Leg-length discrepancy (LLD) is not an unusual condition in the general population. One study reported that up to 70% of the normal adult population showed minor LLD,1 and another demonstrated that 93% of schoolchildren had some degree of LLD.2 For adults with minor LLD less than 2 cm, surgical intervention may not be necessary, whereas larger LLD requires surgical correction, including epiphysiodesis for growing children and bone lengthening for adolescents.3–6 However, for children with minor LLD, the indications for and timing of epiphysiodesis are difficult to determine because the incidence of clinically significant discrepancy is unpredictable. A better understanding of the longitudinal course of minor LLD is required to successfully correct it.
There are many causes of LLD, including paralytic disorders (eg, cerebral palsy and polio), disruption of the physis (eg, trauma, infection, and tumor), and congenital disorders.7,8 Several established pathological conditions, such as Silver-Russell syndrome, proximal focal femoral deficiency, fibular hemimelia, congenital neoplastic diseases (eg, neurofibromatosis, enchondromatosis, and exostosis), syndromes associated with vascular malformations (eg, Klippel-Trénaunay-Weber syndrome), and some overgrowth syndromes (eg, Proteus syndrome), cause congenital LLD.9–12 Other disorders cause congenital LLD with unknown mechanisms, such as idiopathic hemihypertrophy and hemiatrophy.13
For the current study, the authors examined longitudinal changes of LLD in patients with congenital disorders who had minor LLD less than 2 cm at the first presentation and determined predictive values for the incidence of significant LLD more than 2.5 cm at skeletal maturity.
Materials and Methods
After institutional review board approval, the authors retrospectively reviewed the clinical records and radiographs of 52 patients with congenital LLD who had been managed at Nagoya University Hospital since 1987. Inclusion criteria were patients who had been followed up for more than 4 years with a serial medical examination at least every year. Patients who had angular deformities of the long tubular bones, such as congenial pseudoarthrosis of the tibia, were excluded because it causes apparent LLD in addition to anatomical LLD. Patients who had joint contractures of the lower limbs, such as arthrogryposis, were also excluded because the authors could not take appropriate radiographs to measure accurate bone lengths. According to their LLD at latest follow-up, the patients were divided into 2 groups: the significant group, with 25 mm or more of LLD; and the minor group, with less than 25 mm of LLD.
The authors used computed radiography (CR)-based modified radiographs to evaluate LLD.14 A radiograph beam is projected toward a single long cassette placed behind the patient. The lower limbs are positioned in supine with both patellae pointing toward the ceiling, and the radiograph tube is set over the knee at least 180 cm away from the patient to minimize magnification error. The CR-based radiographs of the bilateral lower limbs were manipulated on a digital imaging workstation (the picture archiving and communication system [PACS]) to measure the lengths of the femur and tibia on each side up to the millimeter. Femur length was measured from the top of the femoral head to the distal end of the medial femoral condyle, and tibia length was measured from the intercondylar eminence of the tibia to the midpoint of the ankle mortise. Leg-length discrepancy was then defined as the difference between the total length of the femur and the tibia on each side. Serial assessments of LLD were performed until the latest visit for conservatively managed patients and until surgery for those who underwent surgical correction.
To evaluate longitudinal changes of LLD, the age and LLD of each patient were plotted on a x-axis and y-axis, respectively. The changes in LLD were classified into 5 developmental patterns proposed by Shapiro15: type I, upward slope pattern=LLD increases continually with time at a constant rate; type II, upward slope-deceleration pattern=LLD increases at a constant rate and then shows a diminishing rate of increase; type III, plateau pattern=LLD stabilizes and remains unchanged for a variable period of growth; type IV, upward slope-plateau-upward slope pattern=LLD first increases then stabilizes for a variable but considerable period of time, and then increases again; and type V, upward slope-plateau-downward slope pattern=LLD first increases with time, then stabilizes, then spontaneously decreases without surgical intervention.
All statistical analyses were performed with SPSS version 22 statistical software (IBM Corp, Armonk, New York). Prior to analysis, data normality was tested using the Kolmogorov-Smirnov test. The unpaired t test was used for a comparison of continuous variables between the significant and minor groups. Fisher’s exact test was used to compare distributions of the sex and developmental patterns of LLD (Shapiro classification) between the 2 groups. A P value less than .05 was considered statistically significant. The authors analyzed repeated measures of LLD up to age 10 years using a linear mixed-effects model with random intercepts containing an independent continuous variable of age and the interaction term for age and the group (significant or minor) to compare the longitudinal age-associated LLD changes of both groups. An analysis of residuals was conducted to determine model fit.
Twenty-one patients (12 male, 9 female) fulfilled the inclusion criteria of this study. Demographic information, developmental pattern of LLD (Shapiro classification), and treatment for 21 patients are summarized in Table 1. Fifteen patients had isolated LLD due to idiopathic hemihypertrophy or hemiatrophy and 6 patients had concomitant LLD associated with various etiologies such as fibular hemimelia (n=2), Klippel-Trénaunay-Weber syndrome (n=2), neurofibromatosis type 1 (n=1), and macrodactyly of the foot (n=1).
Mean age at the first and latest examinations was 3.1±2.5 years and 11.3±2.1 years, respectively. Mean LLD at the first and latest examinations was 11±5 mm and 22±9 mm, respectively. Nine patients were treated conservatively, and 12 underwent surgery (epiphysiodesis in 10 patients and bone lengthening in 2). According to Shapiro classification, all patients had type I (n=10), II (n=2), or III (n=7), except for 2 patients who showed an atypical type V pattern. One of these patients had neurofibromatosis type 1 without associated congenial pseudoarthrosis of the tibia, and the other patient had macrodactyly of the left foot, probably associated with an overgrowth syndrome. The patient with neurofibromatosis type 1 exhibited an increasing LLD during the first decade of life due to overgrowth of the right tibia with cortical thickening, but the LLD declined prior to reaching 25 mm after age 10 years (Figure 1). The patient with macrodactyly underwent amputation of the second ray at age 2 years and partial resections of the first and third toes at age 3 years. The patient’s LLD gradually increased until age 10 years, then decreased prior to reaching 20 mm.
Computed radiography–based modified radiographs of a patient with neurofibromatosis type 1 at age 11 years (A) and 16 years (B) showing overgrowth of the right tibia associated with cortical thickening and growth impairment of the ipsilateral femur.
The significant group comprised 11 patients, and the minor group comprised 10 patients. No statistical significance was observed between the 2 groups regarding LLD at the first visit and at follow-up (Table 2). The relationship between predicted LLD calculated by linear mixed-effects analysis vs observed LLD within the first 10 years is shown in Figure 2. A linear regression line within the scatter plot indicates that the model is not perfect, but with a slope of 0.70 and an R2 of 0.76, it is a reasonable fit. Figure 3 illustrates longitudinal age-associated LLD changes in the 2 groups, developed from linear mixed-effects analysis of data from observed LLD within the first 10 years. In the significant group, the mean annual rate of LLD change was 2.1±0.2 mm and was significantly larger than that in the minor group (difference in slope, 1.3 mm; P<.0001).
Comparison of Significant and Minor Groups
Scatter plot of predicted leg-length discrepancy calculated with the use of a linear mixed-effects model vs observed leg-length discrepancy. The model was developed from the longitudinal measures of leg-length discrepancy during the first decade of life. A linear regression line is drawn to evaluate the fit of the model. The slope, intercept, and coefficient of determination are shown within the scatter plot.
Longitudinal age-associated leg-length discrepancy changes in the significant and minor groups, developed from linear mixed-effects analysis of data from longitudinal measures of leg-length discrepancy during the first decade of life. Filled and open circles denote the significant and minor groups, respectively.
The results of this study show that the annual rate of LLD change in the first decade of life can predict the incidence of clinically significant LLD (≥2.5 cm) at skeletal maturity, which usually requires epiphysiodesis during the growth period.
Shapiro15 examined LLD data from 803 patients and developed a classification system. Of 279 patients with congenital disorders, all but 3 represented a type I, II, or III pattern, suggesting that congenital LLD monotonically increases with time, which is in agreement with the current study.
There are currently 4 prevailing methods of predicting LLD at skeletal maturity: the arithmetic method, the growth-remaining method, the straight-line graph method, and the multiplier method.16–19 However, none of the methods can predict the accurate timing of epiphysiodesis because they tend to underestimate the final LLD at skeletal maturity.20,21 Lee et al20 emphasized the importance of assessment of the growth pattern for the accurate prediction of LLD and proposed the use of leg lengths at the initial examination and the final examination before surgery to assess the pattern. The current authors consider the annual rate of LLD change in the first decade of life as a useful complement to the existing methods of determining the optimal timing of epiphysiodesis.
The cutoff point of 2 to 2.5 cm of LLD has widely been accepted as a reasonable guideline on when perform orthopedic surgery8 based on the biomechanical effects of LLD on musculoskeletal disorders, such as low back pain, gait and posture abnormalities, stress fractures, functional scoliosis, hip and knee pain, and joint contractures.22–30 In addition to harmful effects on physical conditions, LLD of 2 cm or more can cause parents to feel guilty.31
The current study has several limitations. First, it was a single-center study that enrolled a small number of patients. A larger number of patients is required to reinforce the results. Second, it excluded patients with acquired disorders, some of which exhibit fluctuating LLD. Additional investigations are needed regarding the growth rate of LLD in acquired diseases. Third, patients with idiopathic hypertrophy and hemiatrophy constituted most of the study population. A change in the proportion of congenital disorders may influence the results.
Assessment of LLD growth pattern is indispensable to the management of minor LLD. For patients with congenital LLD (< 2 cm) at age 10 years or younger, epiphysiodesis can be required during the growth period when mean annual rate of LLD change between the first presentation and age 10 years reaches 2 mm.
- Subotnick SI. Limb length discrepancies of the lower extremity (the short leg syndrome). J Orthop Sports Phys Ther. 1981; 3 (1):11–16. doi:10.2519/jospt.19220.127.116.11 [CrossRef]
- Woerman AL, Binder-Macleod SA. Leg length discrepancy assessment: accuracy and precision in five clinical methods of evaluation. J Orthop Sports Phys Ther. 1984; 5 (5):230–239. doi:10.2519/jospt.1918.104.22.168 [CrossRef]
- Sailhan F. Bone lengthening (distraction osteogenesis): a literature review. Osteoporos Int. 2011; 22 (6):2011–2015. doi:10.1007/s00198-011-1613-2 [CrossRef]
- Hasler CC, Krieg AH. Current concepts of leg lengthening. J Child Orthop. 2012; 6 (2):89–104. doi:10.1007/s11832-012-0391-5 [CrossRef]
- Pendleton AM, Stevens PM, Hung M. Guided growth for the treatment of moderate leg-length discrepancy. Orthopedics. 2013; 36 (5):e575–e580. doi:10.3928/01477447-20130426-18 [CrossRef]
- Ghanem I, Karam JA, Widmann RF. Surgical epiphysiodesis indications and techniques: update. Curr Opin Pediatr. 2011; 23 (1):53–59. doi:10.1097/MOP.0b013e32834231b3 [CrossRef]
- Shailam R, Jaramillo D, Kan JH. Growth arrest and leg-length discrepancy. Pediatr Radiol. 2013; 43 (suppl 1):S155–S165. doi:10.1007/s00247-012-2598-5 [CrossRef]
- Gurney B. Leg length discrepancy. Gait Posture. 2002; 15 (2):195–206. doi:10.1016/S0966-6362(01)00148-5 [CrossRef]
- Alwan S, Tredwell SJ, Friedman JM. Is osseous dysplasia a primary feature of neurofibromatosis 1 (NF1)?Clin Genet. 2005; 67 (5):378–390. doi:10.1111/j.1399-0004.2005.00410.x [CrossRef]
- Stieber JR, Dormans JP. Manifestations of hereditary multiple exostoses. J Am Acad Orthop Surg. 2005; 13 (2):110–120.
- Watanabe K, Tsuchiya H, Sakurakichi K, Yamashiro T, Matsubara H, Tomita K. Treatment of lower limb deformities and limb-length discrepancies with the external fixator in Ollier’s disease. J Orthop Sci. 2007; 12 (5):471–475. doi:10.1007/s00776-007-1163-9 [CrossRef]
- Redondo P, Aguado L, Martinez-Cuesta A. Diagnosis and management of extensive vascular malformations of the lower limb: Part I. Clinical diagnosis. J Am Acad Dermatol. 2011; 65 (5):893–906. doi:10.1016/j.jaad.2010.12.047 [CrossRef]
- Ballock RT, Wiesner GL, Myers MT, Thompson GH. Hemihypertrophy: concepts and controversies. J Bone Joint Surg Am. 1997; 79 (11):1731–1738.
- Sabharwal S, Kumar A. Methods for assessing leg length discrepancy. Clin Orthop Relat Res. 2008; 466 (12):2910–2922. doi:10.1007/s11999-008-0524-9 [CrossRef]
- Shapiro F. Developmental patterns in lower-extremity length discrepancies. J Bone Joint Surg Am. 1982; 64 (5):639–651.
- Anderson M, Green WT, Messner MB. Growth and predictions of growth in the lower extremities. J Bone Joint Surg Am. 1963; 45:1–14.
- Green WT, Anderson M. Experiences with epiphyseal arrest in correcting discrepancies in length of the lower extremities in infantile paralysis: a method of predicting the effect. J Bone Joint Surg Am. 1947; 29 (3):659–675.
- Moseley CF. A straight-line graph for leg-length discrepancies. J Bone Joint Surg Am. 1977; 59 (2):174–179.
- Paley D, Bhave A, Herzenberg JE, Bowen JR. Multiplier method for predicting limb-length discrepancy. J Bone Joint Surg Am. 2000; 82 (10):1432–1446.
- Lee SC, Shim JS, Seo SW, Lim KS, Ko KR. The accuracy of current methods in determining the timing of epiphysiodesis. Bone Joint J. 2013; 95 (7):993–1000. doi:10.1302/0301-620X.95B7.30803 [CrossRef]
- Little DG, Nigo L, Aiona MD. Deficiencies of current methods for the timing of epiphysiodesis. J Pediatr Orthop. 1996; 16 (2):173–179. doi:10.1097/01241398-199603000-00007 [CrossRef]
- Raczkowski JW, Daniszewska B, Zolynski K. Functional scoliosis caused by leg length discrepancy. Arch Med Sci. 2010; 6 (3):393–398. doi:10.5114/aoms.2010.14262 [CrossRef]
- Needham R, Chockalingam N, Dunning D, Healy A, Ahmed EB, Ward A. The effect of leg length discrepancy on pelvis and spine kinematics during gait. Stud Health Technol Inform. 2012; 176:104–107.
- Mahar RK, Kirby RL, MacLeod DA. Simulated leg-length discrepancy: its effect on mean center-of-pressure position and postural sway. Arch Phys Med Rehabil. 1985; 66 (12):822–824.
- Kocaoglu M, Eralp L, Atalar AC, Bilen FE. Correction of complex foot deformities using the Ilizarov external fixator. J Foot Ankle Surg. 2002; 41 (1):30–39. doi:10.1016/S1067-2516(02)80007-2 [CrossRef]
- Kiapour A, Abdelgawad AA, Goel VK, Souccar A, Terai T, Ebraheim NA. Relationship between limb length discrepancy and load distribution across the sacroiliac joint: a finite element study. J Orthop Res. 2012; 30 (10):1577–1580. doi:10.1002/jor.22119 [CrossRef]
- Harvey WF, Yang M, Cooke TD, et al. Association of leg-length inequality with knee osteoarthritis: a cohort study. Ann Intern Med. 2010; 152 (5):287–295. doi:10.7326/0003-4819-152-5-201003020-00006 [CrossRef]
- Golightly YM, Tate JJ, Burns CB, Gross MT. Changes in pain and disability secondary to shoe lift intervention in subjects with limb length inequality and chronic low back pain: a preliminary report. J Orthop Sports Phys Ther. 2007; 37 (7):380–388. doi:10.2519/jospt.2007.2429 [CrossRef]
- Gofton JP, Trueman GE. Studies in osteoarthritis of the hip: II. Osteoarthritis of the hip and leg-length disparity. Can Med Assoc J. 1971; 104 (9):791–799.
- Friberg O. Leg length asymmetry in stress fractures: a clinical and radiological study. J Sports Med Phys Fitness. 1982; 22 (4):485–488.
- Lee KM, Chung CY, Gwon DK, et al. Parental perspectives on leg length discrepancy. J Pediatr Orthop B. 2012; 21 (2):146–149. doi:10.1097/BPB.0b013e32834c67b2 [CrossRef]
|Patient No./Sex||Diagnosis||Age, y||Follow-up, y||LLD, mm||Shapiro Classification||Treatment|
|At First Examination||At Latest Follow-up||At First Examination||At Latest Follow-up|
Comparison of Significant and Minor Groups
|Variable||Significant Group (n=11)||Minor Group (n=10)||P|
|Age at first examination, mean±SD, y||2.0±2.1||4.2±2.5||.047|
|Age at latest follow-up, mean±SD, y||10.3±1.6||12.4±2.2||.019|
|Follow-up, mean±SD, y||7.7±2.7||8.6±2.9||.98|
|LLD at first examination, mean±SD, mm||12±4||9±4||.122|
|Type I–III: other Shapiro classification, No.||11:0||8:2||.21|