Toe Walking

Updated: May 02, 2017
  • Author: Ryan Krochak, MD; Chief Editor: Vinod K Panchbhavi, MD, FACS  more...
  • Print


Toe walking is a gait abnormality characterized by an absence of normal heel-to-floor contact (heel strike) by both feet during gait. It is defined as the inability to make heel contact with the floor during the initial stance phase of the gait cycle and the absence of full foot contact with the ground during the remainder of the gait cycle. The forefoot engages in the majority of floor contact throughout the gait cycle.

Toe walking has multiple etiologies, ranging from idiosyncratic habit to profound neuromuscular disease. The most commonly observed type of toe walking is idiopathic toe walking (ITW). Critical to the management of toe walking is the exclusion of neurologic or muscular diseases as a cause of the perceived gait abnormality. 

Treatment depends on the patient’s age and the severity of the gait abnormality. Although toe walking is commonly seen during development in children who are first learning to walk, a consistent heel-toe pattern of gait usually develops by approximately age 22 months. [1, 2]  If toe walking persists beyond age 2 years, further evaluation is warranted. [3]

Specific treatment options range from simple observation to surgical lengthening of muscles or tendons in the lower extremity. Tenotomy is the commonly used procedure for the treatment of toe walking. It is also one of the oldest procedures in orthopedics. The first tenotomy is believed to have been a multilevel percutaneous Achilles tendon lengthening performed by Delpech in 1823 while he was treating a patient with clubfoot. [4]

For patient education resources, see Cast Care.



The triceps surae muscle-tendon complex is the confluence of the gastrocnemius and soleus muscles and the Achilles tendon or heel cord, the largest tendon in the human body.

The gastrocnemius muscle originates from the posterior medial and lateral femoral condyles and inserts onto the calcaneus through the Achilles tendon. It crosses both the knee and the ankle joints and acts as a major plantarflexor of the ankle and a minor flexor of the knee. It typically has more fast-twitch type II muscle fibers, which are responsible for short, powerful bursts of activity such as running and jumping.

The soleus muscle lies deep (anterior) to the gastrocnemius and originates from the posterior surfaces of the proximal tibia and fibula and inserts into the calcaneus by way of the conjoined Achilles tendon. It crosses only the ankle joint and functions to plantarflex the ankle. The soleus muscle is made up of primarily slow-twitch type I muscle fibers and is responsible primarily for postural control.

The Achilles tendon measures approximately 4-8 cm from the point where the gastrocnemius and soleus muscles join to its insertion on the calcaneus. As the fibers of the tendon traverse this distance, they rotate approximately 90º in the axial plane. The fibers from the more superficial gastrocnemius muscle insert on the posterolateral aspect of the calcaneus, and the fibers from the deeper soleus muscle insert on the posteromedial aspect of the calcaneus. [5] Understanding this rotation of the fibers helps in planning the percutaneous lengthening procedure (see Surgical Therapy).

It is possible to separate the contributions of the gastrocnemius and soleus muscles to an equinus ankle contracture via the Silfverskiöld test. According to this test, increased ankle dorsiflexion with the knee in flexion compared with the knee in extension indicates gastrocnemius tightness. This occurs because the gastrocnemius relaxes with knee flexion as the muscle spans the knee joint, whereas the soleus does not. If there is no difference in dorsiflexion with knee flexion, an Achilles tendon contracture is present. The test assists in deciding the surgical approach to lengthening of a patient’s heel cord.



Although the exact pathophysiology of ITW remains unknown, it is postulated that mild defects in sensory processing exist in affected children and that this gait may result from a vestibular disorder or abnormal sensitivity to touch. [6] However, there is limited research exploring these relations.

ITW likely has a genetic component, in that a positive family history has been reported in many case series. [7] In the initial description of the condition, Hall et al noted that all their patients had congenital shortness of the Achilles tendon, which led to ankle equinus and toe walking. [8] However, subsequent studies found that not all patients with ITW exhibit a congenitally short Achilles tendon as a mechanical explanation for the gait difference and that some patients toe-walk despite a volitional ability to walk flatfooted (“dynamic" or "habitual" toe walkers). [7]

The natural history of ITW remains poorly defined; most studies offering long-term follow-up concurrently report treatment of their cohorts. Opinion on the adult consequences of a persistent toe-walking pattern may be divided into two schools of thought.

The first school of thought holds that regardless of the initial status of the heel cord and ankle range of motion (ROM), children with persistent toe walking eventually develop a fixed ankle contracture and ankle equinus in adulthood, which lead to hindfoot valgus and myriad potential foot disorders. [9, 10, 11, 12, 13] This argument supports aggressive intervention for ITW in early childhood.

The second school of thought holds that ITW can have a benign natural history, with the majority of patients manifesting no particular functional limitations or pathologic sequelae in adulthood, despite variable objective improvement in overall ankle ROM or gait. [14] This argument supports an observational approach to management of this gait abnormality.

The pathophysiology of toe walking in patients with cerebral palsy is clearer and relies on the following two basic mechanisms centered on the underlying spasticity of the lower-extremity musculature [15] :

  • First, spasticity of the foot and ankle muscles can lead to progressive ankle equinus contracture as spastic muscles grow at a slower rate than muscles that are not spastic
  • Second, spasticity and flexion of the more proximal hip and knee joints can result in apparent toe walking as the patient attempts to maintain balance when upright; if the hip and knee are flexed in stance and the ankle is held at a right angle relative to the tibia (plantigrade), the patient bears weight on the toes and forefoot, even though the ankle itself is not in equinus

In Duchenne muscular dystrophy (DMD), the quintessential example of degenerative muscle disease, lower-extremity muscles progressively weaken as they degenerate and are replaced by fibrous tissue. Toe walking results from the relatively greater involvement of the dorsiflexors rather than the plantarflexors of the foot.

In addition, as noted, toe walking also develops to compensate for the weakening quadriceps muscle. As the quadriceps weakens, active knee extension is lost and knee stability during gait is threatened as the knee preferentially buckles into flexion. By walking on the forefoot, the patient generates a knee-extension moment that aids in stabilization. [16]



Various causes of toe walking exist and include both central and peripheral neurologic disorders (eg, tethered cord, diastematomyelia, spina bifida, and cerebral palsy), muscular disorders (eg, congenital muscular dystrophy), and anatomic disorders (eg, limb-length discrepancy). [3] Some rarer causes of toe walking have also been reported in the literature, such as acute toe walking secondary to viral myositis in a previously healthy 4-year-old child. [17]

As noted, toe walking is most commonly idiopathic, which means that no identifiable pathologic process exists to explain the perceived gait abnormality. ITW, first described by Hall in 1967 as “habitual toe walking” and “congenital short Achilles tendon,” is a diagnosis of exclusion. [8] It is best defined as bilateral toe walking with or without Achilles tendon contracture in a child older than age 2 years in the absence of other etiologies. (See Physical Medicine and Rehabilitation for SpasticitySpinal Cord Trauma and Related DiseasesMuscular Dystrophy, and Congenital Myopathies.)



The true prevalence of ITW is unknown because not all children with this condition present to a physician.

In several small studies, ITW was estimated to occur in 7-24% of the childhood population. [18] A large Dutch cross-sectional study from 2011 found a prevalence of 12% in the general population. [18] A large Swedish study from 2012 found the prevalence of ITW to be 4.9% in children aged 5 years 6 months. [19]   A comprehensive literature review published by Ruzbarsky et al in 2016 found the prevalence at 5.5 years to be 2% in normally developing children and 41% in children with a neuropsychiatric diagnosis or developmental delay. [20]

ITW is observed more frequently in patients with autism, developmental delay, and language disorders. [7] A 2011 study determined the incidence of toe walking in children with autistic spectrum disorder to be 20.1%, whereas historical reports estimated it to be as high as 63%. [21] Conversely, children with ITW display an increased prevalence for other pediatric neuropsychiatric disorders, such as attention-deficit/hyperactivity disorder, tics, Tourette syndrome, and obsessive-compulsive disorder. [22]

A 2016 article by Baber et al reviewed the birth characteristics of children with ITW or toe walking due to medical reasons. [23]  They evaluated 95 children diagnosed with ITW (mean age, 5.8 ± 2.9 years).  These children were found to have higher rates of prematurity, higher rates of admission to a special care nursery or neonatal intensive care unit, and lower birth weights than the normative population. Children with a medical reason for toe walking also had greater rates of prematurity than the normative population and more instrumented births than children with ITW.

Baber et al found no association between assisted-birth intervention and the ITW cohort in comparison with the normative population or the medical-cause cohort. [23] ITW gait was associated with higher rates of complications during and after delivery. Such complications have previously been cited as risk factors for neurologic insult affecting motor development.

The most common identifiable etiology of toe walking is cerebral palsy, which affects 1-7 in 1000 children. Reports vary greatly regarding the incidence of toe walking in this population, largely because of the wide spectrum of disease severity. It has been noted, however, that overall, fewer than 50% of these patients walk on their toes. [24]

DMD has an incidence of 1 case per 3500 live male births and is the most common degenerative muscle disease of childhood. [25] The typical child with DMD initially walks with a heel-to-toe pattern but progresses to a consistent toe-walking pattern as a compensation for the progressive weakness of the knee extensors.



The long-term results of treatment for ITW are unclear. Because most studies mix a variety of management methods, critical evaluation of individual treatment options is difficult.

Reported outcomes of treatment with stretching casts have been variable. In a long-term study of 14 children with ITW treated with a stretching regimen with or without casting, only three children required additional treatment for persistent toe walking after several years. [26]  However, in a larger series, no difference in outcome was seen between patients treated with observation and those treated with serial casting: 50% of both groups had improvements in gait. [14]  Other studies showed similar variability, with one finding 66% of patients to have improved ankle dorsiflexion and gait after casting. [14, 27, 28]

Overall, casting management seems to be most successful in patients with an initial ankle dorsiflexion of greater than 0º. In addition, recurrence of toe walking is unlikely in patients who have spontaneously discontinued toe walking or who remain down on their heels for more than 1 year after casting treatment. [3]

Surgical management of ITW, through either isolated gastrocnemius lengthening or Achilles lengthening, appears to be effective in the appropriate clinical setting. [29]  In one study, 72% of surgically treated patients reported a normal or improved gait at follow-up. [14]  In another, all surgically treated patients showed improvement in ankle dorsiflexion, and the reported parental satisfaction rate was 67%. [28]  Long-term outcome studies suggest that surgical management is more effective than nonsurgical management. However, direct comparison is difficult, owing to the variability of patient age and disease severity in these studies. [14]

The prognosis for patients with spasticity or paralytic muscle disease is less predictable. These patients require long-term follow-up, in that recurrence rates are high. [30]  For ambulatory cerebral palsy patients without a fixed ankle contracture, treatment with full-time bracing (ankle-foot orthoses [AFOs]) has been shown to have a positive effect on ankle ROM and power at pushoff. [31]

With regard to cerebral palsy patients who have a rigid deformity, children younger than 7 years at the time of surgery have been shown to be at greater risk for recurrence. Hemiplegic patients have also been found to have higher recurrence rates than diplegic patients. [32]  In one long-term study, surgical lengthening in cerebral palsy patients with a fixed equinus contracture resulted in a significant gain in average dorsiflexion, from –25° preoperatively to 8° postoperatively, and this correction was maintained by stretching and AFOs in 80% of patients 7 years after the index procedure. [33]

Randomized controlled studies have not shown botulinum toxin alone to offer better long-term outcomes than casting alone for dynamic equinus in children with cerebral palsy or ITW. [34, 35]  Furthermore, combining botulinum toxin injections with casting has not been shown to yield improved results for either toe-walking cerebral palsy patients or patients with ITW as compared with casting alone. [34, 35]