Isometric Muscle Torque in Children 5 to 15 Years of Age: Normative Data

Methods 

A total of 149 children (76 boys, 73 girls) 5 to 15 years old participated. Written consent was obtained from the parents, along with information about the child’s health status. The children had no known diseases affecting muscle strength according to their parents’ statements. An additional 41 children received but did not reply to the letter of invitation and 25 children said no to participation in the study.

Measurement Procedure 

We measured height using a measuring rod on the wall and weight was measured on an electronic scale. Anthropometric data were compared with normative data for Swedish children used in the school health care system.22 Hand dominance was assessed when the children wrote their name.

We measured muscle strength using a hand-held electronic dynamometera (fig 1). The dynamometer can measure both traction and compression with a maximum force of about 550N and a precision of 0.5N. The dynamometer was calibrated with known weights before and after the study. We tested the nondominant side, as was done in a previous study.21 A study on healthy adults showed that upper-extremity muscle force values were different between sides but that lower-extremity muscles were not,23 so nondominant side was based on the hand dominance. Eight muscle groups in the leg and 4 in the arm were measured (table 1). The procedure was standardized for every muscle group, subject position, dynamometer position, and stabilization of subject. We used positions previously published3, 20, 21 with minor adjustments. We added measurement of hip flexors and extensors in a second position, which in clinical practice we have found useful for stronger patients. Four of the muscle groups were measured in 2 different positions (elbow extensors, hip extensors and flexors, knee flexors).

View Large Image Download to PowerPoint

Fig 1. Measurement of hip extension with the Chatillon dynamometer. Lever arm indicated with ×.

Table 1. Muscle Groups

	Muscle Group	Position	Stabilization	Resistance	Lever Arm From
	Shoulder abductors	Sitting Shoulder abducted 90°, elbow flexed, opposite hand in knee		Humerus distally	Acromial process
	Elbow extensors 1	Prone Shoulder abducted 90°, elbow flexed 90°	Lower part of humerus	Forearm distally	Lateral humeral epicondyle
	Elbow extensors 2	Supine Shoulder adducted, elbow flexed 90°, forearm in neutral	Humerus	Forearm distally	Lateral humeral epicondyle
	Elbow flexors	Supine Shoulder adducted, elbow flexed 90°, forearm in supination	Shoulder	Forearm distally	Lateral humeral epicondyle
	Wrist dorsiflexors	Supine Shoulder adducted, elbow extended, forearm in pronation, fingers extended	Forearm	Dorsum of hand	Ulnar styloid process
	Hip extensors 1	Supine with legs outside the couch Not tested foot on the floor	Hold on to bench	Femur distally	Greater trochanter
	Hip extensors 2	Supine	Hold on to bench	Femur distally	Greater trochanter
	Hip flexors 1	Sitting	Hold on to bench	Femur distally	Greater trochanter
	Hip flexors 2	Supine Hip and knee flexed 90°	Hold on to bench	Femur distally	Greater trochanter
	Hip abductors	Supine Hip and knee extended	Hold on to bench Stabilization of the other leg	Femur distally	Greater trochanter
	Hip adductors	Supine Hip and knee extended Not tested knee flexed	Hold on to bench	Femur distally	Greater trochanter
	Knee extensors	Sitting	Hold on to bench	Shank distally	Lateral knee joint
	Knee flexors 1	Sitting	Hold on to bench	Shank distally	Lateral knee joint
	Knee flexors 2	Prone Knee flexed 90°	Hold on to bench	Shank distally	Lateral knee joint
	Ankle dorsiflexors	Supine Hip and knee extended, ankle 0°	Hold on to bench	Dorsum of foot	Lateral malleolus
	Ankle plantarflexors	Sitting Knee extended	Hold on to bench	Metatarsal heads	Lateral malleolus

NOTE. Description of position and stabilization of subject, position of the dynamometer’s resistance to the body movement (arrows), and description of point for measuring lever arm.

We started with the tests in sitting position where the child could familiarize with the procedure and practice, leg first then arm muscles. This was followed by testing in supine and prone position. Three attempts were made for each muscle group with the make test technique, where resistance is gradually built up for about 5 seconds. Time for rest was given between trials and we varied between measurements of leg and arm to avoid fatigue. Encouragement to maximum effort was given in a standardized way. The maximum result for each muscle group was used. To allow as long a lever arm as possible, at a location where a strong pressure on the skin did not hurt and prevent a maximal contraction, the HHD was placed distally at the segment tested at a place that was comfortable for the subject (see table 1, fig 1). The position of the HHD head was marked on the skin. The lever arm was measured with a tape measure using bony landmarks (acromial process, lateral humeral epicondyle, ulnar styloid process, greater trochanter, lateral knee joint, lateral malleolus) and the position of the center of the HHD head. Torque was calculated in newton meters. The procedure was performed for 30 to 45 minutes.

A special device was made for measuring ankle plantarflexion. The subject was seated with extended knees and with 1 belt around the hips and 1 around the foot at the level of metatarsal heads. The dynamometer, set in traction mode, was attached between the 2 belts. A piece of wood was placed under the child’s foot inside the belt, to press comfortably on.

The measurements of muscle strength were performed by 3 physical therapists. Three muscle groups (hip flexion, knee extension, knee flexion) were tested by 2 examiners during the same session, to determine interrater reliability.

Results 

Patient Characteristics 

Data on weight and height in the children are presented in Table 2, Table 3. Age, height, and weight were within age norms for Swedish children except for 3 girls (5, 8, 11y) who were above the 3 standard deviation (SD) curve for height or weight.22 There were strong correlations between age, height, and weight (table 4). There was no statistically significant difference in height or weight between boys and girls except at the age of 14 and weight of 55kg. Fourteen (9.4%) children were left handed.

Table 2. Participants’ Height and Weight

	Subjects	Age (y)
		5	6	7	8	9	10	11	12	13	14	15
	Boys (n)	6	7	6	6	6	5	5	5	7	18	5
	Girls (n)	6	5	7	6	6	6	9	6	7	9	6
	Height (cm)
	Boys	113.3±5.3	120.1±4.5	130.4±4.3	129.1±3.6	137.8±4.7	146.2±5.5	148.3±10.8	154.9±11.6	159.4±8.2	172.3±5.9⁎	172.1±11.8
	Girls	117.5±8.8	120.6±2.3	124.2±7.6	136.9±8.2	135.8±6.0	144.7±7.0	151.9±7.3	154.8±4.4	159.7±5.0	165.6±5.6⁎	167.2±5.4
	Weight (kg)
	Boys	20.1±2.7	22.6±1.5	26.2±2.4	28.1±2.0	33.0±2.9	35.8±4.8	37.1±9.2	43.9±5.8	49.0±9.4	59.2±8.8	59.3±9.3
	Girls	22.4±6.5	22.1±0.7	23.2±4.3	33.1±9.0	29.0±3.6	38.2±7.8	37.9±6.1	44.4±6.2	47.8±8.8	57.1±7.6	57.7±7.3

NOTE. Values are n or mean ± standard deviation (SD).

⁎ Statistically significant difference between boys and girls.

Table 3. Number of Subjects Divided Into Weight Classes

	Subjects (n)	Weight (kg)
		15−20	20−25	25−30	30−35	35−40	40−45	45−50	50−55	55−60⁎	60−65	65−70	70−75
	Boys	3	12	12	8	5	8	5	5	5	8	1	4
	Girls	4	13	6	10	7	6	11	5	5	4	2	NA

Abbreviation: NA, not applicable.

⁎ Statistically significant difference between boys and girls.

Table 4. Correlation Among Age, Weight, and Height

	Subjects	Weight	Height
	Age
	Boys	0.91	0.95
	Girls	0.87	0.93
	Weight
	Boys	1.00	0.96
	Girls	1.00	0.93

Muscle Strength 

There was an increase in torque with age and weight in all muscle groups. The correlation between torque and both age and weight was strong for arm muscles (r range, .79−.90) and leg muscles (r range, .84−.95), except plantarflexors (r range, .66−.71) (all P<.001). There were few differences in torque between boys and girls (figs 2A–D). In the arms, there was a statistically significant difference in elbow extensors (age 7, 9, 14, 15y) and in elbow flexors (age 9 and 14y). In the legs, there was a statistically significant difference in hip abductors and adductors (age 7 and 14y) and knee extensors (age 15y). Mean values ± SD for all muscle groups related to age and weight are presented in Table 5, Table 6, Table 7.

View Large Image Download to PowerPoint

Fig 2. Mean muscle strength with error bars ±1 SD in elbow flexors: (A) age related, (B) weight related, hip extensors, (C) age related, and (D) weight related.

Table 5. Mean Age-Related Values for Children 5 to 12 Years Old

	Muscle Group	Age
		5	6	7	8	9	10	11	12
	Shoulder abductors	8.8±2.8	11.1±2.9	14.3±5.0	18.2±3.7	19.3±4.4	21.9±5.8	23.8±4.9	30.5±6.3
	Elbow extensors 1	8.8±2.6	11.3±2.9	11.3±4.0	15.9±3.7	18.6±7.3	16.2±3.8	19.4±4.5	26.2±8.4
	Elbow extensors 2	7.1±1.5	10.3±2.7	10.0±3.0	14.8±3.3	15.7±3.2	15.9±2.4	17.1±6.9	20.5±3.9
	Elbow flexors	8.5±2.5	11.7±2.3	12.9±3.9	16.5±3.2	20.0±4.6	21.6±3.8	24.3±6.5	29.4±6.3
	Wrist dorsiflexors	1.7±0.6	2.4±0.5	2.3±0.8	3.3±1.0	4.0±1.3	3.8±0.8	4.7±1.6	5.1±1.0
	Hip extensors 1	20.1±6.6	32.1±11.0	35.8±16.2	49.3±11.9	56.0±16.6	58.9±14.9	68.7±22.2	95.1±31.4
	Hip extensors 2	16.1±3.7	23.6±5.5	31.5±11.6	43.7±12.4	51.0±13.1	61.2±19.4	67.4±16.9	105.4±30.5
	Hip flexors 1	15.8±3.9	24.1±5.5	27.0±8.9	38.6±11.5	46.3±10.6	54.0±13.8	62.6±23.2	73.7±15.7
	Hip flexors 2	16.2±4.4	23.0±6.0	26.2±8.1	35.3±8.9	39.8±9.5	45.3±8.0	50.3±15.2	64.7±14.5
	Hip abductors	16.6±4.5	22.1±4.4	25.4±5.6	40.7±11.2	45.0±9.4	56.7±13.0	62.4±18.8	72.1±18.3
	Hip adductors	15.5±3.2	20.7±3.8	25.5±9.7	33.6±9.9	40.9±15.3	43.0±13.5	57.6±17.1	73.2±17.4
	Knee extensors	21.0±5.8	26.0±4.0	30.2±8.5	45.4±12.1	42.9±5.9	61.4±14.9	63.3±17.0	74.3±12.7
	Knee flexors 1	15.9±3.5	20.0±3.8	22.7±5.6	32.4±9.4	33.9±5.5	46.8±11.6	48.9±11.6	62.6±17.0
	Knee flexors 2	12.5±3.6	16.9±4.3	17.4±4.0	24.1±5.1	29.1±6.7	34.7±10.2	36.8±11.7	47.0±12.9
	Ankle dorsiflexors	8.1±2.8	10.6±2.1	11.7±3.8	14.4±2.4	21.3±3.8	19.9±3.6	22.5±6.9	27.1±7.1
	Ankle plantarflexors	17.3±5.7	25.8±6.0	21.1±11.5	31.8±5.2	40.2±9.6	ND	ND	ND

NOTE. Values are mean newton meters ± SD. Abbreviations: ND, no data, given that the children were stronger than the device could measure (>500N).

Table 6. Mean Age-Related Values for Children for Boys and Girls 13 to 15 Years Old

	Muscle Group	Age
		13		14		15
		Boys	Girls	Boys	Girls	Boys	Girls
	Shoulder abductors	36.5±10.0	33.9±4.9	42.8±9.8	39.0±7.2	42.2±8.3	36.6±5.0
	Elbow extensors 1	27.7±4.2	27.4±9.2	37.1±8.8	29.8±8.5	31.7±9.8	27.0±7.3
	Elbow extensors 2	25.4±8.9	23.9±5.5	33.0±6.9	25.3±5.0	31.7±6.0	23.6±4.8
	Elbow flexors	35.2±11.9	33.3±7.8	46.0±9.2	35.7±6.1	47.5±12.9	34.0±10.6
	Wrist dorsiflexors	7.2±2.5	6.1±2.4	9.1±3.5	8.0±1.6	11.9±2.7	9.4±3.2
	Hip extensors 1	103.4±44.8	98.6±26.3	134.9±32.2	114.6±23.3	142.1±35.2	141.0±13.0
	Hip extensors 2	105.9±30.8	103.9±23.4	138.0±34.1	127.6±24.6	161.6±15.8	144.1±22.3
	Hip flexors 1	85.5±18.9	84.4±23.1	120.4±25.5	101.6±22.8	124.6±30.5	115.0±16.5
	Hip flexors 2	64.5±19.3	70.2±21.8	78.9±18.2	73.0±13.5	80.4±16.7	75.8±9.8
	Hip abductors	82.5±28.5	82.2±19.0	122.3±27.9	100.9±15.1	120.3±38.2	119.0±29.6
	Hip adductors	83.7±27.7	85.2±28.5	111.9±19.8	92.8±23.5	130.5±36.6	110.8±17.0
	Knee extensors	82.5±18.3	79.9±13.4	110.4±23.2	97.4±18.5	122.1±18.6	98.0±14.9
	Knee flexors 1	67.9±24.9	68.2±17.5	89.2±22.4	79.3±13.2	104.4±36.4	82.7±9.1
	Knee flexors 2	53.4±9.6	53.2±20.5	72.3±19.8	59.8±12.0	63.9±16.4	58.6±10.9
	Ankle dorsiflexors	31.4±8.4	27.7±9.1	34.6±8.5	32.0±6.5	40.3±6.9	34.8±6.6
	Ankle plantarflexors	ND	ND	ND	ND	ND	ND

NOTE. Values are mean newton meters ± SD. Abbreviations: ND, no data, given that the children were stronger than the device could measure (>500N).

Table 7. Mean Weight-Related Values

	Muscle Group	Weight Group (kg)
		15−19	20−24	25−29	30−34	35−39	40−44	45−49	50−54	55−59	60−64
	Shoulder abductors	8.0±2.1	11.0±2.8	17.2±3.6	20.7±3.6	24.0±6.5	28.8±5.0	30.9±6.4	37.9±5.6	42.5±8.8	41.6±8.8
	Elbow extensors 1	9.3±2.2	10.3±3.4	14.9±3.2	17.6±6.6	19.6±5.2	23.0±5.1	26.0±7.9	27.7±4.6	34.1±8.4	36.6±11.1
	Elbow extensors 2	6.8±1.0	9.3±2.8	13.7±2.5	15.6±3.1	16.1±3.3	20.5±3.9	20.8±5.7	26.9±5.0	30.9±6.4	31.9±8.3
	Elbow flexors	7.2±1.3	11.0±2.7	16.9±2.7	20.2±4.0	23.9±4.4	27.8±6.2	29.7±6.3	37.6±4.1	42.5±10.2	44.6±13.5
	Wrist dorsiflexors	1.7±0.5	2.1±0.6	3.2±0.9	4.0±1.1	4.5±1.0	5.2±1.4	5.5±1.6	8.1±2.2	10.0±2.7	9.7±4.1
	Hip extensors 1	ND	29.6±11.7	46.5±10.0	57.7±14.4	70.8±25.3	88.9±33.4	89.6±20.0	125.0±28.7	125.4±28.6	141.0±32.6
	Hip extensors 2	17.1±4.6	22.3±6.1	43.7±8.3	55.3±12.6	68.4±24.4	88.2±22.8	95.9±29.6	127.2±23.6	139.3±23.9	151.2±35.5
	Hip flexors 1	14.8±2.7	22.7±6.9	37.1±9.6	46.1±9.9	57.5±14.7	78.0±17.8	76.2±18.7	104.0±21.2	115.7±26.1	121.2±21.1
	Hip flexors 2	13.3±2.4	21.9±5.7	34.1±4.4	41.6±8.9	48.4±12.1	57.8±11.0	62.2±15.6	79.5±16.9	74.8±12.6	79.5±19.6
	Hip abductors	15.6±2.8	21.3±5.5	36.9±8.5	48.3±9.5	56.6±14.5	73.3±18.7	73.9±12.0	99.4±10.3	115.7±17.8	129.4±28.7
	Hip adductors	14.8±2.2	19.2±4.3	34.5±7.9	42.4±13.3	54.0±16.1	76.8±32.6	73.3±18.4	96.8±17.5	100.6±22.3	113.7±15.7
	Knee extensors	18.2±4.6	25.9±5.5	39.9±5.7	46.6±6.6	60.8±10.6	73.1±10.0	79.1±10.3	100.3±18.8	97.4±16.3	117.8±14.9
	Knee flexors 1	14.0±2.2	19.7±4.0	29.5±5.3	37.2±7.6	49.0±11.4	57.3±7.9	64.3±12.9	75.2±18.3	85.3±15.2	98.2±22.4
	Knee flexors 2	12.1±3.1	16.1±4.6	22.4±5.4	30.5±6.0	36.2±8.1	43.1±10.5	49.5±14.0	54.5±7.6	64.8±13.9	77.0±19.4
	Ankle dorsiflexors	6.8±1.4	10.4±3.0	15.7±3.6	20.4±6.2	20.5±4.2	24.1±4.8	28.2±8.2	34.0±6.3	35.8±7.4	34.6±8.4
	Ankle plantarflexors	15.7±4.4	25.4±9.0	26.8±8.4	40.5±11.8	ND	ND	ND	ND	ND	ND

NOTE. Values are mean newton meters ± SD. Abbreviations: ND, no data, given that the children were stronger than the device could measure (>500N).

Equations for predicted torque with correlation coefficients are presented in table 8. Age and weight were the strongest predictors of torque; sex was included from age of 13. Plantarflexors were not possible to measure in children over the age of 9 years. This was due both to difficulties in stabilizing manually and the maximum measurement range of the HHD.

Table 8. Equation for 3 Independent Variables and Muscle Strength (Nm) and With Coefficient of Correlation

	Muscle Group	Regression Equation	R2
	ln(shoulder abductors)	=−1.00+0.90×ln(age)+0.55×ln(weight)−0.09×(age×sex)	.797
	ln(elbow extensors 1)	=−0.92+0.45×ln(age)+0.78×ln(weight)+0.12×(age×sex)	.774
	ln(elbow extensors 2)	=−0.95+0.51×ln(age)+0.70×ln(weight)+0.11×(age×sex)	.823
	ln(elbow flexors)	=−1.03+0.76×ln(age)+0.64×ln(weight)+0.05×(age×sex)	.865
	ln(wrist dorsiflexors)	=−3.68+0.64×ln(age)+1.02×ln(weight)+0.39×(age×sex)	.709
	ln(hip extensors 1)	=−0.83+1.05×ln(age)+0.69×ln(weight)+0.04×(age×sex)	.812
	ln(hip extensors 2)	=−1.72+1.39×ln(age)+0.71×ln(weight)+0.02×(age×sex)	.914
	ln(hip flexors 1)	=−1.41+1.08×ln(age)+0.79×ln(weight)+0.04×(age×sex)	.898
	ln(hip flexors 2)	=−0.74+0.80×ln(age)+0.73×ln(weight)+0.04×(age×sex)	.866
	ln(hip abductors)	=−1.68+0.86×ln(age)+1.01×ln(weight)+0.03×(age×sex)	.920
	ln(hip adductors)	=−1.56+1.09×ln(age)+0.80×ln(weight)+0.07×(age×sex)	.904
	ln(knee extensors)	=−0.93+0.55×ln(age)+1.01×ln(weight)+0.03×(age×sex)	.931
	ln(knee flexors 1)	=−1.34+0.72×ln(age)+0.95×ln(weight)+0.03×(age×sex)	.922
	ln(knee flexors 2)	=−1.25+0.77×ln(age)+0.82×ln(weight)+0.05×(age×sex)	.888
	ln(ankle dorsiflexors)	=−1.15+0.82×ln(age)+0.62×ln(weight)+0.01×(age×sex)	.823

NOTE. Sex is included when age is ≥13 (boy = 1, girl = 0).

Comparison between different positions 

Four muscle groups were tested in 2 different positions (elbow extensors, hip extensors and flexors, knee flexors; see table 1). The correlations were strong between the 2 positions in each muscle group (r range, .90−.93). In hip flexors, knee flexors, and elbow extensors, there was a statistically significant difference between the 2 positions (P<.001) (figs 3A−D).

View Large Image Download to PowerPoint

Fig 3. Comparison between measurements of a muscle group in 2 different positions: (A) elbow extensors, (B) knee flexors, (C) hip extensors, and (D) hip flexors.

Interrater reliability 

Three muscle groups (hip flexion, knee extension, knee flexion) were tested by 2 examiners in 25 children 6.7 to 13.7 years of age (mean, 11.0±2.2y). There was no statistically significant difference between measurements by the 2 testers and with ICCs (range, .93−.97), which is graded as an excellent correlation. Data are plotted in figures 4A through C.

View Large Image Download to PowerPoint

Fig 4. Comparison between measurements by 2 different testers: (A) hip flexors, (B) knee extensors, and (C) knee flexors.

Discussion 

This study presents reference values for maximal isometric torque in normal children 5 to 15 years old, obtained with an HHD. The 2 main objectives were to obtain normative data for torque and to have weight-related reference values. Childhood is an ongoing process of change in many respects. When working with forces and biomechanics, body proportions and mass are of great importance. Almost all musculoskeletal movements in the human body pivot around a joint and exert a torque, and should thus be measured as a torque. Standardization of placement of resistance to body parts is not sufficient for comparison, because this place gives a longer lever arm when the child has grown. The same goes for comparisons between 2 children if they are not of the same height.8, 19 To calculate torque gives a possibility for better comparison over time and between subjects. The method with applying the HHD at a comfortable place distally on the segment and then measure lever arm is also useful when factors such as pain can prevent a maximal contraction.

In our material, there was a strong correlation between torque and both age and weight up to the age of 12, and there were few differences between girls and boys until that same age. This is consistent with findings of other authors.20, 25 There was a tendency from the age of 13 for the boys to be stronger in some of the muscle groups measured, especially in the arm.

A child with a disability may sometimes be small for age, so comparing with age-related reference values may lead to misinterpretation of muscle weakness. In these cases, we recommend the use of equations for predicted torque that take into account age, weight, and sex, and if this is not possible, to use the weight-related reference values.

It is difficult to find 1 position for measuring that is useful for all patients. The ideal position for measuring, from a mechanical viewpoint, is with the measured segment in a position not affected by gravity.8, 15 However, it is very difficult to find stable gravity neutral positions for hip extensors and hip flexors to use for patients that are strong. When measuring small children and children who cannot fully cooperate, it is easier to measure when the child can see what they are doing.6 There may also be problems such as in patients with contractures in the hips who are not able to lie prone. The conclusion from this is that it is useful to have different positions to choose between for measuring, that can be applied for all patients.

In our study, we chose to measure 4 muscle groups in 2 different positions.

Hip extension was tested both prone with the opposite hip extended, and prone with the opposite hip in flexion and with the opportunity to stabilize on the floor. In our study, there was a strong correlation between the positions; some of the children were stronger in 1 position and some in the other. In the first position the child has to be able to extend the hip over 0°.

Hip flexion was tested in sitting and in supine, with the hip in 90° of flexion. We obtained higher values sitting although the subjects were also lifting the extremity against gravity. This may be because it is easier to stabilize in a sitting position for children with normal muscle strength. Elbow extension was tested in prone with the shoulder abducted and in supine with the shoulder adducted. The values were higher in prone where it may be easier to stabilize and also to use other muscle groups to augment the results. Knee flexion was tested in sitting with the hip and knee in 90° of flexion and in prone with the hip extended and knee in 90° of flexion. The higher values were obtained in sitting which is expected with the muscle in a better length and tension position.19 A few children tended to get a cramp in the muscle when tested in prone. We recommend using the positions that are easiest for the patient.

Plantarflexion is vital for locomotion such as walking, running, and jumping. According to Gage,26 the plantarflexors are estimated to account for about 50% of the propulsion in normal walking—forces needed to move the body forward. The ankle plantarflexors are strong, because they have to move the whole body weight and they have a short lever (the foot) to work with. It is difficult to measure this accurately because of the high force and also because of difficulties in stabilizing both the subject and the HHD. The problem with measuring isometric plantarflexor strength manually is addressed by Broström et al27 who measured it using a microprocessor-controlled torque motor and found that isometric values generated with the torque motor were higher than values measured with an HHD, both in healthy children and in children with juvenile idiopathic arthritis. Data on plantarflexors are often not included in published reference materials, which also may indicate that it is difficult to measure. We tried to measure the plantarflexors sitting with a specially constructed belt. This worked well with the younger children but there were stabilizing problems which may account for the variability in results. There was also another problem—our device could measure forces up to 560N, which were sufficient for children up to 8 or 9 years, but older children were too strong for the device. We still find that it is important to measure plantarflexion in patients. Our data are interesting as guidelines for comparison in spite of the fact that they only cover the earlier ages. If it is possible to measure plantarflexion manually, the child is probably weak.

One crucial requirement when testing torque with an HHD is that the tester has to be stronger than the patient; otherwise one is not measuring the maximum force of the patient but of the examiner. The reliability testing was done on muscle groups in the lower extremities because there are higher forces in the legs than in the arms. Previous studies have shown that reliability is higher in the upper extremities.16 Interrater reliability was tested between 2 female testers and was found to be excellent. We interpret this as meaning that our protocol is standardized enough to give the same result with different testers. In our study, the oldest and strongest children could only be tested by the male tester because the female testers were not strong enough to stabilize the HHD. Difficulty in stabilizing manually is also suggested as the cause of lower values when testing hip musculature with an HHD than with a dynamometer anchoring station in a study that tested healthy participants ages 23 to 44 years.28 These problems when testing with an HHD can many times explain the differences in reliability between upper and lower extremities16 and between affected and nonaffected side on hemiparetic adults.17 The limitation with measuring strong muscle groups is, however, seldom a problem when testing children with muscle weakness, so we find the use of an HHD very useful in this patient population.

Conclusions 

Reference values for torque calculation obtained in this study correlated significantly with age and weight in children 5 to 15 years of age. There were few differences between boys and girls; boys were stronger in some muscle groups after the age of 13. There was a good agreement between examiners.

The study establishes reference values for torque to use in a clinical context. The combination of using torque instead of force and a predicted value based on the child’s age, weight, and sex makes it possible to compare over time and between subjects and provides a tool for evaluation of physical status and efficacy of therapy.