1.1

Introduction

Measurement of trunk muscle strength and endurance contributes to clinical evaluation of the LBP patient; however, their physiological and individual variability effectively complicates appraisal and comparison.

From a practical standpoint, muscle assessment methods using an isokinetic dynamometer are considered to be reliable and reproducible, with correlation coefficients between 0.93 and 0.99 for peak force values and between 0.91 and 0.96 for total workload values . This dynamic assessment technique necessitates written as well as oral information from the patient, who is called upon to follow instructions throughout the procedure (means of installation, strapping, warm-up, initiation to training, etc.).

Most existing studies focusing on isokinetic assessment of trunk muscle function involve adult populations consisting in healthy subjects, LBP subjects or operated subjects . On the other hand, studies assessing trunk muscle strength in children and adolescents have remained scarce .

Through clinical trunk muscle endurance tests, Salminen et al. evaluated the strength of the flexors and extensors in 15-year-old and observed diminished endurance in the LBP group compared to the control group. In the study by Sjolie and Ljunggren , diminished muscle strength in the spinal extensors and increased spinal mobility served as predictive factors for lower back pain in subjects between 14 and 16 years of age.

The studies conducted in children have underscored a number of non-specific factors associated with LBP: relatively advanced age , female sex , past family LBP history , stiffness of the posterior muscle chain , vigorous physical activity , as well as the type and duration of seated position . In a previous study we showed that when using the Sorensen test, the spinal extensors were weaker in LBP patients than in members of the control group, and we wished to know if this remained the case from an isokinetic standpoint.

The main objective of this study consists in comparing by means of an isokinetic dynamometer the respective strength, power and endurance in healthy and LBP children with ages ranging from 11 to 16. The secondary objective, which pertains to the LBP population alone, consists in analyzing the relative influence of various clinical (pain intensity, hours of sports practice per week, muscle extensibility…) and radiologic (pelvic incidence, lumbar lordosis…) with regard to isokinetic performance.

1.2

Patients and method

1.2.4

Central criterion of judgment

Muscle strength assessment by isokinetic dynamometer is carried out with a Cybex Norm ^® apparatus . The spinal flexors and extensors are concentrically explored at speeds of 60°, 90° and 120°/s. The children are set up in a standing position, and their legs, pelvis and chest are kept in place with fastening material ( Fig. 1 ).

Installation for isokinetic muscle strength assessment by the Cybex Norm ^® isokinetic dynamometer.

The patient is informed of the apparatus’s operational principle and is given advice on how to provide maximum repetitive effort throughout his movements, whether they consist in bending or in extending. The isokinetic test is preceded by 20 to 30 minutes of warm-ups on a cycle ergometer and a preliminary phase is carried out with less than maximal exertion. The protocol itself consists in 3 series of 4 consecutive bending-extension movements with trunk anteflexion at 60°, 90° and 120°/sec and at an amplitude of 0–70°; it also involves a series of 20 movements at 120°/s. Rest time between series lasts 1 minute. All of the recordings are carried out by the same examiner. During each series, the subject is encouraged by the examiner.

1.3

Results

1.4

Discussion

The results of our study show that when bracketed by age, our healthy and back pain populations are homogeneous. In our work, we wished to classify the subjects by age brackets in order to take into account the growth of the rachis during the pubertal growth spurt. We have observed that the 14-to-16-year-old practice less sport (30% of them no longer practice) than the 11-to-13-year-old; it is as though the pain had become more of a functional impairment.

For the two classes, radiologic lordosis and kyphosis parameters have largely disappeared (Type 2 in the Roussouly classification), but given the absence of radiological data in the controls , it is difficult to further advance the discussion.

The results of our study do not illustrate a significant difference in isokinetic performances between children with and without back pain in the age brackets of 11 to 13 and 14 to 16 years. With regard to the latter, however, we observed significantly different ratios having to do with loss of capacity in the spinal extensors. Balagué et al. found no significant difference in isokinetic performances between children with and without back pain. It is in terms of patient recruitment that our study differs from Balagué’s; our back pain population, unlike his, is treated during a hospital stay for long-standing pains that interfere with everyday activities on account of their intensity (VAS > 50%) and their functional repercussions; what is more, they are insufficiently responsive to physical therapy. The lower back pain patients in Balagué’s study were all enrolled in school and not under treatment. We have already noted this difference in a previous study involving static assessment of muscle strength in which children going to school, tested in their school setting and presenting with back pain were found to be statistically comparable to healthy children tested on the same site but statistically different in terms of muscle strength from our hospitalized children with back pain . This result means that the pain experienced by a pre-adolescent or an adolescent going to school and having a backache not entailing functional repercussions (school absenteeism, medication, para-medical care…) will be neither as intense nor as long-standing as the pain undergone by children receiving institutional care (physiotherapy, braces…) for spinal pain. This comparison underscores the limits of pain assessment on the sole basis of a visual analog scale and the complexity of analyzing functional repercussions .

The mean power of the extensors for back pain children aged 11 to 13 years is positively correlated with lumbar lordosis; on the average, the lordosis is rather weak in comparison with the control population reported by Mac-Thiong; this finding could mean that back shape (type 2 in the Roussouly classification, with small lordosis) could have a direct impact on extensor strength and consequently on the agonist/antagonist imbalance from which the pains originate. As for the children aged 14 to 16 years, the moments of force and total workload of the extensors are negatively correlated with the sacral base; this result may signify that the greater the lumbar lordosis, the weaker the flexors. It would be interesting to correlate these data with pelvic incidence, which is a fixed parameter, rather than with the sacral base, which constitutes a variable parameter.

In our work, the isokinetic performances of the back pain patients were correlated neither with the modified Saint-Antoine questionnaire score nor with the back pain VAS. The limited number of patients studied, the long-standing nature of their pain and its average intensity and frequency did not cumulatively suffice to produce “muscle deconditioning” and to influence the performances of the spinal musculature during the isokinetic tests. On the other hand, Anne Keller et al. showed that age, sex, trunk muscle density, psychological state and pain as evaluated on the visual analog scale influenced the isokinetic performances of the trunk in adults with chronic lower back pain. Pain as evaluated on the VAS was considered a predictive factor for the isokinetic performances of the trunk in extension. It has also been noted that isokinetic performances are not influenced by the emotional and general mental state . As for Ebermeyer et al. , in a population of 23 female gymnasts they took note of a correlation between pain and small F/E ratio for the rachis in comparison with female gymnasts not in pain.

In our population of pre-teen (11-13-year-old) back pain and control subjects, gender had no influence on their isokinetic performances, which may perhaps mean that the isokinetic dynamometer is not a sufficiently sensitive device or that the children in this age bracket are too young or that their back pain is not intense enough to evaluated. Among the 14-to-16-year-old, on the other hand, whether they are healthy or not, the boys appear stronger than the girls, with higher values for the extensors in the controls and higher values for the flexors in the patients. In a study involving 62 school children with an average age of 12 years, Mérati et al. found that isokinetic performances pertaining to peak torque, total workload and mean power were higher for boys as regards the trunk extensors at 60°/s and at 90°/s. Was the population healthy or not? The authors found that in the back pain population, gender had differing effects on isokinetic performances, particularly with regard to the spinal extensors. However, Balagué et al. showed that in children with back pain, there was no connection between the isokinetic performances of the trunk and the various clinical parameters (age, gender, sports practice). If the conclusions to our study differ from Balagué’s, this is probably due to the fact that the typology of the back pain patients differs. Let us recall that as a general rule, in a healthy adult subject the extensors are stronger than the flexors and that a typical man’s muscle capacity is greater than a woman’s, with differences approximating 25% for the extensors and 20% for the flexors ; moreover, athletic subjects show more muscle capacity than sedentary subjects . A normal flexor/extensor ratio is lower than 1, ranging from 0.80 to 0.85 according to Grémion and from 0.64 to 0.72 according to Voisin , without correction for gravity. Most authors would agree that the trunk extensor/flexor ratio in healthy subjects ranges from 1.2 to 1.4 . In our study we have noted that the control children performed more efficiently in terms of power; this is likely to be explained by a form of inhibition toward high-speed movement existing in the children with back pain. The deficiencies detected at the level of the extensors may be interpreted as a loss of actual force or, more precisely, as a loss of the straightening function due to the proprioceptive deficit of the posterior muscle chain.

The isokinetic performances with regard to endurance (total workload) are strongly correlated in the back pain population with the number of sports practice hours. According to Negrini , isokinetic performances in flexion and in extension at 60°/s are better in children participating in regular physical activity in a competitive setting. He goes on to conclude that participation in regular physical activity has a significant effect on the isokinetic performances of healthy school children. In the adult, the relationship between isokinetic trunk muscle performances and sports activity in LBP patients is controversial . Balagué et al. have observed that peak torque in extension is at its maximum level in 12-year-old girls and that in boys, from the age of 14 years it increases constantly.

Negrini et al. have observed that in LBP school children a diminution of peak torque for the trunk flexors and extensors occurs at 60°/s.

In our study, we have observed weaker performances with regard to power at high speed in LBP children compared to the healthy population; this may be explained by a form of inhibition toward high-speed movement existing in LBP children. In this population, the mean high-speed isokinetic power with regard to trunk extension depends on the value of the radiologic lumbar lordosis. Once the correlation between power and lordosis was found to be positive, we could conclude that the more pronounced the lordosis, the stronger the thrust of the lower limbs as trunk extension motion gets underway. That said, only a complementary and comparative study with a healthy population could allow us to confirm this hypothesis.

Several authors have recommended strengthening of the spinal muscles (rachis extensors) in order to support maintenance of physiological lumbar lordosis, to diminish axial loading , to improve overall posture (or sagittal balance) , to protect the ligamentary structures of the rachis and, finally, to facilitate lifting of a heavy weight .

The main objective of this comparative study was to attempt to discern a connection between trunk muscle strength and LBP occurrence in children. If we decided to perform this study in a pre-teen age bracket (11–13 years), it was largely in order to reduce the influence of pubertal gain in weight and height. This is a necessary limitation to our work as it was not possible to regroup all the children, given that the gains in height and weight vary greatly between boys and girls (including within each of the groups) at the time of puberty, which occurs earlier in girls (11–13 years) than in boys (14–16 years). So it is that we chose 11-to-13-year-old in order to obtain two populations that would be homogeneous in weight and height. Below the age of 11, on the other hand, there would have been a problem of critical height (lower than 150 cm). While our populations are hardly sizable, our results are largely in agreement with those of the literature pertaining to adults, which has reported predominant muscular capacity in men as opposed to women, deficient extensors in LBP patients (especially at high speeds), and flexor/extensor ratios approximating 1.

Longitudinal monitoring of the LBP children would allow us to underline the development of capacities and the possible risk of deconditioning.

Muscle assessments in healthy children should be carried on at the same time as surveillance of LBP children in view of pinpointing a possible relationship between development of symptoms and eventual diminution of muscle strength.

Ours is a preliminary study involving small samples, and the age brackets assessed have been limited so as to diminish any biases due to pubertal growth.

In our populations, we could not deduce reference values for the isokinetic parameters; all we could do in this respect was establish frameworks for values that will be adjusted as our series of tests increase in number:

• •
  

  for the girls:

  
  
  
  
  • ∘
    

    the MMF and MP values for the flexors range from 1.7 to 2 times body weight,

    
    
  • ∘
    

    the MMF and MP values for the extensors range from 2 to 2.8 times body weight.

  
  
  
  
• •
  

  for the boys:

  
  
  
  
  • ∘
    

    the MMF and MP values for the flexors range from 1.4 to 2 times body weight at 11–13 years and from 2.4 to 2.8 times the body weight at 14–16 years,

    
    
  • ∘
    

    the MMF and MP values for the extensors range from 1.7 to 2.7 times body weight at 11–13 years and from 2.4 to 3.5 times body weight at 14–16 years,

    
    
  • ∘
    

    as regards the ratios, they tend to be normalized for the boys, decreasing with growth from 0.72–0.91 to 0.67–0.77,

    
    
  • ∘
    

    the girls in our populations show more elevated ratios, evolving from 0.81–0.94 to 0.75–0.95 for the controls and from 0.75–0.91 to 0.85–1.06 for the LBP patients,

    
    
  • ∘
    

    71% of the LBP girls present inverted ratios in MP and TW at 14–16 years, as opposed to only 9% at 11–13 years.

  
  

1.5

Conclusion

While there exists no significant difference in isokinetic performances between LBP and asymptomatic children in the 11–13 year age bracket, the isokinetic performances in the LBP subjects are positively influenced by the BMI index, the number of hours of sports activity and the radiologic value of lumbar lordosis. Performances in terms of power are weaker in the LBP sub-group. Given these results, as regards 11-to-13-year-old assessment by isokinetic dynamometer does not apparently yield characteristic muscle-related elements liable to explain LBP incidence.

As regards the 14–16 year age bracket, we have observed that muscle strength is correlated with age and that the extensors of the LBP patients are weaker than those of the controls, while the flexors of the LBP patients are stronger. At this stage, the isokinetic dynamometer becomes an interesting means of evaluation. For this age bracket, the secondary objective of the study has been addressed, with results on extensor capacity in LBP patients that confirm the static evaluation carried out in preceding work.

As we have not standardized our criteria of judgment (number of trials, work speed…), it is difficult to compare or results with those reported in the literature. In order to better comprehend the phenomena studied, a comparative and longitudinal study with a larger population remains necessary.

Disclosure of interest

The authors declare that they have no conflicts of interest concerning this article.

Acknowledgements

The authors wish to thank the Association de Recherche Médicale of the Centre Médico-Chirurgical de Réadaptation des Massues and the documentation unit, particularly Pascale Raillard, for the assistance they gave in writing this article. They also wish to thank Rachel Bard-Pondarré for her critical review.