Isometric endurance test of the cervical flexor muscles – reliability and normative reference values

Objective. To obtain reference values for the isometric endurance test (IET) of the cervical flexor muscles, investigate its reproducibility, and compare the results with the maximal isometric strength test (MIST) of the cervical flexor muscles. Design. Cross-sectional non-comparative study with single group repeated measurements. Methods: Altogether 219 healthy females aged 20 to 59 years volunteered to participate in the study. The IET was performed in the supine position and MIST seated. The reproducibility was evaluated by the intraclass correlation coefficient (ICC) and an analysis described by Bland and Altman. The relationship between the two measuring methods was evaluated by Pearson’s correlation coefficient. Results: The mean (SD) IET time was 60 (33) seconds with no significant differences between the age groups of each decade. The ICC for intrarater repeatability was 0.80. However, the Bland-Altman analysis suggested moderate variation in repeated measurements. Pearson’s correlation coefficient between the IET and MIST was 0.56. Conclusion: Normative reference values for the IET are presented. Although the ICC showed good repeatability, one should consider that the change at follow-up visits has to be considerable to be clinically relevant. The correlation between the endurance time and maximal flexion strength was moderate. Thus IET of the cervical flexor muscles may be used in the clinic like the Biering-Sorenson test has been used to assess fatigue of the trunk extensor muscles.


Introduction
Neck pain is common in the adult population. The results of a large survey conducted in Finland showed that 41% of people aged 30 years or more have experienced neck pain during the previous month and 6% suffer from chronic neck pain (1). Similar results have been obtained in other epidemiologic studies in western industrialized countries (2). Neck pain has shown to be the most common reason to visit a physician in primary health care (3). Moreover, people suffering from chronic neck pain use healthcare services twice as often as the general population and thus comprise a considerable burden to the health care system.
In comparison studies, women have shown to have about 60 % of the maximal isometric neck strength (MINS) achieved by men (4)(5). Accordingly, one factor why chronic neck pain is more common among females may be their lower muscle strength, even though the weight of the head that needs to be carried is quite equal between genders. Indeed, several studies have shown that neck pain is related to low MINS among patients, i.e., patients with chronic neck pain have considerably weaker neck muscles compared to their healthy controls (6)(7)(8)(9)(10).
Moreover, randomized controlled studies have shown improvement in neck muscle strength as a result of regular long-term specific exercising in patients with chronic neck pain (11). This improvement has resulted in reduction of neck pain and disability. However, an accurate and precise measurement of the MINS requires equipment that is seldom available in clinical practices. Thus simple, fast, but still precise enough tests for clinical practice are welcome.
The purpose of this study was to evaluate the isometric endurance strength of the cervical flexor muscles in healthy women to obtain reference values for clinical examination and M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT rehabilitation of patients suffering from chronic neck pain. The second aim was to evaluate the repeatability of the isometric endurance test (IET) of the cervical flexor muscles. Finally the results were compared to the maximal isometric strength test (MIST) results of the same study population.

Methods
Volunteers were recruited by the study personnel, who sent information about the study to the personnel of the largest employers in the City of Jyväskylä. The subjects were primarily employees of the City of Jyväskylä, the local hospital, and various industrial facilities, and consisted of both blue-collar and white-collar workers. The youngest age group was primarily made up of students. The subjects completed a questionnaire containing questions on their health status, occupation, and competitive-sports activities. The inclusion criteria were that the subject should be healthy, female, and aged 20 to 59 years. The exclusion criteria were neck and shoulder pain experienced within the previous 6 months, previous or current injuries or other disorders of the neck-shoulder area, arthritis, fibromyalgia, severe depression or mental disorder, or an active competitive sports career. Out of 241 volunteers, 18 were excluded due to neck-shoulder symptoms, three for not giving the information requested and the data of one person was lost. A total of 219 females were enrolled in the study. The purpose of the study and the study protocol was explained to the subjects, after which the subjects gave their written consent. The study was approved by The Ethics Committee of the Central Finland Health Care District.
Anthropometric measurements included body height and mass. The IET was performed in the supine position with craniocervical flexion. Subjects were instructed to tuck in their chins and then to raise their heads approximately 2-3 cm above the plinth and the forehead against a position detector stick. This position has shown to activate both the deep and superficial neck muscles (12). The time when the test position remained stable, until the forehead began to drop from the stick, was measured in seconds (s) with a stopwatch. There was one week between the two measurement sessions. All measurements were made by the same physical therapist who had several years of clinical experience in neck muscle strength tests. The MIST M A N U S C R I P T

A C C E P T E D ACCEPTED MANUSCRIPT
was performed with the subject in the sitting position with the force gauge mounted in a sturdy stand and the subject fastened to it (13). Two warm-up trials were performed, followed by three maximum-effort trials in each direction. If the result of the third trial was 5% or more above the highest of the 2 previous trials in that direction, additional trials were performed until the improvement remained under 5%. We have previously reported the MIST results of the same study population (14).

Statistical analysis
The results are expressed as means with SDs and with 95% confidence intervals (CIs), and as medians with interquartile (25th-75th percentile) ranges. Statistical comparisons among age groups were done by analysis of variance (ANOVA) with Sidak's adjustments for pairwise correlations. Pearson's correlation coefficients with bootstrapped 95 % confidence intervals were calculated between IET and MIST results. The intraclass correlation coefficient (ICC 2,1) was used to examine intra-rater reliability for the IET. The reliability is regarded as acceptable if ICC>0.75 (15). The standard error of measurement (SEM) was used as a parameter of absolute reliability and agreement; it was calculated as the square root of residual variance from oneway analysis of variance. The confidence intervals for SEM were obtained using the degrees of freedom associated with estimated residual variance and percentage points from corresponding chi-square distribution (16). An analysis described by Bland and Altman was also done in which differences between two IET measurements were set against the corresponding mean for each patient, to show the variability of the results at the individual level (17).
To determine the smallest change in each impairment measure that can confidently be considered to exceed measurement error at a 95% confidence level, the Minimum Detectable Change (MDC) was calculated according to the following formula: MDC = 1.96 · √2 · SD · √(1 −test-retest reliability coefficient) (18).

Results
The anthropometric data, IET and MIST results of the participants in the different age groups are presented in Table 1. Although the oldest age group was the shortest in height, this age group was the heaviest. The weight, and thus the body mass index as well, increased with advancing age, leading to the significant difference between the youngest and oldest group. There was a great variation in individual IET among all age groups as shown in Figure 1.
However, there were no statistically discernible differences in endurance time between the different age groups. No significant correlations were found between age or height and endurance time. Only small positive association was observed between body mass and neck flexor endurance time (r = 0.16, p=0.05). In the present study, the IET results were shown to vary individually from a few seconds up to two minutes in all age groups. Very low neck muscle strength affects every day life because one has to support the head while rising up from the supine position. Quick acceleratory movements of the body may cause the head to move into the opposite direction, which may predispose neck muscles to a strain injury if the stabilizing muscles are too weak to protect the vulnerable soft tissues.
One would assume that body size is an important factor for endurance strength -a larger body often means bigger muscles. However, we did not find correlation between endurance time and height, and only weak correlation with body weight. This is in accordance with the previous findings that only very low or no correlation at all between the MINS and anthropometric measures have been found in women (19). In the cadaver study, the cross-sectional areas of neck muscles did not scale proportionately with body height and weight (20). The reason may be that the weight of the head does not vary as much as body size and thus the load on the neck muscles is fairly similar.  (Table 2). Most studies obtained lower mean endurance time compared to the present study. The possible reason may be that in the present study we used a stick to show subjects the level at which they tried to hold their head as long as possible. In previous studies, subjects had to try to keep the head in place without any specific feedback of the position. The considerable variation in results of the previous studies is probably due to the small amount of subjects in each study. Individually, neck flexor muscle endurance time has shown large variation from a few seconds to over two minutes in different studies. Thus study population selection may have had an effect on the results. The range of endurance time in the current study varied from 6 to 246 s. Thus the best time was over 40 times longer than the shortest. The variation in endurance time is larger compared to that found in maximal isometric strength, ranging from 30 N to four times that in the same study population (14). Several studies have reported only the average endurance time for the whole study population, although they have contained both males and females. Due to significant differences in neck strength and endurance times between sexes, shown in many studies, the average endurance  (26). They found no difference between the groups, but the amount of neck pain was not reported, and the mean NDI was only 14 (7). Low neck pain and disability, as well as the short duration of symptoms, may explain why there has been no difference in endurance strength between patients with neck pain and healthy controls in some studies. Males have shown to have significantly longer neck flexor muscle endurance times compared to females (28,29). Men also have more variation in endurance times. However, several studies have analyzed results without separating the sexes, which should be considered as a confounding factor.  (11). Exercise frequency and total exercise dose in the exercise program have also shown to be important factors in the treatment of chronic non-specific neck pain (31).
The Biering-Sorensen test has been used for evaluating the isometric endurance of trunk extensor muscles. Its reproducibility and discriminative validity has shown to be good and the test has been used commonly for evaluating muscle performance in patients with low back pain (32). Similarly the IET may be used for evaluating muscle performance in patients with neck pain before and after rehabilitation programs.

M A N U S C R I P T
A C C E P T E D ACCEPTED MANUSCRIPT Table 2.
Mean neck flexor muscle endurance time (SD) in healthy subjects found in previous studies. The subjects lifted and kept the head approximately 2-3 cm/10° above the plinth during the test.