| Sign In to gain access to subscriptions and/or personal tools. |
Identification of a Sleep Bruxism Subgroup with a Higher Risk of Pain
1 Faculty of Dental Medicine, Université de Montréal, C.P. 6128, succ. Centre-Ville, Montréal, Canada, H3C 3J7; and Correspondence: * corresponding author, gilles.lavigne{at}umontreal.ca
Sleep bruxism research diagnostic criteria (SB-RDC) have been applied since 1996. This study was performed to validate these criteria and to challenge the hypothesis that pain is associated with lower frequencies of orofacial activities. Polygraphic recordings were made of 100 individuals presenting with a clinical diagnosis of sleep bruxism and 43 control individuals. TwoStep Cluster analyses (SPSS) were performed with sleep bruxism variables to reveal groupings among sleep bruxers and control individuals. Participants completed questionnaires during screening, diagnosis, and recording sessions. Cluster analysis identified three subgroups of sleep bruxers. Interestingly, 45 of the 46 sleep bruxers with values below SB-RDC were classified in the low-frequency cluster. These individuals were more likely to complain of pain and fatigue of masticatory muscles than were the higher-frequency sleep bruxers (odds ratios > 3.9, p < 0.01). Sleep bruxers were distributed among three heterogeneous groups. Sleep bruxers with low frequencies of orofacial activities were more at risk of reporting pain.
Key Words: sleep bruxism • polygraphy • cluster analysis • diagnostic criteria • tooth grinding • pain
Sleep bruxism was recently classified as a movement disorder during sleep (AASM, 2005). In 1996, we proposed sleep bruxism research diagnostic criteria (SB-RDC) for polygraphic recording of sleep bruxism motor activity (Lavigne et al., 1996). These criteria were based on a small sample size (18 sleep bruxers and 18 control individuals). They were derived from electromyographic and audio-video recognition of jaw muscle activity in relation to: (1) a positive report of tooth grinding during sleep; and (2) the presence of sleep bruxism in a sleep laboratory. The criteria were: > 4 sleep bruxism episodes/hr of sleep, > 25 sleep bruxism bursts/hr of sleep, and > 1 sleep bruxism episode with tooth-grinding sounds. Over the last 15 years, we have made sleep laboratory recordings of 100 individuals with a positive home history of tooth grinding. Interestingly, half of the persons presented a low frequency of sleep bruxism episodes per hour of sleep (lower than 4 episodes/hr) and few tooth-grinding episodes, in spite of their sleep partners’ complaints of frequent grinding noise. One explanation may be that these persons experienced pain, which caused them to have fewer orofacial activities, in accordance with the pain adaptation model (Lund, 1995; Lavigne et al., 1997). Therefore, the objectives of this paper were to validate the 1996 SB-RDC in a large number of control individuals and sleep bruxers recorded in the sleep laboratory, and to challenge the hypothesis that pain is associated with lower frequencies of orofacial activities.
Participants Data from 100 sleep bruxers and 43 control individuals, recorded in the sleep laboratory since 1990, were used for the analysis. Sleep bruxers were chosen based on: (1) a history of frequent tooth grinding occurring at least 3 nights per week for the preceding 6 mos, as confirmed by a sleep partner; (2) clinical presence of tooth wear; (3) masseter muscle hypertrophy; and (4) report of jaw muscle fatigue or tenderness in the morning (American Sleep Disorders Association, 1997; AASM, 2005; Lavigne et al., 2005). Control individuals were selected on the basis of the absence of a history of tooth grinding during sleep and of other clinical evidence of sleep bruxism (American Sleep Disorders Association, 1997; AASM, 2005; Lavigne et al., 2005). Exclusion criteria for both sleep bruxers and control individuals were TMD as a primary complaint, a medical history of neurological disorders, mental disorders, or sleep disorders (e.g., apnea, periodic leg movements, insomnia). At the time of recordings, none of the participants was taking medication, or was under the influence of alcohol, nicotine, or caffeine. All participants provided informed consent according to the institutional rules (Hôpital du Sacré-Coeur).
Polygraphic Recordings
Sleep and Sleep Bruxism Scoring
Questionnaires
Statistical Analyses
Participants were young, with a mean age around 25 yrs old (bruxers, mean ± SE = 26.5 ± 0.6; control individuals, 24.5 ± 0.9, p = 0.07). There was a significantly higher proportion of womenamong bruxers (60%) and of men among control individuals (62.8%, p = 0.02). Participants were included in or excluded from further studies based on an analysis of results from the second night in the sleep laboratory. Bruxers were included when they displayed values higher than 2 of the 3 previously established cut-offs: 4 sleep bruxism episodes/hr, 25 sleep bruxism bursts/hr, 1 episode with grinding noise. Control individuals were included when they had values lower than or equal to 2 of the 3 cut-offs. Based on these EMG criteria, 54 bruxers were included for further studies, while 46 were excluded. Similarly, 34 control individuals were included for further studies, while nine were excluded.
Sleep and sleep bruxism variables for these four subgroups (included bruxers, excluded bruxers, included control individuals, and excluded control individuals) are presented in Table 1
TwoStep cluster analysis for bruxers identified three subgroups of bruxers who differed in sleep bruxism frequency: low, moderate, and high (Table 2  ). Box plots for the three clusters are shown on the Fig. Dotted lines indicate cut-off values of the SB-RDC. Groups were well-defined, and most individuals in the low cluster (n = 49) had values lower than cut-off for episodes/hr (cut-off = 4), bursts/hr (cut-off = 25), and episodes with noise (cut-off = 1). Interestingly, 45 of the 46 excluded bruxers were classified in this cluster. The importance of each variable in the three clusters is presented in the APPENDIX.
Cluster analysis performed in control individuals identified two clusters (Table 2  ). Most individuals in the high cluster had values above cut-off (Fig.). Seven of the nine control individuals who were excluded were classified in this cluster. The importance of each variable in the two clusters is given in the APPENDIX.
Answers to questionnaires (Table 3
In a study published in 1996, the selection criteria for sleep bruxers included having exhibited grinding sounds during sleep at least 5 nights a wk in the preceding 6 mos (Lavigne et al., 1996). Moderate to severe sleep bruxers were recruited, as shown by their number of sleep bruxism episodes/hr (mean ± SE, 5.4 ± 0.6) and number of bursts/hr (40.7 ± 6.7). These values are comparable with those of sleep bruxers included for further studies and with those of the moderate cluster. Over the years, to have access to a larger pool of potential participants, investigators have reduced the criterion for reported grinding sounds during sleep from 5 to 3 nights a wk. Based on this criterion, 46 sleep bruxers recorded in the sleep laboratory were excluded from further studies, according to the SB-RDC. Consequently, the sensitivity and specificity of the SB-RDC established in 1996 would be lower if re-evaluated based on the total sample of sleep bruxers (n = 100) and control individuals (n = 46), due to the low sleep bruxers and control individuals with high frequencies of activities. For example, the cut-off of 4 episodes/hr would lead to lower sensitivity (55% instead of 72%) and specificity (84% instead of 94%). Lower values of cutoff (2.5 episodes/hr instead of 4.0) would be required to have sensitivity and specificity around 70%. However, the classification established by the SB-RDC closely matches that of the cluster analyses. Sleep bruxers in the low cluster and control individuals in the high cluster are those excluded by the SB-RDC. These persons displayed frequencies of activities different from those typical of their clinical group. Their inclusion in sleep laboratory studies is therefore not suitable. Close to 50% of persons with a clinical history of tooth grinding presented low frequencies of jaw muscle contractions (episodes/hr, bursts/hr) and tooth-grinding events in the sleep laboratory. A high proportion of these participants reported painful jaw and fatigue of masticatory muscles, although they did not complain of or present TMD (temporomandibular muscle or joint pain or dysfunction). Other studies have reported that sleep bruxers frequently present with low pain intensity in jaw and neck muscles, or temporal headaches upon waking (Bader et al., 1997; Lavigne et al., 1997; Camparis et al., 2006; Huynh et al., 2006). The possibility that this episodic pain and headache occurred in relation to the recording in the sleep laboratory may have contributed to a reduction in the motor activity, as suggested by the Pain Adaptation Model (Lund, 1995). Another explanation could be the natural variability in the occurrence of sleep bruxism over time. We noted that the variability in the number of episodes/hr of sleep was 25%, while the variability in the number of episodes with grinding sound was over 50% (Lavigne et al., 2001a). Conversely, approximately 20% (8/42) of control individuals displayed a high frequency of orofacial activities and were classified in the ‘high control’ cluster. Seven of these control individuals had more than 4 episodes/hr of sleep, eight reported more than 25 bursts/hr of sleep, and four presented at least 2 episodes with grinding sounds. This last finding may be surprising, but can be explained by the fact that some control individuals, who are young (mean age = 25 yrs old), may grind their teeth very rarely. Persons may also sleep alone or with a sleep partner who is not disturbed by their grinding sounds, thus providing an unreliable report of absence of tooth grinding at home. A correlation of sleep bruxism with stress and anxiety from situational or psychological sources has been suggested (Rugh and Harlan, 1988; Hicks et al., 1990; AASM, 2005). However, this association remains controversial (Harness and Peltier, 1992; Pierce et al., 1995; Watanabe et al., 2003). The present sleep study does not support this association, since the levels of stress and anxiety did not differ between included sleep bruxers and control individuals. The observation of 3 clusters in sleep bruxers and of 2 in control individuals further supports the suggestion that jaw muscle contractions during sleep are a natural activity, with a wide spectrum of frequency (number of episodes/hr). In a previous study, control individuals presented from 0.1 to 12.6 episodes/hr of jaw muscle contraction during sleep, while sleep bruxers presented from 1.2 to 15.2 episodes/hr (Lavigne et al., 2001b). It was also observed that the probability of jaw muscle contraction during sleep is related to intrinsic physiological cardiac and brain-related arousals called ‘micro-arousals’ (Macaluso et al., 1998; Kato et al., 2001, 2003). From the above observations, we suggested that these contractions are probably distributed over a continuum, from a low-frequency range to intermediate and high ranges, with coincidental tooth grinding (Lavigne et al., 2003). This study provides confirmation that the SB-RDC developed ten years ago facilitates a high level of discrimination between sleep bruxers and control individuals. The SB-RDC distinguishes low sleep bruxers and high control individuals from the other individuals within their group. Furthermore, pain is frequently reported among sleep bruxers who display low frequencies of jaw muscle contractions.
This study was supported by the Canadian CIHR and the Québec FRSQ. The authors thank Mrs. Christiane Manzini and Nelly Huynh for their help in recruiting participants. We appreciate Dr. Alice Petersen’s contribution to editing of this paper.
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org. Received for publication February 14, 2006. Revision received April 18, 2007. Accepted for publication May 8, 2007.
Journal of Dental Research, Vol. 86, No. 9,
837-842 (2007) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
