Assessment of Sleep Benefit in Parkinson's Disease
Assessment of Sleep Benefit in Parkinson's Disease
The existence of SB in PD has been reported repeatedly in the literature. Here, we examined subjective SB in a structured, prospective design using a 7-day diary. Additionally, we compared the diary responses to a more traditional way of assessing SB using a questionnaire. As such, the present study represents the first prospective assessment of sleep benefit over a longer period of time. Using either instrument, we found a prevalence of subjective SB of just over 30% and the SB group had longer medication use and a tendency towards shorter night sleep. Remarkably, the groups with SB according to the diary and according to the self-report questionnaire were overlapping only partially. Many patients showed incongruent results on the diary and the questionnaire; they either showed improvement in the diary, but did not report to experience this improvement in the questionnaire, or they reported to experience an improvement after sleeping in the questionnaire that was not present in the diary. Nevertheless, the mean change in motor function after sleep tended to be higher in patients reporting subjective SB.
We applied a new approach in studying SB, using a diary in which structured questions about motor functioning reflected the subjective PD motor symptom severity at that moment. The diary enabled us to prospectively study patients for more than one night. Furthermore, patients did not have to indicate their general morning function, but assessed their functioning at a specific moment on specific motor domains. This was the first study that addressed over-night functioning change over a longer period of time. We chose to do this, to see whether there was any day-to-day variation in sleep-related symptom severity changes in PD. In previous research, SB has been treated as an "all or nothing" phenomenon, classifying patients as having either SB or not at all. Here we show that this is not necessarily the case. From both the diary and the questionnaire it became clear that in the majority of patients, SB was not experienced after every period of sleep. Although 29% of SB patients (according to the questionnaire) claimed that their SB was 'always' present, none of the patients had a positive change in functioning in all of the diary nights.
Our questionnaire provided a carefully formulated definition of SB to make it as clear as possible for the patients what we were looking for. However, the written description of their experiences indicated that some patients may still have misinterpreted the given definition. Although it was stated that SB is a specific reduction in PD symptoms, some patients confused it with more general and a-specific refreshing effects of sleep, which are not necessarily related to PD. We previously established this problem and attributed this to the ill-defined description of SB in earlier studies. We feel that our strict definition increased the specificity to detect SB, which also fits with the SB prevalence of 31% in our cohort, which is slightly lower than in previous reports.
In our explanation of SB in the questionnaire, it was stated that the decrease in PD symptoms should be as good as feeling "on" (or better). However, 30% of the patients who reported SB in the questionnaire, subsequently stated that a symptom decrease was present after sleep, but not as large as when under the effect of medication. This may have caused an overestimation of SB prevalence. On the other hand, among the patients that answered negatively to the SB question, there were probably also patients who do experience some improvement after sleep, although not as large as on medication. Our data certainly indicate that there is large variation in the degree of symptom change by SB. Here -as well as in previous studies- some of the patients may have (partly) misinterpreted the SB definition, incorrectly stating to experience SB. Nevertheless, the characteristic written descriptions given by some patients, indicate that there may indeed be a group of PD patients that genuinely benefits from sleep.
Even patients with a highly characteristic and convincing description of perceived SB, did not always show sleep related symptoms changes in the diary. Therefore, the incongruence between the results from the diary and the questionnaire may further imply that these instruments assess different aspects of SB. When discussing SB, it seems necessary to make a distinction between 1) a subjective general feeling of improvement after sleep and 2) a specific improvement in actual motor functioning. Both aspects may contribute to the experiences which patients perceive as SB. The latter aspect was mostly assessed by the symptom diary, as we specifically targeted this instrument towards changes in motor function. The definition of SB in the questionnaire could be interpreted in a more general way, including non-motor symptoms of PD and/or a-specific refreshing effects of sleep. Both aspects could for sure be clinically relevant, but the underlying mechanisms are possibly different. Therefore, this distinction should be an important point of focus in future research.
We have assessed possible determinants of SB in our cohort. In both the diary- and questionnaire-determined SB we found that patients with SB had longer disease duration and longer medication use. Using the questionnaire we also found that patients with SB were younger, had a lower age of onset and used higher doses of dopaminergics. Similar results have previously been found for disease duration, medication use, age of onset and LED. Only one other study has reported a difference in age between SB and non-SB patients, but found SB patients to be older.
In the diary data we found a negative correlation with the magnitude of SB and self-reported sleep duration and quality. In addition, in the questionnaire a trend towards shorter sleep duration and lower sleep efficiency was present in the SB group. This possible association with shorter and/or worse nighttime sleep with the presence of SB is in line with a recent polysomnography study in which a shorter total sleep time and a longer sleep latency were found in patients experiencing SB. Together, these data may suggest that SB is in fact not related to sleep, or even that sleep deprivation facilitates SB. This hypothesis has been put forward in the literature before. When a causal relation between (good) sleep and improved motor function is abandoned, the thinking about the putative mechanism underlying SB may also have to change. The currently most common hypothesis states that dopamine storage in nigral neuronal terminals are replenished during sleep as the mediator of SB. However, this would not fit with the association between poorer sleep quality and SB, although it should be noted that these results are all based on the evaluation of nocturnal sleep. SB is also reported after daytime naps; and assessing nap related SB may shed further light on possible underlying mechanisms. Our study however, is unsuitable for making strong inferences on these mechanistic aspects.
This study had some limitations. In the diary we could only record planned naps, as it was essential that the patient completed the diary both before and after a sleep episode. We probably missed some unplanned and unnoticed sleep episodes. However, as our classification of patients having SB relied on an minimum number of naps with a positive change, this could at worst have led to an underestimation of patients with SB based on daytime naps. Furthermore, self-reports of sleep quality are not always reliable and subjective to various factors including affective disorders. However, previous studies on sleep benefit that included assessment of depressive symptoms did not find differences between patients with and without sleep benefit.
As patients dropping out of the study were older and may have been different with respect to disease duration or severity, one should be careful to generalize our findings to the whole PD population.
We used a minimum of 2 nights or naps with any symptom improvement as a cut-off for SB. Because this was the first time that SB was studied using a symptom diary, we had to choose cut-off values. For this particular study, we chose to be relatively sensitive and to classify any amount of 'over-sleep improvement' that occurred for more than one day as indicative of SB. Although arbitrary, this is at least a very clear way of defining SB, allowing comparisons between studies.
Discussion
The existence of SB in PD has been reported repeatedly in the literature. Here, we examined subjective SB in a structured, prospective design using a 7-day diary. Additionally, we compared the diary responses to a more traditional way of assessing SB using a questionnaire. As such, the present study represents the first prospective assessment of sleep benefit over a longer period of time. Using either instrument, we found a prevalence of subjective SB of just over 30% and the SB group had longer medication use and a tendency towards shorter night sleep. Remarkably, the groups with SB according to the diary and according to the self-report questionnaire were overlapping only partially. Many patients showed incongruent results on the diary and the questionnaire; they either showed improvement in the diary, but did not report to experience this improvement in the questionnaire, or they reported to experience an improvement after sleeping in the questionnaire that was not present in the diary. Nevertheless, the mean change in motor function after sleep tended to be higher in patients reporting subjective SB.
We applied a new approach in studying SB, using a diary in which structured questions about motor functioning reflected the subjective PD motor symptom severity at that moment. The diary enabled us to prospectively study patients for more than one night. Furthermore, patients did not have to indicate their general morning function, but assessed their functioning at a specific moment on specific motor domains. This was the first study that addressed over-night functioning change over a longer period of time. We chose to do this, to see whether there was any day-to-day variation in sleep-related symptom severity changes in PD. In previous research, SB has been treated as an "all or nothing" phenomenon, classifying patients as having either SB or not at all. Here we show that this is not necessarily the case. From both the diary and the questionnaire it became clear that in the majority of patients, SB was not experienced after every period of sleep. Although 29% of SB patients (according to the questionnaire) claimed that their SB was 'always' present, none of the patients had a positive change in functioning in all of the diary nights.
Our questionnaire provided a carefully formulated definition of SB to make it as clear as possible for the patients what we were looking for. However, the written description of their experiences indicated that some patients may still have misinterpreted the given definition. Although it was stated that SB is a specific reduction in PD symptoms, some patients confused it with more general and a-specific refreshing effects of sleep, which are not necessarily related to PD. We previously established this problem and attributed this to the ill-defined description of SB in earlier studies. We feel that our strict definition increased the specificity to detect SB, which also fits with the SB prevalence of 31% in our cohort, which is slightly lower than in previous reports.
In our explanation of SB in the questionnaire, it was stated that the decrease in PD symptoms should be as good as feeling "on" (or better). However, 30% of the patients who reported SB in the questionnaire, subsequently stated that a symptom decrease was present after sleep, but not as large as when under the effect of medication. This may have caused an overestimation of SB prevalence. On the other hand, among the patients that answered negatively to the SB question, there were probably also patients who do experience some improvement after sleep, although not as large as on medication. Our data certainly indicate that there is large variation in the degree of symptom change by SB. Here -as well as in previous studies- some of the patients may have (partly) misinterpreted the SB definition, incorrectly stating to experience SB. Nevertheless, the characteristic written descriptions given by some patients, indicate that there may indeed be a group of PD patients that genuinely benefits from sleep.
Even patients with a highly characteristic and convincing description of perceived SB, did not always show sleep related symptoms changes in the diary. Therefore, the incongruence between the results from the diary and the questionnaire may further imply that these instruments assess different aspects of SB. When discussing SB, it seems necessary to make a distinction between 1) a subjective general feeling of improvement after sleep and 2) a specific improvement in actual motor functioning. Both aspects may contribute to the experiences which patients perceive as SB. The latter aspect was mostly assessed by the symptom diary, as we specifically targeted this instrument towards changes in motor function. The definition of SB in the questionnaire could be interpreted in a more general way, including non-motor symptoms of PD and/or a-specific refreshing effects of sleep. Both aspects could for sure be clinically relevant, but the underlying mechanisms are possibly different. Therefore, this distinction should be an important point of focus in future research.
We have assessed possible determinants of SB in our cohort. In both the diary- and questionnaire-determined SB we found that patients with SB had longer disease duration and longer medication use. Using the questionnaire we also found that patients with SB were younger, had a lower age of onset and used higher doses of dopaminergics. Similar results have previously been found for disease duration, medication use, age of onset and LED. Only one other study has reported a difference in age between SB and non-SB patients, but found SB patients to be older.
In the diary data we found a negative correlation with the magnitude of SB and self-reported sleep duration and quality. In addition, in the questionnaire a trend towards shorter sleep duration and lower sleep efficiency was present in the SB group. This possible association with shorter and/or worse nighttime sleep with the presence of SB is in line with a recent polysomnography study in which a shorter total sleep time and a longer sleep latency were found in patients experiencing SB. Together, these data may suggest that SB is in fact not related to sleep, or even that sleep deprivation facilitates SB. This hypothesis has been put forward in the literature before. When a causal relation between (good) sleep and improved motor function is abandoned, the thinking about the putative mechanism underlying SB may also have to change. The currently most common hypothesis states that dopamine storage in nigral neuronal terminals are replenished during sleep as the mediator of SB. However, this would not fit with the association between poorer sleep quality and SB, although it should be noted that these results are all based on the evaluation of nocturnal sleep. SB is also reported after daytime naps; and assessing nap related SB may shed further light on possible underlying mechanisms. Our study however, is unsuitable for making strong inferences on these mechanistic aspects.
This study had some limitations. In the diary we could only record planned naps, as it was essential that the patient completed the diary both before and after a sleep episode. We probably missed some unplanned and unnoticed sleep episodes. However, as our classification of patients having SB relied on an minimum number of naps with a positive change, this could at worst have led to an underestimation of patients with SB based on daytime naps. Furthermore, self-reports of sleep quality are not always reliable and subjective to various factors including affective disorders. However, previous studies on sleep benefit that included assessment of depressive symptoms did not find differences between patients with and without sleep benefit.
As patients dropping out of the study were older and may have been different with respect to disease duration or severity, one should be careful to generalize our findings to the whole PD population.
We used a minimum of 2 nights or naps with any symptom improvement as a cut-off for SB. Because this was the first time that SB was studied using a symptom diary, we had to choose cut-off values. For this particular study, we chose to be relatively sensitive and to classify any amount of 'over-sleep improvement' that occurred for more than one day as indicative of SB. Although arbitrary, this is at least a very clear way of defining SB, allowing comparisons between studies.
Source...