Continuous positive airway pressure treatment impact on memory processes in obstructive sleep apnea patients: a randomized sham-controlled trial

Sleep Medicine 24 (2016) 44–50

Contents lists available at ScienceDirect

Sleep Medicine j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / s l e e p

Original Article

Continuous positive airway pressure treatment impact on memory processes in obstructive sleep apnea patients: a randomized sham-controlled trial Marie Joyeux-Faure a,b,*, Bernadette Naegelé c,d, Jean-Louis Pépin a,b, Renaud Tamisier a,b, Patrick Lévy a,b, Sandrine H. Launois a,b a

HP2 Unit, Inserm U1042, Grenoble Alpes University, Grenoble, France Department of Physiology and Sleep, Grenoble University Hospital, Grenoble, France c Neuro-Vascular Department, Grenoble University Hospital, Grenoble, France d Inserm U836, Grenoble Neurosciences Institut, Grenoble, France b

A R T I C L E

I N F O

Article history: Received 4 March 2016 Received in revised form 20 May 2016 Accepted 6 June 2016 Available online 23 August 2016 Keywords: Obstructive sleep apnea Continuous positive airway pressure Memory processes Episodic memory Procedural memory Working memory

A B S T R A C T

Objective: The aim of this study was to investigate the changes in a large panel of memory processes after six weeks of continuous positive airway pressure (CPAP) in obstructive sleep apnea (OSA) patients. This randomized controlled trial compared the influence of effective CPAP to sham CPAP over six weeks on different memory processes in OSA patients. Methods: The study took place in a sleep laboratory and outpatient sleep clinic in a French tertiary-care university hospital. A total of 36 patients with OSA were randomized to receive either CPAP (n = 18) or sham CPAP (n = 18) for six weeks. Interventions were either effective CPAP or non-effective sham CPAP, for six weeks. All patients underwent an extensive battery of tasks evaluating three separate memory systems, before and after treatment. Verbal episodic memory was tested after forced encoding, procedural memory was tested using simplified versions of mirror drawing and reading tests, and working memory was examined with validated paradigms based on a theoretical model. Results: The study subjects were 55 ± 11 years of age and 72.2% were male. The mean body mass index was 29.5 ± 4.1 kg/m2 and the apnea–hypopnea index was 37.1 ± 16.3/h. Prior to treatment, memory performances of OSA patients were altered. In an intention-to-treat analysis, memory deficits were not significantly improved after six weeks of effective CPAP compared to sham CPAP treatment. Verbal episodic, procedural, and working memory scores were comparable between both groups. Conclusion: Using cautious methodology in comparing effective CPAP to sham CPAP and a well-defined set of memory assessments, we did not find improvement in memory performance after six weeks of treatment. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Patients with obstructive sleep apnea (OSA) exhibit neuropsychological impairment such as cognitive dysfunctions [1,2] that could be due to sleep fragmentation as well as intermittent nocturnal hypoxia [3–5]. Indeed, we and others have previously demonstrated that OSA patients had mild but significant memory impairment affecting episodic, procedural, and working memories [6–10]. The memory system is divided into short-term and longterm memories, which, in turn, can be separated into different

ClinicalTrials.gov identifier: NCT00464659. * Corresponding author. Laboratoire EFCR, CHU de Grenoble, CS10217, Grenoble CEDEX 9 38043, France. Fax: 33 476 765 586. E-mail address: [email protected] (M. Joyeux-Faure). http://dx.doi.org/10.1016/j.sleep.2016.06.023 1389-9457/© 2016 Elsevier B.V. All rights reserved.

processes. Long-term memory includes episodic memory (which refers to the recollection of specific experiences) and procedural memory (which refers to learning skills) [11]. Short-term memory is a multicomponent working-memory system that allows the temporary maintenance of limited information available for immediate access by other cognitive processes [12]. The negative impact of OSA on short-term memory is still unclear [10,13]. Working memory also includes many cognitive processes such as storage, processing, supervision, and coordination. Thus, the different memory systems can be affected independently of one another by OSA and could be differently affected by its treatment. The effect of continuous positive airway pressure (CPAP) treatment, the first-line therapy for OSA, on cognitive decline in OSA patients is still debated. About half the studies report that CPAP improves, at least partially, memory impairment in OSA patients. Two randomized controlled trials (RCT) have shown that two to three

M. Joyeux-Faure et al. / Sleep Medicine 24 (2016) 44–50

45

46 patients assessed for elligibility Excluded (n = 6) for not meeting inclusion criteria

40 OSA patients included and randomized 18 patients received sham CPAP

Study discontinuation for withdrawal (n = 6)

6 weeks

Declining to participate (n = 4, 2 in each group)

18 patients received CPAP

6 weeks

Study discontinuation for withdrawal (n = 3)

Intention-to-treat analysis 12 patients completed sham CPAP treatment

15 patients completed CPAP treatment

(3 patients with low sham CPAP compliance, <4 h/night)

(5 patients with low CPAP compliance, <4 h/night)

Fig. 1. Study flow chart. CPAP, continuous positive airway pressure; OSA, obstructive sleep apnea.

weeks of CPAP is insufficient to show a beneficial effect on overall cognitive dysfunction [14,15]. A more recent RCT [16] suggests that two months of CPAP-use in severely obese sleep apnea patients results in only mild improvement of working memory (executive functions). On the other hand, one study showed that 15 days of CPAP treatment is sufficient to normalize attentive, visuospatial learning, and motor performances [17], and in another study, six months of CPAP normalized most neuropsychological deficits in the areas of memory, attention, and executive tasks [18]. We also showed in ten OSA patients that four to six months of CPAP treatment normalized most of the cognitive executive and learning disabilities without modifying short-term memory impairment [19]. However, the variability in study design (and, in particular, the inclusion or exclusion of a control group receiving sham CPAP), sampling methodology, OSA severity, and comorbidities across studies makes their analysis difficult [20,21]. Consequently, a definite demonstration of the CPAP efficacy on memory dysfunction in OSA is still lacking. Moreover, there is a strong variability in tests used across studies, and few studies have explored more than one aspect of memory processes. To test the hypothesis that patients with OSA syndrome would improve memory performances after effective CPAP treatment, we compared verbal episodic memory, visual–motor and reading procedural memory, auditory and spatial working memory, and the capacity to allocate attentional resources in OSA patients randomized to receive either effective or sham CPAP.

Patients were recruited from the Sleep Laboratory at University Grenoble Hospital and the Outpatients Sleep Clinic in a French tertiary-care university hospital (Grenoble, France). Subjects more than 18 years of age who were diagnosed with OSA (apnea– hypopnea index [AHI] > 15/h) on polysomnography (PSG), were naive of CPAP treatment, and gave written informed consent were eligible. The study was conducted following the Consolidated Standards of Reporting Trials (CONSORT) recommendations [22]. Patients were excluded if they declined to participate or were unable to give informed consent. Patients with any of the following were also excluded: severe depressive disorders (Hospital Anxiety and Depression [HAD] score >16), mild intellectual deterioration (Mini Mental State [MMS] score <28), functional failure of the dominant upper limb to achieving graphomotor task, associated oxygen therapy, current pregnancy or lactation, history of stroke, or uncontrolled cancer (Fig. 1).

2. Methods

2.3. Procedures

2.1. Design and setting

2.3.1. Patient visits and treatments At the baseline visit, patients underwent an overnight sleep study. The Epworth Sleepiness Scale (ESS) was completed, and arterial blood gases analysis was performed to exclude obesity hypoventilation syndrome in subjects with a body mass index above 30 kg/m2. Memory evaluation was then performed, and clinical office blood pressure (BP) was measured. Patients were then randomized by an independent statistician to be treated by CPAP or sham CPAP for six weeks. This treatment duration was similar to that used in a previous RCT [23]. Patients

This study performed at Grenoble University Hospital, France, was a randomized, double-blind, parallel-group, sham-controlled trial. It was conducted in accordance with applicable good clinical practice requirements in Europe, French law, ICH E6 recommendations, and ethical principles of the Declaration of Helsinki (South Africa 1996 and Edinburgh 2000). The study was approved by an independent Ethics Committee (Comité de Protection des Personnes, Grenoble, France, IRB0005578) and registered on the ClinicalTrials.gov

site (NCT00464659). Written informed consent was obtained from all patients. An external data quality control was performed systematically for the following criteria: informed consent, complications, adverse events, and case report forms. The primary endpoint was the change in memory function of OSA patients after six weeks of CPAP treatment, in comparison with sham CPAP treatment. 2.2. Patients

46

M. Joyeux-Faure et al. / Sleep Medicine 24 (2016) 44–50

in the CPAP group were equipped with auto-titrating machines (Remstar Auto; Philips Respironics, Murrysville, PA) provided by a home care company (Bastide Medical, France). Pressure was set between 6 and 14 cm of water in the effective CPAP group. Patients receiving sham CPAP had a similar machine delivering a pressure that was too low to suppress sleep respiratory events, as previously validated [24]. Investigators, patients, and the study team were blinded to treatment allocation. At the end of the treatment (second visit), the same parameters were measured to compare effective to sham CPAP treatment. Compliance data from the CPAP and sham CPAP machines were also recorded. 2.3.2. Sleep studies All patients had an overnight sleep study before treatment with CPAP or sham CPAP as previously described [25]. Overnight sleep studies were scored according to standard criteria, and the AHI was calculated from the number of apneas and hypopneas per hour according to international guidelines [26]. 2.3.3. Memory study The memory study is described in detail in an online supplement. All the specific cognitive tests used were chosen based on our previous work evaluating three types of memory and showing which memory processes are affected in OSA patients [10]. Verbal episodic memory was tested using a serial verbal learning task, with control of encoding and recall, according to the modified procedure of Grober et al. [27], as previously described in detail [10]. Subjects were asked to learn a list of 16 words. Each item belonged to a different semantic category and was chosen so that it was not the most prototypic item of its category. Three parameters were measured: the number of recalled words at the first trial (Immediate recall), the total number of recalled words in the three trials (Free recall) and the difference between the number of recalled words on the third trial and the number of recalled words after a delay of 20 minutes (Delayed free recall forgetfulness). Procedural memory was tested using two tasks: a simplified version of a mirror drawing test and a mirror reading test, as previously described [10]. In each test, we assessed learning between the fifth and the first trials (Pattern learning and List learning) and the performance difference between the sixth and the first trials (Procedural learning). On the sixth trial, the frieze or the list was changed, to evaluate the ability to generalize procedural learning to a new pattern of equivalent difficulty. Working memory was examined with two validated paradigms based on a theoretical model. One paradigm requires maintenance and processing of information (Modified Paced Auditory Serial Addition Test (M-PASAT) and Self-ordered spatial memory tasks); the other requires simultaneous work on two types of information (Dual tasks paradigm) [10]. Thus, auditory and spatial working memory was assessed as well as the capacity to allocate attentional resources. Short-term memory was tested using an auditory digit span task. The test procedure was adapted to the subject’s performance with ten series of digits. The length of each series presented depended on the success or failure on the previous series. The average digit span was recorded. For all tests, two parallel task versions were used at baseline and at the six-week visit, and were counterbalanced between patients and between visits. Examiners and patients were blinded to task version allocation. 2.3.4. Other tests The global mental state of patients was measured through the MMS examination [28]. The mood status was evaluated by the HAD

determination [29] and a test of verbal intelligence quotient (IQ), the French National Adult Reading Test (f-NART) [30]. Clinical office blood pressure was measured using an automated mercury sphygmomanometer. The mean of three measurements was calculated on three occasions, in line with European Society of Hypertension–European Society of Cardiology guidelines [31]. Office systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured and office mean arterial blood pressure (MABP) was calculated. 2.4. Statistical analysis Analysis was performed with SAS Version 9.4 software (SAS Institute Inc, Cary, NC, USA). Data were analyzed in intention-totreat (ITT), which included all patients who signed the informed consent form. In the per-protocol analysis, patients with poor treatment compliance (<4 h/night) were excluded. Baseline data were compared by a Wilcoxon–Mann–Whitney test for continuous data and by a χ2 or Fisher exact test for categorical data. For the analysis of data evolution between baseline and six weeks, a repeated-measures two-way analysis of variance (ANOVA) was performed, followed by a Bonferroni post hoc test when necessary. A p value <0.05 was considered significant. Missing data were replaced by an imputation at the median, using the minimum bias method for baseline data and the maximum bias method for data at six weeks. Study design and data are reported here in accordance with the CONSORT criteria [22]. Sample size calculation was based on a previous study from our group [10] showing significant differences in memory performances in OSA patients compared to controls and on the fact that approximately 30% of OSA patients are not compliant with CPAP therapy. Consequently, we estimated that 20 patients in each group would be sufficient to show differences before and after effective or sham CPAP with an α value of 0.05 and a power of 80%. 3. Results 3.1. Trial design and patient characteristics As shown in the study flow chart (Fig. 1), 46 patients were eligible for this study. Six were excluded because they did not meet inclusion criteria, and four refused to participate after inclusion and randomization (two in each group). As a result, 36 participants were randomized to be treated by sham CPAP (n = 18) or by effective CPAP (n = 18). Six patients in the sham CPAP group and three patients in the CPAP group withdrew from the study; all data from these nine patients were considered to be missing data and were imputed, after six months of treatment. Three of 12 patients in the sham CPAP group and five of 15 patients in the CPAP group showed low compliance (<4 hours of use per night). Patient characteristics are shown in Table 1. Key demographics for the study population included the following: mean age 55 ± 11 years, 72.2% male, mean body mass index 29.5 ± 4.1 kg/m2, mean AHI 37.1 ± 16.3/h, mean 24-hour BP (mean 24-hours SBP/DBP: 126 ± 10 / 79 ± 8 mmHg) and office BP (SBP/DBP: 131 ± 12 / 78 ± 7 mmHg). There was no significant difference regarding baseline demographic data, medical history, education, verbal IQ, Hospital Anxiety and Depression (HAD) score, and MMS score between the two groups. Baseline sleep apnea characteristics did not differ between the two groups in terms of AHI, mean nocturnal oxygen saturation (SpO2), and time spent at SpO2 <90%. We noticed only that arousal index was higher and REM sleep time was lower in the CPAP group than in the sham CPAP group. Baseline memory characteristics did not differ between the groups except for dual auditory and tracking tasks data (μ, working memory

M. Joyeux-Faure et al. / Sleep Medicine 24 (2016) 44–50

47

Table 1 Baseline patient characteristics. Characteristic

All patients

Sham CPAP

CPAP

p

No. of subjects (n) Age (y) Male sex, n (%) Education (y) BMI (kg/m2) Verbal IQ score HAD score MMS score History Smoking, n (%) Alcohol, n (%) Diabetes, n (%) Dyslipidemia, n (%) Sleep studies Total sleep time (min) Sleep efficiency (%) Sleep latency (min) Wake after sleep onset (min) Stages 1 + 2 (% of TST) Stages 3 + 4 (% of TST) Stage REM (% of TST) Arousal index (n/h) AHI (n/h) Mean nocturnal SpO2 (%) SpO2 < 90% (%) ESS (score) 24-hour BP SBP (mmHg) DBP (mmHg) MABP (mmHg) Clinic BP SBP (mmHg) DBP (mmHg) MABP (mmHg)

36 55.4 ± 10.6 26 (72.2) 11.5 ± 4.0 29.5 ± 4.1 109.2 ± 8.9 11.6 ± 5.7 28.9 ± 0.9

18 55.8 ± 9.3 13 (72.2) 11.4 ± 4.2 29.5 ± 4.3 110.5 ± 10.1 10.1 ± 5.1 28.9 ± 1.0

18 54.9 ± 12.0 13 (72.2) 11.6 ± 3.9 29.6 ± 4.1 107.8 ± 7.7 13.2 ± 6.0 28.8 ± 0.9

0.89 1.00 0.94 0.96 0.53 0.28 0.28

21 (58) 9 (25) 4 (11) 12 (33)

11 (61) 5 (28) 1 (6) 7 (39)

10 (56) 4 (22) 3 (17) 5 (28)

0.74 1.00 0.60 0.48

398.9 ± 81.4 85.8 ± 10.2 27.7 ± 22.8 62.4 ± 50.4 69.5 ± 9.7 8.7 ± 7.3 22.6 ± 5.8 35.9 ± 15 37.1 ± 16.3 93.0 ± 1.9 7.1 ± 11.6 10.3 ± 3.4

426.9 ± 48.3 89 ± 6.5 26.4 ± 21.2 46.6 ± 38.5 69.2 ± 9.2 7.5 ± 6.5 25.0 ± 5.1 30.5 ± 14.7 32.6 ± 10.5 93.0 ± 1.6 7.0 ± 11.0 10.2 ± 3.6

370.8 ± 98.1 82.6 ± 12.2 28.9 ± 24.9 78.3 ± 56.7 69.8 ± 10.5 10 ± 8.1 20.3 ± 5.6 41.4 ± 13.5 41.5 ± 19.8 93.0 ± 2.2 7.1 ± 12.6 10.4 ± 3.3

0.12 0.16 0.91 0.07 0.95 0.37 0.03 0.01 0.47 0.90 0.64 0.39

126.3 ± 9.8 78.6 ± 7.8 94.5 ± 8.0

125.9 ± 6.2 79.7 ± 5.2 95.1 ± 4.9

126.6 ± 12.6 77.4 ± 9.8 93.8 ± 10.4

0.73 0.36 0.52

131.3 ± 12.0 78.0 ± 7.4 95.8 ± 7.4

130.4 ± 8.0 79.3 ± 6.0 96.3 ± 5.7

132.3 ± 15.2 76.6 ± 8.5 95.2 ± 8.9

0.39 0.06 0.41

Abbreviations: AHI, apnea–hypopnea index; BMI, body mass index; BP, blood pressure; CPAP, continuous positive airway pressure; DBP, diastolic blood pressure; ESS, Epworth Sleepiness Scale; HAD, Hospital Anxiety and Depression; IQ, intelligence quotient; MABP, mean arterial blood pressure; MMS, Mini Mental State; SpO2, oxygen saturation; SpO2 < 90%, percentage of recording time spent at a SpO2 < 90%; SBP, systolic blood pressure; TST, total sleep time. Data are presented as mean ± SD or as number (%) of patients. Analysis of data by Wilcoxon–Mann–Whitney test for continuous data and by a χ2 or Fisher exact test for categorical data. Significant p < 0.05 are in bold.

parameter) at baseline, which was higher in the sham CPAP group than in the CPAP group (Table 2). Before treatment, memory performances of OSA patients (in particular, episodic and working memories) were altered compared to data from control non-OSA subjects from our previous study [10]. 3.2. Primary outcome analysis In ITT analysis, after six weeks of treatment, changes in memory performances of OSA patients were not different between the sham CPAP and CPAP groups (Table 2). Verbal episodic memory was comparable between groups, with no difference in the Immediate recall test results. The Free recall test results were overall improved after treatment in both groups, compared to baseline. Moreover, in both groups, the Delayed free recall forgetfulness test results were altered after treatment compared to baseline, which is certainly due to a ceiling effect. The Procedural memory test results were not different between the sham CPAP and CPAP groups after treatment. The Procedural learning of Mirror drawing test results were altered after treatment in both groups, compared to baseline, again probably due to a ceiling effect. Indeed, at the second visit, patients progressed less than at the first visit, since they had memorized the procedure (their first trial was better). Working memory was also comparable after CPAP or sham CPAP treatment. The μ test (dual auditory + tracking tasks) results were higher in the sham CPAP group than in the effective CPAP group at baseline, but their evolution after treatment was not different

between both groups. M-PASAT and Self-ordered spatial memory test results were improved after treatment in both groups, compared to baseline. Short-term memory was not different between sham CPAP and CPAP groups after treatment. It was overall improved after treatment in both groups, compared to baseline. The ESS score was also comparable after CPAP or sham CPAP treatment and was improved compared to baseline. In the CPAP group, eight patients had an ESS score ≥10 at baseline, which decreased to a normal score ≤9 after treatment in three cases, whereas in the sham CPAP group, four patients had an ESS score ≥10 at baseline, which decreased to a normal score ≤9 after treatment in one case. 3.3. Change in respiratory events In ITT analysis, AHI was significantly lowered by six weeks of CPAP treatment in comparison with the sham CPAP effect, showing the therapeutic efficacy of CPAP (Table 2). After six weeks of sham CPAP, AHI was also significantly decreased, but overall patients still met the criteria for OSA (only two of 12 patients reported AHI <15/h). Treatment compliance was comparable between the sham CPAP (4.4 ± 1.9 h/night) and CPAP (4.5 ± 2.4 h/night) patients. Three of 12 patients in the sham CPAP group and five of 15 in the effective CPAP group were not compliant. The per-protocol analysis confirmed the results of the ITT analysis (data not shown).

48

M. Joyeux-Faure et al. / Sleep Medicine 24 (2016) 44–50

Table 2 Results of memory testing at baseline (J0) and after six weeks (W6) of sham CPAP or effective CPAP treatment. J0 Episodic memory Immediate recall Sham CPAP (n = 18) 15.00 ± 1.57 CPAP (n = 18) 14.94 ± 1.35 Free recall Sham CPAP (n = 18) 30.22 ± 5.04 CPAP (n = 18) 31.22 ± 7.63 Delayed free recall forgetfulness Sham CPAP (n = 18) −0.78 ± 1.93 CPAP (n = 18) −1.11 ± 1.57 Procedural memory Mirror drawing Pattern learning Sham CPAP (n = 18) 107.19 ± 54.30 CPAP (n = 18) 89.56 ± 78.75 Procedural learning Sham CPAP (n = 18) 96.67 ± 54.12 CPAP (n = 18) 85.47 ± 64.11 Mirror reading List learning Sham CPAP (n = 18) 8.44 ± 3.36 CPAP (n = 18) 9.78 ± 9.07 Procedural learning Sham CPAP (n = 18) 4.33 ± 3.82 CPAP (n = 18) 5.83 ± 7.22 Working memory Maintenance and process M-PASAT Sham CPAP (n = 18) 48.72 ± 12.24 CPAP (n = 18) 49.17 ± 10.19 Self-ordered spatial memorya Sham CPAP (n = 18) 14.39 ± 14.79 CPAP (n = 18) 14.33 ± 19.21 Self-ordered spatial strategya Sham CPAP (n = 18) 33.06 ± 9.58 CPAP (n = 18) 35.56 ± 11.78 Dual tasks μ (auditory task + tracking task) Sham CPAP (n = 18) 98.56 ± 12.59 CPAP (n = 18) 86.35 ± 11.88** Short-term memory Auditory span Sham CPAP (n = 18) 5.57 ± 0.74 CPAP (n = 18) 5.69 ± 0.68 Epworth Sleepiness Scale Sham CPAP (n = 18) 8.9 ± 3.2 CPAP (n = 18) 8.9 ± 5.6 AHI Sham CPAP (n = 18) 32.6 ± 10.5 CPAP (n = 18) 41.5 ± 19.8

W6

Change W6–J0

Difference in change (95% CI) (CPAP–sham CPAP)

p Group

p Visit

p Interaction

15.06 ± 1.16 15.17 ± 1.15

0.06 ± 2.01 0.22 ± 1.56

0.17 (−1.39; 1.05)

0.93

36.78 ± 5.94* 36.69 ± 5.66*

6.56 ± 5.39 5.47 ± 5.52

−1.08 (−2.61; 4.78)

0.80

0.28 ± 1.45* −0.39 ± 0.85*

1.06 ± 2.60 0.72 ± 1.99

−0.33 (−1.24; 1.90)

0.12

0.03

0.67

74.08 ± 25.74 86.17 ± 35.16

−33.11 ± 49.27 −3.39 ± 85.00

29.72 (−77.22; 17.77)

0.83

0.12

0.21

45.33 ± 29.93* 41.14 ± 34.26*

−51.33 ± 64.13 −44.33 ± 68.45

7.00 (−51.93; 37.93)

0.51

<0.001

0.75

8.39 ± 2.91 8.44 ± 5.40

−0.06 ± 3.59 −1.33 ± 5.22

−1.28 (−1.76; 4.31)

0.70

0.36

0.40

4.78 ± 2.76 5.22 ± 5.12

0.44 ± 4.66 −0.61 ± 3.93

−1.06 (−1.86; 3.97)

0.52

0.91

0.47

6.22 ± 11.52 2.00 ± 9.73

−4.22 (−3.00; 11.44)

0.54

0.03

0.24

−7.33 ± 11.25 −5.14 ± 11.20

2.19 (−9.80; 5.41)

0.81

<0.01

0.56

31.11 ± 7.65 30.56 ± 11.30

−1.94 ± 9.40 −5.00 ± 11.63

−3.06 (−4.11; 10.22)

0.74

0.06

0.39

92.42 ± 10.39 88.39 ± 13.74**

−6.14 ± 11.43 2.04 ± 12.81

8.19 (−16.41; 0.04)

0.03

0.32

0.05

5.92 ± 0.49* 5.92 ± 0.70*

0.35 ± 0.51 0.22 ± 0.43

−0.13 (−0.19; 0.44)

0.76

<0.001

0.43

6.4 ± 3.7* 6.7 ± 3.7*

−2.6 ± 2.0 −2.3 ± 3.0

0.3 (−2.0; 1.4)

0.92

<0.001

0.74

19.0 ± 14.6* 9.5 ± 7.8*

−13.6 ± 12.2 −32.0 ± 19.7

−18.4 (7.3; 29.5)

0.93

<0.0001

54.94 ± 4.66* 51.17 ± 9.73* 7.06 ± 6.24* 9.19 ± 12.42*

0.65

<0.0001

0.78

0.56

<0.01

Abbreviations: AHI, apnea–hypopnea index; CPAP, continuous positive airway pressure; M-PASAT, Modified Paced Auditory Serial Addition Test. * p < 0.05 vs baseline data. ** p < 0.05 vs data from sham CPAP group at the same visit. a Lower scores indicate better performance. Data are mean ± SD or percentage. Analysis of data by repeated-measures two-way analysis of variance, followed by a Bonferroni post hoc test when necessary. Significant p < 0.05 are in bold.

4. Discussion In this RCT, although six weeks of effective CPAP improved sleep apnea, it did not improve the memory processes in OSA patients, compared to sham CPAP. The originality of our study is that we explored the effects of CPAP on not one but several mnesic processes known to be altered in OSA patients [10]. The recent literature shows discrepancies regarding CPAP effect on memory processes in OSA patients. About half the studies report that CPAP improved, at least partially, memory impairment of OSA patients after two weeks [17] as well as six months [18,19] and ten years [32]. These studies, however, did not include a control group on sham CPAP. Results from RCTs using sham CPAP have been summarized in two systematic reviews and meta-analyses [33,34]. In

four studies including 174 patients, the CPAP effect on cognitive function in OSA patients is overall limited. No changes in any specific cognitive domain have been reported after one or two weeks of CPAP, in terms of verbal and spatial episodic memory as well as auditory short-term memory [14,35]. Another more recent RCT from the same team, using the same tests, showed that three weeks of CPAP had no overall beneficial cognitive effects except for one attention test [15]. Moreover, Barbé et al. reported no changes in attention, vigilance, verbal episodic memory, and short memory, after six weeks of CPAP treatment in severe and nonsleepy OSA patients [23]. In accordance with these previous studies, we showed here that memory impairment in OSA patients was not improved by six weeks of CPAP, in terms of verbal episodic, procedural, working memory, and shortterm memory. Taken together, these results suggest that a short CPAP

M. Joyeux-Faure et al. / Sleep Medicine 24 (2016) 44–50

treatment duration (one to six weeks) has no beneficial effect on memory processes. However, the recent randomized controlled APPLES study, in more than 1100 OSA patients, showed that a longer exposure (two months) improved working memory but only in severe OSA patients [16]. However, this beneficial effect was observed for only one test and was not present after six months of CPAP. Indeed, no effect was observed after two or six months of CPAP for verbal episodic memory, whereas vigilance was significantly improved in severe OSA patients. Finally, another RCT, using a functional magnetic resonance imaging paradigm and working memory test, reported a positive response in Task Positive and Default Mode networks after two months of CPAP, yet without complete recovery of coexisting behavioral and neuronal deficits [36]. We observed that memory performance in patients was overall improved compared to baseline conditions after six weeks of CPAP or sham CPAP. This improvement was therefore independent of respiratory disturbance correction and could be due to a global nonspecific gain as well as to a task procedural learning at the second visit, even if test versions used were different from those at baseline. Our study has several limitations. Although all OSA patients included in the study had an AHI >15/h, with a mean baseline AHI of 37/h, 44% of patients included had mild OSA (15 < AHI < 30/h) and could have been less responsive to CPAP than more severely apneic patients. Moreover, few patients were sleepy at baseline (36%), with a mean ESS score for the group within normal range (Table 1). The small sample size of our study population represents another limitation of this study as we were not able to stratify the analysis of the CPAP effect on memory according to baseline OSA severity or sleepiness. Finally, patients’ attention and alertness are known to influence memory processes, and it would have been interesting to measure these parameters in our study with tests other than the ESS. Memory impairment may be related to sleepiness severity before treatment and its improvement on CPAP. In the APPLES study [16], CPAP treatment improved sleepiness, especially in individuals with severe OSA. In a subgroup of patients who were sleepy at baseline, working memory improvement after two months of CPAP was correlated with change in objective sleepiness, suggesting that sleepiness may be associated with one domain of OSA-related neurocognition. Moreover, we have previously shown that up to 10% of patients do not normalize sleepiness and fatigue despite adequate CPAP use [37]. CPAP improves symptoms in the whole population, but to a lesser extent in patients with residual excessive sleepiness (ESS score ≥11). Residual symptoms are not limited to sleepiness, suggesting a true “CPAP-resistance syndrome,” which could include persistent memory impairment as well. OSA severity may be important for detecting improvement in neurocognitive outcomes. As measures of disease severity, both AHI [1,38] and oxygen saturation [6,39] have been previously implicated in the etiology of the OSA-associated neurocognitive dysfunction. Memory impairment could result from brain hypoxic damages and thus would not be reversible despite effective CPAP treatment. However, the APPLES study [16] showed that CPAPinduced improvement of working memory only in severe OSA patients but not in mild and moderate OSA patients. This discrepancy could be explained by the cognitive reserve theory: individual differences in how the brain processes tasks may allow some, but not others, to cope with greater insult by using pre-existing cognitive processes or by enlisting compensatory processes before performance is detrimentally affected [40]. 5. Conclusions This study showed that six weeks of CPAP treatment did not improve memory performance in OSA patients in terms of verbal episodic memory, visual–motor, and reading procedural memory

49

as well as auditory and spatial working memory, which are known to be altered in these patients. Memory impairment may be related to sleepiness severity before treatment and its improvement on CPAP. It could also result from brain hypoxic irreversible damage. Finally, further investigations with longer CPAP exposures (≥2 months) or in selective populations (such as severe or non-severe OSA patients) would be of interest. That suggests the existence of a complex OSA–neurocognitive relationship, and that clinicians should consider disease severity, sleepiness, and individual differences in managing their patients with CPAP. Conflict of interest The authors declare that they have no conflicts of interest. The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2016.06.023. Acknowledgements The authors are grateful to Fond de dotation AGIR pour les maladies chroniques; the homecare company Bastide Medical, which provided unrestricted funding for the study; and Philips Respironics, which provided sham CPAP devices. These sponsors had no role in the study design, data analysis, and study conclusions. The authors are also grateful to Sleep Laboratory technicians Martine Selek and Corinne Loiodice for their collaboration during the sleep studies; Marion Perrin for statistical analyses (INSERM U1042, HP2 Laboratory, Grenoble University Hospital, Grenoble, France); and Emilie Granchamp (Pedopsychiatry Department, Grenoble University Hospital, Grenoble, France) and Caroline Poulet (Grenoble Alpes University, Grenoble, France) for data collection. Appendix: Supplementary material Supplementary data to this article can be found online at doi:10.1016/j.sleep.2016.06.023. References [1] Greenberg GD, Watson RK, Deptula D. Neuropsychological dysfunction in sleep apnea. Sleep 1987;10:254–62. [2] Beebe DW, Groesz L, Wells C, et al. The neuropsychological effects of obstructive sleep apnea: a meta-analysis of norm-referenced and case-controlled data. Sleep 2003;26:298–307. [3] Bedard MA, Montplaisir J, Richer F, et al. Obstructive sleep apnea syndrome: pathogenesis of neuropsychological deficits. J Clin Exp Neuropsychol 1991;13:950–64. [4] Van Dongen HP, Maislin G, Mullington JM, et al. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 2003;26:117–26. [5] Verstraeten E, Cluydts R, Pevernagie D, et al. Executive function in sleep apnea: controlling for attentional capacity in assessing executive attention. Sleep 2004;27:685–93. [6] Findley LJ, Barth JT, Powers DC, et al. Cognitive impairment in patients with obstructive sleep apnea and associated hypoxemia. Chest 1986;90:686–90. [7] Decary A, Rouleau I, Montplaisir J. Cognitive deficits associated with sleep apnea syndrome: a proposed neuropsychological test battery. Sleep 2000;23:369–81. [8] Rouleau I, Decary A, Chicoine AJ, et al. Procedural skill learning in obstructive sleep apnea syndrome. Sleep 2002;25:401–11. [9] Thomas RJ, Rosen BR, Stern CE, et al. Functional imaging of working memory in obstructive sleep-disordered breathing. J Appl Physiol 2005;98:2226–34. [10] Naegele B, Launois SH, Mazza S, et al. Which memory processes are affected in patients with obstructive sleep apnea? An evaluation of 3 types of memory. Sleep 2006;29:533–44. [11] Cohen NJ, Eichenbaum H, Deacedo BS, et al. Different memory systems underlying acquisition of procedural and declarative knowledge. Ann N Y Acad Sci 1985;444:54–71. [12] Baddeley A. The fractionation of working memory. Proc Natl Acad Sci USA 1996;93:13468–72. [13] Greneche J, Krieger J, Bertrand F, et al. Effect of continuous positive airway pressure treatment on short-term memory performance over 24 h of sustained

50

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22] [23]

[24] [25]

[26]

M. Joyeux-Faure et al. / Sleep Medicine 24 (2016) 44–50

wakefulness in patients with obstructive sleep apnea-hypopnea syndrome. Sleep Med 2013;14:964–72. Lim W, Bardwell WA, Loredo JS, et al. Neuropsychological effects of 2-week continuous positive airway pressure treatment and supplemental oxygen in patients with obstructive sleep apnea: a randomized placebo-controlled study. J Clin Sleep Med 2007;3:380–6. Lee IS, Bardwell WA, Kamat R, et al. A model for studying neuropsychological effects of sleep intervention: the effect of 3-week continuous positive airway pressure treatment. Drug Discov Today Dis Models 2011;8:147–54. Kushida CA, Nichols DA, Holmes TH, et al. Effects of continuous positive airway pressure on neurocognitive function in obstructive sleep apnea patients: the Apnea Positive Pressure Long-term Efficacy Study (APPLES). Sleep 2012;35: 1593–602. Ferini-Strambi L, Baietto C, Di Gioia MR, et al. Cognitive dysfunction in patients with Obstructive Sleep Apnea (OSA): partial reversibility after Continuous Positive Airway Pressure (CPAP). Brain Res Bull 2003;61:87–92. Bedard MA, Montplaisir J, Malo J, et al. Persistent neuropsychological deficits and vigilance impairment in sleep apnea syndrome after treatment with continuous positive airways pressure (CPAP). J Clin Exp Neuropsychol 1993;15:330–41. Naegele B, Pepin JL, Levy P, et al. Cognitive executive dysfunction in patients with obstructive sleep apnea syndrome (OSAS) after CPAP treatment. Sleep 1998;21:392–7. Kielb SA, Ancoli-Israel S, Rebok GW, et al. Cognition in Obstructive Sleep Apnea-hypopnea Syndrome (OSAS): current clinical knowledge and the impact of treatment. Neuromolecular Med 2012;14:180–93. Ferini-Strambi L, Marelli S, Galbiati A, et al. Effects of continuous positive airway pressure on cognitition and neuroimaging data in sleep apnea. Int J Psychophysiol 2013;89:203–12. Calvert M, Blazeby J, Altman DG, et al. Reporting of patient-reported outcomes in randomized trials: the CONSORT PRO extension. JAMA 2013;309:814–22. Barbe F, Mayoralas LR, Duran J, et al. Treatment with continuous positive airway pressure is not effective in patients with sleep apnea but no daytime sleepiness: a randomized, controlled trial. Ann Intern Med 2001;134:1015–23. Farre R, Hernandez L, Montserrat JM, et al. Sham continuous positive airway pressure for placebo-controlled studies in sleep apnoea. Lancet 1999;353:1154. Mendelson M, Vivodtzev I, Tamisier R, et al. CPAP treatment supported by telemedicine does not improve blood pressure in high cardiovascular risk OSA patients: a randomized, controlled trial. Sleep 2014;37:1863–70. Berry RB, Budhiraja R, Gottlieb DJ, et al. Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and

[27] [28]

[29] [30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38] [39]

[40]

Associated Events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med 2012;8:597–619. Grober E, Buschke H, Crystal H, et al. Screening for dementia by memory testing. Neurology 1988;38:900–3. Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–70. Mackinnon A, Ritchie R, Mulligan R. The measurement properties of a French language adaptation of the National Adult Reading Test. Int J Methods Psychiatr Res 1999;8:27–38. Mancia G, De Backer G, Dominiczak A, et al. 2007 Guidelines for the management of arterial hypertension: the task force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2007;25:1105–87. Crawford-Achour E, Dauphinot V, Martin MS, et al. Protective effect of long-term CPAP therapy on cognitive performance in elderly patients with severe OSA: the PROOF study. J Clin Sleep Med 2015;11:519–24. Sanchez AI, Martinez P, Miro E, et al. CPAP and behavioral therapies in patients with obstructive sleep apnea: effects on daytime sleepiness, mood, and cognitive function. Sleep Med Rev 2009;13:223–33. Kylstra WA, Aaronson JA, Hofman WF, et al. Neuropsychological functioning after CPAP treatment in obstructive sleep apnea: a meta-analysis. Sleep Med Rev 2013;17:341–7. Bardwell WA, Ancoli-Israel S, Berry CC, et al. Neuropsychological effects of one-week continuous positive airway pressure treatment in patients with obstructive sleep apnea: a placebo-controlled study. Psychosom Med 2001;63:579–84. Prilipko O, Huynh N, Schwartz S, et al. The effects of CPAP treatment on task positive and default mode networks in obstructive sleep apnea patients: an fMRI study. PLoS ONE 2012;7:e47433. Gasa M, Tamisier R, Launois SH, et al. Residual sleepiness in sleep apnea patients treated by continuous positive airway pressure. J Sleep Res 2013;22:389– 97. Naegele B, Thouvard V, Pepin JL, et al. Deficits of cognitive executive functions in patients with sleep apnea syndrome. Sleep 1995;18:43–52. Quan SF, Wright R, Baldwin CM, et al. Obstructive sleep apnea-hypopnea and neurocognitive functioning in the Sleep Heart Health Study. Sleep Med 2006;7:498–507. Stern Y. Cognitive reserve. Neuropsychologia 2009;47:2015–28.