Plantar Pressures in Children With and Without Sever’s Disease

Abstract

Background: A case-control study was conducted to compare static plantar pressures and distribution of body weight across the two lower limbs, as well as the prevalence of gastrocnemius soleus equinus, in children with and without calcaneal apophysitis (Sever’s disease). Methods: The participants were 54 boys enrolled in a soccer academy, of which eight were lost to follow-up. Twenty-two boys with unilateral Sever’s disease comprised the Sever’s disease group and 24 healthy boys constituted a control group. Plantar pressure data were collected using pedobarography, and gastrocnemius soleus equinus was assessed. Results: Peak pressure and percentage of body weight supported were significantly higher in the symptomatic feet of the Sever’s disease group than in the asymptomatic feet of the Sever’s disease group and the control group. Every child in the Sever’s disease group had bilateral gastrocnemius equinus, while nearly all children in the control group had no equinus. Conclusions: High plantar foot pressures are associated with Sever’s disease, although it is unclear whether they are a predisposing factor or a result of the condition. Gastrocnemius equinus may be a predisposing factor for Sever’s disease. Further research is needed to identify other factors involved in the disease and to better understand the factors that contribute to abnormal distribution of body weight in the lower limbs.

Apophysitis, or inflammation of an apophysis, is caused by microavulsions at the bone-cartilage junction[1] resulting from repetitive motion and overuse during periods of rapid growth. Calcaneal apophysitis was first described in 1912 by Sever[2] and later given the name “Sever’s disease.” The calcaneal apophysis is a cartilaginous growth center onto which the Achilles tendon inserts.[3,4] Kvist and Heinonen[5] and Kim et al[6] refined the definition of Sever’s disease to indicate a traction epiphysitis. It is most commonly associated with sports that involve running and jumping.[7–9] In a retrospective study of 20 children with Sever’s disease, McKenzie et al[10] found that all of the children participated in such sports, mainly in track and field events and soccer.

Sever’s disease has been reported to be the most common cause of heel pain in athletic children.[8,11] Orava and Puranen[12] and Orava and Virtanen[13] found that it comprises 16.3% and 22.7% of exertion injuries in children, respectively. The heel pain limits physical activity and may interfere with activities of daily living. The condition is self-limiting because the calcaneal cartilage disappears, at about age 14 years in girls and 16 years in boys, to allow for complete calcaneal ossification.[5] The etiology of Sever’s disease is controversial,[14] and proposed contributing factors include participation in high-impact sports, improper footwear, running on hard surfaces, and excessive pressures on the plantar heel.[15] The condition is associated with radiographic changes such as variations in the normal ossification pattern.[8] Cessation of the high-impact activity and thus reducing mechanical overload at the affected heel is a standard component of treatment.[16–18]

We sought to investigate the relationship between mechanical heel overload and development of Sever’s disease by measuring static plantar pressures of young athletes with and without Sever’s disease. We also sought to determine the distribution of body weight between the two limbs and to investigate its association with Sever’s disease. The association between gastrocnemius soleus equinus and Sever’s disease was also evaluated because of the relationship found between gastrocnemius soleus equinus and Achilles tendon thickening.[19] The calcaneal apophysis is a cartilaginous growth center onto which the Achilles tendon inserts.[3,4,20] Haglund[21] titled his article “Concerning Some Rare but Important Surgical Injuries Brought on by Violent Exercise,” but he provided no justification of how violent the exercise was. Many authors[2,5,10,12,18,22–27] have noted anecdotally that affected children were active and vigorous and that symptoms often presented at the beginning of the sporting season and while the child was undergoing a growth spurt. However, none of these studies measured or reported the children’s actual activity levels or provided evidence of growth spurts.

The plantar fascia originates from the medial tubercle on the plantar aspect of the calcaneus, near the attachment of the Achilles tendon to the calcaneus, a secondary bony growth center, or the epiphysis.[7,28–30] Proximal to the epiphysis is the apophysis, where the Achilles tendon actually inserts. The apophysis has its own slower-growing growth plate, separate from the physeal plate.[3,4] The calcaneal growth plate and the apophysis are situated in an area subject to high stress from the plantar fascia and Achilles tendon[31] and may be affected by increased tension on the calcaneus.

The main theory in the literature on the pathogenesis of Sever’s disease is that it is an overuse syndrome from repetitive microtrauma from increased traction on the apophysis. This increased traction is believed to be initiated by running and jumping, which causes avulsion fractures on a tiny scale followed by inflammation.[8,20,32–36] Because of the rapid proliferation of cells in growth plates, the apophysis is thought to be more susceptible to injury.[37] Liberson et al[31] examined calcaneal apophyses histologically and with computed tomography and found fibrous bands in the cartilage perpendicular to os calcis. This finding indicated powerful stresses in the remodeling process, leading Liberson et al[31] to suggest a traction-stress argument, where pain is thought to occur when remodeling exceeds certain rates.

The theory that tight triceps surae cause excessive tension through the Achilles tendon, increasing the traction on the apophysis, has been a commonly cited mechanical factor in much of the literature.[1,7,8,17–19,32,37–44] Studies[8,18,19,32] that looked at patients with Sever’s disease generally involved assessment by multiple raters of foot dorsiflexion on the leg, thereby reducing uniformity of measurement and reliability of results. This omission calls into question the existence of excessive tightness in the triceps surae at all. No studies have compared whether symptomatic patients are tighter in the triceps surae than their asymptomatic counterparts.[45]

References to biomechanical influences in Sever’s disease are scattered throughout the literature. However, neither prospective systematic measurement of the feet of those identified with Sever’s disease nor comparison to an asymptomatic population has been performed.[45] Krantz[32] observed limited ankle dorsiflexion and postulated traumatic inflammation to the apophysis. McKenzie et al[10] noted, retrospectively, that 95% of their Sever’s disease patients had a biomechanical imbalance that produced a whipping action in the Achilles tendon, increasing the stress on the apophysis. However, this theory has no supportive evidence. Szames et al[19] found that of 79 children with Sever’s disease, 65 (82.3%) tested positive for ankle equinus owing to Achilles tendon tightness and concluded that equinus may predispose a child to Sever’s disease.

The exact mechanism by which equinus may influence Sever’s disease is unknown.[46] In a retrospective study, we also found a relationship between equinus and Sever’s disease. This strong relationship between the calcaneal apophysitis and equinus could explain why treating heel lift and limiting strenuous activity seem to relieve symptoms. A child with Sever’s disease usually reports nonradiating pain in the posterior calcaneus with weightbearing activities, which is relieved by rest. The pain is usually accompanied by tight triceps surae, resulting in a reduction of dorsiflexion to 10° or less.[2,8,11,14,17–19] One possible cause of the pain is the tension on the Achilles tendon due to shortening of the gastrocnemius soleus complex and production of excessive traction force on the apophysis.[47]

Patients and Methods

Participants

Participants in this study were boys enrolled in a soccer school who presented for a final health screening of the season. The study protocol conformed to the guidelines set forth in the Declaration of Helsinki, and written parental consent for the evaluation was obtained. All of the boys spent the same amount of time participating in their sport, including after-school and weekend training and competition. The exclusion criterion was a recent history of ankle injury, including sprain, tendinitis, or any pathology other than Sever’s disease. None of the participants had a neurologic condition affecting the lower extremity.

The soccer school enrolled 215 children aged 8 to 15 years. Fifty-four boys were selected randomly from individuals who presented to the study, of which eight were lost to follow-up (two with and six without Sever’s disease). Thus, 46 boys completed the study. They were divided into two groups: 1) those who had been diagnosed (by a podiatrist [R.B.B.V.] affiliated with the study) with unilateral Sever’s disease between September 2006 and August 2007, and 2) those who were healthy, who served as the control group. The asymptomatic foot of each child in the Sever’s disease group was also used as a control. No child had been diagnosed with bilateral Sever’s disease or had bilateral heel pain. For diagnostic purposes, Sever’s disease was defined as pain on mediolateral compression of the calcaneus in the area of the growth plate.[48] The pain had to be of at least 2 months’ duration and coincide with physical activity, as well as be severe enough to sufficiently interfere with normal physical activity such as walking.[11] All of the children were symptomatic when they presented to the clinic.

The characteristics of the boys in the Sever’s disease and control groups are given in

Table 1. Characteristics of Children in the Sever’s Disease and Control Groups.

. All of the children were in the “healthy weight” category as defined by the Centers for Disease Control and Prevention,[49] from the 5th percentile to the 85th percentile, except for one 11-year-old boy in the Sever’s disease group who was underweight (4th percentile). Body mass index was not considered in the inclusion criteria for either group.

Study Protocol and Data Collection

Plantar pressure data have been shown to be sensitive to the data-collection protocol.[50] Using a pressure platform to collect data can be challenging, particularly with children who often have trouble staying still. We collected data as follows: the child stood on the pressure platform and simulated gait by walking in place to accommodate the feet to the platform at the normal angle. After 15 sec, the child was told to stop moving and to stand still in a natural manner, with the entire foot on the mat and the feet in the normal angle of gait, looking straight ahead with the arms close to the body. The plantar foot pressure measurements for both feet during a 30-sec period were taken simultaneously and recorded by an independent observer. If the child moved, the data were discarded, and the trial was repeated until data were obtained with the child remaining still. Aside from these cases, no “practice trials” were conducted.

Gastrocnemius soleus equinus was assessed with the knee extended and flexed. The amount of ankle dorsiflexion was determined by using a goniometer to measure the angle between the plantar aspect of the heel (medially or laterally) and the tibia. Care was taken to maintain the subtalar joint in a neutral position and to measure ankle dorsiflexion and not midfoot dorsiflexion (rocker bottom) or midfoot equinus (pseudoequinus).[46] The Silfverskiöld test was performed to differentiate gastrocnemius equinus from other types of equinus.[46,51] The normal amount of ankle dorsiflexion is approximately 10° with the knee extended and 20° with the knee flexed.[46,52,53] Gastrocnemius equinus is the inability of the ankle to dorsiflex normally with the knee extended but the ability of the ankle to dorsiflex more than 10° with the knee flexed. Gastrocnemius soleus equinus is the inability of the ankle to dorsiflex beyond a neutral position with the knee extended (it remains <0°) or with the knee flexed (it remains <0°).[54,55]

Ankle joint dorsiflexion range of motion was performed by the same clinician with broad experience in foot and ankle physical examinations. The study by Evans and Scutter[56] assessed the intrarater and interrater reliability of sagittal ankle range of motion in children. The results show that measures of ankle dorsiflexion in children are highly variable among examiners and, in general, that gastrocnemius range of motion is more reliable than is soleal range of motion. The most reliable clinical examination of the pediatric ankle sagittal plane range of motion has been found with knee extension, an accepted proxy for gastrocnemius length. Examination of ankle range of motion with knee flexion (soleus length) has been demonstrated to be highly unreliable.

Plantar pressures were measured by an independent observer using a digital portable force plate (EPS-Platform; Loran Engineering, Castel Maggiore, Bologna, Italy). The platform dimensions were 70 × 50 cm, the thickness was 5 mm, the weight was 7 kg, and the number of resistive sensors was 2,304. Measurements were accurate to the nearest 0.001 kPa. The equipment met the CE Declaration of Conformity and was calibrated a few days before the study began. Vertical force was recorded at a frequency of 60 Hz. The platform was linked via an interface unit to a personal computer containing the data-collection software program Foot Checker, version 4.0 for Windows (Loran Engineering). The software produced pressure maps with pressure measured in kilopascals for each incident of data collection.

The following static measurements were obtained for all children in both groups: peak pressure (kPa) under each foot, percentage of body weight supported by each limb, and plantar surface area (cm²) of each foot in contact with the pedobarograph.

Statistical Analysis

The Kolmogorov-Smirnov test was used to test the normality of the data. The results of these tests indicated that the data were normally distributed and that parametric statistical tests were most appropriate. Independent Student t tests were performed to determine whether there were statistically significant differences in height, weight, and age between the two groups. Paired Student t tests were performed to determine differences between the symptomatic and asymptomatic feet in the Sever’s disease group and between the right and left feet in the control group. Independent Student t tests were performed to determine differences between the symptomatic and asymptomatic feet in the Sever’s disease group and the corresponding feet in the control group. The association between gastrocnemius soleus equinus and Sever’s disease was measured by using the Pearson product moment correlation with χ² analysis. In all of the analyses, P < .05 (with a 95% confidence interval) was considered statistically significant. Data analysis was conducted with SPSS software, version 14.0 (SPSS Science, Chicago, Illinois).

Results

No statistically significant differences were found between the two groups for participant height (P < .138), weight (P < .106), or age (P < .924), but one was found for body mass index (P < .021). The results for the variables measured are shown in

Table 2. Peak Pressure, Percentage of Body Weight Supported by Each Foot, and Plantar Surface Contact Area in the Symptomatic and Asymptomatic Feet in the Sever’s Disease Group and the Corresponding Feet in the Control Group.

. In the Sever’s disease group, the symptomatic feet had significantly higher peak pressure values than the asymptomatic feet (P < .001) and supported a significantly higher percentage of body weight (P < .001). The surface area of the foot in contact with the pedobarograph can affect the peak pressure measured, but no significant differences in surface contact area were found between the symptomatic and asymptomatic feet in the Sever’s disease group (P > .05). In the control group, no significant differences were found between the left and right feet in peak pressure, percentage of body weight supported, or plantar surface area in contact with the pedobarograph (P > .05).

In the comparison between the two groups, a statistically significant difference was found in peak pressure between the symptomatic feet in the Sever’s disease group and the corresponding feet in the control group (P < .001). No difference was found between the asymptomatic feet of the Sever’s disease group and the corresponding feet in the control group (P > .05). A significantly higher percentage of body weight was supported by the symptomatic feet in the Sever’s disease group than by the corresponding feet in the control group (P < .001). A significantly lower percentage of body weight was supported by the asymptomatic feet in the Sever’s disease group than by the corresponding feet in the control group (P < .001). No significant difference was found in the plantar surface area of the foot in contact with the pedobarograph between the symptomatic feet in the Sever’s disease group and the corresponding feet in the control group (P > .05), or between the asymptomatic feet in the Sever’s disease group and the corresponding feet in the control group (P > .05).

A statistically significant difference was found between the two groups in the prevalence of gastrocnemius equinus, with all 22 children in the Sever’s disease group having bilateral gastrocnemius equinus and 21 of 24 children in the control group having no equinus (P < .001).

Discussion

The main findings of this study were the statistically significant differences in static peak pressure and percentage of body weight supported between the symptomatic and asymptomatic feet in the Sever’s disease group. Using the asymptomatic feet as controls eliminated many potential confounding variables such as differences in age, weight, and physical activity. Our results indicate that in a child with Sever’s disease, the symptomatic foot supports a greater percentage of body weight than the asymptomatic foot, leading to higher pressure at the heel of the symptomatic foot. The absence of pain in the asymptomatic heel may be due to the smaller load that it supports. The greater load supported by the symptomatic heel could translate into greater injury to the growth cartilage of the heel. However, the asymptomatic feet in the Sever’s disease group supported a lower percentage of body weight than the corresponding feet in the control group. These results indicate that Sever’s disease is associated with abnormal distribution of body weight between the two limbs. The control group demonstrated a more balanced distribution of body weight, with each foot supporting about half the load. This may avoid overloading one heel, decreasing the risk of heel inflammation associated with Sever’s disease.

Overload in tissues may be associated with an inability to adequately attenuate forces during gait. Lower loading rates are widely regarded as less damaging than higher rates. Jahss et al[57] postulated that if force or repetition of force is not attenuated to below a critical level, tissue destruction can result, with healing responses leading to further structural change and alteration of tissue mechanics. This can increase forces on the heel during gait, leading to repetitive microtrauma to the subcalcaneal tissues.[58] Tax,[59] McCrea,[37] and Alexander[60] suggested that Sever’s disease is probably caused by isolated or repeated trauma, and Madden and Mellion,[7] Peck,[4] and Topham and White[61] noted that anecdotal evidence supports major or minor trauma as a probable etiology. This supports the pathophysiological concept most often presented in the literature of Sever’s disease as an overuse syndrome caused by repetitive microtrauma from increased traction on the apophysis.[8,62–64] In a review by Scharfbillig et al,[45] recommended treatments for Sever’s disease included rest or cessation of sports, as well as padding, which helps reduce pressures in the heel. The authors stated that little information had been reported on how these treatment regimens work; further research is needed to determine how the plantar pressures could be relieved or affected by different forms of treatment.

Although many authors have attempted to specify the necessary degrees of ankle dorsiflexion, normative values have been limited.[53,54] Biomechanically, the maximum amount of dorsiflexion in the stance phase of normal gait occurs just before heel lift with the knee extended.[53] The minimum amount of ankle range of motion necessary for normal gait is 10° of dorsiflexion and 20° of plantarflexion.[53–55] The most widely accepted values in the literature for static measurements, and for the purpose of this study, state that the minimum amount of dorsiflexion necessary at the ankle for normal gait is 10° of motion.[46,62–67]

Equinus imparts a major deforming force to the foot and is a causative factor in many foot and ankle pathologic entities, including plantar fasciitis, pes planus, hallux abducto valgus, Achilles tendinosis, Charcot’s midfoot collapse, and diabetic ulcerations.[55] DiGiovanni et al[54] found either gastrocnemius or gastrocnemius soleus equinus in patients with a symptomatic foot and ankle. In asymptomatic patients, gastrocnemius and gastrocnemius soleus equinus are not uncommon (33% and 17%, respectively).[68]

In this study, we found a 100% incidence of bilateral gastrocnemius equinus in the Sever’s disease group, which probably accounts for the mechanical overload in the symptomatic heel. Szames et al[19] evaluated 79 cases of calcaneal apophysitis in 53 patients. They found that 82.3% of the cases had ankle equinus due to muscular retraction, but they did not distinguish between gastrocnemius and soleus contracture in the equinus conditions, unlike in our study. Gastrocnemius equinus may be a predisposing factor, and increased plantar foot pressures at the heel seem to be associated with unilateral Sever’s disease, but further research is needed to identify other factors involved. A recent review of Sever’s disease by Scharfbillig[45] found that no studies have compared whether symptomatic patients are tighter in the triceps surae than their asymptomatic counterparts.

However, one limitation of this study is that the results do not allow determination of whether high plantar pressures are a predisposing factor or a result of Sever’s disease. Further research studies are needed to monitor plantar foot pressures, before or during heel pain and after heel pain has ceased, before making a recommendation about the initiation of preventive or therapeutic action. All 22 children diagnosed with Sever’s disease had symptoms in the left heel, and 20 of these children had right-limb dominance, using the right foot to kick the ball while playing soccer. Thus, it seems that the increased amount of time during which the left foot supports the body in “right-footed” athletes leads to overload of the left heel. It is important to investigate the relationship between limb dominance and development of the disorder in a specific foot, as other authors have done for other pathologic entities,[69–71] to confirm or refute this hypothesis. An important limitation of this study is that only static plantar pressures were measured. This initial study focused on static pressures because the children with Sever’s disease had pain during standing.

Conclusions

Children with unilateral Sever’s disease demonstrate higher static plantar pressures at the affected foot than the unaffected foot and the corresponding foot in healthy children. Thus, high plantar foot pressures are associated with the Sever’s disease symptoms that characterize this condition. Gastrocnemius equinus may be a predisposing factor for the increased plantar pressures at the heel found in patients with Sever’s disease and could serve as a screening tool to indicate the need for pressure measurements. Further research is needed to identify other factors involved in the disease, as well as to better understand the factors that contribute to abnormal distribution of body weight in the lower limbs.