Isolated gastrocnemius tightness has been associated with chronic foot symptoms and conditions such as plantar fasciitis, calf pain, metatarsalgia, Achilles tendinopathy, and pes planus1-6. Studies regarding gastrocnemius tightness in healthy children are limited, but there is evidence that ankle dorsiflexion decreases with increasing age in children7-9. The normal range of ankle dorsiflexion with a flexed and extended knee has not been sufficiently mapped. We do not know whether gastrocnemius tightness is a normal finding in children. More knowledge of gastrocnemius tightness is of interest as recurrent leg pain is not uncommon in children10.
Furthermore, we know that flexible flatfoot is a normal observation in developing children and that the medial longitudinal arch develops during the first decade of life11,12. We also know that the prevalence of high-arch feet is greater in older children9. Both foot morphology and ankle dorsiflexion change in developing children; however, we do not know whether there is an association between the two.
The purpose of the present study was to investigate and describe passive ankle dorsiflexion with a flexed and extended knee, thereby clinically measuring the tension of the gastrocnemius, in healthy schoolchildren of different ages. We also wanted to investigate whether gastrocnemius tightness was a normal finding in this group of children. The association between ankle dorsiflexion and footprints was also investigated.
Materials and Methods
This study was approved by the regional committee for medical and health research ethics (2018/951 Rec Central).
Three Norwegian public schools were contacted for the purpose of recruiting children for participation in a cross-sectional study. School classes from 4 different age groups were invited to participate: 5 to 6 years (group I), 8 to 9 years (group II), 11 to 12 years (group III), and 14 to 15 years of age (group IV). These groups represented the spread from the youngest to the oldest children attending mandatory school. Written parental consent was obtained from all study participants.
The children were examined during school hours, with 3 to 5 children entering an examination area at a time. Child age, weight, and height were recorded. Passive ankle dorsiflexion was examined with the child supine on a padded table. Time and care were taken to ensure that the child was relaxed and comfortable.
Two simple screening questions were asked to get an overview of the child’s activity level and function. The children in the oldest age group answered directly, whereas the parents answered for those in the younger age groups. One question pertained to pain (“Have you had pain in 1 or both legs [knee, leg, or foot] in the last 3 months that stopped you from participating in sports or in physical or recreational activity?”) and 1 question pertained to function (“Are you able to walk 1 kilometer?”).
Gastrocnemius tightness was assessed with use of the Silfverskiöld test by measuring passive ankle dorsiflexion with the knee flexed and extended13. The Silfverskiöld test was considered to be positive when ankle dorsiflexion was reduced by ≥10° from the flexed knee position to the extended knee position, with an equinus contracture. The thresholds for equinus contracture were set at either ≤5° or ≤0° of ankle dorsiflexion when measured with the knee extended1,4,14. The Silfverskiöld test was performed by an experienced orthopaedic surgeon as a 2-handed examination with avoidance of midfoot abduction. Simultaneously, another experienced orthopaedic surgeon measured the ankle dorsiflexion with use of a digital goniometer (Medigauge). The goniometer arms were laid parallel to the long axes of the leg and foot. The goniometer provided a digital output of the angle measurement, thereby reducing operator errors.
Ankle measurements were performed sequentially for both feet, first with the knee extended, then with the knee flexed to 90°, and finally with the knee extended again. To mimic a clinical Silfverskiöld test, the second extension measurements were used. The data from the first extension measurements were used to determine repeatability.
An ink pad (PedoPrint; Bauerfeind) was used to acquire footprints with the child in the bipedal position with bilateral weight-bearing and watching the horizon. The prints were used to calculate the Chippaux-Smirak Index (CSI)15. This index is based on the maximum width of the metatarsals relative to the parallel minimum width of the foot in the area of the arch (Fig. 1). The CSI decreases as the arch develops with age and has been deemed valid16-18 and reliable19 in the pediatric population. Flatfoot is defined as a CSI of >62.7% in children 5 to 6 years of age17,18,20, a CSI of >59% in children 8 to 9 years of age19,20, and a CSI of ≥40% in older children16,20. High-arch morphology was set as CSI = 0%15.
All statistical calculations were performed with use of SPSS (version 27; IBM). Visual inspection of histograms showed the scale parameters to be normally distributed. Generalized mixed linear models were used to account for data dependency caused by the 2 feet nested within each child. Separate analyses were performed for measurements obtained with the knee flexed and extended. Fixed factors included side, sex, and age. Intraclass correlation coefficients (ICCs) were calculated. Age (in months) was first modeled as a continuous variable (termed the “age coefficient”) to describe a possible overall effect of age on dorsiflexion. Side and sex were modeled as categories. Subsequently, the age variable was modeled as a categorical variable (4 age groups) to present model estimates of group means and differences, with corresponding 95% confidence intervals (CIs). The model’s residuals were normally distributed as evaluated on histograms. The level of significance was set at p ≤ 0.05. Repeatability was calculated according to the method of Bland and Altman28. The repeatability coefficient was defined as twice the standard deviation of the differences between the 2 sequential ankle dorsiflexion results measured with knee extended. For each child, a measurement from 1 randomly selected leg was used for analysis.
Source of Funding
The research was partially funded by Trøndelag Orthopaedic Workshop. They played no role in the study design, data collection, data interpretation, or writing and submitting the article for publication.
Two hundred and four children (92.7%) participated in this study, and 16 (7.3%) declined to participate. Demographic data are presented in Table I.
TABLE I -
Demographic Data (N = 204)
||No. of Children
*Group I = 5-6 years of age; group II = 8-9 years of age; group III = 11-12 years of age; and group IV = 14-15 years of age.
†The values are given as the mean and the standard deviation.
‡The values are given as the number of participants unable to perform sports or physical or recreational activity in the last 3 months because of pain in 1 or both legs, with the percentage in parentheses.
The raw ankle dorsiflexion data are presented in Figure 2. Four measurements were obtained for each child: left and right sides with the knee flexed and extended.
The ICC for ankle dorsiflexion was 0.860 with the knee flexed and 0.920 with the knee extended. These high numbers describe small variations between the left and right sides within children and larger variations between children.
With the knee flexed, the mean ankle dorsiflexion was 22.9° on the right and 23.4° on the left, with a difference of −0.5° (95% CI, −0.8° to −0.1°; p = 0.018). The mean ankle dorsiflexion was 22.4° for girls and 23.9° for boys, with a difference of −1.5° (95% CI, −3.1° to 0.1°; p = 0.059). The age coefficient was −0.09°/month (95% CI, −0.11°/month to −0.07°/month; p < 0.001).
With the knee extended, the mean ankle dorsiflexion was 4.9° on the right and 5.0° on the left, with a difference of −0.1° (95% CI, −0.4° to 0.1°; p = 0.380). The mean ankle dorsiflexion was 4.6° for girls and 5.2° for boys, with a difference of −0.6° (95% CI, −1.4° to 0.1°; p = 0.106). The age coefficient was −0.04°/month (95% CI, −0.05°/month to −0.03°/month; p < 0.001).
The mean values of ankle dorsiflexion with the knee flexed and extended, stratified by age groups, and differences between age groups, are presented in Table II. Ankle dorsiflexion decreased with increasing age.
TABLE II -
Ankle Dorsiflexion of All Feet (N = 408)*
||Group I, 5-6 Years
||Group II, 8-9 Years
||Group III, 11-12 Years
||Group IV, 14-15 Years
||Difference, Groups I-II
||Difference, Groups II-III
||Difference, Groups III-IV
|Ankle dorsiflexion with knee flexed
|Ankle dorsiflexion with knee extended
*The values for ankle dorsiflexion measured with a flexed or extended knee are given as the mean, with the 95% CI in parentheses.
When equinus contracture was defined as ankle dorsiflexion of ≤5° when measured with the knee extended (dashed green line, Fig. 2), 238 feet (58.3%) were contracted. Of those, 224 feet (representing 54.9% of the total number of feet in the study) had ≥10° of reduction in dorsiflexion when the measurement in knee flexion was compared with the measurement in knee extension, indicating gastrocnemius tightness.
When equinus contracture was defined as dorsiflexion of ≤0° when measured with the knee extended (solid green line, Fig. 2), 15 feet (3.7%) were contracted. All had ≥10° of reduction in dorsiflexion when the measurement in knee flexion was compared with the measurement in knee extension. None were in the youngest age group. No child had ≤5° of dorsiflexion with the knee flexed.
The repeatability coefficient of the ankle dorsiflexion measurements was 2.8°.
Four hundred and eight feet were examined with footprint CSI analysis (Table III). Of those, 340 (83.3%) had a normal footprint, 53 (13.0%) had a cavus footprint, and 15 (3.7%) had a flat footprint.
TABLE III -
Foot Morphology of All Feet (N = 408)*
||Normal Footprint (N = 340)
||Cavus Footprint (N = 53)
||Flat Footprint (N = 15)
*The values are given as the number of feet in each group, with the percentage of each footprint type in the group in parentheses.
The relationships between CSI and ankle dorsiflexion with the knee flexed and extended are presented in Figures 3 and 4. On visual inspection, there is no clear association between footprints and ankle dorsiflexion, apart from a tendency for the youngest age group to have higher CSI and greater dorsiflexion.
Of the 15 feet with gastrocnemius tightness defined as dorsiflexion of ≤0°, 12 (80%) had a normal footprint.
Pain and Function
Twenty-seven children (13.2%) responded that they had been unable to participate in physical activity at some point during the last 3 months because of pain from the lower extremities (Table I). Only 1 child from this cohort had a tight gastrocnemius defined as ≤0° of ankle dorsiflexion with the knee extended. One boy responded that he currently was unable to walk 1 km; this child had normal gastrocnemius tension.
In this population of schoolchildren, ankle dorsiflexion decreased with increasing age and gastrocnemius tightness was a common finding.
The mean values for dorsiflexion in the different age groups are presented in Table II and are comparable with previous findings7,8. To our knowledge, no previous study has specifically assessed gastrocnemius tension in children; rather, previous studies have examined healthy children to determine normal range of motion in several joints7,8. Some variation in the normative data of ankle dorsiflexion has been reported, but studies have differed in terms of measurement techniques and participant positioning. In some studies, ankle dorsiflexion has been measured with the patient in the prone position, which results in higher measurements21,22. This positioning is not comparable with the typical clinical method used when the Silfverskiöld test is performed.
None of the children had a soleus contracture when defined as ≤0° of dorsiflexion of the ankle with the knee flexed, and only a small proportion (3.7%) had ≤0° of dorsiflexion with the knee extended. A comparable Danish study9 showed similar changes in ankle dorsiflexion with increasing age. The authors of that study found that a majority of older children had dorsiflexion of <0° with the knee extended; however, they did not examine dorsiflexion with the knee flexed for comparison.
If we were to define gastrocnemius tightness with a threshold of ≤5° of ankle dorsiflexion with the knee extended as described by DiGiovanni et al.14, 224 feet (54.9%) in the present study would be included. This value represents a majority of the study population and indicates that gastrocnemius tightness is a common finding in normal children. Consequently, there is a possibility that such children would be considered to have a pathologically tight gastrocnemius, with an associated risk of overtreatment. Some studies have set a stricter threshold for defining a positive Silfverskiöld test (≤0° of ankle dorsiflexion with the knee extended)1,4. In the present study, only 15 feet (3.7%) had dorsiflexion of ≤0° with the knee extended. While the difference between 5° and 0° is subtle, a threshold set at ≤0° would be less likely to include healthy children, and we believe that it would be clinically more useful.
Figure 2 shows the degree of ankle dorsiflexion with the knee flexed and extended in this population of schoolchildren. We found decreased dorsiflexion with increasing age, most prominently between groups I and II. The age coefficient showed a significant reduction in ankle dorsiflexion: for every month, dorsiflexion was reduced by 0.09° and 0.04° with the knee flexed and extended, respectively. This finding indicates that, in schoolchildren over a 10-year period, one can expect a reduction in ankle dorsiflexion of approximately 11° with the knee flexed and 5° with the knee extended. The reduction of dorsiflexion with the knee flexed is greater, indicating that the soleus contributes more to the change with age than the gastrocnemius does.
Footprint analysis was performed with use of the CSI, which is used to assess flatfoot. As the child’s arch develops with age, the ratio decreases accordingly20. Although footprints have long been used for assessing foot morphology15, footprint analysis provides a 2-dimensional representation of the contact area of the 3-dimensional foot and does not necessarily determine the presence of pathology. Despite this limitation, footprints are simple, fast, and inexpensive to obtain, and there is evidence that footprint analysis is as effective as radiographic measurements for determining flatfoot morphology23. We therefore used footprints as a screening method for foot morphology in this population and assessed the findings with use of the CSI, which is validated for children16-18. Of the 408 footprints, 83.3% were found to be normal, 13.0% demonstrated a cavus foot, and 3.7% demonstrated flatfoot. We found no association between footprint analysis and ankle dorsiflexion measurements (Figs. 3 and 4). The youngest age group had a greater dorsiflexion and higher CSI. These findings are consistent with those of previous studies as greater dorsiflexion and a higher CSI are normal in this age group9,16-18. The 53 cavus feet (CSI = 0%) had a spread of dorsiflexion measurements, with no clear association between cavus feet and dorsiflexion. When we specifically looked at the subgroup of 15 feet (3.7%) with gastrocnemius tightness defined as ≤0° of dorsiflexion, 80% had normal footprints. When screening for foot morphology with footprints, we found no association with gastrocnemius tightness or with ankle dorsiflexion in general.
Whole school classes were recruited from Norwegian public schools. Almost all children in Norway attend public school. There are few socioeconomic differences in the society, and schools do not tend to cluster children of a particular ethnicity or socioeconomic class. Our screening questions regarding foot pain and function revealed that 27 children (13.2%) had been unable to participate in sports or physical or recreational activity because of leg or foot pain at some point during the last 3 months. These children were spread across the age groups, and only 1 had a tight gastrocnemius defined as ≤0° of ankle dorsiflexion. The reasons for pain could be multiple, and further subgroup analyses were not intended. Although we did not screen for underlying medical conditions, all children were ambulatory and were examined by the same clinicians. We believe that this non-selected group of schoolchildren reflects the background population.
We are aware of criticism of the Silfverskiöld test as a poorly reliable clinical test. When assessing ankle dorsiflexion, one must address the subtleties of midfoot motion. Correct examination requires locking of the talonavicular joint with an adduction force on the lateral forefoot and reducing the midfoot on the hindfoot14,24. The anterior muscles must also be relaxed; otherwise, they will provide a false increase in dorsiflexion by force of their contraction24. Novel devices have been designed to exactly measure ankle dorsiflexion and gastrocnemius tension25,26; however, to our knowledge, none of those devices have been validated for the evaluation of children. Applying the same predetermined dorsiflexion force is not comparable in younger and older children. As we examined children of varying weight and size, the most accurate and reproducible method to test the gastrocnemius was to let 2 experienced clinicians perform all measurement in a similar manner for all children. It was the clinician’s responsibility to make sure that there was no active plantar flexion or dorsiflexion applied by the child as this is something that one both feels and observes. While controlling the hindfoot and avoiding talonavicular abduction, dorsiflexion force was applied to stretch the gastrocnemius passively to an end point with the knee fully extended, but not past what the child would find uncomfortable, thereby making them unable to relax.
Ankle range of motion was measured by the other clinician with use of goniometry. Previous authors have found goniometry to be a reliable and reproducible method for assessing ankle dorsiflexion, with a mean intraobserver measurement error of 4°, when a strict but simple protocol (similar to our methodology) was applied27. We calculated the repeatability of our measurements as we performed 2 measurements of ankle dorsiflexion with the knee extended. Our calculations showed a repeatability coefficient of 2.8°. It is unlikely that one would be able to distinguish measurements below 2.8° clinically. Furthermore, these measurement variations are at the individual level and become smaller if group means are calculated.
In the present study, we demonstrated the normal range of passive ankle dorsiflexion with the knee flexed and extended in a population of typical schoolchildren grouped by age. There was a clear association between the degree of dorsiflexion and age, highlighting the importance of using age-matched norms. A majority of patients had a positive Silfverskiöld test when the threshold for dorsiflexion was set at ≤5°, indicating that gastrocnemius tightness is common in this population and should not be interpreted as pathological on its own. When the threshold was set at a stricter value of ≤0°, only 3.7% of the population had a positive Silfverskiöld test, indicating that this threshold is more likely to signify pathology in the clinical setting. We found no association between dorsiflexion measurements and footprints.
Note: The authors thank Elin Maria Therès Ljungström for her important contribution in data collection.
1. Molund M, Paulsrud Ø, Ellingsen Husebye E, Nilsen F, Hvaal K. Results after gastrocnemius recession in 73 patients. Foot Ankle Surg. 2014 Dec;20(4):272-5. Epub 2014 Jul 24.
2. Molund M, Lapinskas SR, Nilsen FA, Hvaal KH. Clinical and Functional Outcomes of Gastrocnemius Recession for Chronic Achilles Tendinopathy. Foot Ankle Int. 2016 Oct;37(10):1091-7. Epub 2016 Sep 12.
3. Gurdezi S, Kohls-Gatzoulis J, Solan MC. Results of proximal medial gastrocnemius release for Achilles tendinopathy. Foot Ankle Int. 2013 Oct;34(10):1364-9. Epub 2013 May 17.
4. Abbassian A, Kohls-Gatzoulis J, Solan MC. Proximal medial gastrocnemius release in the treatment of recalcitrant plantar fasciitis. Foot Ankle Int. 2012 Jan;33(1):14-9.
5. Holtmann JA, Südkamp NP, Schmal H, Mehlhorn AT. Gastrocnemius Recession Leads to Increased Ankle Motion and Improved Patient Satisfaction After 2 Years of Follow-Up. J Foot Ankle Surg. 2017 May - Jun;56(3):589-93.
6. Nakale NT, Strydom A, Saragas NP, Ferrao PNF. Association Between Plantar Fasciitis and Isolated Gastrocnemius Tightness. Foot Ankle Int. 2018 Mar;39(3):271-7. Epub 2017 Dec 4.
7. Lee SY, Lee SH, Chung CY, Park MS, Lee KM, Akhmedov B, Choi IH, Cho TJ, Yoo WJ, Sung KH. Age-related changes in physical examination and gait parameters in normally developing children and adolescents. J Pediatr Orthop B. 2013 Mar;22(2):153-7.
8. Soucie JM, Wang C, Forsyth A, Funk S, Denny M, Roach KE, Boone D; Hemophilia Treatment Center Network. Range of motion measurements: reference values and a database for comparison studies. Haemophilia. 2011 May;17(3):500-7. Epub 2010 Nov 11.
9. Reimers J, Pedersen B, Brodersen A. Foot deformity and the length of the triceps surae in Danish children between 3 and 17 years old. J Pediatr Orthop B. 1995;4(1):71-3.
10. Evans AM, Scutter SD. Prevalence of “growing pains” in young children. J Pediatr. 2004 Aug;145(2):255-8.
11. Chen KC, Tung LC, Yeh CJ, Yang JF, Kuo JF, Wang CH. Change in flatfoot of preschool-aged children: a 1-year follow-up study. Eur J Pediatr. 2013 Feb;172(2):255-60. Epub 2012 Nov 7.
12. Evans AM, Rome K. A Cochrane review of the evidence for non-surgical interventions for flexible pediatric flat feet. Eur J Phys Rehabil Med. 2011 Mar;47(1):69-89.
13. Silfverskiöld N. Reduction of the uncrossed two-joints muscles of the leg to one-joint muscles in spastic conditions. Acta Chir Scand. 1924;56:315-30.
14. DiGiovanni CW, Kuo R, Tejwani N, Price R, Hansen ST Jr, Cziernecki J, Sangeorzan BJ. Isolated gastrocnemius tightness. J Bone Joint Surg Am. 2002 Jun;84(6):962-70.
15. Forriol F, Pascual J. Footprint analysis between three and seventeen years of age. Foot Ankle. 1990 Oct;11(2):101-4.
16. Villarroya MA, Esquivel JM, Tomás C, Moreno LA, Buenafé A, Bueno G. Assessment of the medial longitudinal arch in children and adolescents with obesity: footprints and radiographic study. Eur J Pediatr. 2009 May;168(5):559-67. Epub 2008 Aug 27.
17. Chen KC, Yeh CJ, Kuo JF, Hsieh CL, Yang SF, Wang CH. Footprint analysis of flatfoot in preschool-aged children. Eur J Pediatr. 2011 May;170(5):611-7. Epub 2010 Oct 23.
18. Chen KC, Tung LC, Tung CH, Yeh CJ, Yang JF, Wang CH. An investigation of the factors affecting flatfoot in children with delayed motor development. Res Dev Disabil. 2014 Mar;35(3):639-45. Epub 2014 Jan 18.
19. Chang CH, Chen YC, Yang WT, Ho PC, Hwang AW, Chen CH, Chang JH, Chang LW. Flatfoot diagnosis by a unique bimodal distribution of footprint index in children. PLoS One. 2014 Dec 31;9(12):e115808.
20. Banwell HA, Paris ME, Mackintosh S, Williams CM. Paediatric flexible flat foot: how are we measuring it and are we getting it right? A systematic review. J Foot Ankle Res. 2018 May 30;11:21.
21. Mudge AJ, Bau KV, Purcell LN, Wu JC, Axt MW, Selber P, Burns J. Normative reference values for lower limb joint range, bone torsion, and alignment in children aged 4-16 years. J Pediatr Orthop B. 2014 Jan;23(1):15-25.
22. Alanen JT, Levola JV, Helenius HY, Kvist MH. Ankle joint complex mobility of children 7 to 14 years old. J Pediatr Orthop. 2001 Nov-Dec;21(6):731-7.
23. Kanatli U, Yetkin H, Cila E. Footprint and radiographic analysis of the feet. J Pediatr Orthop. 2001 Mar-Apr;21(2):225-8.
24. Pinney SJ, Hansen ST Jr, Sangeorzan BJ. The effect on ankle dorsiflexion of gastrocnemius recession. Foot Ankle Int. 2002 Jan;23(1):26-9.
25. Jastifer JR, Marston J. Gastrocnemius Contracture in Patients With and Without Foot Pathology. Foot Ankle Int. 2016 Nov;37(11):1165-70. Epub 2016 Jul 22.
26. Molund M, Husebye EE, Nilsen F, Hellesnes J, Berdal G, Hvaal KH. Validation of a New Device for Measuring Isolated Gastrocnemius Contracture and Evaluation of the Reliability of the Silfverskiöld Test. Foot Ankle Int. 2018 Aug;39(8):960-5. Epub 2018 Apr 20.
27. Allington NJ, Leroy N, Doneux C. Ankle joint range of motion measurements in spastic cerebral palsy children: intraobserver and interobserver reliability and reproducibility of goniometry and visual estimation. J Pediatr Orthop B. 2002 Jul;11(3):236-9.
28. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986 Feb 8;1(8476):307-10.