INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Continuous positive airway pressure (CPAP) is the treatment of choice for patients with the sleep apnea/hypopnea syndrome (SAHS) (1). Randomized crossover studies have demonstrated that objective daytime sleepiness, cognitive performance, and quality of life are all improved by CPAP therapy (2, 3). However, in a few cases, patients continue to complain of excessive daytime sleepiness, poor concentration, and a decreased quality of life, despite effective CPAP use.

Modafinil, 2-[(diphenylmethyl)-sulfinyl] acetamide, is a novel wake-promoting agent that is currently licensed in the United Kingdom to treat the excessive daytime sleepiness found in the sleep disorder narcolepsy. Previous placebo-controlled trials have found modafinil to reduce objective excessive daytime sleepiness in narcoleptic patients (4), sleep- deprived normal subjects (9), a SAHS animal model (10), and patients with untreated sleep apnea (11, 12). Studies in healthy volunteers (13) and narcoleptics (7, 8) have demonstrated a low level of adverse effects of modafinil. In addition long-term modafinil use demonstrates no tolerance or drug dependence in narcoleptics (14).

The use of modafinil as an adjunct therapy in patients with SAHS who remain sleepy during CPAP has not been established. The aim of the current investigation was to examine the efficacy and safety of modafinil on this group of patients with SAHS who remain objectively and subjectively sleepy, despite effective CPAP usage.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Suitable study patients were identified from consecutive patients attending the Edinburgh Sleep Center outpatient clinic. Patients were identified as "suitable" if they had an Epworth Sleepiness Scale (ESS) (15)  11 during CPAP and a CPAP run time of > 3.5 h/night. These patients were then approached about the study and screened for suitability. Inclusion criteria were: age, 18 to 70 yr; documented diagnosis and treatment for SAHS; SAHS diagnosis defined as an apnea/hypopnea index (AHI) of 15 breathing pauses per hour slept during polysomnography, or more than 30 breathing pauses per hour in bed using a limited sleep study recorder (EdenTrace, Bicester, UK) or AutoSet (ResMed, Abingdon, UK); average CPAP run time of 3.5 h/night since the previous clinic visit; have received CPAP therapy for at least 4 wk; an ESS  11; written informed consent obtained; availability to attend sleep center for required study visits.

Exclusion criteria included: narcolepsy, pregnancy, potentially pregnant or lactating women; moderate to severe hypertension (diastolic blood pressure > 100 mm Hg); arrhythmias, angina within the previous 3 mo, significant cardiac, respiratory, or cerebrovascular diseases, CNS medication within the previous 2 wk, sedating medication or body mass index (BMI)  45. Informed consent was obtained from all participants, and the study was approved by the local Ethics Committee. Patients who fulfilled all the screening criteria were transferred onto CPAP machines to objectively monitor CPAP compliance (mask run time: Sullivan V Elites; ResMed).

Study Design

The study was a 7-wk, single center, randomized, double-blind, placebo-controlled crossover trial (Table 1). Patients who passed the screening criteria then underwent a 2-wk baseline period, during which effective CPAP compliance was monitored. A baseline assessment occurred at the end of this 2-wk period. The baseline assessment consisted of overnight polysomnography and a session of daytime function assessments. To proceed to randomization after the baseline day, patients had to continue to satisfy the inclusion and exclusion criteria, still have an ESS of 11 or more, a Multiple Sleep Latency Test (MSLT) (16) sleep onset latency of 10 min or less, fewer than 15 breathing pauses per hour slept, and an effective CPAP use of 4 h/ night on 10 of the 14 nights.

At the end of the baseline day, all consecutive suitable patients were randomly assigned to one of two possible treatment sequences. Each patient received placebo or 400 mg modafinil for 2 wk (Treatment Period 1), followed by a 1-wk washout period, and then 400 mg modafinil or placebo for the final 2-wk period (Treatment Period 2) (Table 1). The patient and study investigators were blind to the randomization schedule. Patients were evaluated for outcome measures at baseline and at the end of the two treatment periods.

Study Drug Administration

Sixty tablets were supplied by Cephalon Inc. (West Chester, PA) in plastic medicine containers (60 × 100 mg modafinil tablets or matching placebo tablets). Patients were given a 2-wk supply (plus 14 additional tablets) at baseline and at Week 3. The treatment regime was as follows: two tablets were taken orally at breakfast for the first 5 d (placebo or 200 mg modafinil) and then four tablets were taken at breakfast for 9 d (placebo or 400 mg modafinil). Patients were requested to return all unused medication at the end of each treatment period for subsequent compliance tablet counts. Percentage compliance was calculated using the following calculation: (tablets taken/expected tablets taken) * 100, where tablets taken = total tablets minus returned tablets.

Overnight Polysomnography

Patients underwent three polysomnographies, these being on the nights before the day of assessments at baseline, end of Treatment Period 1, and end of Treatment Period 2. Full polysomnography was performed using our standard techniques (17). Sleep was monitored by electroencephalography (EEG; CZ-PZ), electrooculography (EOG), and submental electromyography (EMG). Thoracic and abdominal respiratory movements were measured by inductance plethysmography, and arterial oxygen saturation was measured using pulse oximetry. Electrocardiogram (ECG), microphone snoring sounds, and right and left leg movements (anterior tibialis) were also monitored. In addition, patients wore their CPAP masks and machines were set at their prescribed pressure. The CPAP pressures were determined previously on an overnight CPAP trial after diagnosis of SAHS. All signals were recorded on a computerized system (Compumedics S, Victoria, Australia) using a 16 channel polygraph configuration.

Off-line analysis. Sleep stages were manually scored using the standard scoring guidelines of Rechtschaffen and Kales (18). Hypopneas were defined as a 50% reduction in thoracoabdominal movement for a minimum of 10 s (17). As patients were receiving CPAP, no airflow signal was monitored; therefore, distinguishing between apneas and hypopneas was not possible. The total number of respiratory events was divided by total sleep time (TST) to give the number of breathing pauses per hour slept. Microarousals were scored using the Cheshire definition (19) of a return to theta or alpha on the EEG for more than 1.5 s, with a concurrent rise in EMG tone, however brief.

Efficacy Measures

Outcome measures were assessed at baseline and at the end of Treatments 1 and 2 and consisted of a whole day of daytime function assessments. Tests were chosen to measure a wide range of functions affected by SAHS (2). Patients were instructed to withdraw from caffeine on the evening prior to the daytime testing, and throughout the test day decaffeinated drinks were provided.

Subjective sleepiness. The primary outcome measure was the patient's subjective assessment of their sleepiness using the ESS (15). This questionnaire was adapted so that patients were instructed to imagine their sleepiness in certain situations during the previous 2 wk instead of during the previous month, as is the usual case. The total score from the eight everyday situations was taken as the outcome measure.

Objective sleepiness. Secondary outcome measures were the mean sleep onset latencies (SOL) on the MSLT (16) and the Maintenance of Wakefulness Test (MWT) (20). All naps were terminated after two consecutive epochs of any stage of Rechtschaffen and Kales sleep (18), or if no sleep occurred, after 20 min. The SOL was determined from "lights out" until the first 30-s epoch of any stage of sleep (21). The mean SOL was calculated from the average of the four MSLT naps (performed at 10:00 A.M. and at 12:00, 2:00 and 4:00 P.M.) and the four MWT naps (performed at 11:00 A.M. and at 1:00, 3:00, and 5:00 P.M.), respectively, across the day.

Effective CPAP use. CPAP machines were downloaded at the end of both treatment sessions onto a personal computer, and the mean effective usage over the treatment period was calculated (h/night).

Quality of life. Two self-assessment questionnaires, the 36-item Short-Form health survey (SF-36) (22) and the Functional Outcomes of Sleep Questionnaire (FOSQ) (23) were used to measure self-reported quality of life. These questionnaires were amended to ask the patients to refer to the previous 2 wk.

Cognitive performance. Two computerized performance tasks Steer Clear and a simple unprepared response time (SURT) task were carried out. Steer Clear (24) was run for 30 min and the outcome measure was the number of obstacles hit. SURT (25) was run for 20 min and the outcome measures were the 95th centile of reaction time and mean reciprocal reaction time (second). In addition, Trail Making (26) and Paced Auditory Serial Addition Task (PASAT) (27) were also performed. Time to completion (seconds) was the outcome measure for Trail Making, and number of correct additions for PASAT.

Global evaluation. A global evaluation of the patient's condition was rated by the investigator at the end of both treatment period assessments. The investigator had to rate the patient's condition as "better," "no change," or "worse" than the baseline assessment day.

Measures of Safety

Adverse events. These were defined as untoward medical occurrences that did not necessarily have to have a causal relationship with study treatment. Adverse events were described in terms of seriousness, action taken regarding both study medication and other treatment, outcome, and likely relation to study drug. In addition, patients were advised to contact the Sleep Center if they experienced anything they were concerned about. The occurrence of adverse events were reported, discussed, and logged at each of the three assessment visits. In addition, any changes in medical history or concomitant medication were also discussed and logged at each assessment.

Clinical tests. Routine hematology (full blood count), blood chemistry (liver function tests, urinalysis, and electrolytes), and urine test (drug screen) were performed at each of the three assessments. In addition, BMI, sitting blood pressure, and a 12-lead electrocardiogram were performed at each assessment visit.

Statistical Analyses

All daytime outcome variables were compared for treatment effects using paired t tests. In addition, the primary outcome measure of the ESS was compared for both treatment effect and order effect using a two-way ANOVA. The between-group differences in adverse events were analyzed using the Fisher's two-sided exact test. A probability value of less than 0.05 was accepted as statistically significant. All patients who received any dose of study drug were included in the safety analyses (n = 31), and all those who completed the study (n = 30) were included in the efficacy analyses.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Forty-four patients fulfilled the criteria at the screening visit and consented to take part in the study. Twelve of these patients were not randomized: four withdrew consent between screening and baseline assessments and eight were classified as "baseline failures" because they no longer fulfilled study criteria at the baseline assessment (one had > 15 breathing pauses/h during CPAP, three had MSLT SOL > 10 min, two changed medications, one had CPAP compliance < 4 h/night, and one had ESS < 11). Thirty-two patients were randomized and 30 (27 men) completed the study (Table 2). One withdrew consent after randomization but prior to taking any study medication, and the second patient was withdrawn because of adverse events during Treatment Period 2 while receiving modafinil.

Efficacy Measures

Subjective sleepiness. There was no significant improvement in ESS with modafinil (Table 3 and Figure 1). In addition, no significant carry-over effect was seen with this primary outcome measure (p > 0.2).

View larger version (19K): [in this window] [in a new window]   Figure 1.   A comparison of the Epworth Sleepiness Scale on placebo versus modafinil. Mean values shown as horizontal lines.

Objective sleepiness. There was a significant improvement in daytime sleepiness as measured by the MWT (Table 3 and Figure 2). However, there was no significant improvement in MSLT sleep onset latency with modafinil (Table 3 and Figure 3).

View larger version (16K): [in this window] [in a new window]   Figure 2.   A comparison of the mean sleep onset latency on the MWT on placebo versus modafinil. MWT = Maintenance of Wakefulness Test. Mean values shown as horizontal lines.

View larger version (21K): [in this window] [in a new window]   Figure 3.   A comparison of the mean sleep onset latency on the MSLT on placebo versus modafinil. MSLT = Multiple Sleep Latency Test. Mean values shown as horizontal lines.

Effective CPAP use. CPAP use was significantly lower in subjects receiving modafinil than in those receiving placebo (6.3 ± 1h/night modafinil; 6.5 ± 1 placebo, p = 0.03) (Figure 4).

View larger version (22K): [in this window] [in a new window]   Figure 4.   A comparison of effective CPAP use on placebo versus modafinil. CPAP = continuous positive airway pressure. Mean values shown as horizontal lines.

Quality of life. There were no significant improvements in quality of life measures with modafinil (Table 3). However, there was a trend for patients to score higher on the vigilance subscale of the FOSQ while receiving modafinil (2.9 ± 0.7 modafinil, 2.7 ± 0.8 placebo; p = 0.06). There was also a trend for patients to score higher on the "role limitations-physical" dimension of the SF-36 with modafinil (75 ± 37, 63 ± 40; p = 0.06).

Cognitive performance. There were no significant improvements in cognitive function measures with modafinil (Table 3).

Global evaluation. There was no significant difference in the investigator's rating of the patient's condition after receiving modafinil than after receiving placebo (p = 0.4).

Compliance with medication. Tablet compliance was marginally better with modafinil than with placebo (99.3 ± 2.7% modafinil, 97.3 ± 5.2% placebo; p = 0.047); however, compliance on both treatments was high.

Overnight Polysomnography

There were no significant differences in respiratory events (7 ± 4/h slept modafinil; 7 ± 4 placebo; p = 0.5), arousal frequency (26 ± 8/h slept; 27 ± 7; p = 0.5), or sleep efficiency (81 ± 10%, 79 ± 10; p = 0.4) between modafinil and placebo while receiving CPAP.

Measures of Safety

Adverse events. One patient withdrew from the study because of nonserious adverse events while receiving modafinil (dry cough, difficulty swallowing, lack of energy, difficulty concentrating). There were significantly more patients reporting side effects with modafinil than with placebo (23, 14; p = 0.04). However, there were no statistical differences for any individual side effect (Table 4). There were no serious adverse events or deaths reported during the current trial.

Clinical laboratory tests. There was no increased frequency of abnormal laboratory findings during the administration of modafinil (seven receiving placebo and six receiving modafinil). None was of major clinical significance.

Vital signs. There was no significant effect of modafinil on systolic (133 ± 15 mm Hg modafinil, 131 ± 16 placebo; p = 0.3) or diastolic (82 ± 10 mm Hg, 83 ± 10; p = 0.6) blood pressure.

Electrocardiogram. There were no new changes in ECG during the treatment periods.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study examined the effect of modafinil on sleepy patients with SAHS who were receiving CPAP, and found it had no therapeutic effect on subjective or objective sleepiness as measured by the MSLT, but it did significantly increase the ability to remain awake as assessed by the MWT. There were significantly more side effects with modafinil than with placebo. It is unclear whether the benefit is sufficient to justify the use of modafinil in such patients in clinical practice.

Previous randomized controlled studies in narcoleptics have reported that modafinil produced significant improvements in daytime sleepiness as measured by the ESS, the MSLT, and the MWT (4, 7, 8). Unlike these studies in narcoleptics, the current study found only improvements in the MWT measure of alertness. This may be because the narcoleptics (7, 8: data combined) were much sleepier at baseline (ESS mean range, 17.1 to 18.3; MSLT, 2.2 to 3.3 min, MWT, 5.8 to 6.6 min) than the current patients with SAHS (ESS, 15; MSLT, 6.9 min; MWT, 16.5 min). Modafinil might, therefore, be more effective in the sleepier narcoleptic patients, as there is greater scope for a treatment effect, or, alternatively, modafinil may have a greater effect on the narcoleptic brain. However, an earlier study in an untreated sample of patients with SAHS (12) found treatment-related improvements in MSLT (from 9 min receiving placebo to 15 min receiving modafinil) when given modafinil for two consecutive days. This suggests that modafinil can work in patients with SAHS and that the lack of change in the current study may reflect the relative lack of sleepiness at baseline.

We found that modafinil caused significant improvements in MWT, but not in MSLT or ESS measures. There are a number of possible explanations for this. Placebo effects were observed with the MSLT and ESS, but not with the MWT (Table 3), which could explain the greater probability of detecting treatment-related differences with the MWT. Previous work by Sangal and colleagues (28) also suggested that the MWT is more sensitive to treatment-related changes than the MSLT. In addition, it can be argued that the physiological function in the MWT of "trying to stay awake" is more relevant to everyday life than the ability to "try and fall asleep." Thus, our findings are relevant to tasks such as overcoming residual sleepiness at work, or when driving, in patients receiving CPAP.

In contrast to previous studies in narcoleptics (4, 7, 8), we found no significant improvements in ESS with modafinil in the study population. Earlier studies on patients with untreated SAHS (11, 12) did not measure ESS. Our data would suggest that patients with SAHS had no perceived alerting effects of modafinil. This is in accordance with the lack of improvement in global evaluation by the investigator.

Despite our efforts to encourage patients to use their CPAP "as normal" on both treatment limbs, patients used their machines significantly less while receiving modafinil. CPAP improves daytime function (2, 3). It follows that lower CPAP usage on the modafinil treatment limb will reduce the likelihood of finding any modafinil-related efficacy improvements. In contrast, CPAP usage on the night in the sleep laboratory prior to the daytime assessments at the end of each treatment limb, was not significantly different (8.0 ± 0.6 h modafinil, 8.0 ± 0.5 h placebo; p = 0.9). Decreased CPAP use is concerning as CPAP not only improves alertness but also mood, quality of life (2, 29), driving ability (30), and blood pressure (31). This study showed an average decrease in CPAP use of 12 min on modafinil, and although this is relatively small, it did occur in a carefully monitored situation where optimal CPAP use was being actively encouraged. Further studies are needed to determine whether this is a greater problem in clinical practice. It was reassuring that there was no difference in blood pressure between the placebo and modafinil limbs.

Modafinil had no treatment effect on quality of life in sleep apneics. This is in contrast to studies of narcoleptics, where the SF-36 dimensions of physical and emotional role limitations, vitality, social functioning, and the mental components summary, were all found to be significantly improved by modafinil (32).

Similarly, we found no improvements in cognitive function. Both reaction time (12) and long-term memory (11) have been demonstrated to improve with modafinil in patients with SAHS who had not been treated with CPAP. In addition, modafinil has been shown to reduce reaction time errors and gaps in narcoleptic patients (33).

In agreement with a previous study in untreated sleep apneics (11), modafinil had no effect on the overnight disease severity, microarousals, or sleep efficiency. This study demonstrated very minor adverse events, and very few abnormal laboratory findingsan observation consistent with previous studies in narcoleptics (7, 8, 13).

Limitations in the current study include study power and the short duration of treatment. The study was designed to give a 90% power at the 5% level to detect a 4-unit change in ESS based on data from previous studies (2, 7, 29). However, as is inevitable using a drug in a new area, this involved assumptions, and these may have resulted in an underestimate of the number of patients needed. It is therefore possible that some of the trends in daytime function with modafinil treatment may have been statistically significant with a larger sample size, but their clinical significance would be likely to be relatively small; however, further larger, longitudinal studies are required. Nevertheless, our failure to demonstrate any significant symptomatic improvement in sleepiness and finding only a minor change in objective alertness suggests the role of modafinil in the treatment of patients with SAHS with persisting sleepiness during CPAP may be limited. The 2-wk study limbs were chosen because they were long enough to demonstrate improvements in symptoms and objective sleepiness based on the rapid onset of efficacy and 15-h drug half life (13) while also keeping study duration low, as safety of modafinil in this patient population was unknown. Although we believe the 2-wk limbs were sufficient to achieve steady-state effects on objective sleepiness and cognitive performance, it is possible that the relatively short duration might have meant some patients had not experienced the situations described in the ESS. This might have added to the variance in the responses and contributed to the nonsignificant effect of modafinil on this primary outcome variable. This may also have been the case for the quality of life questionnaires. However, Broughton and colleagues (4) demonstrated significant improvements in the ESS with modafinil, where the Epworth scale had been adapted to ask patients to base their answers on the previous 7 d. Our patients received 200 mg modafinil for 5 d and 400 mg for 9 d. It could be argued that they had effective therapy for only 9 d, but 200 mg is virtually as effective as 400 mg, at least in narcoleptics (4, 7, 8). Indeed the increased frequency of minor side effects with modafinil compared with placebo suggests higher dosages might not be beneficial, although this would have to be tested. Other potentially confounding factors include the possibly diverse causes of daytime sleepiness in the patients studied. This group was being effectively treated with CPAP, with a mean AHI while receiving CPAP of 7/h slept, yet they still felt sleepy. A large number of the normal population, 30% of men in the study of Young and colleagues (34), have AHIs of > 5/h slept. Thus, the relationship between this level of AHI and sleepiness is unlikely to be causal, although persisting resistive events (35) cannot be excluded as a contributor. CPAP improves daytime sleepiness (2), but does not normalize it. Perhaps, some patients with SAHS who have been falling asleep readily for years retain this habit and ability even once effectively treated with CPAP. Perhaps, some have psychological causes for their sleepiness that are not influenced by modafinil.

The results of this study suggest that modafinil may have a limited place as an adjunct therapy in those patients who are sleepy despite using their CPAP machines. However, it would be an inappropriate first-line therapy in sleep apnea, as it does not treat the underlying upper airway collapse or the resulting arousals and blood pressure changes.