, fracture after being hit by bicyclist n = 1) and two in the placebo group (abdominal pain n = 1, nausea n = 1). None of the adverse events were thought to be serious or related to the study drugs. In Kryger 2007, adverse events were reported by three participants in the ramelteon group (headache n = 2, urinary tract infection n = 1). No adverse events were reported with placebo.

DISCUSSION

Summary of main results We identified 14 studies assessing the effects of 10 drugs on sleepdisordered breathing in patients with pre-existing OSA. Most studies were of short duration, involved only a small number of participants and poorly described methodology and outcomes. For some treatments, only a single trial was conducted, and because of the variety of reported outcomes and doses of drugs used, we were able to pool data only for eszopiclone and ramelteon, and only for AHI (Eckert 2011; Gooneratne 2010; Kryger 2007; Rosenberg 2007). For cross-over studies, we were not able to use the data in the meta-analysis, as the exact P values of the confidence interval for paired data were not available. Seventy-five participants were using CPAP therapy before screening but not during the study night, and 37 participants were treated with CPAP during the study night. Therefore, it must be borne in mind that in 112 participants the treatment effect of CPAP could have diminished the effect of the compounds assessed in these studies. Although treatments with CPAP can influence the effects of drugs on OSA, the cohort of participants receiving treatment represents an important group, as some hypnotics have been used temporarily to help with CPAP titration, or during early stages of CPAP initiation. Both studies (Berry 2006; Gooneratne 2010) assessed the effects of hypnotics on the efficacy of CPAP in participants with OSA. Gooneratne 2010 studied the effects of ramelteon on those with concomitant insomnia treated with CPAP. In clinical settings, this patient group may find the prospect of CPAP more disruptive to their sleep; therefore a short prescription of a hypnotic may help with compliance and adjustment to treatment. In six studies (n = 118), it is not clearly described whether study participants received any treatment with CPAP before the study night or in the

past. In two studies (Eckert 2011; Kryger 2007), 43 participants did not receive any CPAP treatment, and in Bernards 2009, 20 participants were off CPAP for longer than seven days; therefore the residual effect of CPAP in these individuals would be expected to be minimal (Kohler 2011). Aside from studies by Gooneratne 2010 of 30 days’ duration and Camacho 1995 of eight to 10 weeks’ duration, most trials lasted a single night only, and therefore long-term therapeutic and symptomatic benefits could not be concluded. Regarding the effects of the drugs on sleep-disordered breathing, 12 trials showed no significant change in AHI or RDI when study drug was compared with placebo. Two drugs-eszopiclone (3 mg) and sodium oxybate (4.5 g)-significantly decreased AHI. Compared with placebo, flurazepam (30 mg) and zolpidem (20 mg) (Cirignotta 1988) caused a significant decrease in both minimum and mean nocturnal SpO2 , and triazolam 0.25 mg in Berry 1995 and remifentanil infusion in Bernards 2009 led to a significant decrease in minimum SpO2 . Trials by Cirignotta 1988 and Berry 1995 were conducted more than 20 years ago and lasted a single night only. Berry 1995 recruited 12 males with severe OSA, and although no baseline AHI was provided in this study, the mean AHI on placebo night was 77.9 ± 6.6 events/h, suggesting that selected participants represented a cohort of patients with very severe OSA, which could influence the final result of this trial. Further limits in studies by Berry 1995 and Cirignotta 1988 include small sample size, short duration and lack of data on power calculation, as well as the potential for risk of bias due to lack of data on allocation of intervention, concealment of allocation and blinding. Larger, methodologically robust studies of longer duration are needed to confirm these results. A single trial comparing an opiate (remifentanil) versus placebo was included. Although this trial did not report an overall significant change in AHI with remifentanil infusion, a significant decrease in obstructive apnoeas was noted. One possible explanation for this finding is that remifentanil reduced the amount of time spent in REM sleep. This effect may be only temporary, as REM suppression is often followed by REM rebound (Endo 1998), which could potentially worsen obstructive events. The single-night design of this trial and the controlled research environment, which does not reflect the usual clinical conditions in which patients receive opiates/opioids, make it impossible for researchers to conclude on the effects of continuous administration of opiates/opioids on sleep apnoea. Remifentanil significantly increased central apnoeas in a subset of participants, who in this trial were older and less obese. This finding may have an implication for clinical practice in that these patients may be less likely identified on preoperative screening for sleep apnoea because of normal or low body mass index (BMI). In addition, remifentanil led to a statistically and clinically significant decrease in minimum SpO2 (MD -7.00, 95% CI -11.95 to -2.05). It must be noted that different OSA phenotypes, such as patients with more pronounced hypoxia, may respond differently to opiates/opioids. It was recently reported that intermittent hypoxia in patients at risk of OSA may


17

enhance sensitivity to opioid analgesia (Doufas 2013), and important work by Brown 2006 showed that total opiate dose to achieve analgesia in children with OSA and recurrent hypoxaemia was half the dose required in children without such a diagnosis. In addition, opioid-related ventilatory depression was reported to be affected by sex in a study by Dahan 1998. Until further studies clarify the effects of opiates/opioids on patients with OSA, caution should be exercised when prescribing opioids to patients with OSA. Sodium oxybate is a standard treatment for patients with narcolepsy, a large proportion of whom may have concurrent OSA (Chokroverty 1986). The effect of acutely administered sodium oxybate (9 g) on sleep-disordered breathing in 42 adults with mild or moderate OSA was assessed in a cross-over trial by George 2010. In this study, treatment did not significantly worsen OSA as measured by change in AHI and mean SpO2 . However, a significant increase in central apnoeas (P value = 0.005) was seen with sodium oxybate. In a parallel-group trial with 48 adults with mild to moderate OSA, short-term use (two weeks) of sodium oxybate (4.5 g) did not have a respiratory depressant effect as measured by AHI and SpO2 (George 2011). Although these results show no significant worsening in AHI and SpO2 with sodium oxybate, given the small sample size, short duration, worsening of CSA in a cross-over trial by George 2010 and potential night-to-night variability in the severity of sleep-disordered breathing, larger studies are warranted to evaluate the long-term effect (over three months) of larger doses of sodium oxybate (6 to 9 g) on sleep-disordered breathing. As a combination of sleep apnoea and insomnia may pose a difficult clinical management challenge, Camacho 1995 and Gooneratne 2010 assessed temazepam and ramelteon, respectively. for treatment of insomnia, in people over 60 years of age with OSA and insomnia. In comparison with placebo, ramelteon (8 mg) administered for four weeks to 21 elderly participants with insomnia and OSA, starting CPAP treatment, did not significantly alter the severity of OSA and improved objective but not subjective sleep onset latency (Gooneratne 2010). Similarly, temazepam (15 mg to 30 mg) administered for eight weeks did not significantly worsen sleep apnoea in a study of 15 older adults with mild OSA (Camacho 1995). Although the sample size was small in both studies, both ramelteon and temazepam were well tolerated and did not adversely affect respiration. Two trials showed a beneficial effect of medication on OSA. Eckert 2011 showed that a single administration of eszopiclone (3 mg) significantly decreased AHI compared with placebo (24 ± 4 vs 31 ± 5; P value ≤ 0.05), and a parallel trial by George 2011 showed that, compared with placebo (n = 22), sodium oxybate (4.5 g) (n = 26) significantly decreased AHI (MD -7.41, 95% CI 14.17 to -0.65). In Eckert 2011, study authors postulated that the increased arousal threshold caused by eszopiclone led to a decrease in AHI, most markedly in a subgroup of participants with a low arousal threshold before the study night. These drugs may be of

therapeutic benefit for a subgroup of patients with OSA. Interpretation of the results in this review is limited by small sample size, short duration and poorly described methodology in some studies. Although none of the studied drugs had a significant impact on OSA as measured by changes in AHI and RDI, studies with remifentanil infusion (Bernards 2009), zolpidem (20 mg), flurazepam (30 mg) (Cirignotta 1988) and triazolam (0.25 mg) (Berry 1995) reported significantly lowered minimum SpO2 . Remifentanil (Bernards 2009) and sodium oxybate (9 g) (George 2010) did not significantly impair sleep-disordered breathing as measured by AHI and SpO2 , but both of these compounds caused a significant increase in central apnoeas in a subset of participants, with possible clinical consequences. Given the small sample size of these trials, it is not possible to identify reliable clinical indicators as to which patients with OSA are at increased risk of developing worsening central apnoeas; larger and longer studies are needed to clarify these results. In addition, only a single trial examined the effect of an opioid (remifentanil); therefore it is not possible to draw conclusions on the effects of opiates/opioids on OSA; further RCTs are needed to confirm these results. Eszopiclone and sodium oxybate may represent a potential therapeutic option for a subset of patients with OSA; further, larger trials may be warranted. In conclusion, our review shows that currently no evidence indicates that the compounds assessed worsen OSA as evidenced by a change in AHI, but a decrease in minimum SpO2 was observed with remifentanil, zolpidem and triazolam. These results must be interpreted with caution, as only a single trial assessed the effects of an opioid (remifentanil), and in several studies, it was not clear how many participants were treated with CPAP in the past, and whether a residual treatment effect of CPAP could have lessened any deleterious impact of the drugs studied. Overall, the low quality of evidence with potential risk of bias due to lack of data on allocation of intervention, concealment of allocation and blinding must be borne in mind when the results of this review are interpreted. Prescribing such agents for patients with OSA, especially outside the doses and severity of the OSA cohorts studied in individual trials, merits caution.

Overall completeness and applicability of evidence Results of this review cannot be extrapolated to patients outside the severity of the OSA cohorts in the individual studies and the corresponding doses of compounds studied. Some drugs have been studied in patients with ongoing treatment with CPAP. Inter-participant variability is apparent in terms of the effects of the drugs administered in some studies. A subset of patients with a certain OSA phenotype may be more susceptible to potential harm and adverse effects associated with such agents. Owing to the small size and short duration of the studies included in this review, as well as the aforementioned individual variability in drug response,


18

we cannot comment on long-term tolerability and likelihood of adverse events.

Quality of the evidence Most included studies were of short duration (one to two nights only) and were small with indiscernible methodological quality. Only one RCT included in this review examined the effects of an opoid, remifentanil. In addition, data on power calculations were lacking in almost half of the studies, and potential risk of bias was recognised with lack of data on allocation of intervention, concealment of allocation and blinding. Although OSA was diagnosed using laboratory PSG, lack of baseline demographics suggests that some studies may have allowed recruitment of patients with central or mixed sleep apnoea.

Potential biases in the review process We did not contact study authors to enquire about further details that would allow us to assess the risk of bias. We believed that this would not have altered our judgement of how the risk of bias might affect the treatment effects observed in these studies. We tried to minimise bias by screening references in duplicate, performing double extraction of data and entering the data into RevMan in pairs, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions.

Agreements and disagreements with other studies or reviews Larger, longer-term studies with uniform reporting of outcomes would offer a clearer picture. To our knowledge, no other systematic review has addressed this question. A systematic review by Ankichetty 2014 examined effects of perioperative anaesthetics and sedatives on medication requirements, pain, respiratory events and hospital stay in patients with OSA. This review included 18 studies, with only one RCT meeting inclusion criteria. No formal baseline sleep study has been conducted to assess the severity of OSA, and none of the included studies assessed the effects of drugs on OSA by change in AHI/ODI from baseline study to the study night. The combination of midazolam and fentanyl led to uvular oedema, intraoperative snoring and impaired airway patency in patients with OSA. A significant decrease in oxygen saturation during the postoperative period with propofol and isoflurane was noted in patients with OSA undergoing upper airway surgery.

AUTHORS’ CONCLUSIONS

Implications for practice Upon examining the current data, we conclude that opiates/opioids, sedatives and hypnotics examined in this review have no deleterious effect on the severity of OSA, although we would emphasise the limitations of the review, in particular possible residual effects of CPAP in some trials and the limited number of trials investigating effects of opioids (remifentanil). Drugs including remifentanil, zolpidem and triazolam may cause a significant decrease in minimum nocturnal SpO2 in patients with pre-existing OSA. Until more studies have examined the effects of opiates/opioids, and until the results of larger, longer and methodologically robust studies become available, caution has to be exercised when these drugs are prescribed for patients with known OSA.

Implications for research Whether drugs with the potential to cause respiratory depression can be safely used in patients with OSA must be determined. Larger, methodologically robust studies of longer duration including patients naive to or without additional CPAP treatment are needed to examine the effects of opioids/opiates, sedatives and hypnotics on patients with pre-existing OSA. Currently only one RCT has examined the effects of an opiate (remifentanil) on patients with OSA. Results of this trial provide some cause for concern regarding prescribing for patients with OSA. Further studies are needed to clarify these results, including the significance of increased central apnoeas, to help with planning of optimal perioperative care for patients with known OSA requiring opioid analgesia after major surgery. The apparently favourable effects of eszopiclone and sodium oxybate on severity of OSA may be due to chance, but further studies examining these agents as potential therapeutic options for patients with OSA could prove valuable.

ACKNOWLEDGEMENTS John White was the editor for this review and gave valued critical input. The review authors would like to thank Katie Chong (KC), who helped with data extraction and screening of references to identify studies meeting inclusion criteria. The background and methods sections of this review are based on a standard template used by the Cochrane Airways Group. This project was supported by the National Institute for Health Research, via Cochrane Infrastructure funding to the Cochrane Airways Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Systematic Reviews Programme, NIHR, NHS or the Department of Health.


19

REFERENCES

References to studies included in this review Bernards 2009 {published data only} Bernards CM, Knowlton SL, Schmidt DF, DePaso WJ, Lee MK, McDonald SB, et al. Respiratory and sleep effects of remifentanil in volunteers with moderate obstructive sleep apnea. Anesthesiology 2009;110:41–9. Berry 1995 {published data only} Berry RB, Kouchi K, Bower J, Prosise G, Light RW. Triazolam in patients with obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine 1995;151: 450–4. Berry 2006 {published data only} Berry RB, Patel PB. Effect of zolpidem on the efficacy of continuous positive airway pressure as treatment for obstructive sleep apnea. Sleep 2006;29:1052–6. Camacho 1995 {published data only} Camacho ME, Morin CM. The effect of temazepam on respiration in elderly insomniacs with mild sleep apnea. Sleep 1995;18:644–5. Cirignotta 1988 {published data only} Cirignotta F, Mondini S, Zucconi M, Gerardi R, Farolfi A, Lugaresi E. Zolpidem-polysomnographic study of the effect of a new hypnotic drug in sleep apnea syndrome. Pharmacology, Biochemistry, and Behavior 1988;29:807–9. Cirignotta 1992 {published data only} Cirignotta F, Mondini S, Gerardi R, Zucconi M. Effect of brotizolam on sleep-disordered breathing in heavy snorers with obstructive apnea. Current Therapeutic Research, Clinical and Experimental 1992;51:360–6. Eckert 2011 {published data only} Eckert DJ, Owens RL, Kehlmann G, Wellman A, Rahangdale S, Yim-Yeh S, et al. Eszopiclone reduces obstructive sleep apnea severity in patients with a low respiratory arousal threshold: a randomized controlled study [Abstract]. American Journal of Respiratory and Critical Care Medicine 2010;181:A5562. Eckert DJ, Owens RL, Kehlmann G, Wellman A, Rahangdale S, Yim-Yeh S, et al. The sedative medication eszopiclone increases sleep duration and reduces sleep apnea severity in patients with a low respiratory arousal threshold [Abstract]. 24th Annual Meeting of the Associated Professional Sleep Societies; 2010 June 5-9; San Antonio. 2010:A160 [0472]. Eckert DJ, Owens RL, Kehlmann GB, Wellman A, Rahangdale S, Yim-Yeh S, et al. Eszopiclone increases the respiratory arousal threshold and lowers the apnoea/ hypopnoea index in obstructive sleep apnoea patients with a low arousal threshold. Clinical Science 2011;120:505–14.

George 2010 {published data only} George C, Feldman N. Effects of sodium oxybate (Xyrem (r)) on sleep-disordered breathing [Abstract]. Sleep Medicine 2006;7:S73 [P366]. George CFP, Feldman N, Inhaber N, Steininger TL, Grzeschik SM, Lai C, et al. A safety trial of sodium oxybate in patients with obstructive sleep apnea: acute effects on sleep-disordered breathing. Sleep Medicine 2010;11:38–42. NCT00086281. Trial effects of oral Xyrem and Zolpidem on sleep-disordered breathing in obstructive sleep apnoea patients. https://clinicaltrials.gov/ct2/show/NCT00086281 (accessed 3 March 2015). George 2011 {published data only} George CFP, Feldman N, Zheng Y, Steininger TL, Grzeschik SM, Lai C, et al. A 2-week, polysomnographic, safety study of sodium oxybate in obstructive sleep apnea syndrome. Sleep and Breathing 2011;15:13–20. Gooneratne 2010 {published data only} Gooneratne NS, Gehrman P, Gurubhagavatula I, AlShehabi E, Marie E, Schwab R. Effectiveness of ramelteon for insomnia symptoms in older adults with obstructive sleep apnea: a randomized placebo-controlled pilot study. Journal of Clinical Sleep Medicine 2010;6:572–80. Höijer 1994 {published data only} Hoijer U, Ejnell H, Elam M, Hedner J. Nitrazepam in patients with sleep apnea: a double-blind placebo controlled randomized evaluation [abstract]. Fifth Annual Meeting of the Australasian Sleep Association. 1992:29. Höijer U, Hedner J, Ejnell H, Grunstein R, Odelberg E, Elam M. Nitrazepam in patients with sleep apnoea: a double-blind placebo-controlled study. European Respiratory Journal 1994;7:2011–5. Kryger 2007 {published data only} Kryger M, Wang-Weigand S, Roth T. Safety of ramelteon in individuals with mild to moderate obstructive sleep apnea. Sleep and Breathing 2007;11:159–64. Sainati S, Tsymbalov S, Demisse S. Phase II safety study of ramelteon (TAK-375) in subjects with mild to moderate obstructive sleep apnoea [Abstract]. American Thoracic Society International Conference; 2005 May 20-25; San Diego, California. 2005:[A29] [Poster: 422]. Sainati S, Tsymbalov S, Demisse S, Roth T. Double blind placebo controlled two way crossover study of ramelteon in subjects with mild to moderate chronic obstructive pulmonary disease (COPD) [Abstract]. American Thoracic Society International Conference; 2005 May 20-25; San Diego. 2005; Vol. 28:A162. Sainati S, Tsymbalov S, Demissie S, Roth T. Doubleblind, single-dose, two-way crossover study of ramelteon in subjects with mild to moderate obstructive sleep apnea [Abstract]. Sleep 2005;28:A163. Rosenberg 2007 {published data only} Rosenberg R, Roach JM, Scharf M, Amato DA. A pilot study evaluating acute use of eszopiclone in patients with


20

mild to moderate obstructive sleep apnea syndrome. Sleep Medicine 2007;8:464–70. Wang 2011 {published data only} Wang D, Marshall NS, Duffin J, Yee BJ, Wong KK, Noori N, et al. Phenotyping interindividual variability in obstructive sleep apnoea response to temazepam using ventilatory chemoreflexes during wakefulness. Journal of Sleep Research 2011;20:526–32.

References to studies excluded from this review Bradshaw 2006 {published data only} Bradshaw DA, Ruff GA, Murphy DP. An oral hypnotic medication does not improve continuous positive airway pressure compliance in men with obstructive sleep apnea. Chest 2006;130:1369–76. Crowe 1992 {published data only} Crowe McCann C, Quera-Salva MA, Boudet J, Barthouil P, Frisk M, Meyer P. Effect of zolpidem on sleep architecture ventilation and blood pressure in heavy snorers [Abstract]. European Respiratory Journal 1992;5:240s. Deflandre 2014 {published data only} Deflandre E, Courtois AC, Degey S, Brichant JF, Hans P, Poirrier R, et al. Influence of alprazolam premedication on the occurrence of postoperative obstructive apnea. Chest. Conference: CHEST World Congress 2014 Annual Meeting Madrid Spain. 2014;145(3 meeting Abstract). Dolly 1982 {published data only} Dolly FR, Block AJ. Effect of flurazepam on sleepdisordered breathing and nocturnal oxygen desaturation in asymptomatic subjects. American Journal of Medicine 1982; 73:239–43. Guo 2003 {published data only} Guo X, Chen W, Hongyu Z, Weimin K, Li A, Li L, et al. Effects of zolpidem on nocturnal breathing and sleep in normal and SAHS subjects [Abstract]. Sleep Medicine 2003; 4:S15. Lofaso 1997 {published data only} Lofaso F, Goldenberg F, Thebault C, Janus C, Harf A. Effect of zopiclone on sleep, night-time ventilation, and daytime vigilance in upper airway resistance syndrome. European Respiratory Journal 1997;10:2573–7. Malish 2009 {published data only} Malish S, Huang P, Dickel M, Gray D, Hungs M, Sassoon CS. Efficacy of eszopiclone and ramelteon during polysomnography for suspected sleep-disordered breathing [Abstract]. Chest 2009;136:23S–f. Noseda 2002 {published data only} Noseda A, Nouvelle M, Lanquart JP, Kempenaers Ch, De Maertelaer V, Linkowski P, et al. High leg motor activity in sleep apnea hypopnea patients: efficacy of clonazepam combined with nasal CPAP on polysomnographic variables. Respiratory Medicine 2002;96:693–9. Park 2013 {published data only} Park JG, Olson EJ, Morgenthaler TI. Impact of zaleplon on continuous airway pressure therapy compliance. Journal of Clinical Sleep Medicine 2013;9:439–44.

Pawlik 2005 {published data only} Pawlik MT, Hansen E, Waldhauser D, Selig C, Kuehnel TS. Clonidine premedication in patients with sleep apnea syndrome: a randomized, double-blind, placebo-controlled study.[Erratum appears in Anesth Analg. 2006 Jul;103(1): 85]. Anesthesia and Analgesia 2005;101:1374–80. Sharf 1994 {published data only} Scharf MB, Brannen DE, Berkowitz DV, McDannold M. Effects of buspirone and alprazolam on snoring and sleep apnea in geriatric patients. Consultant Pharmacist 1994;9: 665–8. Sun 2015 {published data only} Sun H, Palcza J, Rosenberg R, Kryger M, Siringhaus T, Rowe J, et al. Effects of suvorexant, an orexin receptor antagonist, on breathing during sleep inpatients with chronic obstructive pulmonary disease. Respiratory Medicine 2015;109:416–26. Wang 2013a {published data only} Wang D, Somogyi AA, Yee BJ, Wong KK, Kaur J, Wrigley PJ, et al. The effects of a single mild dose of morphine on chemoreflexes and breathing in obstructive sleep apnea. Respiratory Physiology and Neurobiology 2013;185:526–32. Webster 2012 {published data only} Webster L, Iverson M, Smith M, Steffens J. Comparison of morning and evening dosing of extended-release hydromorphone on sleep-disordered breathing in patients with low back pain. Advancing the Science and Practice of Pain Medicine. 28th Annual Meeting of the American Academy of Pain Medicine; 2012 Feb 23-26; Palm Springs. Malden, Mass, 2012:345.

References to studies awaiting assessment Coyle 2005 {published data only} Coyle MA, Mendelson WB, Derchak PA, James SP, Wilson MG. Ventilatory safety of zaleplon during sleep in patients with obstructive sleep apnea on continuous positive airway pressure. Journal of Clinical Sleep Medicine 2005;1:97.

References to ongoing studies Sun 2013 {published data only} Sun H, Palcza J, Siringhaus T, Rowe J, Card DJ, Gipson A, et al. Effects of the orexin receptor antagonist suvorexant on respiration during sleep in patients with compromised respiratory function. Sleep. Conference: 27th Annual Meeting of the Associated Professional Sleep Societies, LLC, SLEEP 2013 Baltimore, MD United States. Conference Start: 20130601 Conference End: 20130605. 2013;36: A222–A223.

Additional references Ankichetty 2014 Ankichetty S, Wong J, Chung F. A systematic review of the effects of sedatives and anesthetics in patients with obstructive sleep apnea. Journal of Anaesthesiology and Clinical Pharmacology 2011;27(4):447–58.


21

Berry 1996 Berry RB, Kouchi KG, Der DE, Dickel MJ, Light RW. Sleep apnea impairs the arousal response to airway occlusion. Chest 1996;109:1490–6. BNF 2013 BNF. British National Formulary (BNF). 6th Edition. London: British Medical Association and the Royal Pharmaceutical Society, 2013. Bonora 1985 Bonora M, St John WM, Bledsoe TA. Differential elevation by protriptyline and depression by diazepam of upper airway respiratory motor activity. American Review of Respiratory Disease 1985;131:41–5. Brown 2006 Brown KA1, Laferrière A, Lakheeram I, Moss IR. Recurrent hypoxemia in children is associated with increased analgesic sensitivity to opiates. Anesthesiology 2006;105(4):665–9. Chokroverty 1986 Chokroverty S. Sleep apnea in narcolepsy. Sleep 1986;9: 250–3. Dahan 1998 Dahan A1, Sarton E, Teppema L, Olievier C. Sex-related differences in the influence of morphine on ventilatory control in humans. Anesthesiology 1998;88(4):903–13. Doufas 2013 Doufas AG1, Tian L, Padrez KA, Suwanprathes P, Cardell JA, Maecker HT, et al. Experimental pain and opioid analgesia in volunteers at high risk for obstructive sleep apnea. PLoS One 2013;8(1):e54807. Drummond 1983 Drummond GB. Comparison of decreases in ventilation caused by enflurane and fentanyl during anaesthesia. British Journal of Anaesthesia 1983;55:825–35. Eckert 2007 Eckert DJ, McEvoy RD, George KE, Thomson KJ, Catcheside PG. Genioglossus reflex inhibition to upperairway negative-pressure stimuli during wakefulness and sleep in healthy males. Journal of Physiology 2007;581: 1193–205. Eckert 2014 Eckert DJ, Younes MK. Arousal from sleep: implications for obstructive sleep apnea pathogenesis and treatment. Journal of Applied Physiology 2014;116(3):302–13. Endo 1998 Endo T, Roth C. Selective REM sleep deprivation in humans: effects on sleep and sleep EEG. American Journal of Physiology 1998;274:R1186–94. Gottlieb 1999 Gottlieb DJ, Whitney CW, Bonekat WH, Iber C, James GD, Lebowitz M, et al. Relation of sleepiness to respiratory disturbance index: the Sleep Heart Health Study. American Journal of Respiratory and Critical Care Medicine 1999;159 (2):502–7.

Gross 2006 Gross JB, Bachenberg KL, Benumof JL, Caplan RA, Connis RT, Cote CJ, et al. Practice guidelines for the perioperative management of patients with obstructive sleep apnea: a report by the American Society of Anesthesiologists Task Force on Perioperative Management of patients with obstructive sleep apnea. Anesthesiology 2006;104:1081–93. Haas 2005 Haas DC, Foster GL, Nieto FJ, Redline S, Resnick HE, Robbins JA, et al. Age-dependent associations between sleep-disordered breathing and hypertension: importance of discriminating between systolic/diastolic hypertension and isolated systolic hypertension in the Sleep Heart Health Study. Circulation 2005;111(5):614–21. Higgins 2011 Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1 [updated March 2011]. The Cochrane Collaboration, 2011. www.cochrane-handbook.org. Jordan 2011 Jordan AS, Eckert DJ, Wellman A, Trinder JA, Malhotra A, White DP. Termination of respiratory events with and without cortical arousal in obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine 2011;184:1183–91. Kohler 2011 Kohler M1, Stoewhas AC, Ayers L, Senn O, Bloch KE, Russi EW, Stradling JR. Effects of continuous positive airway pressure therapy withdrawal in patients with obstructive sleep apnea: a randomized controlled trial. American Journal of Respiratory and Critical Care Medicine 2011 Nov;184(10):1192–9. Kuniyoshi 2008 Kuniyoshi FH, Garcia-Touchard A, Gami AS, RomeroCorral A, van der Walt C, Pusalavidyasagar S, et al. Daynight variation of acute myocardial infarction in obstructive sleep apnea. Journal of the American College of Cardiology 2008;52:343–6. Lalley 2008 Lalley PM. Opioidergic and dopaminergic modulation of respiration. Respiratory Physiology and Neurobiology 2008; 164:160–7. Leino 1999 Leino K, Mildh L, Lertola K, Seppala T, Kirvela O. Time course of changes in breathing pattern in morphine- and oxycodone-induced respiratory depression. Anaesthesia 1999;54:835–40. Leiter 1985 Leiter JC, Knuth SL, Krol RC, Bartlett D Jr. The effect of diazepam on genioglossal muscle activity in normal human subjects. American Review of Respiratory Disease 1985;132: 216–9. Mason 2013 Mason M, Welsh EJ, Smith I. Drug therapy for obstructive sleep apnoea in adults. Cochrane Database


22

of Systematic Reviews 2013, Issue 5. [DOI: 10.1002/ 14651858.CD003002.pub3] Mezzanotte 1996 Mezzanotte WS, Tangel DJ, White DP. Influence of sleep onset on upper-airway muscle activity in apnea patients versus normal controls. American Journal of Respiratory and Critical Care Medicine 1996;153:1880–7. Peppard 2013 Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep-disordered breathing in adults. American Journal of Epidemiology 2013 May;1(177 (9)):1006–14. Punjabi 2009 Punjabi NM, Caffo BS, Goodwin JL, Gottlieb DJ, Newman AB, O’Connor GT, et al. Sleep-disordered breathing and mortality: a prospective cohort study. PLoS Medicine 2009; 6(8):e1000132. [DOI: 10.1371/journal.pmed.1000132] RevMan 2012 [Computer program] The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). Version 5.2. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2012. Robinson 1987 Robinson RW, Zwillich CW, Bixler EO, Cadieux RJ, Kales A, White DP. Effects of oral narcotics on sleep-disordered breathing in healthy adults. Chest 1987;91:197–203. Santiago 1985 Santiago TV, Edelman NH. Opioids and breathing. Journal of Applied Physiology 1985;59:1675–85. Shneerson 2005 Shneerson JM. Sleep Medicine, A Guide to Sleep and Its Disorders. 2nd Edition. New York: Wiley-Blackwell, 2005. Volkow 2011 Volkow ND, McLellan TA, Cotto JH, Karithanom M, Weiss SR. Characteristics of opioid prescriptions in 2009. JAMA 2011;305:1299–301.

Walker 2007 Walker JM, Farney RJ, Rhondeau SM, Boyle KM, Valentine K, Cloward TV, et al. Chronic opioid use is a risk factor for the development of central sleep apnea and ataxic breathing. Journal of Clinical Sleep Medicine 2007;3:455–61. Weil 1975 Weil JV, McCullough RE, Kline JS, Sodal IE. Diminished ventilatory response to hypoxia and hypercapnia after morphine in normal man. New England Journal of Medicine 1975;292:1103–6. Wellman 2004 Wellman A, Jordan AS, Malhotra A, Fogel RB, Katz ES, Schory K, et al. Ventilatory control and airway anatomy in obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine 2004;170:1225–32. Younes 2004 Younes M. Role of arousals in the pathogenesis of obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine 2004;169:623–33. Young 1993 Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. New England Journal of Medicine 1993; 328:1230–5. Young 1997 Young T, Evans L, Finn L, Palta M. Estimation of the clinically diagnosed proportion of sleep apnea syndrome in middle-aged men and women. Sleep 1997;20:705–6. Yue 2010 Yue HJ, Guilleminault C. Opioid medication and sleepdisordered breathing. Medical Clinics of North America 2010;94:435–46. ∗ Indicates the major publication for the study


23

CHARACTERISTICS OF STUDIES

Characteristics of included studies [ordered by study ID] Bernards 2009 Methods

Randomised, double-blind, parallel-group trial. Statistical test: Student’s unpaired t test (normally distributed data) or Mann-Whitney U test (non-normally distributed data)

Participants

19 adults with moderate OSA diagnosed on PSG within previous 12 months (5 males in control group and 6 males in remifentanil group), mean age in remifentanil group 50 ± 12 and placebo group 49 ± 11; BMI in remifentanil group 31 ± 8 and placebo group 36 ± 7; AHI in remifentanil group 24 ± 5 and placebo group 23 ± 6 Exclusion criteria: history of acute or chronic pulmonary disease, coronary artery disease, current pregnancy or breast feeding, long-term opioid or sedative-hypnotic use, CPAP use during sleep within 7 days before the study, change in weight > 10 pounds between baseline and research PSG study, worsening OSA symptoms

Interventions

intravenous remifentanil 0.075 mcg/kg/h vs placebo (normal saline) for 1 night

Outcomes

AHI, sleep characteristics, arousal indices

Notes

Funding source: Anesthesia Patient Safety Foundation, Indianapolis, Indiana, USA

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Described as randomised but no further information given

Allocation concealment (selection bias)

Comment: not described

Unclear risk

Blinding of participants and personnel Low risk (performance bias) All outcomes

Participants and outcome assessors were blinded to intervention given

Blinding of outcome assessment (detection Low risk bias) All outcomes

Outcome assessors were blinded to intervention given.

Incomplete outcome data (attrition bias) All outcomes

Low risk

All but 1 participant completed the study. 1 participant from placebo group withdrew from the study with no reason stated

Selective reporting (reporting bias)

Low risk

Appeared to report the outcomes recorded


24

Berry 1995 Methods

Randomised, double-blind, cross-over trial. Statistical tests used: paired t test, ANOVA

Participants

12 males with severe OSA but no baseline AHI reported, mean age 46.6 ± 14.1 (32 to 72 years), mean weight 84.0 ± 18.1 kg Exclusion criteria: benzodiazepine or alcohol use for at least 2 weeks before the study

Interventions

Triazolam 25 mg vs placebo for 1 night with 1 night washout period

Outcomes

Duration of apnoea, minimum SpO2 , arousal response to airway occlusion

Notes

No funding details reported

Risk of bias Bias






Unclear risk


Blinding of participants and personnel Unclear risk (performance bias) All outcomes


Blinding of outcome assessment (detection Unclear risk bias) All outcomes



Unclear risk



Low risk


Berry 2006 Methods

Randomised, double-blind, cross-over trial. Statistical tests used: paired t test, nonparametric data compared using Wilcoxon rank sign test

Participants

16 adults (14 males) with severe OSA (AHI > 30/h) on CPAP ≥ 6 months. No baseline AHI reported, mean age 49.4 ± 12.4, BMI 36.1 ± 4.8 Exclusion criteria: on other hypnotics, non-compliance with CPAP (not defined in the article), ESS > 12. OSA diagnosed on PSG

Interventions

Zolpidem 10 mg vs placebo for 1 night with 7-day washout period

Outcomes

AHI and other sleep parameters


25

Berry 2006

(Continued)

Notes

Non-industry-sponsored study, supported by Veterans Administration Research Service. Dr Berry has received research support from ResMed, Dymedix, Cypress Bioscience

Risk of bias Bias






Unclear risk







Unclear risk



Low risk


Camacho 1995 Methods

Multi-centre, randomised, double-blind, parallel-group trial. Statistical method: repeated-measure analysis of variance (ANOVA)

Participants

15 elderly participants (9 males) 65 ± 5.8 years old with insomnia and mild OSA (RDI 5-15/h). 7 participants randomly assigned to treatment arm and 8 to placebo. The study was part of the large multi-centre trial comparing pharmacological and behavioural treatments for insomnia Exclusion criteria: insomnia secondary to physical illness and other sleep disorders

Interventions

Temazepam 15-30 mg ± CBT vs placebo ± CBT for 8-10 weeks

Outcomes

RDI, TST and sleep efficiency (SE) measured by PSG in sleep lab at beginning and at end of treatment

Notes

This work was supported in part by mental health grant (no. MH 47020) to C.M

Risk of bias Bias




26

Camacho 1995

(Continued)




Unclear risk







Unclear risk

Unclear


Unclear risk

Unclear

Cirignotta 1988 Methods

Randomised, double-blind, cross-over trial. Statistical method used: ANOVA

Participants

12 adults (11 males) with mild OSA, mean age 49 ± 5, OSA diagnosed on PSG Exclusion criteria not stated

Interventions

Zolpidem 20 mg, flurazepam 30 mg vs placebo for 1 night with 6-day washout period

Outcomes

Sleep parameters, snoring, AHI

Notes

Funding source not stated

Risk of bias Bias




“Randomised” Comment: not described


Unclear risk



“Double blind” Comment: not described




27

Cirignotta 1988

(Continued)


Unclear risk

Unclear


Unclear risk

Unclear

Cirignotta 1992 Methods

Randomised, double-blind, cross-over trial. Statistical method used: Friedman test

Participants

12 males with mild OSAS, mean age 49 (28-62 years), BMI 28.2 (24.3-36), mean baseline AHI not stated, but 4 and 8 participants were stage 0 and stage 1, respectively, of Lugaresi’s classification Exclusion criteria: SpO2 < 70% during apnoea, significant co-morbidity

Interventions

Flurazepam 30 mg and brotizolam 0.25 mg vs placebo for 2 nights with 10-day washout period

Outcomes

Mean SpO2 , minimum SpO2 , RDI, mean sleep parameters evaluated on second night of treatment by PSG

Notes

Sponsored by Boehringer Ingelheim, Italia

Risk of bias Bias






Unclear risk







Unclear risk

Unclear


Low risk



28

Eckert 2011 Methods

Randomised, double-blind, cross-over trial. Statistical test: for normally distributed data - paired Student’s t test, for non-normally distributed data - Mann-Whitney U test

Participants

17 adults (10 males), mean age 45 ± 4 years (19-62), BMI 33 ± 2 kg/m2 , AHI 31 ± 5 events per hour sleep, nadir SaO2 > 70%. OSA was diagnosed on PSG Exclusion criteria: severe OSA defined as SpO2 < 70% or AHI > 60/h, high arousal threshold (-25 cmH2 O to -63 cmH2 O)

Interventions

Eszopiclone 3 mg vs placebo for 1 night.

Outcomes

AHI, total sleep time, sleep-onset latency, mean overnight oxygen saturations, arousal threshold

Notes

Funding source: “This study was supported by the National Institutes of Health (NIH) [grant number P01 HL095491-01 A1] and an unrestricted investigator-initiated research grant from Sepracor Pharmaceuticals. Other support includes the NIH [grant numbers HL73146 R01 HL085188-01A2, R01 HL090897-01A2, K24 HL 093218-01 A1]. D. J.E. is supported by an Overseas Based Biomedical (CJ Martin) Fellowship from the National Health and Medical Research Council of Australia [grant number 510392], and the American Heart Association [grant number 10SDG3510018]. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript”

Risk of bias Bias



Random sequence generation (selection Low risk bias)

“Randomised” “Service in performing the preparation, randomisation and blinding code for the study intervention. D.J.E. is a consultant for Apnex Medical” To ensure that the arousal threshold comprised a random allocation of arousals across the night, arousal selection was performed by allocating each individual arousal a sequential number and using a random number generator to select the 20 arousals to be analysed


“Randomised” “Service in performing the preparation, randomisation and blinding code for the study intervention. D.J.E. is a consultant for Apnex Medical”

Low risk


29

Eckert 2011

(Continued)


“Each patient received placebo or 3 mg of eszopiclone in random order immediately prior to sleep with only a research pharmacist knowing the treatment assignment”


“Each patient received placebo or 3 mg of eszopiclone in random order immediately prior to sleep with only a research pharmacist knowing the treatment assignment”


Low risk

All participants completed study


Low risk


George 2010 Methods

2-centre, randomised, double-blind, 4-way cross-over study. Statistical tests: ANOVA model included fixed effect for treatment sequence, period and treatment, as well as random effects for subject-within-treatment sequence

Participants

42 adults (76.7% males), median age 50 years (24 -77), Median BMI 31.9 (23.8-44.6) , mean AHI 24.7 ± 9.2. 16 participants were using CPAP therapy before screening but not during the study Exclusion criteria: recent history of substance abuse, shift work, major illness including unstable cardiovascular, endocrine, neoplastic, gastrointestinal, haematological, hepatic, immunological, metabolic, neurological, pulmonary and/or renal disease that would place the patient at risk during the trial or would compromise the objectives outlined in the protocol, psychiatric disorders, other problems that in the investigator’s opinion would preclude participation in and completion of this trial or would compromise reliable representation of subjective symptoms

Interventions

1. Sodium oxybate 9 g; 2. Sodium oxybate 9 g and modafinil 200 mg; 3. zolpidem 10 mg; 4. placebo

Outcomes

Change in AHI from baseline, mean SpO2 , polysomnographically measured sleep characteristics

Notes

Study undertaken as post-marketing safety commitment. Funding not clearly stated

Risk of bias Bias



Support for judgement Described as randomised but no further information given


30

George 2010

(Continued)


Unclear risk







Unclear risk

1 section mentioned - 6 participants withdrew from study after receiving sodium oxybate and modafinil, 2 withdrew after receiving sodium oxybate, 2 withdrew after receiving placebo, 1 withdrew after receiving zolpidem. However, another section mentioned that 9 participants withdrew because of adverse events and 1 for protocol violation It was also mentioned that participants receiving any treatment outside the protocol (n = 10) were not included in the final analysis. It is unclear whether this occurred before randomisation


Low risk


George 2011 Methods

2-centre, randomised, double-blind, parallel-group trial. Statistical tests: 2-sample t test. If the normality assumption was not met, Wilcoxon’s rank sum test was used

Participants

48 adults (sodium oxybate: n = 26/21males, placebo: n = 22/20 males); age: sodium oxybate 49.7 (8.0), placebo 49.7 (11.1); BMI: sodium oxybate 33.0 (5.4), placebo 33. 2 (5.0); AHI: sodium oxybate 23.5 (10.0), placebo 25.0 (8.1); ESS: sodium oxybate 12.2 (5.1), placebo 11 (5.8). 16 participants were using CPAP before screening and discontinued use during the study Exclusion criteria: recent history of substance abuse or shift work, experiencing any major illness, including unstable cardiovascular, endocrine, neoplastic, gastrointestinal, haematological, hepatic, immunological, metabolic, neurological, pulmonary and/or renal disease

Interventions

4.5 g/night sodium oxybate vs placebo. 4.5 g per night for 13 days and 9 g for day 14, taken in divided doses at bedtime and 2.5-4 hours after first dose

Outcomes

AHI change from baseline, mean SpO2 , mean changes in sleep architecture variables, TST and ESS from baseline


31

George 2011

(Continued)

Notes

Sponsored by Jazz Pharmaceutical.Grzeschik. Zheng and Inhaber were former employees of Jazz Pharmaceutical. Lai and Steininger are current employees of Jazz Pharmaceutical. All above have own stock options in Jazz Pharmaceutical. George and Feldman had received support from the company for a study, and Fledman is a member of the speakers’ bureau for Jazz Pharmaceutical and Cephalon, Inc

Risk of bias Bias







Unclear risk


Although this was described to be the double-blind phase, treatment arm was receiving medications once a night for 13 nights, which was different from the placebo arm, which had received the medications twice a night for the whole period




Low risk

All participants completed the study. Sodium oxybate: 1 participant withdrew before sodium oxybate because of upper respiratory tract infection, viral gastroenteritis; placebo: 1 participant withdrew before receiving placebo because of nausea


Low risk


Gooneratne 2010 Methods

Randomised, double-blind, parallel-group trial. Statistical method: Study comparisons between treatment arms were performed using generalised linear modelling (PROC GLM) Effect size estimates were derived using the Centre for Evaluation and Monitoring (Durham University) Effect Size Calculator

Participants

21 adults (ramelteon n = 8 (5 males) and placebo n = 13 (8 males)); mean age ramelteon 73.6 (5.6), placebo 70.6 (3.5); BMI ramelteon 24.7 (4.9), placebo 28.7 (5.8); AHI ramelteon 13.9 (7.5), placebo 10.9 (12.9); ESS ramelteon 9.8 (3.8), placebo 10.4 (7.0) Inclusion criteria: insomnia, > 60 years Exclusion criteria: potential participants with insomnia due to caffeine use, inadequate


32

Gooneratne 2010

(Continued)

sleep hygiene, major depression or anxiety disorder, active use of sedative-hypnotics, restless legs syndrome, PLMI > 5/h on PSG, alcohol abuse, cognitive impairment (dementia), liver abnormalities, prior history of CPAP use, active use of fluvoxamine (potential interaction with ramelteon), COPD. All study participants (ramelteon and placebo) received auto CPAP therapy Interventions

Ramelteon 8 mg vs placebo for 30 days

Outcomes

Change in sleep onset latency determined from PSG at 4 weeks, sleep efficiency, AHI, periodic limb movement index, PSQI, ISI, ESS, FOSQ, RAND, SF-36, CIRS

Notes

Investigator-initiated pilot grant from Takeda Pharmaceuticals North America, Inc. Takeda Pharmaceuticals provided all study research medications (ramelteon and placebo tablets). Takeda Pharmaceuticals had no role in study design, research study participant enrolment, data collection, data storage, data interpretation and manuscript preparation; no role in decision to submit the manuscript for publication nor to edit the final manuscript

Risk of bias Bias




Randomisation was performed by the Investigational Drug Service of the University of Pennsylvania CTSA using a computergenerated central system and opaque, sequentially numbered sealed envelopes


Low risk

Randomisation was performed by the Investigational Drug Service of the University of Pennsylvania CTSA using a computergenerated central system and opaque, sequentially numbered sealed envelopes


Person responsible for participant care (study investigators): yes Participant: yes




Low risk

Intent-to-treat comparisons were used (6 participants dropped out, n = 4 ramelteon and n = 2 placebo)


Low risk



33

Höijer 1994 Methods

Randomised, double-blind, cross-over trial. Statistical test: ANOVA

Participants

11 adults (12 males), age 47 (± 7.5), BMI 27 (±3.3), AHI 15 (± 5.0) Exclusion criteria: regular intake of benzodiazepines, theophylline, antipsychotic agents or alcohol (> 30 mg/d), impaired renal or hepatic function Initial diagnosis of OSA by respiratory polygraphy in the sleep laboratory, full PSG on study night

Interventions

Nitrazepam 5 mg or 10 mg vs placebo for 3 nights with 1-week washout period

Outcomes

AHI, mean SpO2 , minimal SpO2 , sleep characteristics including TST measured by PSG

Notes

This study was supported by grants from the Swedish Medical Research Council (grant No. 9862), the Swedish Heart and Lung Foundation and Inga-Britt and Arne Lundbergs Forskningsstiftelse

Risk of bias Bias






Unclear risk







Unclear risk

Unclear


Low risk


Kryger 2007 Methods

Randomised, double-blind, cross-over, multi-centre trial (5 centres). Statistical methods used: ANOVA mixed model, a non-parametric analysis: Wilcoxon rank sum test

Participants

26 adult patients (18 males), mean age 47.4 ± 9.45 (21-64 years), BMI 30.2 ± 4.76, AHI 12.2 ± 7.69

Interventions

Ramelteon 16 mg vs placebo for 1 night with 5-12 days’ washout


34

Kryger 2007

(Continued)

Outcomes

AHI, mean SpO2 , % SpO2 < 80%

Notes

Sponsored by TAKEDA Pharmaceuticals

Risk of bias Bias






Unclear risk







Low risk

All participants completed the study


Low risk


Rosenberg 2007 Methods

Randomised, double-blind, cross-over study

Participants

22 adults (15 males), mean age 48.4 ± 9 (35-64), mean BMI 32 ± 42.1, AHI 17.2 ± 8. 8 events per hour sleep, on CPAP ≥ 3 months with good reported compliance but off treatment during the study night. OSA confirmed on PSG Exclusion criteria: severe OSA, CSA, PLM, RLS, mental disorder, other significant medical co-morbidity, excessive alcohol, shift work, pregnancy

Interventions

Eszopiclone 3 mg vs placebo for 2 consecutive nights and 5-7 days’ washout

Outcomes

Mean AHI on 2 nights, AI, HI, average duration of apnoea and hypopnoea episodes, longest duration of apnoea and hypopnoea episodes, average oxygen saturation during apnoea and hypopnoea episodes, lowest oxygen saturation during apnoea and hypopnoea episodes. TST, arousal index. Analyses of all endpoints used a mixed model with the 2night mean as the dependent variable; treatment, period and sequence as fixed effects; and participant nested within sequence as a random effect

Notes

This study was funded by Sepracor. Russell Rosenberg and Martin Scharf have received grant support from


35

Rosenberg 2007

(Continued)

Sepracor. James Roach and David Amato are employees of Sepracor Risk of bias Bias




“Randomised” Comment: not described



Unclear risk


“Double blind, placebo controlled” Comment: not described




Low risk

All but 1 (protocol violation) participant completed the study


Low risk


Wang 2011 Methods

Randomised, double-blind, cross-over study. Statistical tests: paired t tests, correlations were tested by Pearson or Spearman tests

Participants

20 male adults, mean age 44 ± 11.6, BMI 27.5 ± 7.5, AHI 16.8 ± 14.1, ESS 9.4 ± 4.3 Exclusion criteria: uncontrolled concurrent medical or psychiatric illness, use of any concurrent medications known to affect sleep respiration or interact with temazepam, medical conditions that would contraindicate temazepam, other major sleep disorders including PLMS, irregular sleep patterns such as occur with shift work

Interventions

Temazepam 10 mg vs placebo for 1 night with 7 days’ washout

Outcomes

AHI, ODI, minimal SpO2 , ventilatory chemoreflex testing, polysomnographically measured sleep characteristics

Notes

Funding source: none stated. Declaration of interest: none

Risk of bias Bias




36

Wang 2011

(Continued)


“Randomisation 1:1 ratio. The randomizations sequence was generated by a research pharmacist at Royal Prince Alfred Hospital Pharmacy, who also dispensed treatment and who never met any patient and only disclosed the sequence to us after we had completed data collection, processing including all sleep scoring”


Low risk

As above


As above


As above


Low risk

All but 2 participants completed the study, 1 was excluded because of significant PLMS that became evident only on PSG after randomisation and 1 for moving interstate


Low risk


AHI: apnoea-hypopnoea index; BMI: body mass index; CBT: cognitive-behavioural therapy; CIRS: Cumulative Illness Rating Scale; COPD: chronic obstructive pulmonary disease; CPAP: continuous positive airway pressure; CSA: central sleep apnoea; ESS: Epworth Sleepiness Scale; FOSQ: Functional Outcomes of Sleepiness Scale; ISI: Insomnia Severity Index; ODI: oxygen desaturation index; OSA: obstructive sleep apnoea; OSAS: obstructive sleep apnoea syndrome; PLMI: periodic limb movement index; PSG: polysomnography; PSQI: Pittsburgh Sleep Quality Index; SF-36: Short Form (36) Health Survey; RDI: respiratory disturbances index; SpO2 : arterial oxygen saturation; RLS: restless leg syndrome; TST: total sleep time.

Characteristics of excluded studies [ordered by study ID]

Study

Reason for exclusion

Bradshaw 2006

Primary outcome CPAP compliance, only 1 sleep study performed without subsequent sleep study on zaleplon or placebo, therefore not meeting inclusion criteria

Crowe 1992

No formal diagnosis of OSA before the study night

Deflandre 2014

Effect of alprazolam on OSA after colonoscopy, no comparison between formal sleep study and study on study drug/placebo


37

(Continued)

Dolly 1982

No formal diagnosis of OSA before the study

Guo 2003

No diagnosis of OSA before the study

Lofaso 1997

No OSA, UARS only

Malish 2009

1-night sleep study only, no comparison between formal sleep study and study on study drug/placebo, therefore not meeting inclusion criteria

Noseda 2002

No RCT

Park 2013

1-night study only without formal diagnostic sleep study and further study on study drug/placebo, therefore not meeting inclusion criteria

Pawlik 2005

Clonidine, not sedative drug

Sharf 1994

Patients without formal diagnosis of OSA before the study

Sun 2015

Effect of suvorexant on patients with COPD

Wang 2013a

No RCT

Webster 2012

Patients with chronic lower back pain and no formal OSA diagnosis

Characteristics of studies awaiting assessment [ordered by study ID] Coyle 2005 Methods

Randomised, placebo-controlled trial, cross-over design

Participants

15 patients with mild to moderate OSA

Interventions

Zaleplon 10 mg vs placebo over 5 nights

Outcomes

AHI

Notes

Published as letter to editor with no significant difference in AHI with zaleplon compared with placebo


38

Characteristics of ongoing studies [ordered by study ID] Sun 2013 Trial name or title

Effects of Suvorexant in Participants With Obstructive Sleep Apnea (MK-4305-036)

Methods

Double-blind, cross-over RCT examining suvorexant vs placebo

Participants

26 patients with mild to moderate OSA

Interventions

Participants administered a 40-mg dose of suvorexant and placebo for 4 days

Outcomes

Primary: mean apnoea-hypopnoea index (AHI) day 4, number of participants with an adverse event (AE), number of participants who discontinued study drug because of an AE Secondary: mean arterial oxygen saturation, percentage of total sleep time that arterial SaO2 is less than 90%

Starting date

February 17, 2011

Contact information

Merck Sharp & Dohme Corp

Notes

Results published as abstract in 2014, full-text publication awaited


39

DATA AND ANALYSES

Comparison 1. Remifentanil vs placebo

Outcome or subgroup title 1 AHI (events/h) 2 Obstructive apnoea (events/h) 3 Central apnoea (events/h) 4 Arousal Index 5 Adverse events 6 Minimum SpO2 (%)

No. of studies

No. of participants

1 1 1 1 1 1

Statistical method Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Mean Difference (Fixed, 95% CI)

Effect size Totals not selected Totals not selected Totals not selected Totals not selected Totals not selected Totals not selected

Comparison 2. Eszopiclone vs placebo

Outcome or subgroup title 1 AHI (events/h) 2 Apnoea index (events/h) 3 Hypopnoea index (events/h) 4 Arousal index 5 Adverse events

No. of studies

No. of participants

2 1 1 1 2

77

77

Statistical method Mean Difference (Random, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)

Effect size -3.40 [-10.36, 3.56] Totals not selected Totals not selected Totals not selected 0.0 [0.0, 0.0]

Comparison 3. Nitrazepam vs placebo

Outcome or subgroup title 1 Apnoea index (events/h) 1.1 5 mg vs placebo 1.2 10 mg vs placebo 2 Minimum SpO2 (%) 2.1 5 mg vs placebo 2.2 10 mg vs placebo

No. of studies 1 1 1 1 1 1

No. of participants

Statistical method Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI)

Effect size Totals not selected 0.0 [0.0, 0.0] 0.0 [0.0, 0.0] Totals not selected 0.0 [0.0, 0.0] 0.0 [0.0, 0.0]


40

Comparison 4. Temazepam vs placebo

Outcome or subgroup title 1 AHI (events/h) 1.1 10 mg 2 Minimum SpO2 (%) 3 Arousal Index 3.1 10 mg 4 RDI (events/h) 4.1 Temazepam 15-30 mg 5 Adverse events

No. of studies

No. of participants

1 1 1 1 1 1 1 2

70

Statistical method Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)

Effect size Totals not selected 0.0 [0.0, 0.0] Subtotals only Totals not selected 0.0 [0.0, 0.0] Totals not selected 0.0 [0.0, 0.0] 0.0 [0.0, 0.0]

Comparison 5. Triazolam vs placebo

Outcome or subgroup title 1 Minimum SpO2 (%) 1.1 NREM 1.2 REM 2 Adverse events

No. of studies

No. of participants

1 1 1 1

24

Statistical method Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)

Effect size Totals not selected 0.0 [0.0, 0.0] 0.0 [0.0, 0.0] 0.0 [0.0, 0.0]

Comparison 6. Sodium oxybate vs placebo

Outcome or subgroup title 1 AHI (events/h) 2 Mean SpO2 (%) 3 Minimum SpO2 (%) 4 Arousal index 5 ESS 6 Adverse events

No. of studies 1 1 1 1 1 1

No. of participants

Statistical method Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)

Effect size Totals not selected Totals not selected Subtotals only Totals not selected Totals not selected Totals not selected


41

Comparison 7. Ramelteon vs placebo

Outcome or subgroup title 1 AHI (events/h) 1.1 8 mg 1.2 16 mg 2 Mean SpO2 (%) 3 ESS

No. of studies

No. of participants

2 1 1 1 1

47 21 26

Statistical method

Effect size

Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI) Mean Difference (Fixed, 95% CI)

0.26 [-2.05, 2.57] 0.5 [-11.60, 12.60] 0.25 [-2.10, 2.60] Totals not selected Totals not selected

Analysis 1.1. Comparison 1 Remifentanil vs placebo, Outcome 1 AHI (events/h). Review:

Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea

Comparison: 1 Remifentanil vs placebo Outcome: 1 AHI (events/h)

Study or subgroup

Bernards 2009 (1)

Remifentanil

Placebo

N

N

10

9

Mean Difference

Mean Difference (SE)

Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

10 (10.1156)

10.00 [ -9.83, 29.83 ]

-50

-25

Favours remifentanil

0

25

50

Favours placebo

(1) Change from baseline


42

Analysis 1.2. Comparison 1 Remifentanil vs placebo, Outcome 2 Obstructive apnoea (events/h). Review:


Comparison: 1 Remifentanil vs placebo Outcome: 2 Obstructive apnoea (events/h)

Study or subgroup

Bernards 2009

Remifentanyl

Placebo

N

N

10

9

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-9 (4.2847)

-9.00 [ -17.40, -0.60 ]

-20

-10

0

Favours remifentanyl

10

20

Favours placebo

Analysis 1.3. Comparison 1 Remifentanil vs placebo, Outcome 3 Central apnoea (events/h). Review:


Comparison: 1 Remifentanil vs placebo Outcome: 3 Central apnoea (events/h)

Study or subgroup

Bernards 2009

Remifentanyl

Placebo

N

N

10

9

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

16 (9.2933)

16.00 [ -2.21, 34.21 ]

-50

-25

Favours remifentanyl

0

25

50

Favours placebo


43

Analysis 1.4. Comparison 1 Remifentanil vs placebo, Outcome 4 Arousal Index. Review:


Comparison: 1 Remifentanil vs placebo Outcome: 4 Arousal Index

Study or subgroup

Remifentanil

Placebo

N

N

10

9

Bernards 2009

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

34 (13.6234)

34.00 [ 7.30, 60.70 ]

-50

-25

0


25

50

Favours placebo

Analysis 1.5. Comparison 1 Remifentanil vs placebo, Outcome 5 Adverse events. Review:


Comparison: 1 Remifentanil vs placebo Outcome: 5 Adverse events

Study or subgroup

Bernards 2009

Remifentanil

Placebo

Odds Ratio

Odds Ratio

n/N

n/N

M-H,Fixed,95% CI

M-H,Fixed,95% CI

1/10

0/9

3.00 [ 0.11, 83.36 ]

0.01

0.1


1

10

100

Favours placebo


44

Analysis 1.6. Comparison 1 Remifentanil vs placebo, Outcome 6 Minimum SpO2 (%). Review:


Comparison: 1 Remifentanil vs placebo Outcome: 6 Minimum SpO2 (%)

Study or subgroup

Remifentanil

Placebo

N

N

10

9

Bernards 2009

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-7 (2.5254)

-7.00 [ -11.95, -2.05 ]

-10

-5

0

Favours placebo

5

10


Analysis 2.1. Comparison 2 Eszopiclone vs placebo, Outcome 1 AHI (events/h). Review:


Comparison: 2 Eszopiclone vs placebo Outcome: 1 AHI (events/h)

Study or subgroup

Mean Difference

Placebo

N

N

Eckert 2011

17

17

-7 (1.3776)

49.3 %

-7.00 [ -9.70, -4.30 ]

Rosenberg 2007

21

22

0.1 (1.0943)

50.7 %

0.10 [ -2.04, 2.24 ]

38

39

100.0 %

-3.40 [ -10.36, 3.56 ]

Total (95% CI)


Mean Difference

Eszopiclone

Weight

IV,Random,95% CI

IV,Random,95% CI

Heterogeneity: Tau2 = 23.66; Chi2 = 16.29, df = 1 (P = 0.00005); I2 =94% Test for overall effect: Z = 0.96 (P = 0.34) Test for subgroup differences: Not applicable

-10

-5

Favours Eszopiclone

0

5

10

Favours placebo


45

Analysis 2.2. Comparison 2 Eszopiclone vs placebo, Outcome 2 Apnoea index (events/h). Review:


Comparison: 2 Eszopiclone vs placebo Outcome: 2 Apnoea index (events/h)

Study or subgroup

Eszopiclone

Placebo

N

N

Rosenberg 2007

21

22

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

1.05 (0.6992)

1.05 [ -0.32, 2.42 ]

-4

-2

0

Favours Eszopiclone

2

4

Favours Placebo

Analysis 2.3. Comparison 2 Eszopiclone vs placebo, Outcome 3 Hypopnoea index (events/h). Review:


Comparison: 2 Eszopiclone vs placebo Outcome: 3 Hypopnoea index (events/h)

Study or subgroup

Rosenberg 2007

Eszopiclone

Placebo

N

N

21

22

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-0.95 (1.1855)

-0.95 [ -3.27, 1.37 ]

-4

-2

Favours eszopiclone

0

2

4

Favours Placebo


46

Analysis 2.4. Comparison 2 Eszopiclone vs placebo, Outcome 4 Arousal index. Review:


Comparison: 2 Eszopiclone vs placebo Outcome: 4 Arousal index

Study or subgroup

Eszopiclone

Placebo

N

N

Rosenberg 2007

21

22

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-1.85 (1.1247)

-1.85 [ -4.05, 0.35 ]

-4

-2

Favours Eszopiclone

0

2

4

Favours placebo

Analysis 2.5. Comparison 2 Eszopiclone vs placebo, Outcome 5 Adverse events. Review:


Comparison: 2 Eszopiclone vs placebo Outcome: 5 Adverse events

Study or subgroup

Eszopiclone

Placebo

Odds Ratio

n/N

n/N

M-H,Fixed,95% CI

Eckert 2011

0/17

0/17

Not estimable

Rosenberg 2007

0/21

0/22

Not estimable

38

39

Not estimable

Total (95% CI)

Weight

Odds Ratio M-H,Fixed,95% CI

Total events: 0 (Eszopiclone), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable

0.01

0.1

Favours Eszopiclone

1

10

100

Favours placebo


47

Analysis 3.1. Comparison 3 Nitrazepam vs placebo, Outcome 1 Apnoea index (events/h). Review:


Comparison: 3 Nitrazepam vs placebo Outcome: 1 Apnoea index (events/h)

Study or subgroup

Mean Difference


Mean Difference

Nitrazepam

Placebo

N

N

11

11

-0.64 (1.7041)

-0.64 [ -3.98, 2.70 ]

11

11

-1 (1.7041)

-1.00 [ -4.34, 2.34 ]

IV,Fixed,95% CI

IV,Fixed,95% CI

1 5 mg vs placebo ¨ Hoijer 1994 2 10 mg vs placebo ¨ Hoijer 1994

-10

-5

0

Favours nitrazepam

5

10

Favours placebo

Analysis 3.2. Comparison 3 Nitrazepam vs placebo, Outcome 2 Minimum SpO2 (%). Review:


Comparison: 3 Nitrazepam vs placebo Outcome: 2 Minimum SpO2 (%)

Study or subgroup

Mean Difference


Mean Difference

Nitrazepam

Placebo

N

N

11

11

2.55 (5.5868)

2.55 [ -8.40, 13.50 ]

11

11

-6.185 (5.6047)

-6.19 [ -17.17, 4.80 ]

IV,Fixed,95% CI

IV,Fixed,95% CI

1 5 mg vs placebo ¨ Hoijer 1994 2 10 mg vs placebo ¨ Hoijer 1994

-20

-10

Favours Placebo

0

10

20

Favours Nitrazepam


48

Analysis 4.1. Comparison 4 Temazepam vs placebo, Outcome 1 AHI (events/h). Review:


Comparison: 4 Temazepam vs placebo Outcome: 1 AHI (events/h)

Study or subgroup

Temazepam

Placebo

N

N

20

20

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

1 10 mg Wang 2011

1.18 (1.5)

1.18 [ -1.76, 4.12 ]

-4

-2

0

Favours temazepam

2

4

Favours placebo

Analysis 4.2. Comparison 4 Temazepam vs placebo, Outcome 2 Minimum SpO2 (%). Review:


Comparison: 4 Temazepam vs placebo Outcome: 2 Minimum SpO2 (%)

Study or subgroup

Wang 2011

Subtotal (95% CI)

Temazepam

Placebo

N

N

20

20

0

0

Mean Difference


Mean Difference

Weight

IV,Fixed,95% CI

IV,Fixed,95% CI

1.7 (1.45)

1.70 [ -1.14, 4.54 ]

0.0 [ 0.0, 0.0 ]

Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P < 0.00001) Test for subgroup differences: Not applicable

-4

-2

Favours placebo

0

2

4

Favours temazepam


49

Analysis 4.3. Comparison 4 Temazepam vs placebo, Outcome 3 Arousal Index. Review:


Comparison: 4 Temazepam vs placebo Outcome: 3 Arousal Index

Study or subgroup

Temazepam

Placebo

N

N

20

20

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

1 10 mg Wang 2011

1.34 (1.25)

1.34 [ -1.11, 3.79 ]

-4

-2

0


2

4

Favours placebo

Analysis 4.4. Comparison 4 Temazepam vs placebo, Outcome 4 RDI (events/h). Review:


Comparison: 4 Temazepam vs placebo Outcome: 4 RDI (events/h)

Study or subgroup

Temazepam

Placebo

N

N

7

8

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

1 Temazepam 15-30 mg Camacho 1995

-1.6 (2.1477)

-1.60 [ -5.81, 2.61 ]

-4

-2

Favours temazepam

0

2

4

Favours placebo


50

Analysis 4.5. Comparison 4 Temazepam vs placebo, Outcome 5 Adverse events. Review:


Comparison: 4 Temazepam vs placebo Outcome: 5 Adverse events

Study or subgroup

Temazepam

Placebo

n/N

n/N

Camacho 1995

0/15

0/15

Not estimable

Wang 2011

0/20

0/20

Not estimable

35

35

Not estimable

Total (95% CI)

Odds Ratio

Weight

M-H,Fixed,95% CI


Total events: 0 (Temazepam), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable

0.01

0.1

1

10

Favours temazepam

100

Favours placebo

Analysis 5.1. Comparison 5 Triazolam vs placebo, Outcome 1 Minimum SpO2 (%). Review:


Comparison: 5 Triazolam vs placebo Outcome: 1 Minimum SpO2 (%)

Study or subgroup

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

1 NREM Berry 1995 (1)

-10.2 (3)

-10.20 [ -16.08, -4.32 ]

-14 (4)

-14.00 [ -21.84, -6.16 ]

2 REM Berry 1995 (2)

-20

-10

Favours placebo

0

10

20

Favours triazolam

(1) Imputed from p=0.02 (2) Imputed from p=0.02


51

Analysis 5.2. Comparison 5 Triazolam vs placebo, Outcome 2 Adverse events. Review:


Comparison: 5 Triazolam vs placebo Outcome: 2 Adverse events

Study or subgroup

triazolam

Placebo

Odds Ratio

n/N

n/N

M-H,Fixed,95% CI

0/12

0/12

Not estimable

12

12

Not estimable

Berry 1995

Total (95% CI)

Weight


Total events: 0 (triazolam), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: not applicable Test for subgroup differences: Not applicable

0.01

0.1

1

Favours

10

100

Favours placebo

Analysis 6.1. Comparison 6 Sodium oxybate vs placebo, Outcome 1 AHI (events/h). Review:


Comparison: 6 Sodium oxybate vs placebo Outcome: 1 AHI (events/h)

Study or subgroup

George 2011

Sodium oxybate

Placebo

N

N

26

22

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-7.41 (3.4477)

-7.41 [ -14.17, -0.65 ]

-20

-10

Favours sodium oxybate

0

10

20

Favours placebo


52

Analysis 6.2. Comparison 6 Sodium oxybate vs placebo, Outcome 2 Mean SpO2 (%). Review:


Comparison: 6 Sodium oxybate vs placebo Outcome: 2 Mean SpO2 (%)

Study or subgroup

Mean Difference


Mean Difference

IV,Fixed,95% CI George 2011

IV,Fixed,95% CI

0.5 (0.5198)

0.50 [ -0.52, 1.52 ]

-2

-1

0

Favours placebo

1

2


Analysis 6.3. Comparison 6 Sodium oxybate vs placebo, Outcome 3 Minimum SpO2 (%). Review:


Comparison: 6 Sodium oxybate vs placebo Outcome: 3 Minimum SpO2 (%)

Study or subgroup

Mean Difference


IV,Fixed,95% CI George 2011

Mean Difference

Weight

IV,Fixed,95% CI

-0.1 (0.1408)

-0.10 [ -0.38, 0.18 ]

Subtotal (95% CI)

0.0 [ 0.0, 0.0 ]

Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P < 0.00001) Test for subgroup differences: Not applicable

-0.5

-0.25

Favours placebo

0

0.25

0.5



53

Analysis 6.4. Comparison 6 Sodium oxybate vs placebo, Outcome 4 Arousal index. Review:


Comparison: 6 Sodium oxybate vs placebo Outcome: 4 Arousal index

Study or subgroup

George 2011

Sodium oxybate

Placebo

N

N

26

22

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-1.2 (1.4505)

-1.20 [ -4.04, 1.64 ]

-4

-2

0


2

4

Favours placebo

Analysis 6.5. Comparison 6 Sodium oxybate vs placebo, Outcome 5 ESS. Review:


Comparison: 6 Sodium oxybate vs placebo Outcome: 5 ESS

Study or subgroup

George 2011 (1)

Sodium oxybate

Placebo

N

N

26

22

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

-2.9 (1.68)

-2.90 [ -6.19, 0.39 ]

-4

-2


0

2

4

Favours placebo

(1) Imputed from p=0.09


54

Analysis 6.6. Comparison 6 Sodium oxybate vs placebo, Outcome 6 Adverse events. Review:


Comparison: 6 Sodium oxybate vs placebo Outcome: 6 Adverse events

Study or subgroup

Sodium oxybate

Placebo

Odds Ratio

Odds Ratio

n/N

n/N

M-H,Fixed,95% CI

M-H,Fixed,95% CI

9/26

6/22

George 2011

1.41 [ 0.41, 4.87 ]

0.1 0.2

0.5

Favours Sodium oxybate

1

2

5

10

Favours placebo

Analysis 7.1. Comparison 7 Ramelteon vs placebo, Outcome 1 AHI (events/h). Review:


Comparison: 7 Ramelteon vs placebo Outcome: 1 AHI (events/h)

Study or subgroup

Ramelteon

Placebo

N

N

Gooneratne 2010

8

13

Subtotal (95% CI)

8

13

Mean Difference


Mean Difference

Weight

IV,Fixed,95% CI

IV,Fixed,95% CI

1 8 mg 0.5 (6.1736)

3.6 %

0.50 [ -11.60, 12.60 ]

3.6 %

0.50 [ -11.60, 12.60 ]

96.4 %

0.25 [ -2.10, 2.60 ]

Heterogeneity: not applicable Test for overall effect: Z = 0.08 (P = 0.94) 2 16 mg Kryger 2007 (1)

Subtotal (95% CI)

13

13

0.25 (1.199)

13

13

96.4 %

0.25 [ -2.10, 2.60 ]

26

100.0 %

0.26 [ -2.05, 2.57 ]

Heterogeneity: not applicable Test for overall effect: Z = 0.21 (P = 0.83)

Total (95% CI)

21

Heterogeneity: Chi2 = 0.00, df = 1 (P = 0.97); I2 =0.0% Test for overall effect: Z = 0.22 (P = 0.83) Test for subgroup differences: Chi2 = 0.00, df = 1 (P = 0.97), I2 =0.0%

-10

-5

Favours ramelteon

0

5

10

Favours placebo


55

(1) Cross over study with 26 participants

Analysis 7.2. Comparison 7 Ramelteon vs placebo, Outcome 2 Mean SpO2 (%). Review:


Comparison: 7 Ramelteon vs placebo Outcome: 2 Mean SpO2 (%)

Study or subgroup

Kryger 2007

Ramelteon

Placebo

N

N

13

13

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

0.5 (0.2551)

0.50 [ 0.00, 1.00 ]

-1

-0.5

0

Favours placebo

0.5

1

Favours ramelteon

Analysis 7.3. Comparison 7 Ramelteon vs placebo, Outcome 3 ESS. Review:


Comparison: 7 Ramelteon vs placebo Outcome: 3 ESS

Study or subgroup

Gooneratne 2010

Ramelteon

Placebo

N

N

8

13

Mean Difference


Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI

0.1 (2.5511)

0.10 [ -4.90, 5.10 ]

-4

-2

Favours ramelteon

0

2

4

Favours placebo


56

ADDITIONAL TABLES Table 1. Summary of results

Drug class

Study

Control

N

Change in AHI

Change in mean Change in miniSpO2 mum SpO2

Opioids/ opiates

Bernards 2009 Remifentanil Placebo infusion (0.075 mcg/kg/h)

19

No change

Not reported

Decrease

’Z drugs’

Rosenberg 2007

Eszopiclone 3 Placebo mg

22

No change

No change

No change

Eckert 2011

Eszopiclone 3 Placebo mg

17

Decrease

No change

No change

Berry 2006

Zolpidem 10 Placebo mg

16

No change

Not reported

No change

Cirignotta 1988

Zolpidem 20 mg

Placebo

12

No change

Decrease

Decrease

Cirignotta 1992

Brotizolam 0. Placebo 25 mg

12

No change in RDI No change (number of oxygen desaturations > 4% /h sleep)

No change

Cirignotta 1988

Flurazepam 30 mg

Placebo

12

No change

Decrease

Cirignotta 1992

Flurazepam 30 mg

Placebo

12

No change in RDI No change (number of oxygen desaturations > 4% /h sleep)

No change

Höijer 1994

Nitrazepam 5 Placebo mg or 10 mg

11

No change in AI

No change

Camacho 1995)

Temazepam 15-30 mg

Placebo

15

No change in RDI Not reported (number apnoeas/hypopnoeas/h sleep)

Not reported

Wang 2011

Temazepam 10 mg

Placebo

20

No change

Not reported

No change

Berry 1995

Triazolam 0. Placebo 25 mg

12

No change

Not reported

Decrease

Benzodiazepines

Intervention

Decrease

Not reported


57

Table 1. Summary of results

Sodium oxy- George 2010 bate George 2011 Melatonin and melatoninrelated drugs

(Continued)

SXB 9 g

Placebo

42

No change

No change

Not reported

SXB 4.5 g

Placebo

48

Decreased

No change

No change

Gooneratne 2010)

Ramelteon 8 Placebo mg

21

No change

Not reported

Not reported

Kryger 2007

Ramelteon 16 Placebo mg

26

No change

No change

Not reported

AHI: apnoea-hypopnoea index; AI: apnoea index; mean SpO2 : mean nocturnal oxygen saturation; minimum SpO2 : minimum nocturnal oxygen saturation; RDI: respiratory disturbance index; SBX: sodium oxybate.

APPENDICES Appendix 1. Sources and search methods for the Cochrane Airways Group Specialised Register (CAGR)

Electronic searches: core databases

Database

Frequency of search

CENTRAL

Monthly

MEDLINE (Ovid)

Weekly

EMBASE (Ovid)

Weekly

PsycINFO (Ovid)

Monthly

CINAHL (EBSCO)

Monthly

AMED (EBSCO)

Monthly

Handsearches: core respiratory conference abstracts


58

Conference

Years searched

American Academy of Allergy, Asthma and Immunology (AAAAI) 2001 onwards American Thoracic Society (ATS)

2001 onwards

Asia Pacific Society of Respirology (APSR)

2004 onwards

British Thoracic Society (BTS) Winter Meeting

2000 onwards

Chest Meeting

2003 onwards

European Respiratory Society (ERS)

1992, 1994, 2000 onwards

International Primary Care Respiratory Group Congress (IPCRG) 2002 onwards Thoracic Society of Australia and New Zealand (TSANZ)

1999 onwards

MEDLINE search strategy used to identify trials for the CAGR

Sleep apnoea search

1. exp Sleep Apnea Syndromes/ 2. (sleep$ adj3 (apnea$ or apnoea$)).mp. 3. (hypopnoea$ or hypopnoea$).mp. 4. OSA.mp. 5. SHS.mp. 6. OSAHS.mp. 7. or/1-6

Filter to identify randomised controlled trials (RCTs)

1. exp “clinical trial [publication type]”/ 2. (randomised or randomised).ab,ti. 3. placebo.ab,ti. 4. dt.fs. 5. randomly.ab,ti. 6. trial.ab,ti. 7. groups.ab,ti. 8. or/1-7 9. Animals/ 10. Humans/ 11. 9 not (9 and 10) 12. 8 not 11 The MEDLINE strategy and RCT filter are adapted to identify trials in other electronic databases


59

Appendix 2. Search strategy for identifying trial reports from the Cochrane Airways Group Specialised Register (CAGR) #1 SLP:MISC1 #2 MeSH DESCRIPTOR Sleep Apnea, Obstructive #3 sleep near3 (apnea* or apnoea*) #4 (hypopnoea* or hypopnoea*) #5 (OSA OR SHS OR OSAHS:TI,AB) #6 (#1 OR #2 OR #3 OR #4 OR #5) #7 non-benzodiazepine* or nonbenzodiazepine* #8 Zaleplon #9 Zolpidem #10 Eszopiclone #11 Zopiclone #12 MeSH DESCRIPTOR Benzodiazepines Explode All #13 Triazolam #14 Nitrazepam #15 Flurazepam #16 Loprazolam #17 Lormetazepam #18 Temazepam #19 Estazolam #20 Oxazepam #21 Alprazolam #22 Lorazepam #23 Clonazepam #24 Quazepam #25 Flurazepam #26 Diazepam #27 Midazolam #28 MeSH DESCRIPTOR Analgesics, Opioid #29 Buprenorphine #30 Dipipanone #31 Diamorphine #32 Alfentanyl #33 Remifentanyl #34 Fentanyl #35 Methadone #36 Oxycodone #37 Papaveretum #38 Pentazocine #39 Pethidine #40 Tramadol #41 Codeine #42 Meptazinol #43 “sodium oxybate” #44 ramelteon or Rozerem #45 Gabapentin #46 Pregabalin #47 Melatonin #48 hypnotic* #49 opiate* or opioid* #50 analgesic* Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

60

#51 sedative* #52 #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or # 25 or #26 or #27 or #28 or #29 or #30 or #31 or #32 or #33 or #34 or #35 or #36 or #37 or #38 or #39 or #40 or #41 or #42 or #43 or #44 or #45 or #46 or #47 or #48 or #49 or #50 or #51 #53 #6 and #52 [Note: in search line #1. MISC1 denotes the field in the record where the reference has been coded for condition, in this case, sleep apnoea]

Appendix 3. CENTRAL, MEDLINE and EMBASE search strategies

CENTRAL (CRS Online) #1 MESH DESCRIPTOR Sleep Apnea, Obstructive EXPLODE ALL TREES #2 (sleep near3 (apnea* or apnoea*)):TI,AB,KY #3 (hypopnoea* or hypopnoea*):TI,AB,KY #4 (OSA OR SHS OR OSAHS):TI,AB,KY #5 #1 OR #2 OR #3 OR #4 #6 (non-benzodiazepine* or nonbenzodiazepine*):TI,AB,KY #7 Zaleplon:TI,AB,KY #8 Zolpidem:TI,AB,KY #9 Eszopiclone:TI,AB,KY #10 Zopiclone:TI,AB,KY #11 MESH DESCRIPTOR Benzodiazepines EXPLODE ALL TREES #12 Triazolam:TI,AB,KY #13 Nitrazepam:TI,AB,KY #14 Flurazepam:TI,AB,KY #15 Loprazolam:TI,AB,KY #16 Lormetazepam:TI,AB,KY #17 Temazepam:TI,AB,KY #18 Estazolam:TI,AB,KY #19 Oxazepam:TI,AB,KY #20 Alprazolam:TI,AB,KY #21 Lorazepam:TI,AB,KY #22 Clonazepam:TI,AB,KY #23 Quazepam:TI,AB,KY #24 Flurazepam:TI,AB,KY #25 Diazepam:TI,AB,KY #26 Midazolam:TI,AB,KY #27 MESH DESCRIPTOR Analgesics, Opioid EXPLODE ALL TREES #28 Buprenorphine #29 Dipipanone:TI,AB,KY #30 Diamorphine:TI,AB,KY #31 Alfentanyl:TI,AB,KY #32 (Remifentanyl or remifentanil):TI,AB,KY #33 Fentanyl:TI,AB,KY #34 Methadone:TI,AB,KY #35 Oxycodone:TI,AB,KY #36 Papaveretum:TI,AB,KY #37 Pentazocine:TI,AB,KY #38 Pethidine:TI,AB,KY #39 Tramadol:TI,AB,KY #40 Codeine:TI,AB,KY Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

61

#41 Meptazinol:TI,AB,KY #42 (“sodium oxybate”):TI,AB,KY #43 (ramelteon or Rozerem):TI,AB,KY #44 Gabapentin:TI,AB,KY #45 Pregabalin:TI,AB,KY #46 Melatonin:TI,AB,KY #47 hypnotic*:TI,AB,KY #48 (opiate* or opioid*):TI,AB,KY #49 analgesic*:TI,AB,KY #50 sedative*:TI,AB,KY #51 #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR # 21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35 OR # 36 OR #37 OR #38 OR #39 OR #40 OR #41 OR #42 OR #43 OR #44 OR #45 OR #46 OR #47 OR #48 OR #49 OR #50 #52 #5 AND #51

MEDLINE (Ovid) 1. exp Sleep Apnea, Obstructive/ 2. (sleep adj3 (apnea* or apnoea*)).tw. 3. (hypopnoea$ or hypopnoea$).tw. 4. 1 or 2 or 3 5. (non-benzodiazepine$ or nonbenzodiazepine$).tw. 6. Zaleplon.tw. 7. Zolpidem.tw. 8. Eszopiclone.tw. 9. Zopiclone.tw. 10. exp Benzodiazepines/ 11. Triazolam.tw. 12. Nitrazepam.tw. 13. Flurazepam.tw. 14. Loprazolam.tw. 15. Lormetazepam.tw. 16. Temazepam.tw. 17. Estazolam.tw. 18. Oxazepam.tw. 19. Alprazolam.tw. 20. Lorazepam.tw. 21. Clonazepam.tw. 22. Quazepam.tw. 23. Flurazepam.tw. 24. Diazepam.tw. 25. Midazolam.tw. 26. exp Analgesics, Opioid/ 27. Buprenorphine.tw. 28. Dipipanone.tw. 29. Diamorphine.tw. 30. Alfentanyl.tw. 31. (Remifentanyl or remifentanil).tw. 32. Fentanyl.tw. 33. Methadone.tw. 34. Oxycodone.tw. 35. Papaveretum.tw. 36. Pentazocine.tw. Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

62

37. Pethidine.tw. 38. Tramadol.tw. 39. Codeine.tw. 40. Meptazinol.tw. 41. “sodium oxybate”.tw. 42. (ramelteon or Rozerem).tw. 43. Gabapentin.tw. 44. Pregabalin.tw. 45. Melatonin.tw. 46. hypnotic$.tw. 47. (opiate$ or opioid$).tw. 48. analgesic$.tw. 49. sedative$.tw. 50. or/5-49 51. 4 and 50 52. (controlled clinical trial or randomised controlled trial).pt. 53. (randomised or randomised).ab,ti. 54. placebo.ab,ti. 55. dt.fs. 56. randomly.ab,ti. 57. trial.ab,ti. 58. groups.ab,ti. 59. or/52-58 60. Animals/ 61. Humans/ 62. 60 not (60 and 61) 63. 59 not 62 64. 51 and 63

EMBASE (Ovid) 1. exp Sleep Apnea, Obstructive/ 2. (sleep adj3 (apnea* or apnoea*)).tw. 3. (hypopnoea$ or hypopnoea$).tw. 4. 1 or 2 or 3 5. (non-benzodiazepine$ or nonbenzodiazepine$).tw. 6. Zaleplon.tw. 7. Zolpidem.tw. 8. Eszopiclone.tw. 9. Zopiclone.tw. 10. exp Benzodiazepines/ 11. Triazolam.tw. 12. Nitrazepam.tw. 13. Flurazepam.tw. 14. Loprazolam.tw. 15. Lormetazepam.tw. 16. Temazepam.tw. 17. Estazolam.tw. 18. Oxazepam.tw. 19. Alprazolam.tw. 20. Lorazepam.tw. 21. Clonazepam.tw. 22. Quazepam.tw. Effects of opioid, hypnotic and sedating medications on sleep-disordered breathing in adults with obstructive sleep apnoea (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

63

23. Flurazepam.tw. 24. Diazepam.tw. 25. Midazolam.tw. 26. exp Analgesics, Opioid/ 27. Buprenorphine.tw. 28. Dipipanone.tw. 29. Diamorphine.tw. 30. Alfentanyl.tw. 31. (Remifentanyl or remifentanil).tw. 32. Fentanyl.tw. 33. Methadone.tw. 34. Oxycodone.tw. 35. Papaveretum.tw. 36. Pentazocine.tw. 37. Pethidine.tw. 38. Tramadol.tw. 39. Codeine.tw. 40. Meptazinol.tw. 41. “sodium oxybate”.tw. 42. (ramelteon or Rozerem).tw. 43. Gabapentin.tw. 44. Pregabalin.tw. 45. Melatonin.tw. 46. hypnotic$.tw. 47. (opiate$ or opioid$).tw. 48. analgesic$.tw. 49. sedative$.tw. 50. or/5-49 51. 4 and 50 52. Randomized Controlled Trial/ 53. randomisation/ 54. controlled clinical trial/ 55. Double Blind Procedure/ 56. Single Blind Procedure/ 57. Crossover Procedure/ 58. (clinica$ adj3 trial$).tw. 59. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (mask$ or blind$ or method$)).tw. 60. exp Placebo/ 61. placebo$.ti,ab. 62. random$.ti,ab. 63. ((control$ or prospectiv$) adj3 (trial$ or method$ or stud$)).tw. 64. (crossover$ or cross-over$).ti,ab. 65. or/52-64 66. exp animals/ or exp invertebrate/ or animal experiment/ or animal model/ or animal tissue/ or animal cell/ or nonhuman/ 67. human/ or normal human/ or human cell/ 68. 66 and 67 69. 66 not 68 70. 65 not 69 71. 51 and 70


64

CONTRIBUTIONS OF AUTHORS Martina Mason (MM) developed the protocol with guidance from Ian Smith (IS), screened the references, entered data into RevMan and wrote the review. Chris Cates co-extracted the data from included studies and calculated variances for entry and meta-analysis in RevMan 5.0, along with MM. He also contributed to interpretation of the results and to preparation of the ’Summary of findings’ table, and provided input into the text of both the protocol and the review. Ian Smith (IS) proposed the review and helped to develop the protocol and write the review.

DECLARATIONS OF INTEREST IS has received a travel bursary from UCB Pharma, manufacturer of an opioid-based antitussive medication.

SOURCES OF SUPPORT

Internal sources • The review authors declare that no internal funding was received for this systematic review, Other.

External sources • The review authors declare that no external funding was received for this systematic review, Other.

DIFFERENCES BETWEEN PROTOCOL AND REVIEW Added mean nocturnal oxygen saturation (mean SpO2 ) outcomes. As many studies did not report on time spent with arterial saturation less than 90% (TSpO2 < 90%) but presented data on mean nocturnal saturation, this variable was added to the secondary outcomes .

INDEX TERMS Medical Subject Headings (MeSH) Analgesics, Opioid [administration & dosage; ∗ adverse effects]; Apnea [chemically induced; diagnosis]; Benzodiazepines [administration & dosage; adverse effects]; Disease Progression; Hypnotics and Sedatives [administration & dosage; ∗ adverse effects]; Oxygen Consumption [drug effects]; Randomized Controlled Trials as Topic; Sleep Apnea, Obstructive [∗ chemically induced]

MeSH check words Adult; Humans; Middle Aged


65

Obstructive sleep apnoea.

Obstructive sleep apnoea and arthrogryposis.

Cough in obstructive sleep apnoea.

Update on paediatric obstructive sleep apnoea.

Obstructive sleep apnoea associated with syringomyelia.

Effects of continuous positive airway pressure on blood pressure in hypertensive patients with obstructive sleep apnoea.

Adverse Effects of Hypnotic Medications.

Effects of maxillomandibular advancement on systemic blood pressure in patients with obstructive sleep apnoea.

Effects of continuous positive airway pressure on blood pressure in hypertensive patients with obstructive sleep apnoea.

Neurogenic effects on the palatopharyngeal muscle in patients with obstructive sleep apnoea: a muscle biopsy study.

Obstructive sleep apnoea syndrome and its management.

Obstructive sleep apnoea: new associations and approaches.

Obstructive sleep apnoea and cardiovascular disease.

New frontiers in obstructive sleep apnoea.

Obstructive Sleep Apnoea and Atrial Fibrillation.

Erythropoietin concentrations in obstructive sleep apnoea.

Obstructive sleep apnoea, insulin resistance and adipocytokines.

Impact of rail medical standard on obstructive sleep apnoea prevalence.

Oxidative stress in patients with obstructive sleep apnoea syndrome.

Effect of theophylline on sleep and sleep-disordered breathing in patients with chronic obstructive pulmonary disease.

Obstructive sleep apnoea in children with craniofacial syndromes.

Association of obstructive sleep apnoea with subclinical coronary atherosclerosis.

Effects of CPAP on energy expenditure in obese obstructive sleep apnoea patients: A pilot study.

Hot tonsillectomy for paediatric obstructive sleep apnoea.