Abstract
Increased pericranial muscle tenderness is connected with tension-type headache in adults. In children, the importance of muscle tenderness in the pericranial or neck-shoulder region in the pathogenesis of different types of headache is unknown. The present study evaluated muscle tenderness in the pericranial and neck-shoulder region in children with migraine, those with tension-type headache and those without headache. An unselected population-based questionnaire study concerning headache was carried out in 1135 Finnish schoolchildren aged 12 years. Of them, 183 children were randomly selected for a face-to-face interview and a clinical examination. Muscle tenderness was recorded by manual palpation and dolorimeter. Children with migraine had increased overall tenderness, recorded by manual palpation, compared with those without headache. They also self-reported tenderness in the neck-shoulder region during daily activities more often than the children of the other groups. Muscle tenderness was not associated with paediatric tension-type headache. The mean pressure pain thresholds did not differ among the three groups. However, a negative correlation between the total tenderness score and the dolorimeter score was found in each group. In conclusion, children with migraine had increased muscle tenderness at palpation of the pericranial and neck-shoulder muscles and they also reported pain symptoms in the neck-shoulder region most frequently. Instead, increased pericranial and neck-shoulder muscle tenderness was not associated with tension-type headache in children.
Introduction
In adults, muscular factors are of importance in tension-type headache, even though their initiating role in headache is not certain (1). In the International Headache Society (2) criteria, both episodic and chronic tension-type headache can be but, not necessarily are, associated with increased tenderness of pericranial muscles demonstrated by manual palpation or using a pressure algometer.
Eighty-seven percent of adults with chronic tension-type headache, and 66% of those with the episodic form have been reported to have a muscular disorder (3). Migraineurs also have more muscle tenderness during headache attack than do nonheadache controls (4). In addition, tender spots are found in subjects without headache (5).
The present study is the first to evaluate muscle tenderness in the pericranial and neck-shoulder region in different types of paediatric headache. Our main purpose was to examine whether muscle tenderness in the pericranial and neck-shoulder region is associated with tension-type headache in children. Other purposes were to study whether muscle tenderness is different in different headache groups, to evaluate the correlation of self-reported symptoms with tenderness measured by manual palpation or dolorimeter, and to study whether the presence of headache has an influence on pericranial muscle tenderness in children.
Study population and methods
Study design
The study had two steps. In the first step, in 1998, the original study population was collected by sending a structured questionnaire concerning headache and its predisposing factors to all of the 1409 children on the sixth grade in 35 primary schools in the city of Turku. Altogether, 1135 (81%) of these children participated in the original study. The mean age of the children was 12.6 years. Different types of headache were classified using the IHS criteria. Of 1135 children, 407 did not suffer from headache, 154 had migraine and 138 had episodic tension-type headache. Nobody had chronic tension-type headache.
The second step was the present study, where part of the representative nonselected child population was interviewed face-to-face and clinically examined. The present study was performed at the paediatric outpatient clinic in the city of Turku from November 1998 to February 1999. Seventy children from the migraine group, 70 children from the episodic tension-type headache group and 70 children from the headache-free group were randomly selected for face-to-face interview and further examination. With this sample, we did calculations to achieve a power of 85% (α= 0.05) to find out the 25% difference in neck-shoulder symptoms between groups. If anyone was unable to participate, another randomly selected child of the same sex and group was invited for further examination instead (24% of children). Most nonparticipants had had no headache. The mean age of the children was 13.4 years.
The headache type was the same in both the questionnaire and face-to-face interview as follows: 59 children (32 girls and 27 boys) with migraine (84%); 65 children (21 girls and 44 boys) with episodic tension-type headache (93%); and 59 children (37 girls and 22 boys) without headache (84%). The present study population consisted of these 183 children.
Ten children with migraine (17%) had a migraine attack and two children with tension-type headache (3%) reported a headache episode on the day of the examination. Sixteen children with migraine (27%) had migraine attacks and 11 children with tension-type headache (25%) had headache episodes in the 72 h prior to the examination.
Five children with migraine (8%) also had episodic tension-type headache. These children were included in migraine group. Of them, four children had suffered from migraine attacks three times a month and from tension-type headache episodes two times a month on average. One child had had two migraine attacks and one tension-type headache episode a month. Children with the no headache group reported that they never had had a headache in their entire life.
The frequency of migraine attacks was on average three times a month in children with migraine. These children had suffered from attacks for five years. The children with episodic tension-type had headache episodes on average three times a month. These children had had headache episodes for 4.5 years.
Seventy-six percent of children with migraine reported pulsating pain, 64% unilateral pain, 58% nausea during attack, 24% vomiting during attack, 78% photophobia, 73% phonophobia and 70% aggravation of pain due to routine physical activity. Of children with episodic tension-type headache, 84 percent reported pressing pain, 83% bilateral pain, 2% nausea during headache episode, 41% photophobia, 23% phonophobia and 6% aggravation of pain due to routine physical activity.
The face-to-face interview, the clinical examination and the classification of the present headache types were conducted for all the 183 children (90 girls and 93 boys) by one of the authors (PA). A blinded examination of tender points with manual palpation and pressure pain threshold measurements was carried out in these children by one trained physiotherapist (SV).
The face-to-face interview included self-reported tenderness in the neck-shoulder region during daily activities (‘never’, ‘sometimes’, ‘often’) and the occurrence of the last headache episode or attack. Seven pericranial and neck-shoulder tender points (frontal muscle, temporal muscle, suboccipital muscle insertion, anterior aspect of C5-7, upper trapezius muscle, supraspinatus muscle insertion, levator scapulae muscle insertion) on the both sides of the body were palpated. We used the modified version of the Total Tenderness Score System (5), excluding some temporomandibular points, because all the children were sent to a dentist for further temporomandibular examination. We decided not to test temporomandibular joints twice, because it is possible that repetition of a test influences the answers. The reaction of the child was classified into four categories: 0 = no pain, 1 = no visible reaction but reports mild pain, 2=reports pain, distorting the face, 3=reports considerable pain and withdraws, and the total tenderness score was counted (score max. 42). A Fischer dolorimeter (6) was employed for the measurement of pressure pain threshold. The rubber tip of the instrument has a surface of 1 cm2, and pain was measured in kg/cm2 units. One and the same physiotherapist performed all the measurements. The pain threshold was measured from 5 bilateral points (frontal muscle, temporal muscle, suboccipital muscle insertion, upper trapezius muscle, levator scapulae muscle). The mean dolorimeter score was counted. We excluded supraspinatus muscles and anterior aspects of C5-7, because their measurement is difficult and therefore unreliable in children. Muscle palpation and the Fischer dolorimeter can not differentiate between skin, muscles and bone.
Statistical methods
Variables with normal (Gaussian) distributions were expressed as means with 95% confidence intervals (95% CI) and standard deviations (SD). Statistical comparison between the groups was made using analysis of variance (
The study design was approved by the Joint Ethics Committee of the University of Turku Medical School and Turku University Central Hospital.
Results
Table 1 shows the mean dolorimeter scores of the five bilateral tender points and the median tender points score of the seven bilateral tender points in children with migraine, those with tension-type headache and those without headache. The median total tenderness score was highest in children with migraine, and the difference between the groups approached, but did not reach the statistical significance (P = 0.054). However, children with migraine differed significantly from those without headache (P = 0.02). In the mean dolorimeter scores, no difference between the groups was found.
Dolorimeter scores and total tenderness scores in children with migraine, tension-type headache and no headache
∗
†Kruskall-Wallis test.
At palpation, more children with migraine reported tenderness in the right suboccipital insertion (P = 0.007) and in the right supraspinatus (P = 0.018) and left supraspinatus (P = 0.033) than children with tension-type headache or those without headache. At the remaining tender points, no difference between the groups was found.
The total tenderness score and the dolorimeter score had negative correlations in children with migraine (r =− 0.49, 95% CI −0.27 to −0.67), in children with episodic tension-type headache (r =− 0.52, 95% CI −0.32 to −0.68) and in those without headache (r =− 0.54, 95% CI −0.32 to −0.70), in other words, the higher the subjective tenderness at palpation, the lower the mean pain thresholds measured with a dolorimeter.
Table 2 presents the frequency of self-reported tenderness in the neck-shoulder region during daily activies. The difference between the group distribution was statistically significant. Children with migraine reported tenderness more frequently in the neck-shoulder region than the other groups. Neither the total tenderness score nor the dolorimeter score correlated with the frequency of self-reported tenderness. The correlation coefficients (Spearman) are shown in Table 3.
The frequency of self-reported tenderness in the neck-shoulder region during daily activies in children with migraine, tension-type headache and no headache
P-value for the difference between the group distributions=0.024.
The Spearman correlations of self-reported tenderness in the neck-shoulder region during daily activities with total tenderness score or dolorimeter score in children with migraine, tension-type headache and no headache
Ten children with migraine and two with tension-type headache reported headache on the day of the examination. The median of the total tenderness score of these 10 children with migraine was 13.5 and the mean dolorimeter score was 3.91 kg/cm2, whereas the median of the total tenderness score of the other children with migraine was 11 and the mean dolorimeter score 3.87 kg/cm2. The differences were not statistically significant.
Sixteen children with migraine had migraine attacks and 11 children with tension-type headache had headache episodes in the 72 h prior to the examination. No significant differences were found in the median of the total tenderness score (P = 0.62) or the mean dolorimeter score (P = 0.49) between these 16 children with migraine in the 72 h prior to the examination and other children with migraine. In addition, no significant differences in the median of the total tenderness score were found in children with tension-type headache who had headache episodes in the 72 h prior to the examination in comparison to those children with tension-type headache who had last headache episode over the 72 h prior to the examination (P = 0.56). However, the mean dolorimeter score was significantly lower in children with tension-type headache who reported headache episodes in the 72 h prior to the examination (3.26 kg/cm2) than in children with tension-type headache who had last headache episode over the 72 h prior to the examination (4.14 kg/cm2) (P = 0.006).
Discussion
The present study is the first population-based randomly selected cohort study on the occurrence of muscle tenderness in the pericranial and neck-shoulder region both in childhood tension-type headache and migraine. Contrary to our expectations, we found no association between muscle tenderness and tension-type headache in 13-year-old-children. However, children with migraine had increased overall tenderness compared with controls. They also had increased tenderness in some special pericranial and neck-shoulder muscles compared with children having tension-type headache or no headache.
Increased muscle tenderness recorded by manual palpation both in subjects with episodic and those with chronic tension-type headache compared with subjects without headache has been shown in adults (7). In addition, the pathogenetic importance of muscular factors in tension-type headache has been shown by the reports of spontaneous and induced headache (8, 9). Subjects with episodic or chronic tension-type headache have had more muscle tenderness than subjects with migraine or no headache (7).
On the basis of adult studies, it is still uncertain whether myofascial tenderness is primary or secondary to the pain in tension-type headache (10). It seems, however, that myofascial pain has prospective importance. In 16-year-old adolescents, tenderness in neck and shoulder muscles was a risk factor for cervical pain from nine to 12 years later (11). Absence of increased tenderness in our children with tension-type headache suggests that interictal myofascial pain is not a significant associated factor, not primary nor secondary, in tension-type headache at least in the majority of children.
As children with migraine had increased tenderness in the pericranial and neck-shoulder region, this may indicate that myofascial pain sensitivity increases especially in association with severe headache, such as migraine. In addition, children with migraine may be more sensitive to overall pain than children with no headache or children with tension-type headache.
Pain is always subjective, but different algometers have been developed to quantify it. The reliability and validity of these algometers have been reported to be acceptable (12–14). However, in adults with headache, palpation seems to be a better method for detecting pain in pericranial muscles than algometers (7).
Jensen et al. (7) found no difference in the mean pressure pain thresholds between adults with migraine, tension-type headache or no headache. In children, the results are analogous to those of adults (7). In the present study, children had episodic form of tension-type headache. In some adult studies, pressure pain thresholds were decreased especially in patients with chronic tension-type headache (15, 16). Children's mean pressure pain thresholds also showed large interindividual variation in each group, as in previous healthy adults (17, 18). However, a negative correlation between the total tenderness score and the dolorimeter score was found in each headache group. This finding is logical and expected.
Children with migraine reported tenderness in the neck-shoulder region during daily activities more often than children in the other groups. On the other hand, there was no correlation between self-reported subjective symptoms and pain findings measured by manual palpation or using a pressure algometer. Children with migraine may be more sensitive to pain or, due to frequent pain attacks, may report pain more easily than children with tension-type headache or headache-free controls.
In the present study, the proximity of a migraine attack to the examination had no influence on either muscle tenderness recorded by manual palpation or pressure pain thresholds. However, the sample containing those with headache on the day of the examination was small. In adults, an increased pericranial muscle tenderness in the temporal region has been reported during migraine attacks compared with the headache-free state (19). On the other hand, neither in the general population (7) nor in a clinic-based study of headache patients (8), no relation existed between pressure pain thresholds and headache episodes.
In conclusion, children with migraine had increased muscle tenderness at palpation of pericranial and neck-shoulder muscles, and they also reported pain symptoms in the neck-shoulder region most frequently. Instead, increased pericranial and neck-shoulder muscle tenderness is not associated with tension-type headache in children.
Footnotes
Acknowledgements
This study was financially supported by the Turku University Foundation.