ural of Pedarie Orhopaccs Part 8 406 So Lippnc Witham & Wikis, ne, Piadelyia Idiopathic Toe-Walking: Does Treatment Alter the Natural History? Deborah M. Eastwood, F.R.C.S.,* Malcolm B. Menelaus, F.R.A.CS., D. Robert V. Dickens, F.R.A.C.S., Nigel S. Broughton, F.R.C.S., and William G. Cole, M.Sc., Ph.D., F.R.C.S.C.+ Department of Orthopaedics, The Royal Children's Hospital, Victoria, Australia; *Department of Orthopaedics, Royal Free ‘Hospital, London, United Kingdom: and ;Division of Orthopaedics, The Hospital for Sick Children, Toronto, Canada cit a ee es a ‘Summary: Outcomes from observation or cast or surgical treatment of idiopathic toe-walking were determined in 136 children. With patient-determined outcomes, for the observa- tion group, gait was normal in 6%, improved in 45%, and. ‘unchanged in 49%, Physician-determined outcomes demon- strated normal gait in 12% of children. Outcomes were similar in the cast group. With patient-determined outcomes in the ‘surgical group, 22% walked normally, 50% had improved, 26% were unchanged, and 2% had deteriorated; with physician- determined outcomes, 37% walked normally. The natural his- tory, determined from the observation group, was for idiopathic toe-walking to persis, albeit with improvement in 50%. Cast treatment did not alter the natural history. Surgical treatment ‘may influence the outcome, but indications for surgery need 10 be clarified. Key Words: Achilles tendon—Equinus—Toe- walking a ee ee Once independent walking has been achieved, most children have a normal heel-toe gait, but some persis- tently walk on their toes. Persistent toe-walking can be produced by neuromuscular disorders such as cerebral palsy and Duchenne muscular dystrophy and has been Associated with a Variety of psychological conditions, Some of which may have a subtle neurologic abnormality (1,5.7.8,10,12,14). Children with bilateral toc-walking who lack features of these diseases are classified as id- iopathic or habitual toe-walkers, “The etiology of idiopathic toe-walking (ITW) remains unclear. Congenital shortening of the Achilles tendon has been proposed (4), but not all patients demonstrate @ tendon contracture. In some patients a genetic influence hhas been established (6,9). Overall, subtle neurologic changes may be more common (2). The natural history of TTW isalso unclear, and the influence of treatment on the final outcome is ill defined (3.4.6,13). This study reports the outcomes of observation alone, the use of plaster casts, or surgical lengthening of the Achilles tendon in a group of 136 children with ITW. PATIENTS AND METHODS From January 1968 to December 1990, 185 children were referred to the orthopaedic staff of the Royal Chil- ‘Address correspondence and reprint requests to Deborah M. Eastwood, Department of Orthopaedies, Royal Free Hospital, London NW3 206, United Kingdom 47 dren’s Hospital in Melbourne with ITW. One hundred thirty-nine children were reviewed by a single examiner (DME) in special research clinics. In three, the toe~ walking was noted to be asymmetric: alternative diaz- noses were established and the patients were excluded from the study. The remaining 136 children, representing 76% of the initial study population, had bilateral sym- metric toe-walking of unknown etiology; they form the basis of this report. Forty-nine children had simply been observed. The parents were reassured that there was no clinical cause for concem, and the child’s progress was reviewed in the clinic as necessary. No further intervention, such as physiotherapy, took place. Forty-one children had been treated with serial below-knee walking plaster casts over a period of 6 weeks. Forty-six children had been treated by a Baker type of aponeurotic lengthening of the Achil- les tendon (11), and a below-knee walking cast was worn after surgery for 6 weeks. No further splinting was used. ‘Although all children in the surgical group had lost ‘movement at the ankle joint, a fixed equinus contracture was not a prerequisite for inclusion in the group. “The demographic and clinical data at initial presenta- tion of the 46 children who did not take part in the review ddd not differ significantly from those who were included in the review. Nineteen had been observed, 13 had been treated by casts, and 14 had undergone surgery Parent-determined assessments of the severity of toe- walking at presentation and at late review were made48 D. M, EASTWOOD ET AL. using two criteria, A linear analog scale was used to record the percentage of walking time thatthe child spent ‘on the toes. In addition, parents were asked to classify the severity of the toe-walking. In grade 1, children walked ‘on the tips of their toes (en pointe), in grade 2 weight was. taken on the metatarsal heads, and in grade 3 the hee! was off the ground during gait. A normal gait was with a heel-toe strike. Outcome could then be classified as a return to normal heel-toe gait, an improvement in gait pattern by one or more grades, an unchanged, or a wors- ‘ened gait. Physician-determined outcomes were made at late review. Wet footprint analyses were undertaken to determine whether the child had a he gait. Late reviews were undertaken 2 to 22 years after pre- sentation. The minimum period of 2 years was consid: cred to be sufficient to determine the response to casting, oor surgery (3). Data were recorded separately for each limb of each child, but because the measurements did not differ significantly, at presentation or at review. only one set of data was analyzed for each child. Statistical com- parisons were made for continuous data using the Mann- Whitney or Kruskal-Wallis tests and for categorical data using Fisher's exact two-tailed tes. RESULTS Observation Group Details of the 49 children in this group are shown in Table 1. The patients presented at a median age of 4 years, at which time they were spending a median 90% Of their walking time on their toes. This had improved significantly by the time of review, but even then the children spent a median 60% of their walking time toe- walking. The patient-determined outcomes indicated that approximately half of the children were better than at presentation, but a normal gait was rare. The physician- determined outcomes showed that only 12% of the ck dren had achieved a normal heel-toe gait Cast Group, Details of the 41 children in the cast group are shown in Table 1, The children in this group were the same as those in the observation group in terms of age at presen- tation and percentage time spent on their toes, Like the children in the observation group, the children treated with a cast showed a statistically significant improve- ment in the time spent walking on their toes at late re- view, although as a group they still spent a median 70% of their time toe-walking. Similarly, the patient-deter- mined outcomes showed that only approximately half the children had an improved pattern of walking; a normal ‘gait was rare, The physician-determined outcomes indi- cated that 22% of the children had achieved a normal heel-toe gait at late review. No complications of cast treatment were noted. ‘There was no significant difference between the ob- servation and cast-treated groups in either patient- or physician-determined outcomes or percentage of time spent toe-walking at presentation or at review. Surgical Group ‘The details of the 46 children treated by aponeurotic lengthening of the Achilles tendon are shown in Table 1 ‘The median age at presentation was 6.5 years; these pa- tients were significantly older than the patients in either the observation or the cast group. Achilles tendon lengthenings took place shortly after presentation (see Table 1). At late review, there were clinically and statistically significant reductions in the time spent toe-walking. Patient-determined outcomes showed that 72% had improved, with almost one third of these reporting a normal gait. The physician-determined ‘outcomes confirmed that 37% of children had a normal TABLE 1, Patient details and results Patent details No.of patents Male-Female ratio Age at presentation Median (range) (9) AAze at treatment (9) Dasation of follow-up () Observation cat Surgery 9 a 6 ana 25:16 26:20 41540 33010 6s@549¢ 3505-103) 6925-185) wens 370215) 196 % Time toe-walking at presentation: Median (Range) 909-1081) 10080-1003" 100(7S-100) & Time toe-walking at review 6010-100)" 70K0-100)" 250-100)" Patent-dtermined outcome Normal gait (%) 306) 440% 10@2) Improved, n (2) 2245) nn) 23150) Unchanged, n () 24049) 20149) Worse, (5) 0 0 10) Pysician-determined outome ‘Normal Gait n (5) 642) 902) cul Toe-walking, n(%) 8188) 208) 29163) 7 The surgical group was significantly older at presentation and at weatment and had a longer follow-up. Kruskal-Wallis tes, P— 0.081 "For each of the groups, tere was significant reduction inthe time spent toewalking a review. Mann-Whitney test, P = 00001 ‘The iferences beween the observation andthe cas groups were not significant when the normal and improved results were combined and compared tothe unchanged result. Feber’ exact test. P oups Were not significant. Fisher's exact test, P= 0.26, "The differences between the observation an cast,IDIOPATHIC TOE-WALKING 9 gait at late review. The age at which the surgery was undertaken did not appear to affect the patient- determined outcome (Mann-Whitney test, P = 0.59). For this analysis, the ages of the children with a normal ‘or improved gait were compared with those with an un- changed or worse gait. ‘The child whose parents believed he was worse had no evidence of overlengthening of his Achilles tendon, and his gait was normal according to the physician- determined outcomes. ‘Because the patients in the surgical group were sig- nificantly older at presentation than the other two groups, their outcomes were not compared with statistical tests, DISCUSSION ‘This current study, the largest reported in the litera- ture, provides new insights into the natural history of ITW and the influence of treatment. The observation group provided the basis for the authors’ comments on the natural history. Toe-walking persisted in most of the ‘observed children for atleast 13 years after presentation, ‘Although the parents reported an improvement in gait in 45% of the children, only 6% of the children were con ‘sidered by their parents to have a truly normal gait at all times. As judged by physicians, only 12% of the children hhad a normal heel-toe gait. The authors therefore con- clude that despite a general belief that these children will ‘outgrow their toe-walking with time, this is true only in terms of some improvement in the severity of toc~ ‘walking in approximately half of the children “The children treated with casts were comparable to the group who were simply observed. The authors cannot ‘comment objectively on the severity of the toe-walking in these two groups at presentation but accept that it may hhave been greater in those treated by casts, although the percentage of time spent toe-walking was the same in both groups, as was the parents’ assessment of the se~ verity of the toe-walking. Although a previous study by Griffin et al. (3) suggested that there was a short-term benefit from cast treatment, the current patients treated by a similar regimen showed no apparent gain over the natural history of the condition. ‘The children in the surgical group were older at pre- sentation and had a longer duration of follow-up than the ther two groups, suggesting that during the early years of this study, children were referred at an older age. It would appear that later referral patterns may have changed, The results showed that after surgical interven- tion, there was a marked reduction in the percentage of time spent toe-walking, from 100% at presentation to 25% at review. Parents reported an improvement in gait in 50% of children and a normal gait in 22%, Objec- tively, 37% of children in this group walked normally ‘Although direct comparisons cannot be made between the surgical and nonsurgical groups because of the dif- ferences in age at presentation and duration of follow-up, it is likely that the outcome after surgical intervention ts fan improvement on the natural history of the condition. “The results in the present study are in accordance with those presented recently by Stricker and Angulo (13) after a smaller study with a shorter follow-up. In their study, surgical lengthening of the Achilles tendon was performed only in the presence of a fixed equinus con- fracture; despite this, one third of the children in their surgical group spent <50% of their time on their toes. ‘Surgery was performed at a mean age of 3.9 years. In the current study, the children in the surgical group were significantly older and spent much more time on their toes. In both studies, the patient-determined outcomes after surgery were similar. ‘A randomized trial comparing the outcomes of obser- vation, cast treatment, and surgery would better control for biases that could not be controlled in the present study. The authors conclude that TW persists in most children if left untreated or treated by serial casts. Sur- gical treatment appears to improve the outcome, but the indications for surgery require further study REFERENCES 1. Colbert EG, Koegler RR. Toe-walhing in childhood schizophrenia. 7 Pediatr 1958:53:219-20. 2. Eastwood DM, Dennett X, Shield LK. Dickens DRV. Muscle ab rormalites in idiopathic toe-walkers. J Petr Orthop Part B 1997:6215-8. 4. Griffin PP, Wheethouse WW, Shiavi R, Bass W. Habitual te Sralkers A clinical and electromyographic gait analysis. J Bone ‘Joint Surg Am 197738997-101 4 HAIIJE, Salter RB, Bhalla SK. Congenital shor tendo-caleaneus J Bone Joint Surg Br 1967:89:698-7, 5. Hicks R, Durinick N, Gage JR. Differentiation of idiopathic te walking and cerebral palsy. J Pediatr Orthop 198838:160-3. 6, Kate MM, Mubarak SJ. Hereditary tendo Achilis contractures. J Pediatr Orthop 1984:4°711-4 7. Kalen V, Adler N, Bleck EE. Electomyor walking J Pediatr Orthop 1986:631-3. 8. Kelly I, Jenkinson A, Stephens M, O'Brien T. The kinematic patterns of toe-walkers J Pediatr Orthop 1997:17478-80. «9, Levine MS, Congenital short tendo-calcaneus. Am J Dis Child 1973: 125:858-9, 10, Monigomery P, Gauger 1. Sensory dysfunction in children who tocswalk: Phys Ther 1978:58:1195-1204 11, Olney BW, Williams PF, Menelaus MB. Treatment of spastic equ- tus By apomeurosis lengthening. J Pediatr Orthop 19888 12, Papuricllo SG, Skinner SR. Dynamic electromyography analy habitual toe-walkes. J Pediatr Orthop 1985;5:171-5. 13, Sucker SI, Angulo JC. Idiopathic toe walking: a comparison of treatment methods. J Pediatr Orthop 1998:18.289-98, 14, Weber D."Toe walking” in children with early childhood aus Acta Paedopsyehiat I978:43:73-83. ny ofiopathic toe 4 Palit On