Residual excessive daytime sleepiness in patients treated for obstructive sleep apnea: guidance for assessment, diagnosis, and management

ABSTRACT

Excessive daytime sleepiness (EDS) affects approximately half of patients with obstructive sleep apnea (OSA) and can persist in some despite normalization of breathing, oxygenation, and sleep quality with primary OSA therapy, such as continuous positive airway pressure (CPAP). EDS is often overlooked and under discussed in the primary care setting and in the follow-up of CPAP-treated patients due to difficult assessment of such a multi-dimensional symptom. This review aims to provide suggestions for procedures that can be implemented into routine clinical practice to identify, evaluate, and manage EDS in patients treated for OSA, including how to appropriately use various self-report and objective assessments along the clinical pathway and options for pharmacotherapy. In addition, examples of when it is appropriate to refer a patient to a sleep specialist for evaluation are discussed.

KEYWORDS:

• Excessive daytime sleepiness
• assessment
• pharmacotherapy
• clinical practice
• sleep medicine

1. Introduction

1.1. Prevalence of excessive daytime sleepiness in obstructive sleep apnea

The prevalence of obstructive sleep apnea (OSA) continues to increase, likely due, in part, to the growing obesity epidemic and heightened awareness by the public and healthcare providers. Recent estimates suggest that nearly 1 billion adults worldwide have OSA [1]. Excessive daytime sleepiness (EDS) is a key symptom of OSA, but estimations of its prevalence have been inconsistent [2]. Some reports approximate that 50 to 80% of patients with OSA report EDS prior to initiating therapy [3,4], whereas others claim that up to 60% do not report EDS [5]. These discrepancies could be due to a number of reasons, such as under-recognition in the clinic due to the multidimensional nature of EDS and consequential ambiguity in patient reporting of symptoms, prevalence overlap with other comorbid etiologies of EDS, insufficient screening questionnaires, and low agreement among various EDS assessment tools [6]. Indeed, there is an increasing recognition that OSA is a heterogeneous disorder with different clinical presentation patterns [7].

Among those who do report EDS, studies have shown that the severity of EDS does not correlate with the severity of OSA (as determined by apnea hypopnea index [AHI]), suggesting that factors other than respiratory events and arousal (eg, hypoxia- and/or sleep fragmentation-related brain injury [8]) contribute to EDS in this population [9-11]. Indeed, EDS has been shown to persist in some patients despite the normalization of breathing, oxygenation, and sleep quality with continuous positive airway pressure (CPAP) (the gold standard treatment for OSA), known as residual EDS [12,13]. Population-based studies have estimated that 9 to 22% of CPAP-treated patients experience residual EDS, even when other potential causes of sleepiness (eg, depression, other sleep disorders, medications, comorbidities, and inadequate sleep duration) are controlled [14,15]. Notably, there is a dose-response relationship between objectively-measured CPAP adherence and residual EDS, such that the prevalence of residual EDS is significantly lower among those who use CPAP for >6 hours per night compared with those who use CPAP for ≤5 hours per night [14,16,17].

To complicate matters, prevalence rates of EDS based on self-report measures are often inconsistent with rates based on objective measures. Prospective studies have suggested that 22 to 34% of patients self-report EDS after CPAP use; however, objective measures, such as the Maintenance of Wakefulness Test (MWT), which measures the ability to maintain alertness, or Multiple Sleep Latency Test (MSLT), which is a test of sleep propensity, indicate that, within those populations, up to 65% of patients may experience residual EDS [16-18]. This discrepancy suggests that some patients may not be aware of their sleepiness and/or may reflect that these assessments measure different aspects of EDS.

1.2. Consequences associated with EDS in OSA

The adverse effects of EDS are multifactorial and extend beyond the individual patient, also impacting the patient’s family, workplace, and society. At the individual level, EDS diminishes mood, quality of life, and cognitive functioning, and can negatively affect marital satisfaction [19-22]. In the workplace, EDS debilitates daily functioning, productivity, and social interactions, impairing time management, work quantity and quality, and on-the-job social interactions [19,23]. At the societal level, EDS is associated with an increased risk of motor vehicle and occupational accidents, thereby posing a safety risk to patients and the community [24-27]. This is particularly concerning because some patients may be unaware of their sleepiness or related impairments [16-18,28]. In addition, patients with OSA and EDS have greater healthcare resource utilization, including more healthcare provider interactions and emergency department visits, than those without EDS [29].

1.3. Unmet need and objective

Residual EDS is often overlooked and under discussed in clinical consultations [30]. Given the potential deleterious consequences that EDS may have on patient and public safety, it is critical for physicians to recognize residual EDS so patients have the opportunity to receive a diagnosis and treatment. Furthermore, the high prevalence of OSA may exceed the capacity of tertiary sleep centers. With advances in assessment and treatment technologies, it is being advocated that primary care manage simpler cases of follow-up with CPAP-treated patients, reserving referrals to sleep specialists for more complex cases [31]. Thus, it is vital that clinicians be informed on how to manage residual EDS along the clinical pathway, including how to appropriately assess for and diagnose residual EDS, when to intervene to optimize primary OSA treatment, when to use pharmacotherapy, how to choose a suitable agent, and when to refer to a sleep specialist.

Recent reviews have provided thorough reports of the definition, prevalence, pathophysiology, assessment tools, and treatment options for EDS associated with OSA [32,33]; however, there remains a need for step-by-step guidance for practice procedures to help clinicians who encounter patients with sleep disorders (eg, primary care physicians [PCPs], respiratory/sleep specialists, cardiologists, otolaryngologists, and nurse practitioners) appropriately evaluate and manage residual EDS. This review aims to provide suggestions for procedures that can be implemented into routine practice to assess for, identify, and manage residual EDS in patients treated for OSA. These strategies are summarized in Figure 1 and discussed in detail below.

Figure 1. Residual EDS in adults with OSA: diagnosis and management. CPAP, continuous positive airway pressure; EDS, excessive daytime sleepiness; OSA, obstructive sleep apnea

2. Recognizing and assessing residual EDS

2.1. Risk factors for residual EDS in OSA

Risk factors for residual EDS in patients with OSA are not well defined, but some have been identified (Table 1) [4,5,18]. Those with severe EDS at diagnosis are more likely to experience residual EDS after CPAP treatment [4,14,15,18]. Other possible risk factors include younger age and depression [14,15]. Interestingly, baseline AHI and body mass index (BMI) are not risk factors for residual EDS in patients who are adherent to CPAP [4,14,15,18]. It is inconclusive whether comorbid diabetes and hypertension influence the risk of residual EDS, as some [4], but not all [14,15,18], studies have reported an association.

Table 1. Risk factors for EDS in patients with OSA based on a combination of available studies.^a

Risk Factor	Odds Ratio	p-value
Diabetes [4]	6.9	<0.001
Baseline^b EDS [4,18]	1.3–5.1^c	<0.001
Not getting enough sleep [5]	4.6	<0.001
Heart disease [4]	2.9	0.004
Awaken with leg cramps [5]	2.4	<0.001
Chronic pain [18]	2.3	0.008
COPD [5]	2.0	0.01
Wake up too early [5]	1.8	0.002
Wake up during night [5]	1.8	0.001
Respiratory disease [5]	1.8	0.003
Depression [18]	1.8	0.059
Habitual snorer [5]	1.6	0.005
Trouble falling asleep [5]	1.5	0.055
Asthma [5]	1.2	0.015

^aBudhiraja et al. [18], n = 1,105; Koutsourelakis et al. [4], n = 208; Kapur et al. [5], n = 6,440. ^bPre-CPAP treatment. ^cOdds of experiencing residual EDS after CPAP treatment.

COPD, chronic obstructive pulmonary disease; CPAP, continuous positive airway pressure; EDS, excessive daytime sleepiness; OSA, obstructive sleep apnea.

2.2. Patient complaints of EDS

In theory, the easiest way for a clinician to identify whether a patient has residual EDS is through recognition of subjective symptoms. Unfortunately, patients’ complaints are not always straightforward and illuminating. A patient may use terms other than EDS to express their sleepiness or they may describe associated consequences, such as feeling tired, unrefreshed awakening, lack of energy, feeling fatigued, napping, morning headaches, irritability, or difficulty concentrating [34,35]. In fact, patients are more likely to report fatigue, tiredness, and lack of energy than sleepiness [34]. Recent evidence has demonstrated that fatigue, anxiety, depression, and negative affect are dimensions associated with EDS that influence the perception and emotional experience of the symptom, suggesting that discrepancies in clinical presentation may reflect EDS’ multidimensional nature [6]. As patient complaints may be inconsistent, initial assessment should begin with a clinical consultation, during which the clinician can obtain a past medical history and physical examination.

As previously discussed, some patients may simply be unaware of their sleepiness or may not perceive it to be troublesome [16-18,28]. In these cases, symptomatic indications of EDS may be even more subtle [6]. To reduce ambiguity, the clinician should utilize self-report EDS assessments. All patients treated for OSA should be assessed for residual EDS, regardless of whether they complain of sleepiness.

2.3. Self-report assessments of EDS

Self-report questionnaires are brief, inexpensive tools, making them convenient for use in routine clinical practice (Table 2) [36,37].

Table 2. Tools for evaluating EDS

Tools	Characteristics Assessed	Findings Suggestive of EDS	Utility	Administrator
Self-report measures
Epworth Sleepiness Scale (ESS)	Sleep propensity in daily situations (trait sleepiness)	Score >10	To assess for and/or diagnose EDS	Any HCP
Stanford Sleepiness Scale (SSS)	Degree of sleepiness at point in time (state sleepiness)	Score >3	Not ideal for clinical purposes	Any HCP
Karolinska Sleepiness Scale (KSS)	Degree of sleepiness at point in time (state sleepiness)	Score ≥7	Not ideal for clinical purposes	Any HCP
Objective measures
Multiple Sleep Latency Test (MSLT)	Ability to fall asleep	Mean sleep latency <8 minutes^a	To make a differential diagnosis (e.g. rule out EDS associated with narcolepsy) To confirm severity of EDS to justify pharmacotherapy	Sleep specialist
Maintenance of Wakefulness Test (MWT)	Ability to stay awake	Mean sleep latency ≤19 minutes^a	To characterize response to wake-promoting treatment To evaluate functional severity of EDS (i.e. if patient’s ability to remain awake is a safety risk)	Sleep specialist
Oxford Sleep Resistance (OSLER) Test	Ability to stay awake	N/A^b	Simplified behavioral-based version of the MWT	Sleep specialist
Psychomotor Vigilance Test (PVT)	Sustained vigilance (reaction time; lapses in attention)	N/A^c	To evaluate functional severity of EDS (i.e. if patient’s ability to remain awake is a safety risk)	Sleep specialist
Actigraphy	Graphical summary of sleep and wakefulness patterns; estimates of sleep parameters (e.g. sleep latency, TST, WASO, SE)	N/A	To improve diagnostic accuracy (e.g. rule out insufficient sleep as cause of EDS)	Sleep specialist

^aThresholds to determine EDS based on mean sleep latency on the MWT and MSLT are not well established as mean sleep latency can vary with methodological differences and individual differences in sleep tendency [36]. ^bNo standard cutoff; however, OSLER test mean sleep latency has been shown to be consistent with MWT mean sleep latency. ^cNo standard cutoff; however, patients with EDS have been shown to have lower reaction times, greater variability in reaction times across a task, and longer and more frequent lapses [37].

EDS, excessive daytime sleepiness; HCP, health care provider; SE, sleep efficiency; TST, total sleep time; WASO, wake after sleep onset.

The Epworth Sleepiness Scale (ESS) is the most commonly used self-administered assessment for EDS. The ESS assesses the propensity to fall asleep in real-world situations [38]. Patients are asked to rate, on a scale of 0 to 3 (with higher numbers indicating more sleepiness), how likely they would be to doze off or fall asleep in 8 situations, such as while reading, watching television, or riding as a passenger in a car. Total scores range from 0 to 24, with higher scores representing greater levels of sleepiness. To identify a patient with EDS, scores ≤10 are considered within the normal range, whereas scores >10 indicate EDS [38,39].

Advantages of the ESS include its brevity, accessibility, and widespread use in clinical and research studies. It takes ~2 to 3 min to complete and has been translated, tested, and validated in many languages and versions (e.g. a pictorial version for patients with diminished literacy skills), making it accessible worldwide. In addition, the ESS has high internal consistency in measuring EDS in patients with OSA [11,40]. In general, studies have found the ESS to have good test‒retest reliability in controlled clinical trial settings or when administered in similar clinical settings, and less reliability when administered across different primary and secondary care settings (e.g. between a PCP and sleep specialist visit) [11,41-45]. Finally, the widespread use of the ESS has produced extensive evidence for normative scores in clinical and nonclinical samples, facilitating interpretation of results.

In addition to these strengths, there are several limitations. The main disadvantage of the ESS is its reliance on self-report, making it susceptible to patient’s self-awareness of their sleepiness. Indeed, ESS scores tend to be higher when scored by the patient’s bed partner than when scored by the patient, suggesting that patients may underestimate their sleepiness. Accuracy may be enhanced when the ESS is completed as a consensus between the patient and partner [46]. Recent data have questioned the reliability of the ESS [43,44,47,48]. One study found variability in ESS scores even when the assessment was repeated on the same day, suggesting that situational sleepiness (i.e. time of test administration) may influence ESS scores [48].

The Stanford Sleepiness Scale (SSS) [49] and Karolinska Sleepiness Scale (KSS) [50] are other self-report assessment tools that can be used to quantify EDS; however, both measure state EDS at a specific point in time [49]. As such, these measures are more often used in research settings and are, for the most part, impractical for clinic settings [49,50].

2.4. Key points

• The ESS is the best option available for clinicians to implement into day-to-day practice to assess EDS, although it is limited by a patient’s perception of their sleepiness and the potential for lower test‒retest reliability along a clinical pathway from primary care to specialist settings

  • Some of these limitations can be overcome if the ESS is completed by both the patient and patient’s partner [46]

• The ESS should be administered with the clinical interview to all patients treated for OSA

3. Differential diagnosis of residual EDS

If a patient who is treated for OSA has an ESS score >10, there are several potential explanations for EDS, including suboptimal OSA treatment (including inadequate adherence), lifestyle factors, comorbidities, and concomitant medications.

3.1. Suboptimal OSA therapy

The first step to making a differential diagnosis of residual EDS associated with OSA is to ensure that EDS is not due to inadequate adherence to primary OSA therapy or suboptimal OSA treatment (i.e. non-normalization of AHI and associated sleep fragmentation or ineffective device). The clinician should first assess whether the patient is adherent to his/her therapy. Although some patients who are adherent to CPAP continue to experience EDS, residual EDS has been shown to significantly improve with increased CPAP use [14]. However, an estimated 30 to 40% of patients with OSA do not use CPAP as recommended (i.e. are nonadherent) [51,52]. To determine whether a patient is using their device as recommended, telemonitoring devices are available for those using CPAP that can provide patients and clinicians with objective data on daily usage, AHI, and leaks [31]. The American Thoracic Society has described how to interpret CPAP tracking data [53]. The patient can be advised to review these data (typically available on an app) as a tool to monitor daily use. When objective data are not available, the clinician can question the patient and bed partner on the frequency and duration of use (i.e. how many hours/night and nights/week the device is used), although self-report is less reliable. While insurance-mandated adherence is considered use ≥4 h/night on ≥70% of nights, patients should be educated that longer use per night is associated with better improvements in EDS in a dose-response manner [14,16,17]. A variety of interventions are available to help improve adherence, such as troubleshooting technical issues (e.g. airway leaks, pressure adjustments, humidity settings), improving side effects (e.g. mask discomfort, dry mouth), patient engagement tools, telemedicine, and cognitive-behavior therapies (e.g. framing) [54]. Detailed approaches to implementing these interventions have been published [54,55]; involvement of a psychologist, patient’s family/friends, or road traffic authorities (i.e. to suspend driving license) may be helpful to improve adherence. Some patients may not be sufficiently adherent to CPAP despite best efforts by both patient and clinician. In these cases, referral to a sleep specialist may be appropriate where second-line, alternative, and experimental therapies can be considered.

After adherence to therapy has been confirmed, the clinician should determine whether the patient is receiving optimal treatment for OSA. For patients using CPAP, this can be assessed via telemonitoring report of AHI. Determining the cause of an elevated AHI may require technical troubleshooting (e.g. airway leaks) and referral to a sleep specialist (e.g. in cases of complex sleep apnea/treatment-emergent central sleep apnea, which may occur in up to 20% of CPAP-treated patients [56-58]). It is also important to determine whether the therapy being used has been adequately evaluated for efficacy. This may be particularly important for non-CPAP therapies, such as mandibular advancement devices or positional therapies which typically do not have telemonitoring capabilities. Adherence and effectiveness may also be improved by switching to a more appropriate ventilatory support [59].

3.2. Lifestyle factors

After ensuring the underlying airway obstruction is being optimally treated, the clinician should evaluate whether lifestyle factors may be contributing to EDS. The most common cause of EDS in the general population is insufficient sleep, with more than one-third of American adults sleeping less than is recommended to prevent cognitive, health, and safety-related impairments [60,61]. Indeed, patients with OSA who do not get enough sleep have a 4.6-fold increased risk of having residual EDS (Table 1) [4,5,18]. Clinicians should advise patients to:

• Get ≥7 h of sleep each night [61]

• Maintain a consistent sleep/wake schedule (i.e. go to bed/wake up at the same time every day, including weekends)

• Create a comfortable sleep environment (e.g. dark, quiet bedroom with a cooler temperature; electronics, including cell phones, turned off/silenced)

• Practice good sleep hygiene (e.g. avoid electronics, arousing/stressful activities [e.g. social media, e-mail, finances, action movie], alcohol, caffeine, nicotine, meals, and exercise prior to bed) Notably, there are inter-individual differences in sleep need. Recommending a sleep extension trial, even in patients who get ≥7 hours of sleep, may help eliminate insufficient sleep as the cause [62]. Sleep specialists and psychologists can help manage poor sleep habits. The clinician should also evaluate a patient’s diet (e.g. caffeine consumption) and exercise habits. Self-reported sleepiness has been shown to worsen after both high-carbohydrate and high-fat meals [63,64]. It has been proposed that a high-fat, high-calorie diet could contribute to regularly occurring EDS [64]. Moreover, several studies have demonstrated an association between obesity and EDS and the ability of exercise to significantly reduce EDS [64,65].

3.3. Comorbidities/concomitant medications

Beyond lifestyle factors, EDS can have many other etiologies, such as medications, substance abuse, psychiatric disorders, medical disorders, or other sleep disorders (Table 3) [66]. For instance, EDS is a common side effect of prescription (e.g. sedating antidepressants, opioids) and nonprescription medications (e.g. first-generation antihistamines) [67]. Depression is another common cause of EDS. In addition to EDS, OSA and depression can have many overlapping symptoms, such as poor concentration, irritability, psychomotor impairments, and weight gain [68].

Table 3. Differential diagnosis of EDS

Suboptimal treatment of OSA
Inadequate adherence
Non-normalization of breathing (AHI)
Inappropriate ventilatory support
Lifestyle factors
Insufficient sleep
Diet
Exercise
Sleep disorders
Narcolepsy (type 1 or 2)
Idiopathic hypersomnia
Kleine-Levin syndrome
Circadian rhythm sleep-wake disorders
Restless legs syndrome
Periodic limb movement disorder
Psychiatric disorders
Depression
Anxiety
Substance use
Medical disorders
Diabetes
Hypothyroidism
Renal disease
Hepatic encephalopathy
Cancer
Inflammatory conditions
Encephalitis
Neurodegenerative conditions
Head trauma
Stroke
Medications
Alpha-2 antagonists^a
Antihistamines^b
Anxiolytics
Antidepressants^c
Anticonvulsants
Antidiarrhea agents
Antiemetics
Antimuscarinics/antispasmodics
Antiparkinsonian agents
Antipsychotics
Antitussives
Barbiturates
Benzodiazepines
Genitourinary smooth muscle relaxants
Mood stabilizers
Opioids
Sedative-hypnotics
Skeletal muscle relaxants

^aClonidine and methyldopa [67].

^bFirst-generation antihistamines; sedation is infrequent with second-generation antihistamines [67].

^cSedating antidepressants include tricyclics (amitriptyline, imipramine, nortriptyline, etc), tetracyclics (trazadone and mirtazapine), paroxetine, citalopram, fluvoxamine, brexanolone, and esketamine [67].

AHI, apnea hypopnea index; EDS, excessive daytime sleepiness; OSA, obstructive sleep apnea.

EDS is also associated with other sleep disorders that can co-occur with OSA. Insomnia is commonly comorbid with OSA, occurring in 39–55% of patients [69,70]. Narcolepsy, which can occur with (type 1) or without (type 2) cataplexy, idiopathic hypersomnia, and Kleine-Levin syndrome are central disorders of hypersomnolence where the primary complaint is EDS. The presence of cataplexy can help distinguish narcolepsy type 1 from other hypersomnolence disorders, whereas idiopathic hypersomnia and Kleine-Levin syndrome can be identified by consistent or episodes of long sleep times, respectively. Circadian rhythm sleep-wake disorders are characterized by misalignment between one’s biological clock and desired/required sleep-wake schedule (e.g. shift work, jet lag); wakefulness during the biological nighttime can lead to EDS. Finally, restless legs syndrome and periodic limb movement disorder, characterized by an urge to move the legs during rest/inactivity and frequent limb movements during sleep, respectively, can cause sleep disturbance-induced EDS [62].

As these etiologies require different treatment approaches, it is critical for the clinician to determine the cause of EDS to ensure each patient receives the most appropriate and comprehensive care. In most cases, a differential diagnosis can be made based on a clinical consultation; however, further testing with questionnaires, laboratory tests, or objective assessments may be needed in other cases. If another sleep disorder is suspected, the clinician should refer the patient to a sleep specialist, who can determine what additional testing is needed. Involvement of a psychologist may be valuable in other cases, particularly if depression or insomnia are suspected.

3.4. Objective assessments for EDS

Several objective assessments are available to better characterize EDS (Table 2). Unfortunately, most are time consuming, expensive, and complex to administer. For these reasons, it is not feasible to use these assessments to assess EDS in routine practice; however, they may be helpful when attempting to make a differential diagnosis [71,72], justify use of pharmacotherapy, or evaluate response to treatment. These assessments are typically conducted and interpreted by sleep specialists in a sleep center and, therefore, are only briefly described below.

The Multiple Sleep Latency Test (MSLT) and Maintenance of Wakefulness Test (MWT) are the most commonly used objective assessments for characterizing an individual’s ability to fall asleep in a sleep-inducing environment or stay awake while seated in a dark, quiet environment, respectively (Table 2). For both assessments, the outcome of interest is mean sleep onset latency (the average amount of time it takes the patient to fall asleep) as determined by electroencephalography. The MSLT is often used to rule out other sleep disorders as the etiology for EDS, such as narcolepsy. The American Academy of Sleep Medicine (AASM) advises that the MWT should not be used for diagnostic purposes [71]. Instead, it may be useful to evaluate the response to treatment or assess whether an individual’s ability to remain awake poses a public or personal safety risk, such as those employed in public transportation [72].

Although less standardized, there are tests that use methods other than EEG to objectively quantify EDS. The Oxford Sleep Resistance (OSLER) test [73] and Psychomotor Vigilance Task (PVT) [74] use simple behavioral-based methods to measure sleep latency and sustained vigilance, respectively [73,74]. In addition, home-based wearables, such as wrist-worn activity monitors, can provide insight into sleep and wakefulness patterns in the home environment and may serve as a useful tool when attempting to rule out insufficient sleep as a cause for EDS, although there is uncertainty about their ability to accurately estimate sleep [75].

Subjective symptoms and self-report measures of EDS often do not correlate with objective measures. In patients with OSA, complaints of sleepiness, tiredness, lack of energy, and fatigue have been found to have no association with MSLT mean sleep latency [34]. In addition, the ESS and MWT are only moderately associated, whereas the ESS and MSLT are weakly associated [11]. There are several possible explanations for these low agreements. As previously discussed, patients may simply be unaware of the severity of their sleepiness, especially if their baseline reference for feeling ‘normal’ has gradually shifted. On the other hand, patients may be reluctant to report negative feelings associated with disease, for fear of work- or driving-related implications. Indeed, research has shown that close relatives often describe the patient’s sleepiness as more severe than the patient reports [76]. Another reason may be that these assessments measure different dimensions of EDS. For example, it has been postulated that the PVT and ESS may predict impaired performance, whereas the MSLT may predict physiologic sleep propensity [77].

3.5. Key points

• If a patient has EDS, the clinician should first determine if the patient is adherent to primary OSA therapy via self-report or telemonitoring data and evaluate effectiveness of device

  • Referral to a sleep specialist may be needed to determine cause of elevated AHI in adherent patients, confirm effectiveness of current treatment in adherent patients, or explore alternative treatments in nonadherent patients

• Next, the clinician should evaluate lifestyle factors and other conditions that could be causing EDS

  • Referral to a sleep specialist may be needed for cases with multiple potential causes

• Objective assessments can be useful when attempting to make a differential diagnosis, justify pharmacotherapy, or document response to treatment

  • If the clinician suspects their use is necessary, the patient should be referred to a sleep specialist

4. Pharmacotherapy for residual EDS

After a diagnosis of residual EDS and altered quality of life associated with OSA has been confirmed and other contributing conditions have been ruled out, pharmacotherapy with a wake-promoting agent or stimulant can be considered. The decision to prescribe a trial of pharmacotherapy should typically be made by a sleep specialist and can be based upon EDS severity, as characterized by the ESS or objective assessments. In general, ESS scores >10 indicate that the patient’s EDS is burdensome and interfering with daily life; however, this threshold is not absolute [38,39]. Clinical judgment should be made on an individual basis, especially for patients whose occupations may pose a personal and/or public safety risk (e.g. public transportation, healthcare, childcare). If pharmacologic treatment is determined to be appropriate, it is important to note that pharmacotherapy for EDS should not replace primary treatment of the underlying airway obstruction. Clinicians should monitor patients’ CPAP use after initiating pharmacotherapy to confirm adherence. Although pharmacotherapy for residual EDS may be prescribed more often in a specialist setting than in primary care, it is useful for clinicians to be familiar with the available options. Key information for the treatment options discussed below is summarized in Table 4 [78-86].

Table 4. Pharmacotherapy options for EDS in OSA

Agent	Dosing	Side Effects	Drug Interactions/Contraindications
FDA-approved
Modafinil (Provigil®) [78]	200 mg/day Administer once daily in morning	Serious rash, including Stevens-Johnson syndrome Headache Nausea Nervousness Rhinitis Diarrhea Back pain Anxiety Insomnia Dizziness Dyspepsia	Steroidal contraceptive Cyclosporine CYP2C19 substrates (e.g. omeprazole, phenytoin, diazepam)
Armodafinil (Nuvigil®) [79]	150 or 250 mg/day Administer once daily in morning	Serious rash, including Stevens-Johnson syndrome Headache Nausea Dizziness Insomnia	Steroidal contraceptive Cyclosporine CYP2C19 substrates (e.g. omeprazole, phenytoin, diazepam)
Solriamfetol (Sunosi™) [80]	37.5, 75, or 150 mg/day Initiate at 37.5 mg/day Administer once daily on waking	Headache Nausea Decreased appetite Insomnia Anxiety	Monoamine oxidase inhibitors Drugs that increase blood pressure and/or heart rate Dopaminergic drugs
Off-label
Pitolisant (Wakix®) [81]	17.8 to 35.6 mg/day^a Initiate at 8.9 mg/day Administer once daily in morning	Insomnia Nausea Anxiety QT interval prolongation	Patients with severe hepatic impairment CYP2D6 inducers CYP3A4 substrates, including hormonal contraceptives
Methylphenidate (Ritalin® or Concerta®) [82,83]	Ritalin® 5, 10, or 20 mg/day Administer 20–30 mg 2 or 3 times daily, preferably 30–45 min before meals (maximum dose of 60 mg/day) Concerta® extended-release 18, 27, 36, and 54 mg/day Initiate at 36 mg/day (adults) Administer once daily in the morning	Tachycardia Palpitations Headache Insomnia Anxiety Hyperhidrosis Weight loss Decreased appetite Dry mouth Nausea Abdominal pain	Monoamine oxidase inhibitors Antihypertensive drugs Halogenated anesthetics
Amphetamines (e.g. Adderall® [84] and Dexedrine® [85])	5 to 60 mg/day Initiate at 10 mg/day Administer at the lowest effective dose and adjust on an individual patient basis Administer in divided doses; first dose upon awakening with subsequent doses at intervals of 4 to 6 h Avoid late evening doses due to resulting insomnia	Hypertension Tachycardia Myocardial infarction Stroke New/worsened psychiatric symptoms Peripheral vasculopathy Seizures Serotonin syndrome Dizziness Insomnia Tremor Headache Vision blurred Dry mouth Gastrointestinal disturbances Anorexia/weight loss	Patients with advanced arteriosclerosis, symptomatic cardiovascular disease, moderate to severe hypertension, hyperthyroidism, known hypersensitivity or idiosyncrasy to the sympathomimetic amines, glaucoma, or agitated states Patients with history of drug abuse Monoamine oxidase inhibitors Acidifying agents, adrenergic blockers, alkalinizing agents, tricyclic antidepressants, CYP2D6 inhibitors, serotonergic drugs, antihistamines, antihypertensives, chlorpromazine, ethosuximide, haloperidol, lithium carbonate, meperidine, methenamine therapy, norepinephrine, phenobarbital, phenytoin, propoxyphene, proton pump inhibitors, veratrum alkaloids

^aNote that these doses are FDA-approved for patients with narcolepsy. In patients with OSA, pitolisant has been tested in Europe at doses of 5, 10, and 20 mg/day; treatment was initiated at 5 mg/day [86]. EDS, excessive daytime sleepiness; FDA, Food & Drug Administration; OSA, obstructive sleep apnea.

4.1. Modafinil and armodafinil

Modafinil (Provigil®; Teva Pharmaceuticals, North Wales, PA) and armodafinil (Nuvigil®; Teva Pharmaceuticals) are approved in the United States, but not in Europe, to improve wakefulness in adults with EDS associated with narcolepsy, OSA, or shift work disorder [78,79]. Both agents have demonstrated efficacy in reducing EDS and improving wakefulness in patients with OSA in short-term randomized controlled trials (RCTs) [87-90] and long-term open-label extension trials [91,92], improving ESS scores by ~2 points and MWT sleep latency by ~3 minutes compared with placebo [91-94]. Modafinil had previously been approved for OSA in the European Union; however, in 2011, the European Medicines Agency (EMA) withdrew this indication from the marketing authorization due to a poor benefit/risk profile. Specifically, the EMA concluded that the risk of serious cardiovascular disorders, neuropsychiatric disorders, and skin and hypersensitivity disorders outweighed any potential benefit for EDS in patients with OSA [78,79,95].

The most common side effect experienced with both agents is headache, although armodafinil tends to be associated with a lower rate of headache than modafinil [96]. Other common side effects include nausea, dizziness, insomnia, upper respiratory tract infection, and anxiety [96]. Serious skin rashes, such as Stevens-Johnson syndrome and toxic epidermal necrolysis, and hypersensitivity, including angioedema and anaphylactoid reaction, have been reported [78,79]. The EMA withdrew approval due to potential adverse cardiovascular effects and the FDA issued a warning for use in patients with known cardiovascular disease [78,79,95]. A recent meta-analysis concluded that modafinil and armodafinil may slightly increase blood pressure [93]. Clinicians should monitor patients’ blood pressure prior to initiating and routinely during treatment. Modafinil and armodafinil can reduce the effectiveness of birth control and have been associated with fetal congenital malformations, including congenital cardiac anomalies [97]; if prescribed to premenopausal women, alternative methods of contraception are recommended and these risks should be discussed.

4.2. Solriamfetol

Solriamfetol (Sunosi™; Jazz Pharmaceuticals, Inc, Palo Alto, CA) is a dopamine/norepinephrine reuptake inhibitor approved in the United States and European Union to improve wakefulness in adults with EDS associated with narcolepsy or OSA [80,98]. Solriamfetol has demonstrated efficacy in reducing EDS and improving wakefulness for up to 9 h in patients with OSA in short-term RCTs [99,100] and a long-term open-label extension trial [101], improving ESS scores by ~2 to 5 points and MWT sleep latency by ~5 to 13 min compared with placebo [80,101].

The most common side effects with solriamfetol treatment include headache, nausea, decreased appetite, insomnia, and anxiety [80]. In addition, solriamfetol has been associated with small increases in heart rate and blood pressure in patients with OSA, although increases were greatest for the 300 mg/day dose, which is not FDA- or EMA-approved [99]. Patients’ hypertension should be controlled prior to initiating solriamfetol, and heart rate and blood pressure should be monitored periodically during treatment. The FDA advises avoiding use in patients with unstable cardiovascular disease, serious heart arrhythmias, or other serious heart problems. Solriamfetol is contraindicated with monoamine oxidase inhibitors [80].

4.3. Pitolisant (off-label)

Pitolisant (Wakix®; Harmony Biosciences, LLC, Plymouth Meeting, PA) is a histamine-3 receptor antagonist approved in the United States to improve wakefulness and cataplexy in adults with narcolepsy and in the European Union for the treatment of narcolepsy with or without cataplexy in adults [81,102]. Recent phase 3 RCTs suggest pitolisant may reduce EDS in patients with OSA [86,103], but it is not currently approved for use in OSA [81,102].

4.4. Stimulants (off-label)

Stimulants, such as methylphenidate (Ritalin®; Novartis Pharmaceuticals Corporation, East Hanover, NJ) and amphetamines (Adderall® [Teva Select Brands, Horsham, PA] or Dexedrine® [Teva Pharmaceuticals]), have also been used off-label to treat EDS associated with OSA. Importantly, no RCTs or published data exist to support their use in this population. Their extensive side-effect profiles and potential for abuse/addiction limit their use.

4.5. Follow-up

After initiating a trial of pharmacotherapy, the sleep specialist or PCP should follow-up with the patient to evaluate efficacy, side effects, and adherence to primary OSA therapy. To determine efficacy, the clinician can re-administer the ESS and evaluate improvement in quality of life and productivity. Determining whether a patient has responded to treatment based on ESS scores can be considered based on the absolute score or the degree of change. In the first case, ESS scores may have decreased within the normal range (≤10); however, improvement this significant may not be observable until after many weeks of stable treatment [86,88,104]. In the second case, the clinician can assess the percent reduction in ESS scores relative to baseline, using a reduction in ESS score ≥25% from baseline as a threshold for a clinically meaningful response to treatment [105,106]. Alternatively, some data suggest that a 2- to 3-point reduction in ESS score, corresponding to an effect size of 0.5, represents the minimal clinically important difference [86,107].

In addition to evaluating efficacy, the clinician should monitor for adverse events, particularly cardiovascular side effects (e.g. hypertension). Heart rate and blood pressure should be monitored prior to initiating and periodically throughout treatment with any wake-promoting agent or stimulant. With modafinil and armodafinil, some side effects, such as rashes, may occur weeks after initiation. With solriamfetol, some common early-onset side effects (those that occur during the first week of treatment) subside within ~10 days of treatment (e.g. headache and nausea), although others can persist for >2 months (e.g. decreased appetite) [108].

4.6. Key points

• Choosing a pharmacologic agent should be individualized to the patient, taking into account the evidence available regarding an agent’s efficacy and safety profile, cost, and patient characteristics

• Regardless of which agent is chosen, the clinician, with support from a sleep specialist, should assess a patient’s response to treatment during follow-up visits

  • Efficacy with the ESS

  • Tolerability with adverse events

  • Blood pressure (every 6 months) and heart rate (a 12-lead ECG annually)

  • Adherence to primary OSA therapy

Whether to treat patients who struggle with adherence to primary OSA therapy with a pharmacologic agent warrants discussion. Treatment of EDS with a wake-promoting agent or stimulant is not a replacement for treatment of the underlying airway obstruction. For every patient, best efforts should be made to optimize adherence to CPAP or other airway therapies. Even in patients who are adherent, pharmacologic treatment has the potential to reduce motivation to use CPAP. Modafinil and armodafinil have been associated with a trend toward reduction in CPAP use, although both short- and long-term data suggest that adherence remains stable during treatment with solriamfetol [96,99,109]. It is unknown whether pitolisant impacts primary OSA therapy adherence [103]. Sometimes, despite best efforts by clinician and patient, patients remain nonadherent to primary OSA therapy. In these cases, the clinician should use his/her best clinical judgment and, in this specific situation, it is recommended to seek guidance from a sleep specialist. In some cases, it may not be feasible to optimally treat contributing causes, and it may be appropriate to temporarily prescribe a pharmacologic agent while working to improve adherence.

5. When to refer to a sleep specialist

In cases when the clinician may be unable to evaluate or manage a patient’s EDS, referral to a sleep specialist for further evaluation and/or testing may be necessary (see Figure 1). This could be when a patient is not sufficiently adherent to CPAP, despite best efforts by both clinician and patient, and the clinician suspects other therapies should be considered. Another appropriate case for referral would be if a clinician needs help determining the cause of an elevated AHI or wants to confirm that OSA is being optimally treated before prescribing wake-promoting pharmacotherapy. Finally, referral may be necessary when a clinician is uncertain about the cause of EDS (e.g. in cases that may have multiple etiologies) or suspects the cause is related to another sleep disorder. In these cases, additional testing (e.g. MSLT, MWT, overnight polysomnography) in a sleep clinic would likely be warranted.

6. Conclusions

EDS negatively impacts patients’ ability to function in daily life and is a risk factor for patient and public health. As such, it is important for any physician to be able to recognize, evaluate, and manage EDS in patients with OSA. Due to the high prevalence, it falls to the clinician to assess patients treated for OSA for residual EDS with the ESS. If a patient is suspected of having EDS, the clinician should first confirm the underlying airway obstruction is being optimally treated. If the patient continues to experience EDS despite adherence to OSA therapies, the patient should be reviewed in the clinic and, when appropriate, with questionnaires, physical examination, laboratory tests, or objective assessments to rule out other potential causes of EDS. After a differential diagnosis of residual EDS associated with OSA has been confirmed, pharmacologic treatment using a wake-promoting agent may be considered. Consistent use of this proposed assessment and management approach can help clinicians identify patients with EDS, improve diagnostic accuracy, enhance management strategies in the primary care setting, and make optimal use of secondary and tertiary resources in specialist centers.

Transparency

Declaration of funding

This review was supported by Jazz Pharmaceuticals. Jazz Pharmaceuticals has worldwide development, manufacturing, and commercialization rights to solriamfetol, excluding certain jurisdictions in Asia. SK Biopharmaceuticals, the discoverer of the compound (also known as SKL-N05), maintains rights in 12 Asian markets, including Korea, China, and Japan.

Declaration of financial/other relationships

R Rosenberg has received consultancy fees from Eisai; honoraria from Merck; research funding from Jazz Pharmaceuticals, Merck, Actelion, Eisai, and Philips Respironics; and has served on the speakers’ bureau for Merck and as an advisory board member for Jazz Pharmaceuticals. PK Schweitzer has received consultancy fees from Jazz Pharmaceuticals, Her institution has received research funding from Jazz Pharmaceuticals, Apnimed, Balance Therapeutics, Avadel-Flamel, Harmony Biosciences, Inspire Medical Systems, and Suven Life Sciences. J Steier has received consultancy fees from Jazz Pharmaceuticals, Sanofi, travel support from Fisher Paykel, GSK, and research funding from ResMed, British Lung Foundation, Actelion, Phillips Respironics. J Steier’s contributions were partially supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London, UK. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. J-L Pepin has received lecture fees or conference traveling grants from Resmed, Perimetre, Philips, Fisher and Paykel, AstraZeneca, Jazz Pharmaceuticals, Agiradom, and Teva, and has received unrestricted research funding from ResMed, Philips, GlaxoSmithKline, Bioprojet, Foundation de la Recherche Medicale (Foundation for Medical Research), Direction de la Recherche Clinique du CHU de Grenoble (Research Branch Clinic CHU de Grenoble), and fond de dotation ‘Agir pour les Maladies Chroniques’ (endowment fund ‘Acting for Chronic Diseases’). Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Biographical notes

Russell P. Rosenberg, PhD, is currently the Chief Science Officer and CEO of NeuroTrials Research, Inc. and the Director of the Atlanta School of Sleep Medicine and Technology in Atlanta, GA. Paula K. Schweitzer, PhD, is director of research at St. Luke’s Sleep Medicine and Research Center in Chesterfield, MO. Professor Joerg Steier works as Consultant Respiratory Physician at the Lane Fox Unit, a tertiary service for weaning and noninvasive ventilation, and the British Sleep Society (BSS) accredited Sleep Disorders Centre at King’s Health Partners; he is an executive committee member and honorary secretary of the BSS, and a task force member of the European Respiratory Society. Jean-Louis Pepin, MD, PhD, is currently a Professor of Clinical Physiology at University of Grenoble-Alpes (UGA) (exceptional class), a member of the national council of universities (CNU, 44-02: Clinical Physiology), Vice-Dean of the Grenoble School of Medicine in charge of research, and Director of the HP2 Laboratory (Inserm U1300; Hypoxia Pathophysiology).

Acknowledgments

Under the direction of the authors, Hannah K. Ritchie, PhD, Jeannette Fee, and Christopher Jaworski of Peloton Advantage, LLC, an OPEN Health company, provided medical writing and editorial support for this review article, which was funded by Jazz Pharmaceuticals.