Symptoms and Physiological Responses to ... - Wiley Online Library

1 downloads 0 Views 818KB Size Report
Noboru Ohata, DDS, PhD; Peter Svensson, DDS, PhD, Dr Odont ... Japan (T. Takeuchi, T. Arima, and T. Yamaguchi); Section of Orofacial Pain and Jaw Function ...
ISSN 0017-8748 doi: 10.1111/head.12528 Published by Wiley Periodicals, Inc.

Headache © 2015 American Headache Society

Research Submissions Symptoms and Physiological Responses to Prolonged, Repeated, Low-Level Tooth Clenching in Humans Tamiyo Takeuchi, DDS; Taro Arima, DDS, PhD; Malin Ernberg, DDS, PhD; Taihiko Yamaguchi, DDS, PhD; Noboru Ohata, DDS, PhD; Peter Svensson, DDS, PhD, Dr Odont

Background.—The traditional view contends bruxism, such as tooth grinding/clenching, is part of the etiology of temporomandibular disorders (TMD) including some subtypes of headaches. The purpose of this study is to investigate if a low-level but long-lasting tooth-clenching task initiates TMD symptoms/signs. Methods.—Eighteen healthy participants (mean age ± SD, 24.0 ± 4.3 years) performed and repeated an experimental 2-hour tooth-clenching task at 10% maximal voluntary occlusal bite force at incisors (11.1 ± 4.6 N) for three consecutive days (Days 1-3). Pain and cardiovascular parameters were estimated during the experiment. Results.—The task evoked pain in the masseter/temporalis muscles and temporomandibular joint after 40.0 ± 18.0 minutes with a peak intensity of 1.6 ± 0.4 on 0-10 numerical rating scale (NRS) after 105.0 ± 5.0 minutes (Day 1). On Day 2 and Day 3, pain had disappeared but the tasks, again, evoked pain with similar intensities. The onset and peak levels of pain were not different between the experimental days (P = .977). However, the area under the curve of pain NRS in the masseter on Day 2 and Day 3 were smaller than that on Day 1 (P = .006). Cardiovascular parameters changed during the task but not during the days. Conclusions.—Prolonged, low-level tooth clenching evoked short-lived pain like TMD. This intervention study proposes that tooth clenching alone is insufficient to initiate longer lasting and self-perpetuating symptoms of TMD, which may require other risk factors. Key words: tooth clenching, masseter muscle, maximal voluntary occlusal bite force, trigeminal physiology Abbreviations: BO blood oxygenation, deOXY deoxygenated, Hb hemoglobin, HF high-frequency, HR heart rate, HRV heart rate variability, LF low-frequency, MANOVA multivariate analysis of variance, MVOBF maximal voluntary occlusal bite force, NRS numerical rating scale, OXY oxygenated, POMP persistent orofacial muscle pain, PPT pressure pain threshold, StO2 tissue blood oxygen saturation, TMD temporomandibular disorder, TMJ temporomandibular joint (Headache 2015;55:381-394)

From the Department of Crown and Bridge Prosthodontics, Graduate School of Dental Medicine, Hokkaido University, Sapporo, Japan (T. Takeuchi, T. Arima, and T. Yamaguchi); Section of Orofacial Pain and Jaw Function, Department of Dental Medicine, Karolinska Institutet, Huddinge, Sweden (M. Ernberg and P. Svensson); Scandinavian Center for Orofacial Neurosciences (SCON) (M. Ernberg and P. Svensson); Graduate School of Dental Medicine, Hokkaido University, Sapporo, Japan (N. Ohata); Section of Clinical Oral Physiology, Department of Dentistry, Aarhus University, Aarhus, Denmark (P. Svensson). Address all correspondence to T. Arima, Department of Crown and Bridge Prosthodontics, Graduate School of Dental Medicine, Hokkaido University, North13 West7, Kita-ku, 060-8586, Sapporo, Japan, email: [email protected] Accepted for publication November 17, 2014. Conflict of Interest: None.

381

382 The International Classification of Headache Disorders (ICHD-III) and the new Diagnostic Criteria for Temporomandibular Disorders (DC/TMD) have recently identified headache and facial pain attributed to temporomandibular disorders (TMD).1,2 This is in recognition of the clinical observation that the symptoms of headache and TMD often overlap,3-9 and previous studies have described the pathways and mechanisms for pain referral from the head to the temporomandibular joint (TMJ) and vice versa.10 It is important to identify common mechanisms and pathophysiological processes. Persistent orofacial muscle pain (POMP) associated with myalgia and myofascial TMD has traditionally been linked to hyperactivity or abnormal contraction of masticatory muscles such as “bruxism.”11-15 Based on this assumption, several different types of contraction-induced jaw-muscle pain models have been proposed and developed. For tooth grinding, Christensen (1971) first investigated the effect of experimental tooth grinding on healthy individuals and reported long-lasting jaw-muscle pain.16 Arima et al (1999) made the first attempt to induce jaw-muscle pain by a standardized tooth-grinding exercise.17 The results supported the hypothesis that overloading of muscle activity was related to the development of jaw-muscle pain and soreness; however, the pain and soreness did not last during the following days as expected from the clinical symptoms of POMP or as predicted by the vicious cycle proposed by Laskin (1969).18 For tooth clenching, some studies had used maximal voluntary tooth clenching,19 and others had used tooth clenching at submaximal levels of maximal voluntary occlusal bite force (MVOBF).20-25 Svensson et al (1996) made the first attempt to evoke POMP by standardized and repeated experimental tooth clenching and reported that the tooth-clenching procedure failed to induce a progressive increase in masticatory muscle pain.26 It did not seem that experimental bruxism activity could reproduce POMP or support the preceding hypothesis about masticatory muscle pain. Recently, it has been found that low-level tooth-clenching tasks (≤10% MVOBF) cannot produce long-lasting pain and fatigue,20,25,27 but perceived levels of fatigue were significantly higher after low-level tooth-clenching

March 2015 tasks than after high-level tooth-clenching tasks (≥15% MVOBF).25 Furthermore, there remain questions about potential gender differences in the painful effects of tooth clenching.22 Thus, the overall aim of this study was to investigate the effects of a prolonged, repeated low-level tooth-clenching task on jaw-muscle symptoms/signs in healthy men and women. Moreover, some studies have indicated that localized intramuscular hemodynamics28,29 and mental stress30 may play an important role in the development of muscle pain and tenderness. Therefore, we also monitored cardiovascular and autonomic responses, such as heart rate (HR) and heart rate variability (HRV).31

MATERIALS AND METHODS Participants.—Nine healthy men (mean age ± standard deviation [SD], 25.8 ± 5.0 years old) and nine healthy women (22.2 ± 2.6 years old) were enrolled. This number of participants was based on previous low-level tooth-clenching studies with male and female participants.22,24,27,32 All participants were university or graduate students from Aarhus University in Denmark, recruited by advertisements on campus or through the website http://www.forsoegsperson .dk/. They also had no history of TMD, which was ascertained according to the Research Diagnostic Criteria for Temporomandibular Disorders33 history and examination. None of the participants took medication that could influence psychological and/or cardiovascular responses. Informed consent was obtained from each participant, and the experimental protocol was approved by the local ethics committee in accordance with the Declaration of Helsinki. Study Design.—The participants sat on a comfortable chair and performed an experimental toothclenching task (Day 1). Perceived levels of their pain and fatigue using numerical rating scales (NRS) and sensitivity changes to pressure pain stimulation (pressure pain threshold [PPT]) were obtained before, during, and after the task as assessment of TMD symptoms and signs. Intramuscular blood oxygenations (BO) in the masseter muscle and HRV were also recorded simultaneously during the experiment. The participants repeated the same tasks and measurements as Day 1 for the following two days (Day 2

Headache and Day 3). On the last day of the experiment (Day 4), only NRS and PPT measurements were performed. All experiments were conducted at Aarhus University, and the temperature in the laboratory was controlled (24 degrees Celsius). Experimental Tooth-Clenching Task.—Two hours of 10% MVOBF tooth-clenching tasks were adopted in this study. First of all, MVOBF was recorded with a bite force transducer system (41.0 × 12.0 × 5.0 mm, length × width × height, Aalborg University, Aalborg, Denmark) connected to an amplifier with peak-hold facility.34 The analog output of the amplifier was connected to a chart recorder (BNC-2090, National Instruments, Texas, USA). MVOBF was measured three times and averaged,35 and then 10% MVOBF was calculated. The duration of each effort was 1-2 seconds with 30-second rest intervals. In the experimental sessions, the participants performed the predefined levels of the tooth-clenching task using visual feedback. To hold the center of the transducer between the upper and lower teeth, a layer of acrylic resin with 2-mm thickness and silicone impression material (President Putty, Coltene, Altstaetten, Switzerland) was added on the blocks and fitted to the occlusion. Incisors were chosen for the placement of transducer. The reasons for this were to reduce the interocclusal distance of the tooth-clenching task; to keep the transducer in the same place during the tooth-clenching task; and to avoid interferences with the cardiovascular recordings. The bite blocks caused a jaw opening at the incisors ranging from 11.5 mm to 13.9 mm (mean ± SD, 12.7 ± 0.9 mm). There was no force transducer or support on the posterior teeth. Numerical Rating Scales.—An 11-point NRS from 0 to 10 was used for assessment of orofacial pain and fatigue, with 0 corresponding to “no pain/fatigue at all” and 10 to “the worst pain/fatigue imaginable.”36 The participants separately scored their pain and fatigue intensity at both the masseter and temporalis muscles and TMJ before, during, and after the toothclenching task on NRS every 5 minutes. For further assessment of the pain and fatigue NRS scores during the tasks, the area under the curve (AUC) of NRS was calculated and used for the analysis.

383 Pressure Pain Thresholds.—An electronic pressure algometer (ALGOMETER, Somedic AB, Horby, Sweden) was used with a probe diameter of 10 mm and a constant application rate of 30 kPa/s. The measurements were performed bilaterally on the central parts of the masseter muscles, left anterior temporalis muscle, and left forefinger (extra-trigeminal control site). The participants were instructed to keep their teeth slightly apart (about 1 to 2 mm) to avoid contraction of the jaw-closing muscles during pressure stimulation. PPT was defined as the pressure that the participants first regarded as painful.37 PPT was calculated by averaging the three measurements obtained with about 30-second intervals.38 Intramuscular Blood Oxygenations.—Nearinfrared spectroscopy was used to evaluate continuously tissue blood oxygen saturation in the left masseter muscle with the use of a laser tissue blood oxygen monitor (BOM-L1TRW, OMEGAWAVE, Tokyo, Japan).The instrument used three laser diodes (780, 810, and 830 nm), measuring the absolute amounts of oxygenated hemoglobin (OXYHb) and deoxygenated hemoglobin (deOXYHb) within the target tissue according to the Beer–Lambert law.39 The absorption coefficient of hemoglobin at each wavelength was based on the data reported by Matcher et al.40 The light probe and the central detector were placed 10 mm apart and the distant detector was placed 20 mm apart from the light probe. Tissue blood oxygen saturation (StO2 = [OXYHb/TotalHb] × 100) and total hemoglobin (TotalHb) were calculated based on the OXYHb and deOXYHb values. The ratios of each parameter from their baselines were calculated as ΔOXYHb, ΔdeOXYHb, ΔTotalHb and ΔStO2, respectively. Heart Rate Variability.—HR and HRV were assessed using a sphygmograph (TAS9 Pulse Analyzer Plus, YKC, Tokyo, Japan) through the right forefinger and analyzed by frequency-domain methods.41 The power spectral components of the R–R interval (R wave to R wave interval in electrocardiogram) in the range of 0.04-0.15 Hz was defined as lowfrequency (LF) components, and those in the range of 0.15-0.40 Hz was defined as high-frequency (HF) components.42 The values of HR were averaged, and the values of LF and HF power were obtained by integrating each frequency band.

384 Statistics.—Potential confounding factors were excluded (side of clenching task, level of bite force, and use of medication). Parametric distribution of data was checked with Kolmogorov–Smirnov tests; logarithmic transformation was used for BO, and square root transformation was used for the area under the curve (AUC) of NRS. Multivariate analysis of variance (MANOVA) with repeated measures was then used to test the data with the factors: gender, time, and day. There was no missing data. Each sample of BO and HRV data was averaged every 5 minutes to observe changes over time. All data were expressed with mean ± SD. Each parameter was analyzed relative to its baseline value, which was the 5-minute time-period before the toothclenching task. We also used the first-15-minute value and the last-15-minute value for analysis, which were the values for 15 minutes on each day immediately after the participants started the tooth clenching or immediately before they stopped the tooth clenching, respectively. Possible gender differences in NRS, PPT, BO, and HRV were also tested with MANOVA with repeated measures. The levels of significance were adjusted for multiple pairwise comparisons with the Tukey honest significant difference test. To further test the relationships between pain, fatigue, and ΔStO2, the values were plotted, fitted with near quadratic equations, and the coefficient of determination (R2) was calculated. The STATISTICA software (StatSoft, Tulsa, OK, USA) was used for all analyses. Significance was accepted at P ≤ .050 (two-tailed).

RESULTS Experimental Tooth-Clenching Task.—The average MVOBF was 111.3 ± 46.0 N without gender differences (gender: P = .128, interaction gender × time × day: P = .572, MANOVA). All participants completed the tooth-clenching tasks on the three consecutive days. Pain and Fatigue.—Pain and fatigue in the masseter and temporalis muscles as well as TMJ are shown for each gender separately in Figure 1. On Day 1, from the overall analyses, already after the beginning of the tooth-clenching task (at 15 minutes), fatigue in the masseter muscles was significantly increased (NRS: 1.2 ± 1.5, P < .001) and reached a peak at 85 minutes (NRS: 2.9 ± 3.0). After the onset of fatigue

March 2015 (at 20 minutes), the participants reported significant levels of pain (NRS: 0.9 ± 1.6, P = .005), which then reached a peak at 110 minutes (NRS: 2.1 ± 2.8, P < .001). However, both fatigue and pain did not last more than 5 minutes after the completion of the tooth-clenching task. The same tendencies observed in the masseter muscle were also seen in the temporalis muscle and TMJ. MANOVA showed that males had higher fatigue NRS scores in the masseter muscles as compared to females during all toothclenching task days (gender: P = .050, interaction gender × time × day: P = .028, MANOVA). On the following days (Day 2 and Day 3), the tooth-clenching tasks, again, evoked significant levels of pain and fatigue reaching similar levels as the preceding task. On Day 4, all NRS pain and fatigue scores were zero except for fatigue in the masseter (0.2 ± 0.5, in two out of 18 participants).The onset and peak levels of pain and fatigue were not different between the experimental days (P = .977); however, the quantitative analysis showed that AUC of NRS in the masseter on Day 1 was larger than those of Day 2 and Day 3 (P = .006). Furthermore, the AUC of the fatigue NRS was always larger than that of the pain NRS in the masseter and TMJ (P = .045, MANOVA). Pressure Pain Thresholds.—The tooth-clenching task did not change the masseter/temporalis muscle PPTs after the task in any days (P = .071; Fig. 2). However, the PPTs on the left forefinger were decreased on Day 2 and Day 3 (P = .007; Fig. 2). MANOVA showed no significant gender differences (P = .369, MANOVA) or interactions between gender, time, and day (P = .925, MANOVA). PPTs on the forefinger were always larger than those on masticatory muscles, and there was a tendency that PPTs on the temporalis were larger than those on the masseter (P < .001, MANOVA). Intramuscular Blood Oxygenations.—Cardiovascular changes in the left masseter are shown for each gender separately in Figure 3. From the overall analyses, as soon as the tooth-clenching task started, ΔOXYHb, ΔTotalHb, and ΔStO2 increased and remained increased until the task ended. Immediately after the completion of the tooth-clenching task, ΔOXYHb and ΔTotalHb quickly decreased and returned to baseline levels. There were significant

Headache

385

Fig 1.—Pain and fatigue (0-10 numerical rating scale) in the masseter, temporalis, and temporomandibular joint (TMJ). Each value represents the mean ± SD. Small figures with arrows show the peak timing and small figures indicate the onset of the pain or fatigue in each gender (solid: male and grey: female). *Significant changes from baseline and †significant difference in gender by post-hoc test (Tukey: P < .050).

386

March 2015

Fig 2.—Pressure pain thresholds (PPT) during the experiment. Each value represents the mean ± SD. *Significant difference by post-hoc test (Tukey: P < .050).

differences between the first 15-minute values and the last 15-minute values for ΔStO2 on Day 2 and Day 3 (P = .001, MANOVA). ΔdeOXYHb showed significant differences only during the 5-10 minutes and 120-130 minutes epoch on Day 3, and it decreased during clenching on Day 3. There was no significant difference between the values obtained on the different experimental days (ΔdeOXYHb: P = .441, ΔOXYHb: P = .306, ΔTotalHb: P = .735 and ΔStO2: P = .062, MANOVA). There were significant gender differences in ΔOXYHb, ΔTotalHb and ΔStO2 (Table). The tooth-clenching task increased ΔOXYHb, ΔTotalHb, and ΔStO2 in males during the experimental days, and decreased ΔdeOXYHb on Day 3, while in females only ΔStO2 on Day 3 was increased (Fig. 3). Heart Rate Variability.—From the overall analyses, LF and HF increased during the tooth-clenching tasks (P < .001, MANOVA; Fig. 4), in addition the last 15-minute values were greater than the first 15-minute values for LF and HF (P < .001, MANOVA). There was no significant difference between the values in LF/HF on each day (P = .966, MANOVA; Fig. 4). In contrast, HR decreased linearly during the tooth-clenching task (P < .001, MANOVA; Fig. 4) with greater values during the first 15-minute than the last 15-minute on Day 2 and Day

3 (Table 1). There was no significant difference between the values on each day (LF: P = .410, HF: P = .532, LF/HF: P = .966, and HR: P = .200, MANOVA). The gender differences in LF/HF were only observed at four time points (during 10-15 minutes, 60-65 minutes, and 75-80 minutes on Day 1, and during 0-5 minutes on Day 2, Table 1). ΔStO2–NRS Curve.—To further explore the relationship between pain, fatigue, and oxygen saturation, NRS pain and fatigue, and StO2 were plotted and tested. There were near quadratic relationships (R2 ≥ 0.600) between ΔStO2 in 5-minute epochs and NRS of pain and fatigue from the masseter immediately after the former 5-minute epoch (Fig. 5). NRS increased after the participants started the toothclenching task with ΔStO2 increasing, reaching a peak, and decreasing rapidly after they stopped the tooth clenching. On Day 1, ΔStO2-NRS curve reached and stayed with a peak after about 60 minutes, while on Day 2 and Day 3, the curves reached a peak just before they stopped the tooth clenching. The curves tended to be less steep as the tooth-clenching task was repeated (Fig. 5).

DISCUSSION This systematic and controlled study for the first time investigated the effects of prolonged low-level

Headache

387

Fig 3.—Effects of the tooth-clenching tasks on hemodynamics in left masseter muscle in each gender (solid: male and grey: female). Changes in deoxygenated hemoglobin (ΔdeOXYHb), oxygenated hemoglobin (ΔOXYHb), total hemoglobin (ΔTotalHb), and tissue blood oxygen saturation (ΔStO2). Each value represents the mean ± SD. *Significant changes from baseline and †significant difference in gender by post-hoc test (Tukey: P < .050).

tooth clenching for 3 consecutive days on TMD symptoms/signs, blood flow responses in the masseter muscle, and autonomic responses. The main findings were that all participants, indeed, could perform the low-level tooth-clenching task continuously for 2 hours and repeat it for 3 days, and that although pain and fatigue were evoked by the task, they were short lasting. Effect of the Tooth-Clenching Tasks on TMD Symptoms and Signs.—In this study, as expected, prolonged low-level tooth clenching evoked pain and fatigue in the jaw-closing muscles and TMJ. However, the task did not initiate any longer lasting or selfperpetuating muscle pain similar to headaches or

TMDs (Fig. 1). Furthermore, composite measures of NRS of pain and fatigue in the masseter muscle using AUC on Day 1 were larger than on Day 2 and Day 3. This indicates that the masseter muscle in healthy participants may indeed be resistant to pain and fatigue in response to repeated tooth-clenching tasks. The gender analyses showed that males had higher levels of fatigue than females. This result is in accordance with a previous study which showed that the fatigue sensation was higher in males than in females during tooth clenching at 10% MVOBF for 30 minutes.22 The reasons for this observation are not known but could be related to gender differences in muscle metabolism during sustained contractions,43

388

March 2015

Table.—Comparison Between the First 15-minute Value and the Last 15-minute Value and Analyses of Gender Differences

P Value First vs Last (Tukey)

Male

ΔdeOXYHb ΔOXYHb ΔTotalHb ΔStO2 LF HF LF/HF HR

P Value (MANOVA)

Female

Day 1

Day 2

Day 3

Day 1

Day 2

Day 3

Gender

Interaction Gender × Time × Day

1.000 .845 .937 .929