3 Diagnosis of Sleep-Disordered Breathing
Abstract
This chapter focuses on the workup of a patient with sleep-disordered breathing. As in all domains of medicine, the workup starts with the medical history (anamnesis). Clinical evaluation includes a general as well as an OSA-specific evaluation, evaluation of the mouth and upper airway. Next, a sleep study is needed. There are several forms of sleep studies, all with their own pros and cons. They can be performed in the sleep lab as well as at home. After the sleep study, a differential diagnosis can be established. In case treatment other than CPAP is considered, a sleep study is often not sufficient for definitive treatment planning. Clinical investigations also include questionnaires. For sleep surgeons specifically, an insight of awake endoscopy, Müller’s maneuver, and drug-induced sleep endoscopy (DISE) is pivotal. To score DISE findings, several systems have been proposed. The VOTE system is the mostly wide used system and is discussed in detail, with regards to organization, logistics and staff, indication, patient selection, sedation protocol, and how to translate the various findings into treatment planning.
3.1 History Taking
The medical history, as in all domains of medicine, is always pivotal and determines the level of suspicion in obstructive sleep apnea (OSA) to a large extent. Although the most striking features of the history as socially unacceptable snoring and witnessed apneic events are already often known from the referral letter from the general physician, it remains important to have a preferably standardized comprehensive own assessment.
It is beneficial if the bed partner accompanies the patient to double check the complaints. It is not unusual if the complaints noticed by the bed partner and the complaints reported by the patient vary considerably, where the bed partner often scores a twice as high severity. For example, the irritability, length and loudness of snoring, and the presence and length of apneas are best noticed by the bed partner. The position also influences the sleep: are the snoring and apneic events present in all sleep positions, or are they only present or more pronounced in supine position? In particular men are inclined to downgrade their snoring level in particular and severity of complaints in general. For surgical purposes it has been strongly recommended to let the bed partner score his/her grade of being disturbed by his/her bed partner’s snoring. For this grading a 10-point visual analogue (VAS) scale is used. VAS grading provides the possibility of comparing pre- and postoperative results, providing the fact, that the patient still has the same bed partner. We all know that the annoyance of snoring depends on the particular individual. Another fact that needs to be kept in mind is that most patients are compelled by their bed partners to undergo painful surgical treatment for snoring, else they would get expelled from the joint bedroom. After a painful palatal surgery we ask the bed partner in presence of the patient about the success. This often affects the answers, resulting in a lower postoperative VAS score. Maybe this is why many scientific papers present highly significant VAS scores about snoring intensity. However, VAS provides a quick and easy-to-perform instrument to measure surgical success.
Many sleep clinics routinely use specific questionnaires regarding sleepiness and different levels of quality of life (QoL) such as the Epworth Sleepiness Scale (ESS) 1 and Functional Outcomes of Sleep Questionnaire (FOSQ). 2 ▶Table 3.1 summarizes the recently used QoL test instruments in sleep surgery. It should be considered that patients filled in these questionnaires electronically in soft copy before the first clinic visit so that their responses can be recorded in electronic files. This system of electronic data collection before the actual visit works only if patients have access to Internet and have no language barriers. As an alternative, patients and bed partners may fill in these questionnaires in the waiting room before meeting the physician. Based on the level of suspicion, a first triage for further diagnostics can be considered, for example, in case of high pretest likelihood of OSA only polygraphy might be ordered first, while in case of lower levels of suspicion, full polysomnography should be considered. It has been noted from the earlier studies that more than 50% of patients visiting sleep centers because of severe snoring and possible OSA, had confirmation of OSA in the sleep registration. Sometimes the history makes OSA less likely (such as in case of tiredness in young lean individuals, no history of snoring, etc.) and no sleep study is performed, or, in case of suspicion of narcolepsy, another form of sleep study is ordered.
It is important to realize that the medical history obtained either from the patient or as noted by the bed partner is far from a 100% reliable positive or negative predictor of the presence of OSA. The possibility of OSA cannot be ruled out if the bed partner does not notice apneic events. In particular, hypopneas (and respiratory-related arousals) can easily be missed. The bed partner is not awake during the night and has no information about the various sleep stages and sleep positions or combinations of sleep stage and sleep position. Bed partners may provide a proof of OSA through video recordings; however, subsequent confirmation by a sleep study is mandatory. Oftentimes the reverse also happens where the medical history told by patient and bed partner makes the presence of witnessed apneas possible, but the subsequent sleep study does not confirm sufficient hypopneic and apneic events to establish the diagnosis.
Nowadays there are various applications that promise to analyze sleep; however, all of them are not validated and require verification by sleep studies.
Many patients do not have the full-blown syndrome of socially unacceptable snoring, witnessed apneas, extreme tiredness, and hypersomnolence. This is in particular true for early stage disease, mild and moderate OSA; but even in case of severe OSA, and in case of apnea–hypopnea index (AHI) above 60, patients might experience remarkably little complaints. Others such as professional drivers, who are concerned about their permission to drive, might deliberately minimize their complaints of sleepiness. Therefore sleep studies remain the gold standard to confirm the clinical suspicion of OSA.
Signs and symptoms that should be included in (electronic) checklists can be divided into complaints during wakefulness (▶Table 3.2) and during sleep (▶Table 3.3).
Sleepiness |
Falls asleep easily |
Tiredness |
Painful muscles and joints |
Loss of concentration/focus |
Difficulty with staying awake when driving |
Lack of energy |
Painful feeling in the chest |
Easily irritated |
Decreased memory and cognitive functioning |
Severe snoring |
Witnessed apneic events by patient or bed partner |
Awakenings with feeling of choking |
Decreased sexual performance and/or impotence |
Transpiration during sleep |
Nycturia |
Dry mouth after awakening |
Morning headache |
Socially unacceptable loud snoring is the most suspicious alarming symptom, but apneas and hypopneas are also very ominous signs but not always noticed. In particular, nycturia is typical for OSA. It is due to a lack of sufficient deep sleep.
Sleepiness and never feeling rested after sleep are suspicious symptoms. Many other signs and symptoms during the day are atypical and might be due to other causes as well. This is one of the reasons that the diagnosis of OSA is often missed, or patients are misdiagnosed as having a depression or burnout. OSA is a gradual progressive disease and many complaints might gradually become worse if OSA is left untreated. A subtle change in character may be noticed by partners, family members, or colleagues; for example, irritability, loss of focus, and less intellectual and sexual performance. These symptoms are often attributed to increasing age and social issues, while they in fact are due to the development of OSA. In retrospect many patients have had complaints for several years before the diagnosis is established. It is the combination of nocturnal symptoms with signs occurring during the day that should lead to the suspicion of OSA.
Daytime sleepiness is mostly scored by means of ESS (▶Table 3.4).
Situation | 0 | 1 | 2 | 3 |
Sit and read | ||||
Watching television | ||||
In a theatre or meeting | ||||
As passenger in a car during a hour without pause | ||||
Resting during the afternoon | ||||
Sitting and having a conversation | ||||
After dinner without alcohol | ||||
In a car, before a traffic light waiting | ||||
Total score |
The ESS consists of eight questions with four answer categories. Scores vary from 0 (never, or not applicable), 1 (sometimes), 2 (usually), to 3 (always). The maximum score is 24, and a score of 10 or more is regarded as abnormal. The average score in healthy people is around 6. It is clear that the positive and negative predictive values of the ESS are moderate, but it does give a first indication of the level of sleepiness. One should be aware of its limitations; for example, patients who do not drive a motor vehicle or have never visited a theatre or meeting cannot reach maximal high scores. With all its limitations, the ESS is still often used in the clinical setting as well as in research projects.
Other questions that should be included concern sleep hygiene and rhythm, lifestyle, alcohol consumption and smoking, medication, and weight loss and weight gain. In case of work shifts, irregular working hours, other sleeping disorders, and too little sleeping hours might be present. Some patients have more than one reason for being sleepy. Alcohol consumption before bedtime induces muscle relaxation and thereby OSA. Smoking is thought to induce edema of the upper airway and is regarded as a cofactor. Its role is not as important as alcohol consumption. Medication, such as sleeping pills and muscle relaxants, should be asked for while taking medical history.
In the authors’ opinion, it is very important to collect the biometric data of each patient such as patient’s weight (and height for calculation of body mass index [BMI]) and history of recent weight gain. Weight gain might lead to the development or worsening of OSA, but the reverse should also be considered: OSA might lead to weight gain.
In authors’ opinion, patients should be asked about their sleep habits before treatment. Do the patient and his/her bed partner still sleep together or does the bed partner go to bed earlier to avoid not being able to fall asleep because of the loud snoring? Does the bed partner use ear plugs, or is sleeping together no longer possible at all? Do other family members complain? How is the sleep in hotels, on camping grounds, and in airplanes? Does the snoring wake up the patient? How do the complaints exist? Are the complaints always present or present in specific situations and circumstances? Are the complaints due to specific sleeping positions (mostly supine) or present in all sleeping positions? Is the patient forced to sleep on the back because of specific medical conditions such as hip, back, or shoulder complaints?
Another crucial issue is to record patient’s comorbidities. The patient’s health condition is relevant to decide, whether the patient may undergo a surgical treatment, and how to organize the perioperative care (please refer to Chapter 10 for more details). Please check for high blood pressure.
Patients should be asked about other sleep disturbances. OSA is only 1 out of about 80 diseases that may cause restless sleep. One should refer to the International Classification of Sleep Disorders (ICSD, Quelle), that is updated by the American Academy of Sleep Disorders every few years. Is the patient less focused, does he/she experience loss of concentration, less intellectual performance, mood changes, transpiration during sleep with several awakenings for urinating during the night, does the patient have a dry mouth when waking up, doe he/she has a morning headache, does he/she feel refreshed/rested after sleep at all? Is the patient less alert, irritable, and depressive? Has the sexual performance deteriorated or is there overt sexual impotence?
It is important to know if there is a history of teeth grinding (bruxism?) or other oral diseases to consider treatment options. This fact might interfere with an oral appliance treatment.
Another crucial thing to know is what has already been done to treat the patient’s OSA or snoring. Has the patient had other previous attempts to treat the snoring and OSA complaints, such as attempts to improve nasal breathing either by means of medication or surgery? Are the complaints worse in case of blocked nose as because of hay fever, seasonal rhinitis, other allergies, during common colds, or because of nasal polyps, or anatomical reasons such as septal deviation or hypertrophic inferior turbinates. In case of nasal pathology, specific questionnaire for nasal pathologies, such as the Nose Obstruction Symptom Evaluation (NOSE) might be helpful. 4 Have there been any attempts to solve the snoring problem either with over-the-counter antisnoring measures, of boil and bite or custom-made oral devices; have there been previous sleep studies, sleep surgery, CPAP use?
The variety of questions addressed in previous sections makes us think that every sleep surgery center should use a comprehensive and surgery-specific OSA questionnaire. Unfortunately there is no standard questionnaire, which means that every center should create its own. In authors’ opinion, the questionnaire should at least include biometric data, questions about sleep habits, comorbidities, dental health, and past sleep-related treatments. The authors expect that smartphone applications will gain relevance within the next couple of years. Earlier they did not enter daily routine.
3.1.1 Conclusion
The medical history determines the level of suspicion in OSA to a large extent. The most striking features of the history are socially unacceptable snoring and witnessed apneic events, but there are many nocturnal and wakeful signs and symptoms that can be attributed to OSA as well. The use of standardized comprehensive OSA-specific questionnaires is strongly recommended.
3.2 Clinical Evaluation of Patients with (Suspected) Sleep-Disordered Breathing
3.2.1 Introduction
Physical and clinical evaluations are essential for the assessment of the patient with SDB. The findings contribute to the treatment advice to the patient. The doctor and patient can only take a shared decision about a treatment plan after all information has been obtained about the severity (mild, moderate, severe) and the type (obstructive, mixed, central) of the OSA. As obstruction is almost always in the upper airway, level(s), severity, and configuration (or direction) of the obstruction(s) are also analyzed. OSA, the most common type of SDB, is a complex syndrome due to many different factors, both anatomical and functional. It can originate from different and multiple segments of the upper airway; therefore, the entire upper airway from the entrance of the nose to the level of the vocal cords should be examined during the evaluation.
An ideal clinical evaluation would be predictive of the site(s) of obstruction and would be cost effective and reproducible. Unfortunately, clinical evaluation is subjective and it is usually performed while awake, which does not represent the situation when asleep. While it provides valuable information in a glance, some general anatomical facts can be easily obtained (e.g., an edentulous patient will not be treated with a mandibular advancement device [MAD], patients with a high body mass index [BMI] are not the best surgical candidates, the combination of large tonsils and small tongue is more attractive from a surgical point of view than vice versa). General ear, nose, and throat (ENT) examination in awake situation is not the preferred workup for a patient in whom surgery is considered. In this chapter OSA-specific ENT examination, as performed in the examination room, is discussed keeping its reservations and limitations in mind.
3.2.2 Anamnesis
A thorough medical history is essential. The use of standardized OSA-specific questionnaires, to be filled in by the patient and bed partner, can be extremely helpful in this regard. Questions should include snoring (frequency, loudness, sleep position), witnessed apneas, sleepiness, tiredness, morning headache, character changes, sexual dysfunction, etc. Often, these questions are better answered by the bed partner. Bed partners can provide information about witnessed apneas and sleep characteristics.
There is also a need to collect data on the surgical history, specially related to the upper airway (e.g., previous tonsillectomy, nasal surgery, etc.) including orthodontic treatment, tobacco and alcohol consumption, as well as comorbidities that can increase cardiovascular risk (hypertension, dyslipidemia, diabetes, arrhythmias, previous stroke) and the use of medication that could alter sleep or decrease upper airway muscle tone while sleeping (sedatives, muscle relaxants, sleeping medication). Family history of snoring and sleep apnea should be evaluated as well, as there might be familial aggregation.
Sleep habits, shift-work, driving performance, occupation of the patient, and questions related to sleep quality and fatigue/sleepiness are also mandatory. The ESS, a self-administered questionnaire consisting eight questions (▶Table 3.5), is the most common scale used to assess sleepiness. The patient scores from 0 to 3 depending on the chance of dozing. A score of more than 10 is considered excessive daytime sleepiness. OSA is not the only cause of sleepiness: sleep deprivation, shift-work, occupation, and lifestyle are also common factors. Although not every patient with OSA is sleepy, and sleepiness is not related to the severity of sleep apnea, if present, it is one of the symptoms that usually improve in case of successful treatment.
Questions about quality of life (QoL) should be included as well. Snoring, sleepiness, and QoL scales help monitor any improvement in the patient after treatment.
3.2.3 General Evaluation
As in all diseases, after taking the history, a clinical evaluation follows. OSA-specific examination differs from standard ENT examination. It is essential that the exploration is systematic and that one pays attention to all the details.
It starts with the general aspect of the patient. Males have a higher probability of suffering from OSA than females. The male:female ratio is approximately 2:1. Prevalence increases with BMI. 9 Height, weight, and BMI (measured as weight in kilogram divided by height squared in meters) are essential. Neck circumference is related to the probability of having OSA. There are increased chances of having OSA if neck circumference is more than 42 cm in men and more than 37 cm in women.
Some studies suggest that neck–height ratio correlates better with AHI than BMI. 10 , 11 Fat deposition both around the pharynx and within upper airway dilator muscles, such as the genioglossus, decreases airway lumen size and causes detrimental changes to upper airway muscle function. Abdominal obesity compresses the abdomen and thoracic cavities, reducing lung volume and possibly causing rostral shifting of the diaphragm, which reduces tracheal tension and thus impairs pharyngeal mechanism. Therefore, fat deposition around the pharynx and trunk both increase airway collapsibility. 12
Face profile may indicate box problems. A person has retrognathia if the menton is more than 2 mm behind the vertical line that starts at the anterior nasal spine. A patient with a long face, also called dolichofacial, will also have a smaller upper airway. In such patients activation of the genioglossus muscle during sleep after apneic or hypopneic events will not widen the lumen of the upper airway as compared with patients having a wide face. (▶Fig. 3.1) A lower position of the hyoid bone will lead to a longer upper airway that is more collapsible than a shorter one.
3.2.4 Upper Airway Evaluation
A systematic upper airway evaluation starts with examination of the nose. Most of the resistance in the upper airway is generated in this area. One looks for nasal valve collapse, septal deviation, possible septal perforation, synechiae, hypertrophy of the inferior turbinates, a concha media bullosa, mucosal swelling due to allergy or aspecific hypersensitivity, nasal polyps, and hypertrophy of the adenoid. Although correction of decreased nasal breathing, either medically or surgically, might improve quality of sleep, it will usually not lead to a significant improvement of the AHI or other sleep-study parameters. A patent nose might also improve CPAP and MAD adherence. Surgery of the nose is very rarely the first-line treatment in OSA. The role of the nose in OSA is less important than previously thought. That said, an obstructed nose should be fixed; although one should not expect a resolution of the snoring or apneas, it is true that SDB will be cured in some rare cases. Please refer to Chapter 7.2 for more information.
Subsequently, the surgeon should examine the patient’s mouth to assess the dental status. Gums and teeth should be checked. Occlusion is classified with Angle’s classification (▶Fig. 3.2). Type 2 occlusion is often linked to retrognathia and denotes a smaller upper airway at the level of the tongue base. Type 3 patients are not good candidates for treatment with a MAD (see Chapter 4). In order to use an MAD, the gums and teeth must be in good condition. Surgeons should also check the maximum mouth opening. If it is limited then problems may arise during intubation and surgery. For example, if a patient cannot open his/her mouth, then it is best to use the laser for lingual tonsillectomy instead of the coblator. Sometimes it is better to intubate nasally.
The hard palate can be narrow. This is frequently seen in people with a history of oral breathing during childhood and will lead to a narrow upper airway and a long palate. This might have consequences for performing surgery. One can suspect a narrow palate if the patient has a long face.
Presence of crowded teeth can make the surgeons think about a small box problem. If the tongue has edge crenations (▶Fig. 3.3) caused by teeth pressure, this can be related to a small box, a (relative) macroglossia (big tongue), or both. The presence of these tongue crenations has been related to presence of obstructive sleep apnea syndrome (OSAS). 13
The relative position of the palate and the tongue can be assessed using two scales: the Mallampati classification (▶Fig. 3.4) and Friedman palate classification, previously called modified Mallampati index (MMI) (▶Fig. 3.5). The patient is asked to breathe through the nose to avoid elevation of the palate that would decrease the degree. In the Mallampati classification, the patient is asked to protrude the tongue, whereas in the Friedman classification, the tongue has to stay inside the mouth without elevating it and without depressing it with a tongue spatula. Therefore, both classifications are not equal and might have a different degree.
The Friedman staging system is a combination of two classifications and BMI (▶Table 3.6). One classification is the MMI; the other is the tonsil hypertrophy classification from 0 (if a previous tonsillectomy was performed) to 4 (kissing tonsils touching in the midline) (▶Fig. 3.4, ▶Fig. 3.5, ▶Fig. 3.6, ▶Fig. 3.7).
Stage 1 is any patient with big tonsils (grade 3 and 4) and a small tongue (grade 1 or 2). In this type of patients, which occurs in less than 15% of patients with OSA, 14 surgical success of uvulopalatopharyngoplasty (UPPP) assessed by Sher’s criteria (AHI <20, and more reduction of the AHI >50%) is more than 80%. 7 Patients categorized in stage 2 share two characteristics: big tonsils with big tongues or small tonsils with small tongues. In this group, 40% of the patients achieve success after UPPP. Stage 3 patients have a big tongue and small tonsils or tonsillectomies and UPPP success is 8%. 15 Stage 4 is for patients with a BMI greater than 40 or with craniofacial deformities such as severe retrognathia. Surgery is not recommended for stage 4 patients. In a later study, Friedman reported higher surgical success rates, performing tongue base surgery, especially in stage 2 patients but not in stage 3. 16
Although earlier studies have given first information about which patients are suitable for sleep surgery, there are some serious concerns about the use of these forms of clinical examination in an awake situation. The authors are concerned whether the clinical assessment alone in an awake situation is sufficient for a careful surgical planning or not. Some of the concerns and limitations are discussed further.
3.2.5 Limitations
First, stage 1 patients can be cured with a simple tonsillectomy without performing UPPP. To the best of our knowledge, there are no studies comparing tonsillectomy with tonsillectomy and UPPP, but there is a meta-analysis showing that tonsillectomy alone can be useful. 17
Second, in these studies, classical UPPP or Friedman’s Z-palatoplasty was performed. There are studies suggesting that more recent pharyngeal techniques addressing the lateral pharyngeal walls in a different way have better surgical results. 18 , 19 These new techniques are discussed in Chapter 8.
Third, the tongue base surgery performed was interstitial radiofrequency treatment (RFT). RFT turned out to be less effective in reducing the AHI. Nowadays, with the use of new technologies, such as coblation, midline glossectomy, or submucosal minimally invasive lingual excision (SMILE), results are better in patients with grade 3 or 4 Mallampati. 20 , 21 , 22
Besides, although Friedman argues that by using his classification, there is less subjectivity and a good agreement in classifying patients; however, in authors’ experience, patients often unintentionally elevate the tongue when they are asked to open their mouth. The size of the tongue can thus be overestimated, while the contrary also happens. The patient might depress the tongue leading to underestimation of its size. This experience has been recently corroborated by Sundman et al. 23 They recruited 4 ENT specialists and 11 ENT residents who scored themselves and there were 210 observations. The median kappa was 0.36 (first and third quartile, 0.23 and 0.42), corresponding to only a slight agreement. This study concluded that the Friedman tongue position demonstrated only a slight inter-examiner agreement among 15 medical doctors, indicating that the method is difficult to perform and could be an uncertain method to select patients for UPPP.
For demonstration purposes, an example of the same individual is shown (▶Fig. 3.8), in which all four stages are demonstrated depending on the tongue and palate position.
Moreover, a higher MMI does not always imply more collapse at the tongue base during sedation or natural sleep. 16 , 17 , 18 , 19 , 24 , 25 , 26 , 27 A big tongue can push the palate against the posterior wall causing a retropalatal obstruction. Small tongues can have retroglossal collapse due to lingual tonsil hypertrophy or loss of muscle tone. This has been discussed in other sections of this chapter (see Chapter 3.4.2).
In both the Friedman and Mallampati classifications the shape, position, and length of palate and uvula are not further specified. In authors’ view this is another important limitation of both systems.
▶Fig. 3.9 shows a “normal” palate. Although a clear consensus is lacking on the definition of the normal palate. If the length of the uvula is less than 15 mm and the width is less than 10 mm, the palate is considered normal.
In contrast, ▶Fig. 3.10 shows a long palate and uvula.
▶Fig. 3.11 shows examples of isolated long uvulae (uvula elongata), whereas ▶Fig. 3.12 shows a uvula which is both wide and long.
The mildest form of cleft palate, uvula bifida (▶Fig. 3.13), has been occasionally seen as well. Palatal surgery has the risk of velopharyngeal insufficiency and rhinolalia aperta.
In the Middle East and North Africa, there is a traditional practice to remove uvula soon after the birth of a child (ritual uvulectomy, ▶Fig. 3.14). This intervention is thought to protect against throat disease. 27 This procedure, in addition to many other complications, might accidentally lead to palatal stenosis.
▶Fig. 3.15 shows a status after UPPP, with a visual good result, with a wide opening and sufficient distance between the palate and the posterior pharyngeal wall.
▶Fig. 3.16 shows the situation after Zetaplasty. The fibrotic tissue in the anterolateral direction (“the scar tissue works for you”) can be clearly seen.
▶Fig. 3.17, ▶Fig. 3.18, and ▶Fig. 3.19 show several degrees of palatal stenosis after incorrectly performed palatal techniques. Such unfortunate situations might occur if the surgeon places his/her suture under too much tension or leaves the tonsils in situ, hoping to reduce pain for the patient. These iatrogenic complications should be avoided at all costs.
Palatal stenosis is scored in 3 grades (▶Fig. 3.20). Often grades I and II can be improved by a modified Zetaplasty to some extent, while grade II needs to be treated by (LASER) excision and the long-term use of an obturator. This situation is extremely rare.
A possible collapse of the epiglottis can be missed with awake examination. Isolated epiglottis collapse is only assessed by DISE and will be discussed in the next section.
3.2.6 Conclusion
The workup of a patient with suspected OSA begins with taking a medical history and performing a physical examination. General examination of the head and neck area and trunk, (height, length, BMI, neck circumference), external examination of the head and neck area, (shape of the face, position of the mandible), and an internal examination of the upper airway, including dental status, are performed. The Friedman and Mallampati classifications are used to assess the upper airway, providing a first general impression of tonsil and tongue size. However, both are insufficient for a careful assessment, in particular when surgery is considered. They both provide common information, with a rough first impression of the anatomical situation of the upper airway. In particular, they do not provide information on dental status, or of the length, shape, and position of the palate and uvula, neither of possible epiglottis collapse. There may be practical reasons to rely on clinical examination only, such as when insurance companies do not reimburse drug-induced sleep endoscopy (DISE). But in our philosophy, in the ideal situation, in case other treatment than CPAP (surgery, MAD, combined treatment) is considered, additional DISE is indicated (discussed in Chapter 3.4.2). A patient who considers upper airway surgery for OSA has a right to a careful and comprehensive workup.
3.3 Sleep Studies
Sleep studies summarize various parameters on sleep, ventilation, oxygen saturation, and limb movement. They help evaluate the sleep profile and identify sleep-associated diseases. Sleep studies are typically divided into four categories based on the number of channels and presence or absence of attendance during the procedure. The attended PSG (Level 1) is the gold standard for diagnosis of sleep-related breathing disturbances (SRBDs), and allows a reliable diagnosis taking into account all aspects obtained in the differential diagnosis. Attended PSG is recommended in patients with comorbidities such as chronic obstructive pulmonary disease, heart failure, or renal failure, in patients with low pretest probability, and in patients suspected of sleep-related diseases other than OSA. A third-level device (polygraphy) is sufficient to confirm the diagnosis in patients with a high pretest probability for OSA. Third and fourth level devices are screening tools to evaluate the pretest probability of the presence or absence of SRBD, but they are not suitable for final diagnosis.
SRBDs are present in three major entities, differing substantially in pathophysiological and therapeutical aspects. They comprise obstructive and central sleep apnea (CSA) as well as alveolar hypoventilation syndromes. The detection, differentiation, and clinical evaluation of SRBD requires profound anamnesis and physical examination along with extensive medical tests, including specific questionnaires, as well as electrophysiological and instrument-based tests.
3.3.1 Differential Diagnosis in Sleep-Related Breathing Disturbances
With an incidence of 90%, OSA is the most common phenotype of SRBD. The prevalence of OSA increases with age and BMI and is associated with a restriction of the upper airways. 28 There is a higher prevalence in men (13%) as compared to women (6%). 29 Skeletal dysmorphias of the facial skull, enlarged tonsils and tongue, fat deposits at the soft tissues of the neck as well as cranial movement of the pulmonic-bronchial-tracheal system predispose to upper airway obstruction. Dilating muscles compensate upper airway obstruction during daytime, but muscle relaxation during sleep supports a reduction or complete interruption of respiratory flow. 30 , 31 One major aspect in the distinction toward CSA is the continuation of respiratory effort in case of OSA.
CSA represents about 10% of SRBD. Within the group of patients suffering from neurological diseases (e.g., stroke), renal failure, and especially systolic heart failure, prevalence of CSA is estimated at 21 to 37%. 32 , 33 CSA is characterized by the cessation of airflow and accompanying absence of respiratory effort. The pathogenesis of CSA is still object of research, so by now the classification of CSA is based on the etiology (e.g., CSA due to medication, high altitude, systolic heart failure, etc.). 34 However, the “loop gain” idea in patients with periodic breathing due to systolic heart failure represents a key concept in understanding the phenomenon of CSA. 35 The loop gain is based on a model known from engineering, defining the response of the respiratory system to changes of PaCO2. Patients with systolic heart failure show an increased left atrial pressure and an accompanying tendency to chronic hyperventilation. In addition, overreacting chemoreceptors support an overshoot of ventilation in case of slight fluctuations of PaCO2, which are quite common during the transition from wakefulness to sleep. An additional reduction of PaCO2 is followed by a decrease of respiratory drive, which leads to apneas or hypopneas and an increase of PaCO2 again. Thus, overreacting chemoreceptors result in alternating conditions of hyper- and hypoventilation (waxing and waning). 35
Alveolar hypoventilation syndromes are characterized by a reduction of minute ventilation leading to chronic hypercapnia. Various diseases induce alveolar hypoventilation including diseases of respiratory drive, for example, obesity hypoventilation syndrome (OHS), failure of the neuromuscular or skeletal system (e.g., muscular dystrophy, diaphragm paralysis, or scoliosis), and pulmonary disorders such as the chronic obstructive lung disease. As minute ventilation during sleep is reduced in healthy persons by 10 to 15%, chronic ventilatory failure unmasks at first during sleep. Typical characteristics include an increase of PaCO2 with a subsidiary sustained decrease of oxygen saturation. Alveolar hypoventilation syndromes often remain undetected in sleep medicine, as a continuous reduction of flow and effort may be overseen in PSG and necessary PCO2 measurements (e.g., transcutaneous capnometry) are not performed routinely.
Especially, OHS often coexists with OSA (9–14%) and is defined by a BMI greater than or equal to 30 kg/m2 and PaCO2 greater than or equal to 45 mm Hg when awake. 36 , 37 , 38 Other diseases causing hypoventilation need to be excluded. OHS is estimated in about 10% of patients with a BMI between 30 and 35 kg/m2. The prevalence rises to 50% in patients with a BMI greater than 50 kg/m2. 36 , 37 , 39 Obesity requires increased respiratory work due to the larger body mass. Obese patients present with reduced diaphragm mobility, increased oxygen demand, restriction of the upper airways, and increased ventilation/perfusion-mismatch. OHS patients are not able to maintain the increased respiratory work and present with dampened ventilatory response to hypoxia and changes of PaCO2 levels. 40 , 41 , 42
Multiple studies have shown that all entities of SRBD are associated with an increased mortality and morbidity. 39 , 43 , 44 , 45 , 46 OSA has proven to be an independent risk factor of arterial hypertension, atrial fibrillation, systolic and diastolic heart failure as well as cerebrovascular events. 47 , 48 Patients suffering from OHS have poorer quality of life and have a higher risk of developing pulmonary hypertension (PH) 49 compared with OSA patients. The consequences of CSA on the cardiovascular system are less clear, but besides an increased mortality there is evidence that patients with CSA show a higher risk for heart transplantation than patients without SRBD. 39
An early and profound distinction between the entities of SRBD is of crucial importance due to the differences in pathophysiology and therapeutic approach.
3.3.2 Clinical Investigations
Patients’ Approach
The ICSD includes 60 specific diagnoses and 6 major categories including SRBD, insomnia, parasomnia, and sleep-related movement disorders. The latest edition, ICSD-3, was released in 2014. 34
SRBD are associated with cyclic oxygen desaturations and catecholamine release, inducing arousals, systemic inflammation, and oxidative stress. 50 Arousals result in fragmented sleep and therefore prevent slow wave and rapid eye movement (REM) sleep. As a consequence the recovery effect is reduced and sleepiness during the day is increased. Typical symptoms of OSA and OHS include snoring, witnessed apneas, daytime sleepiness, reduced concentration, momentary nodding off, morning headaches, depression, and nocturnal dyspnea (▶Table 3.7). CSA is not associated with any specific symptoms, but can be accompanied by unspecific symptoms such as fatigue and daytime sleepiness. 39
During sleep | Awake |
Nonrestorative sleep Apneas witnessed by bed partner Awaking with choking Nocturnal restlessness Vivid dreams Gastroesophageal reflux Nocturia Diaphoresis Hypersalivation Insomnia with frequent awakenings | Daytime sleepiness Lack of concentration Cognitive deficits Changes in mood Morning headaches Dry mouth Impotence or decreased libido |
The ICSD-3 defines insomnia as “a repeated difficulty with sleep initiation, duration, consolidation, or quality that occurs despite adequate opportunity and circumstances for sleep, and results in some form of daytime impairment.” 34 Anxiety, frustration, irritability, and mood changes during daytime are accompanying phenomena.
Parasomnia includes episodes of dream-enacting behavior during REM sleep as well as episodes of inconsolable screaming and amnesia during the first third of the night. Information from the bed partner on parasomnia can be of crucial importance for the evaluation of abnormal events during sleep.
Sleep-related movement disorders comprise restless legs syndrome (RLS) and periodic limb movements (PLM). Typical symptoms include discomfort or pain shortly before sleep onset (RLS) in addition to daytime sleepiness and fatigue (PLM). 34
Physical Examination
Physical examination includes evaluation of the oral cavity, skeletal status, and soft tissue factors. The Mallampati score identifies patients with restrictions of the oropharyngeal cavity. 51 Liistro et al found that the visualization of the soft palate (Grade I–IV) is an associated risk factor for OSA (▶Fig. 3.4). 52
Maxillary and/or mandibular hypoplasia as well as the retroposition and inferior displacement of the hyoid reduce the diameter of the posterior airway space. Cephalometry (▶Fig. 3.21) visualizes and evaluates the anatomy of the upper airways.
Abnormalities of the soft tissue include adenotonsillar hypertrophy, macroglossia, fat deposition, pharyngeal inflammation, and edema. CT of head and throat 31 as well as acoustic pharyngometry quantify the geometry of oral cavity.
Thus, Young et al showed that body weight, neck circumference, and BMI correlate with the presence and severity of OSA. 53 Moreover, any signs of heart failure (e.g., edema of the lower leg) or neurological diseases (e.g., sensorimotor deficits) can be a first indicator for CSA.
Questionnaires
Questionnaires help detect and quantify SRBD as well as other entities of sleep-associated diseases. Although not predictive, they can be used as a first indicator for SRBD. The most important questionnaires used today are the ESS, the Berlin questionnaire, and the STOP-BANG questionnaire.
The ESS is the most common instrument to evaluate sleepiness. It assesses the patient’s probability of falling asleep in eight different everyday situations. Numerical values ranging from 0 (“I never fall asleep”) to 3 (“I have a high probability of falling asleep”) are used to quantify sleepiness. A score above or equal to 10 is indicative for daytime sleepiness and requires further investigation. 1
The Berlin questionnaire identifies OSA patients in primary care and general population. It is characterized by 10 questions divided into 3 categories investigating the severity of snoring, daytime sleepiness, and history of high blood pressure or obesity. Predefined questions are assigned to a certain score. SRBD becomes very likely if two categories are classified positive. 54 The specificity and sensitivity of the Berlin questionnaire to detect OSA is 80% and 40%, respectively. 55
The STOP-BANG questionnaire was originally developed for preoperative situations. 56 It consists of two domains: A self-assessment questionnaire evaluating snoring, tiredness, witnessed apneas, and treatment of arterial hypertension. The second part collects anthropometric data including BMI, age, neck circumference, and sex. The score is calculated by assigning one point to each positive answer (BMI >35 kg/m2, male gender, neck circumference >40 cm, and age >50 years). A score above or equal to 3 indicates a high probability for OSA. Compared to the Berlin questionnaire, STOP-BANG is more sensitive (87%) in identifying subjects with moderate to severe SRBD and is gaining clinical acceptance. 57
Electrophysiological Tests
The most common electrophysiological tests to objectify central nervous activation and daytime sleepiness are the multiple sleep latency test (MSLT), the multiple wakefulness test (MWT), and the Oxford Sleep Resistance (OSLER) test. 58
The MSLT evaluates daytime sleep latency at five standardized times during the day, based on electroencephalogram (EEG), electro-oculogram (EOG), and electromyogram (EMG), assuming that sleep latency decreases with sleepiness. The American Sleep Disorders Association (ASDA) evaluated degrees of severity based on experience and defined age-dependent norms. 58 While mean sleep latency at the age of 20 is 10.4 minutes, it is 12.1 minutes at the age of 50, and 15.2 minutes at the age of 80. REM sleep less than 10 minutes after sleep onset in combination with a reduced sleep latency (<8 minutes) indicates narcolepsy (see ICSD3). 34
The MWT evaluates sleepiness that overpowers a patients’ ability to stay awake in a sleep-promoting environment. Its results correlate with the severity of OSA and reflect improvement during treatment. 59 , 60 The Atlas Task Force of American Academy of Sleep Medicine (AASM) defined guidelines for the protocol of the procedure. In four 40-minute sessions separated by 2-hour intervals, patients are observed by EEG, EOG, and EMG. Sleep latency, total sleep time (TST), and sleep stages are evaluated. 61 The AASM refers to data of healthy people in a clinical trial performed by Doghramji et al from 1997 to define standard values. 62 Sleep onset below 13 minutes is defined pathological.
The OSLER test is a modification of the MWT. In addition to the setup of the MWT, patients are instructed to respond to a dim light flash at 3-second intervals during a 40-minute period by hitting a button on a portable device. 63 Missing responses for 21 seconds (e.g., 7 consecutive flashes) suggest that the person has fallen asleep and the test is finished. 64
These three tests are of crucial relevance in the evaluation of daytime sleepiness and central nervous activation. Although there is a limited validity of each test alone, they represent a beneficial tool in the diagnostical algorithm in addition to a comprehensive patient history and PSG.
3.3.3 Computer-Based Tests of Vigilance and Attention
Computer-based tests such as Carda, driving simulation tests, or the Pupillographic Sleepiness Test evaluate qualities of attention and vigilance.
Carda is a computer-assisted response test to evaluate sustained attention. The image of a road is projected on a black background and patients are required to respond to obstacles. They are visible for 20 milliseconds and randomly appear on the road about 100 times in 10 minutes. 65
Driving-simulation tests focus on different components of attention, especially divided attention or sustained attention. Reaction time and directional stability represent target parameters. 66 Risser et al found higher error rates in OSA patients as compared with healthy controls. Sufficient therapy reduced failures significantly. 67
The Pupillographic Sleepiness Test is based on the correlation of fluctuations of the pupil width and daytime sleepiness. 58 The pupil behavior is recorded by an infrared-sensitive camera. Target variables are the pupillary unrest index (PUI) in millimeters per minute and the amplitude spectrum. Based on the data of 349 healthy people, unusual values were defined starting from a PUI more than or equal to 1.89 mm/min, pathological values from a PUI more than or equal to 2.28 mm/min. 68
In the setting of the Quatember Maly test, patients observe a moving point inside a circle and respond to double jumps. Seven missed or false reactions are considered pathological. Rühle et al found that the Quatember Maly test shows the best sensitivity (61.9%) and specificity (72.2%) as compared to the PUI (47.6%/57.9%) or the Carda test (23.8%/21.1%). 69
Although instrument-based tests have no specific validity, they help evaluate the ability to concentrate, reaction time, attention during monotonous activities, and objectify therapy responsiveness during the course of the disease.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
