PMC:2953451 / 2011-8071 JSONTXT

{"project":"2_test","denotations":[{"id":"20948585-8165390-38615982","span":{"begin":210,"end":214},"obj":"8165390"},{"id":"20948585-8990055-38615983","span":{"begin":326,"end":330},"obj":"8990055"},{"id":"20948585-8423911-38615984","span":{"begin":364,"end":368},"obj":"8423911"},{"id":"20948585-12499479-38615985","span":{"begin":431,"end":435},"obj":"12499479"},{"id":"20948585-10845816-38615986","span":{"begin":537,"end":541},"obj":"10845816"},{"id":"20948585-17040005-38615987","span":{"begin":619,"end":623},"obj":"17040005"},{"id":"20948585-16459136-38615988","span":{"begin":1063,"end":1067},"obj":"16459136"},{"id":"20948585-7436384-38615989","span":{"begin":1733,"end":1737},"obj":"7436384"},{"id":"20948585-8235243-38615990","span":{"begin":1762,"end":1766},"obj":"8235243"},{"id":"20948585-9693228-38615991","span":{"begin":1784,"end":1788},"obj":"9693228"},{"id":"20948585-19109055-38615992","span":{"begin":1804,"end":1808},"obj":"19109055"},{"id":"20948585-1798881-38615993","span":{"begin":2437,"end":2441},"obj":"1798881"},{"id":"20948585-8723379-38615994","span":{"begin":2454,"end":2458},"obj":"8723379"},{"id":"20948585-15511708-38615995","span":{"begin":2476,"end":2480},"obj":"15511708"},{"id":"20948585-15884036-38615996","span":{"begin":2498,"end":2502},"obj":"15884036"},{"id":"20948585-17557446-38615997","span":{"begin":2510,"end":2514},"obj":"17557446"},{"id":"20948585-17557447-38615998","span":{"begin":2526,"end":2530},"obj":"17557447"},{"id":"20948585-18246987-38616001","span":{"begin":3242,"end":3246},"obj":"18246987"},{"id":"20948585-19109055-38616003","span":{"begin":3653,"end":3657},"obj":"19109055"},{"id":"20948585-16796208-38616004","span":{"begin":3854,"end":3858},"obj":"16796208"},{"id":"20948585-14611622-38616005","span":{"begin":4811,"end":4815},"obj":"14611622"}],"text":"Introduction\nPeriodic leg movements (PLM) during sleep are a sleep related phenomenon with periodic episodes of repetitive stereotypical movements of the lower extremities (American Sleep Disorder Association, 1993). PLM are more frequent in several sleep disorders, including restless legs syndrome (RLS, Montplaisir et al., 1997), narcolepsy (Moscovitch et al., 1993), rapid eye movement sleep behavior disorder (Fantini et al., 2002) as well as in disorders not primarily affecting sleep such as end stage renal disease (Benz et al., 2000). PLM during sleep are also documented in normal subjects (Pennestri et al., 2006) giving rise to the debate about functional relevance.\nTo determine the functional significance of PLM, the criteria used to score PLM during sleep are pivotal. Thus the criteria of detecting single leg movements (LM) as well as determining what constitutes periodicity may have a significant impact on the final result. According to the new standards PLM occur in a series of at least four LM and are separated by 5–90 s (Zucconi et al., 2006). An increased number of PLM during sleep with a PLM index (PLMI, i.e., the number of PLM per hour of total sleep time) of five or more is considered abnormal (American Academy of Sleep Medicine, 2005). Though the PLMI is taken to reflect periodicity in PLM in most studies, periodicity is not always obvious and PLMI may fail to report it. The definition of a period implicit in the PLMI encompasses inter movement intervals (IMI) from 5 to 90 s. This range of IMI may not be sufficiently restrictive to identify periodicity in LM and could allow randomly occurring LM to be considered periodic. On the other hand, IMI may change across the night (Coleman et al., 1980; Pollmächer and Schulz, 1993; Nicolas et al., 1998; Ferri et al., 2009) and PLM will be missed if period criteria are too narrow. Varying IMIs during night may lead to broad IMI distributions or to a distribution showing several local maxima. This makes a wide range of periods appear similarly prevalent, and thus hides intrinsic periodicity. Another way intrinsic periodicity may become difficult to detect is when pooling data from patients with different average periods of PLM. The definition of periodicity of LM is problematic and may not be sensitive enough. This lack of sensitivity could be one of the reasons for the ongoing debate about their functional relevance (Ancoli-Israel et al., 1991; Mendelson, 1996; Hornyak et al., 2004; Carrier et al., 2005; Högl, 2007; Mahowald, 2007).\nA new statistical approach to analyze PLM was developed by Ferri et al. (2006b) who recommend to study the entire distributions of IMI rather than parametric measures as mean and standard deviation. They examined periodicity with a metric, termed the Periodicity Index (PI) measuring the percentage of LM appearing in series of periodic events. Instead of focusing on IMI frequency distributions the PI accounts for certain dependencies between subsequent IMIs. The PI was investigated for homogenous patient groups with RLS, narcolepsy/cataplexy and rapid eye movement sleep behavior disorder and without further potentially sleep relevant comorbidities or medication (Ferri et al., 2006a,b; Manconi et al., 2007). Based on different levels of the PI patients were divided into subgroups. For patients with narcolepsy/cataplexy and RLS a significantly different median of the PI was found (Ferri et al., 2006a). Very recently increased PI was also detected in a subgroup of patients with insomnia, for which no apparent cause could be found and enabled diagnosis of periodic leg movements disorder (PLMD, Ferri et al., 2009). Quantifiable aspects of PLM such as periodicity may serve to differentiate between clinical conditions and may eventually help capturing the nature of the underlying biological phenomenon (Rye, 2006). However, as afore mentioned, quantitative assessment of the periodicity of LM/PLM as measured by IMI is difficult because the empirical probability distribution may obscure underlying periodicity.\nTo date no study uses methods to enhance and reveal intrinsic periodicity before analyzing the statistics of IMI. We here employ the so-called unfolding procedure, which potentially uncovers intrinsic periodicity of LM data that is confounded by external influences. The unfolding procedure was originally developed in a branch of statistical physics called Random Matrix Theory (RMT) to study and compare the correlation between energy levels of heavy atomic nuclei and other complex quantum systems (see Brody et al., 1981; Guhr et al., 1998; Mehta, 2004, for comprehensive reviews). In the context of neurophysiology the concept was recently applied to the eigenvalue spectrum of the cross-correlation matrix calculated from multi-channel EEG (Seba, 2003; Müller et al., 2006; Baier et al., 2007).\nIn the present study we use the unfolding concept to assess the intrinsic periodicity of IMI of PLM examining a typical clinical patient cohort with PLM during sleep and different comorbidities from our database. Instead of studying the absolute IMI sizes, IMI are locally normalized to unit mean event spacing. Under these conditions the distribution of the spacings of randomly distributed events decays exponentially (Poisson distribution). In contrast, for strictly periodic systems the events are equally spaced. Fitting intermediate distributions, which interpolate between these extreme cases, to unfolded IMI distributions we quantify PLM periodicity.\nSpecifically, the goal of our study is to answer the following questions:\nDoes applying the unfolding procedure enhance and reveal periodicity that is not obvious from the measured IMI distribution?\nDoes periodicity differentiate between the clinical conditions of RLS, of OSA and of PLM without RLS or OSA?\nDoes data-driven clustering separate patients into different groups and are such groups clinically homogeneous?\nAnswering these questions could potentially provide further information about the clinical relevance of PLM during sleep."}

{"project":"0_colil","denotations":[{"id":"20948585-8165390-754333","span":{"begin":210,"end":214},"obj":"8165390"},{"id":"20948585-8990055-754334","span":{"begin":326,"end":330},"obj":"8990055"},{"id":"20948585-8423911-754335","span":{"begin":364,"end":368},"obj":"8423911"},{"id":"20948585-12499479-754336","span":{"begin":431,"end":435},"obj":"12499479"},{"id":"20948585-10845816-754337","span":{"begin":537,"end":541},"obj":"10845816"},{"id":"20948585-17040005-754338","span":{"begin":619,"end":623},"obj":"17040005"},{"id":"20948585-16459136-754339","span":{"begin":1063,"end":1067},"obj":"16459136"},{"id":"20948585-7436384-754340","span":{"begin":1733,"end":1737},"obj":"7436384"},{"id":"20948585-8235243-754341","span":{"begin":1762,"end":1766},"obj":"8235243"},{"id":"20948585-9693228-754342","span":{"begin":1784,"end":1788},"obj":"9693228"},{"id":"20948585-19109055-754343","span":{"begin":1804,"end":1808},"obj":"19109055"},{"id":"20948585-1798881-754344","span":{"begin":2437,"end":2441},"obj":"1798881"},{"id":"20948585-8723379-754345","span":{"begin":2454,"end":2458},"obj":"8723379"},{"id":"20948585-15511708-754346","span":{"begin":2476,"end":2480},"obj":"15511708"},{"id":"20948585-15884036-754347","span":{"begin":2498,"end":2502},"obj":"15884036"},{"id":"20948585-17557446-754348","span":{"begin":2510,"end":2514},"obj":"17557446"},{"id":"20948585-17557447-754349","span":{"begin":2526,"end":2530},"obj":"17557447"},{"id":"20948585-18246987-754352","span":{"begin":3242,"end":3246},"obj":"18246987"},{"id":"20948585-19109055-754354","span":{"begin":3653,"end":3657},"obj":"19109055"},{"id":"20948585-16796208-754355","span":{"begin":3854,"end":3858},"obj":"16796208"},{"id":"20948585-14611622-754356","span":{"begin":4811,"end":4815},"obj":"14611622"}],"text":"Introduction\nPeriodic leg movements (PLM) during sleep are a sleep related phenomenon with periodic episodes of repetitive stereotypical movements of the lower extremities (American Sleep Disorder Association, 1993). PLM are more frequent in several sleep disorders, including restless legs syndrome (RLS, Montplaisir et al., 1997), narcolepsy (Moscovitch et al., 1993), rapid eye movement sleep behavior disorder (Fantini et al., 2002) as well as in disorders not primarily affecting sleep such as end stage renal disease (Benz et al., 2000). PLM during sleep are also documented in normal subjects (Pennestri et al., 2006) giving rise to the debate about functional relevance.\nTo determine the functional significance of PLM, the criteria used to score PLM during sleep are pivotal. Thus the criteria of detecting single leg movements (LM) as well as determining what constitutes periodicity may have a significant impact on the final result. According to the new standards PLM occur in a series of at least four LM and are separated by 5–90 s (Zucconi et al., 2006). An increased number of PLM during sleep with a PLM index (PLMI, i.e., the number of PLM per hour of total sleep time) of five or more is considered abnormal (American Academy of Sleep Medicine, 2005). Though the PLMI is taken to reflect periodicity in PLM in most studies, periodicity is not always obvious and PLMI may fail to report it. The definition of a period implicit in the PLMI encompasses inter movement intervals (IMI) from 5 to 90 s. This range of IMI may not be sufficiently restrictive to identify periodicity in LM and could allow randomly occurring LM to be considered periodic. On the other hand, IMI may change across the night (Coleman et al., 1980; Pollmächer and Schulz, 1993; Nicolas et al., 1998; Ferri et al., 2009) and PLM will be missed if period criteria are too narrow. Varying IMIs during night may lead to broad IMI distributions or to a distribution showing several local maxima. This makes a wide range of periods appear similarly prevalent, and thus hides intrinsic periodicity. Another way intrinsic periodicity may become difficult to detect is when pooling data from patients with different average periods of PLM. The definition of periodicity of LM is problematic and may not be sensitive enough. This lack of sensitivity could be one of the reasons for the ongoing debate about their functional relevance (Ancoli-Israel et al., 1991; Mendelson, 1996; Hornyak et al., 2004; Carrier et al., 2005; Högl, 2007; Mahowald, 2007).\nA new statistical approach to analyze PLM was developed by Ferri et al. (2006b) who recommend to study the entire distributions of IMI rather than parametric measures as mean and standard deviation. They examined periodicity with a metric, termed the Periodicity Index (PI) measuring the percentage of LM appearing in series of periodic events. Instead of focusing on IMI frequency distributions the PI accounts for certain dependencies between subsequent IMIs. The PI was investigated for homogenous patient groups with RLS, narcolepsy/cataplexy and rapid eye movement sleep behavior disorder and without further potentially sleep relevant comorbidities or medication (Ferri et al., 2006a,b; Manconi et al., 2007). Based on different levels of the PI patients were divided into subgroups. For patients with narcolepsy/cataplexy and RLS a significantly different median of the PI was found (Ferri et al., 2006a). Very recently increased PI was also detected in a subgroup of patients with insomnia, for which no apparent cause could be found and enabled diagnosis of periodic leg movements disorder (PLMD, Ferri et al., 2009). Quantifiable aspects of PLM such as periodicity may serve to differentiate between clinical conditions and may eventually help capturing the nature of the underlying biological phenomenon (Rye, 2006). However, as afore mentioned, quantitative assessment of the periodicity of LM/PLM as measured by IMI is difficult because the empirical probability distribution may obscure underlying periodicity.\nTo date no study uses methods to enhance and reveal intrinsic periodicity before analyzing the statistics of IMI. We here employ the so-called unfolding procedure, which potentially uncovers intrinsic periodicity of LM data that is confounded by external influences. The unfolding procedure was originally developed in a branch of statistical physics called Random Matrix Theory (RMT) to study and compare the correlation between energy levels of heavy atomic nuclei and other complex quantum systems (see Brody et al., 1981; Guhr et al., 1998; Mehta, 2004, for comprehensive reviews). In the context of neurophysiology the concept was recently applied to the eigenvalue spectrum of the cross-correlation matrix calculated from multi-channel EEG (Seba, 2003; Müller et al., 2006; Baier et al., 2007).\nIn the present study we use the unfolding concept to assess the intrinsic periodicity of IMI of PLM examining a typical clinical patient cohort with PLM during sleep and different comorbidities from our database. Instead of studying the absolute IMI sizes, IMI are locally normalized to unit mean event spacing. Under these conditions the distribution of the spacings of randomly distributed events decays exponentially (Poisson distribution). In contrast, for strictly periodic systems the events are equally spaced. Fitting intermediate distributions, which interpolate between these extreme cases, to unfolded IMI distributions we quantify PLM periodicity.\nSpecifically, the goal of our study is to answer the following questions:\nDoes applying the unfolding procedure enhance and reveal periodicity that is not obvious from the measured IMI distribution?\nDoes periodicity differentiate between the clinical conditions of RLS, of OSA and of PLM without RLS or OSA?\nDoes data-driven clustering separate patients into different groups and are such groups clinically homogeneous?\nAnswering these questions could potentially provide further information about the clinical relevance of PLM during sleep."}