Original Paper Neuropsychobiology 2014;70:189–194 DOI: 10.1159/000365486
Received: November 12, 2013 Accepted after revision: June 16, 2014 Published online: November 4, 2014
Sleepiness and Performance Is Disproportionate in Patients with Non-Organic Hypersomnia in Comparison to Patients with Narcolepsy and Mild to Moderate Obstructive Sleep Apnoea Nicole C. Kofmel a Wolfgang J. Schmitt c Christian W. Hess a Matthias Gugger b Johannes Mathis a Departments of a Neurology and b Pneumology, Inselspital, Bern University Hospital, and c Department of Psychiatry, University Hospital of Psychiatry, University of Bern, Bern, Switzerland
Abstract Background/Aims: Clinical differentiation between organic hypersomnia and non-organic hypersomnia (NOH) is challenging. We aimed to determine the diagnostic value of sleepiness and performance tests in patients with excessive daytime sleepiness (EDS) of organic and non-organic origin. Methods: We conducted a retrospective comparison of the multiple sleep latency test (MSLT), pupillography, and the Steer Clear performance test in three patient groups complaining of EDS: 19 patients with NOH, 23 patients with narcolepsy (NAR), and 46 patients with mild to moderate obstructive sleep apnoea syndrome (OSAS). Results: As required by the inclusion criteria, all patients had Epworth Sleepiness Scale (ESS) scores >10. The mean sleep latency in the MSLT indicated mild objective sleepiness in NOH (8.1 ± 4.0 min) and OSAS (7.2 ± 4.1 min), but more severe sleepiness in NAR (2.5 ± 2.0 min). The difference between NAR and the other two groups was significant; the difference between NOH and OSAS was not. In the Steer Clear
© 2014 S. Karger AG, Basel 0302–282X/14/0703–0189$39.50/0 E-Mail [email protected] www.karger.com/nps
performance test, NOH patients performed worst (error rate = 10.4%) followed by NAR (8.0%) and OSAS patients (5.9%; p = 0.008). The difference between OSAS and the other two groups was significant, but not between NOH and NAR. The pupillary unrest index was found to be highest in NAR (11.5) followed by NOH (9.2) and OSAS (7.4; n.s.). Conclusion: A high error rate in the Steer Clear performance test along with mild sleepiness in an objective sleepiness test (MSLT) in a patient with subjective sleepiness (ESS) is suggestive of NOH. This disproportionately high error rate in NOH may be caused by factors unrelated to sleep pressure, such as anergia, reduced attention and motivation affecting performance, but not conventional sleepiness measurements. © 2014 S. Karger AG, Basel
Introduction
The differential diagnosis in patients with excessive daytime sleepiness (EDS) is broad, including sleep insufficiency syndrome, inadequate sleep hygiene, drug-related reasons, obstructive sleep apnoea syndrome (OSAS), periodic leg movement disorder, narcolepsy (NAR) with or without cataplexy, idiopathic hypersomnia, and nonorganic hypersomnia (NOH) [1]. However, the differenProf. Dr. med. Johannes Mathis Department of Neurology, Sleep-Wake-Centre, Inselspital Bern University Hospital, University of Bern CH–3010 Bern (Switzerland) E-Mail Johannes.mathis @ insel.ch
Downloaded by: University of Aberdeen139.133.11.2 - 5/22/2015 12:48:38 PM
Key Words Excessive daytime sleepiness · Depression · Non-organic hypersomnia · Narcolepsy · Obstructive sleep apnoea syndrome · Atypical depression · Hypersomnolent depression
190
Neuropsychobiology 2014;70:189–194 DOI: 10.1159/000365486
with objective measures of sleep [10, 17, 18]. In the ICSD2 [1] the discrepancy between the subjective feeling of sleepiness in NOH patients and the often normal objective latency in MSLT is mentioned as an ‘unresolved issue’ calling for further research on the question why the majority of patients with mental disorders complain of insomnia while a minority complain of hypersomnia and/or EDS. In order to distinguish EDS from tiredness and fatigue, the interviewing physician will primarily use the Epworth Sleepiness Scale (ESS) [19], which contains questions referring to the likelihood of falling asleep under various conditions. However, a differentiation of various causes of EDS is not possible with this clinical tool. The purpose of this pilot study was first to find out whether EDS in NOH, as indicated by an increased ESS, can be objectively confirmed and specified by using the MSLT, the Steer Clear performance test, and pupillography. Secondly, we analysed whether this multimodal approach can help to differentiate NOH from other causes of EDS. We chose NAR and OSAS rather than idiopathic hypersomnia for this pilot study because these former entities can be diagnosed with more certainty. Methods Patients with EDS who had been systematically studied with PSG, ESS, and a battery of sleepiness and vigilance tests (MSLT, Steer Clear performance test, and pupillography) were recruited from the outpatient clinic of the Sleep-Wake-Centre at the University of Bern. Stimulants, hypnotics and other drugs on demand were routinely paused during the assessments, but thyroxine, antidepressants and cardiovascular drugs were continued. Patients were referred to a psychiatrist with extensive experience in sleep medicine (W.J.S.) if no diagnosis of OSAS, periodic leg movement disorder, or NAR could be established and after medical conditions causing sleepiness or drug abuse, or other causes of EDS such as inadequate sleep hygiene or insufficient sleep had been ruled out by testing for at least 1 week with prolonged night-time sleep and regular bedtimes. Patients with a depressive episode, recurrent depressive disorder, and dysthymia according to ICD-10 criteria were included in the NOH group if EDS was their major complaint and the ESS was above 10. All had PSG, MSLT, and a Steer Clear performance test, and the majority also underwent pupillography and actigraphy. Patients with schizoaffective disorders, organic depression, other psychiatric co-morbidity, or any other detectable cause of EDS were not included. The NAR group included consecutive patients who, apart from the typical clinical picture of NAR, had a mean sleep latency under 8 min and two or more sleep onset REMs in the MSLT [8]. However, NAR patients with co-morbid OSAS, periodic leg movement disorder, or psychiatric disorders were not included.
Kofmel /Schmitt /Hess /Gugger /Mathis
Downloaded by: University of Aberdeen139.133.11.2 - 5/22/2015 12:48:38 PM
tiation between some of those entities remains a major challenge in spite of clinical studies with day-night protocols and objective assessments with actigraphy, polysomnography (PSG), and multiple sleep latency test (MSLT). In particular, the clinically relevant discrimination of NAR without cataplexy and idiopathic hypersomnia with or without prolonged sleep from NOH is often difficult. Organic hypersomniacs will usually be treated by stimulants, whereas in NOH antidepressants will be considered first, making the diagnostic distinction important. One of the earliest clinically based attempts to differentiate hypersomnia of organic from hypersomnia of nonorganic origin was that of Bedrich Roth [2], who coined the term ‘neurotic hypersomnia’ for hypersomnic patients with psychiatric disorders. Interestingly, this very experienced expert and pioneer in sleep-wake disorders altered some of the terminology from his first book in 1962 [3] to his second book in 1980 [4], i.e., changing the diagnoses ‘neurotic hypersomnia’ into ‘idiopathic hypersomnia’, which underlines the ambiguity in differentiating between these entities. The diagnosis of NOH associated with depression or other psychiatric disorders is complicated by small but relevant differences in the three major classification systems ICSD-2, ICD-10 and DSM-IV [1, 5, 6]. Unlike the other classifications, ICSD-2 requires reduced sleep efficacy in nocturnal PSG. An excellent overview of this issue has been published by Kaplan and Harvey [7], and they rightly state that surprisingly little research is available on hypersomnia in mood disorders. Patients with major depression usually complain of tiredness, fatigue, lack of energy, and insomnia. However, up to 40% rather complain of hypersomnia or EDS [8, 9]. In atypical depression, hypersomnia is one of the key features. Moreover, hypersomnia in mood disorders is not only a frequent complaint but seems particularly resistant to antidepressant treatment. Note that the term ‘hypersomnia’ – in its narrow sense indicating prolonged sleep over 24 h – within the diagnostic entity of ‘non-organic hypersomnia’ also refers to EDS. Objective data from nocturnal PSG in depressed patients usually show an increased sleep latency [8, 10–13], sleep fragmentation [8, 10–12, 14, 15], reduced sleep efficiency [8, 10–13], or early awakening [8, 12]. In the MSLT, depressed patients usually show normal sleep latencies [16]. When describing their subjective complaints, patients do not always differentiate reliably between ‘sleepiness’ on the one hand and ‘tiredness’ or ‘fatigue’ on the other. This might explain why subjective reports of sleepiness in depressed patients do not always correlate
Table 1. Mean values (SD) of vigilance tests within groups and corresponding p values of the univariate analysis of variance
n Female, % Age, years ESS score MSLT, min MSLT 38% PSG, sleep efficiency
Received: November 12, 2013 Accepted after revision: June 16, 2014 Published online: November 4, 2014
Sleepiness and Performance Is Disproportionate in Patients with Non-Organic Hypersomnia in Comparison to Patients with Narcolepsy and Mild to Moderate Obstructive Sleep Apnoea Nicole C. Kofmel a Wolfgang J. Schmitt c Christian W. Hess a Matthias Gugger b Johannes Mathis a Departments of a Neurology and b Pneumology, Inselspital, Bern University Hospital, and c Department of Psychiatry, University Hospital of Psychiatry, University of Bern, Bern, Switzerland
Abstract Background/Aims: Clinical differentiation between organic hypersomnia and non-organic hypersomnia (NOH) is challenging. We aimed to determine the diagnostic value of sleepiness and performance tests in patients with excessive daytime sleepiness (EDS) of organic and non-organic origin. Methods: We conducted a retrospective comparison of the multiple sleep latency test (MSLT), pupillography, and the Steer Clear performance test in three patient groups complaining of EDS: 19 patients with NOH, 23 patients with narcolepsy (NAR), and 46 patients with mild to moderate obstructive sleep apnoea syndrome (OSAS). Results: As required by the inclusion criteria, all patients had Epworth Sleepiness Scale (ESS) scores >10. The mean sleep latency in the MSLT indicated mild objective sleepiness in NOH (8.1 ± 4.0 min) and OSAS (7.2 ± 4.1 min), but more severe sleepiness in NAR (2.5 ± 2.0 min). The difference between NAR and the other two groups was significant; the difference between NOH and OSAS was not. In the Steer Clear
© 2014 S. Karger AG, Basel 0302–282X/14/0703–0189$39.50/0 E-Mail [email protected] www.karger.com/nps
performance test, NOH patients performed worst (error rate = 10.4%) followed by NAR (8.0%) and OSAS patients (5.9%; p = 0.008). The difference between OSAS and the other two groups was significant, but not between NOH and NAR. The pupillary unrest index was found to be highest in NAR (11.5) followed by NOH (9.2) and OSAS (7.4; n.s.). Conclusion: A high error rate in the Steer Clear performance test along with mild sleepiness in an objective sleepiness test (MSLT) in a patient with subjective sleepiness (ESS) is suggestive of NOH. This disproportionately high error rate in NOH may be caused by factors unrelated to sleep pressure, such as anergia, reduced attention and motivation affecting performance, but not conventional sleepiness measurements. © 2014 S. Karger AG, Basel
Introduction
The differential diagnosis in patients with excessive daytime sleepiness (EDS) is broad, including sleep insufficiency syndrome, inadequate sleep hygiene, drug-related reasons, obstructive sleep apnoea syndrome (OSAS), periodic leg movement disorder, narcolepsy (NAR) with or without cataplexy, idiopathic hypersomnia, and nonorganic hypersomnia (NOH) [1]. However, the differenProf. Dr. med. Johannes Mathis Department of Neurology, Sleep-Wake-Centre, Inselspital Bern University Hospital, University of Bern CH–3010 Bern (Switzerland) E-Mail Johannes.mathis @ insel.ch
Downloaded by: University of Aberdeen139.133.11.2 - 5/22/2015 12:48:38 PM
Key Words Excessive daytime sleepiness · Depression · Non-organic hypersomnia · Narcolepsy · Obstructive sleep apnoea syndrome · Atypical depression · Hypersomnolent depression
190
Neuropsychobiology 2014;70:189–194 DOI: 10.1159/000365486
with objective measures of sleep [10, 17, 18]. In the ICSD2 [1] the discrepancy between the subjective feeling of sleepiness in NOH patients and the often normal objective latency in MSLT is mentioned as an ‘unresolved issue’ calling for further research on the question why the majority of patients with mental disorders complain of insomnia while a minority complain of hypersomnia and/or EDS. In order to distinguish EDS from tiredness and fatigue, the interviewing physician will primarily use the Epworth Sleepiness Scale (ESS) [19], which contains questions referring to the likelihood of falling asleep under various conditions. However, a differentiation of various causes of EDS is not possible with this clinical tool. The purpose of this pilot study was first to find out whether EDS in NOH, as indicated by an increased ESS, can be objectively confirmed and specified by using the MSLT, the Steer Clear performance test, and pupillography. Secondly, we analysed whether this multimodal approach can help to differentiate NOH from other causes of EDS. We chose NAR and OSAS rather than idiopathic hypersomnia for this pilot study because these former entities can be diagnosed with more certainty. Methods Patients with EDS who had been systematically studied with PSG, ESS, and a battery of sleepiness and vigilance tests (MSLT, Steer Clear performance test, and pupillography) were recruited from the outpatient clinic of the Sleep-Wake-Centre at the University of Bern. Stimulants, hypnotics and other drugs on demand were routinely paused during the assessments, but thyroxine, antidepressants and cardiovascular drugs were continued. Patients were referred to a psychiatrist with extensive experience in sleep medicine (W.J.S.) if no diagnosis of OSAS, periodic leg movement disorder, or NAR could be established and after medical conditions causing sleepiness or drug abuse, or other causes of EDS such as inadequate sleep hygiene or insufficient sleep had been ruled out by testing for at least 1 week with prolonged night-time sleep and regular bedtimes. Patients with a depressive episode, recurrent depressive disorder, and dysthymia according to ICD-10 criteria were included in the NOH group if EDS was their major complaint and the ESS was above 10. All had PSG, MSLT, and a Steer Clear performance test, and the majority also underwent pupillography and actigraphy. Patients with schizoaffective disorders, organic depression, other psychiatric co-morbidity, or any other detectable cause of EDS were not included. The NAR group included consecutive patients who, apart from the typical clinical picture of NAR, had a mean sleep latency under 8 min and two or more sleep onset REMs in the MSLT [8]. However, NAR patients with co-morbid OSAS, periodic leg movement disorder, or psychiatric disorders were not included.
Kofmel /Schmitt /Hess /Gugger /Mathis
Downloaded by: University of Aberdeen139.133.11.2 - 5/22/2015 12:48:38 PM
tiation between some of those entities remains a major challenge in spite of clinical studies with day-night protocols and objective assessments with actigraphy, polysomnography (PSG), and multiple sleep latency test (MSLT). In particular, the clinically relevant discrimination of NAR without cataplexy and idiopathic hypersomnia with or without prolonged sleep from NOH is often difficult. Organic hypersomniacs will usually be treated by stimulants, whereas in NOH antidepressants will be considered first, making the diagnostic distinction important. One of the earliest clinically based attempts to differentiate hypersomnia of organic from hypersomnia of nonorganic origin was that of Bedrich Roth [2], who coined the term ‘neurotic hypersomnia’ for hypersomnic patients with psychiatric disorders. Interestingly, this very experienced expert and pioneer in sleep-wake disorders altered some of the terminology from his first book in 1962 [3] to his second book in 1980 [4], i.e., changing the diagnoses ‘neurotic hypersomnia’ into ‘idiopathic hypersomnia’, which underlines the ambiguity in differentiating between these entities. The diagnosis of NOH associated with depression or other psychiatric disorders is complicated by small but relevant differences in the three major classification systems ICSD-2, ICD-10 and DSM-IV [1, 5, 6]. Unlike the other classifications, ICSD-2 requires reduced sleep efficacy in nocturnal PSG. An excellent overview of this issue has been published by Kaplan and Harvey [7], and they rightly state that surprisingly little research is available on hypersomnia in mood disorders. Patients with major depression usually complain of tiredness, fatigue, lack of energy, and insomnia. However, up to 40% rather complain of hypersomnia or EDS [8, 9]. In atypical depression, hypersomnia is one of the key features. Moreover, hypersomnia in mood disorders is not only a frequent complaint but seems particularly resistant to antidepressant treatment. Note that the term ‘hypersomnia’ – in its narrow sense indicating prolonged sleep over 24 h – within the diagnostic entity of ‘non-organic hypersomnia’ also refers to EDS. Objective data from nocturnal PSG in depressed patients usually show an increased sleep latency [8, 10–13], sleep fragmentation [8, 10–12, 14, 15], reduced sleep efficiency [8, 10–13], or early awakening [8, 12]. In the MSLT, depressed patients usually show normal sleep latencies [16]. When describing their subjective complaints, patients do not always differentiate reliably between ‘sleepiness’ on the one hand and ‘tiredness’ or ‘fatigue’ on the other. This might explain why subjective reports of sleepiness in depressed patients do not always correlate
Table 1. Mean values (SD) of vigilance tests within groups and corresponding p values of the univariate analysis of variance
n Female, % Age, years ESS score MSLT, min MSLT 38% PSG, sleep efficiency
