*3.4. E*ff*ects of Characteristic Factors*

The fixed effects of the patient's age, sex, synostosis and initial ankle alignment on subsequent tibial and fibular lengths, and the tibiotalar angle, are listed in Table 4. The amount of change per a month in the fibular and tibial lengths and the ratio of the fibular and tibial lengths decreased significantly (Figure 4). Sex differences were found to not affect the changes in the tibiofibular length ratio or tibiotalar angle. In the presence of synostosis, the rate of growth per month of the tibia and fibula decreased significantly (tibia, −0.125 mm/month; fibula, −0.138 mm/month); however, it did not affect the changes in the ratio of the fibular and tibia lengths or tibiotalar angle. When the initial ankle alignment was valgus, the ratio of the fibular and tibial length increased by 0.0005 per month (*p* = 0.004). For ankles with an initial varus alignment, significant increases in the tibiotalar angle were demonstrated (0.044 mm per month, *p* = 0.034) (Figure 5). − − − −

**Figure 4.** The serial radiographs of a male patient demonstrated progression of valgus deformity on the left ankle. Despite exostoses in bilateral legs, valgus angulation of the left ankle accompanied by shortening of the fibula was observed, while the neutral alignment was maintained on the right ankle. In the last measurement, the left tibial length was also about 9 mm shorter than the right tibia.

**Figure 5.** A male patient had prominent exostosis on the left proximal fibula and an increased gap of the proximal tibia and fibula at initial evaluation. Although the right lower extremity initially maintained a normal tibiofibular relationship, after the growth was completed, the right ankle joint developed valgus deformity to be equivalent to the left ankle. Finally, the leg length discrepancy was approximately 12 mm due to the 10 mm shortened length of the left tibia.


**Table 4.**Interactive effects between the parameters at initial visit and the tibiofibular alignment.

Values are presented as mean (SE, standard error) or as numbers only. Effects were analyzed using the linear mixed model and calculated as amount per month. \* Statistical significance was noted. † The reference value is set to 0 to compare the effect of each factor among the parameters of tibiofibular alignment.

#### **4. Discussion**

In this study, we compared the longitudinal changes between the tibiofibular relationship and ankle alignment in HME patients. When the valgus angulation of the tibiotalar angle progressed, we found that the tibial growth was relatively greater than the fibular growth and was accompanied by significantly relative fibular shortening in the proximal and distal portions. This is in line with previous observations that have reported that HME of the lower extremities induces a disproportionate shortening of the fibula with respect to the tibia, leading to progressive valgus deformities of the ankle joint [10,11,15]. Additionally, the sex of the patient was identified to not affect the growth of the tibia or fibula, and synostoses was shown to slow the growth of both the tibia and fibula.

Although the absolute growth rates of the fibula cannot be directly compared, it may be assumed that the relative retardation of fibular growth caused the ankle valgus deformity and fibular shortening [4,13]. For children aged 7–10 years, during physiological growth, the proximal and distal tibiofibular physis differences are usually measured as 1.2 to 1.7 cm and 0.7 to 1.0 cm, respectively. However, in our study, the proximal and distal tibiofibular physis differences were greater than this, and, therefore, we assumed that the fibular growth retardation was apparent [13]. A similar phenomenon may also occur in the forearm, as deformities are caused by relative ulnar shortening [16]. The size of the osteochondromas in the ulna has been inversely correlated with the rate of ulnar shortening, and the location of osteochondromas on the distal ulnar metaphysis has been shown to restrict longitudinal bone growth to under 20% of their expected growth [17]. Using a longitudinal observation study design, we found that the fibula growth slopes were similar regardless of the direction of the change in the ankle alignment. However, for the tibia, significantly greater growth slopes were noted when ankle valgus angulation was present. Therefore, these changes in the tibiofibular relationship of HME patients support the effectiveness of surgical treatment of the tibia with procedures such as medial malleolar screw epiphysiodesis [8,9].

In studies investigating tibial and fibular growth in normal children, progressive distal migration of the distal fibular epiphysis and physis occurred with increasing age [3,13,18,19]. In cases of pediatric ankle fractures or iatrogenic tibiofibular synostosis, it has been reported that the tibiofibular synostosis blocks the physiological distal migration of the fibula, resulting in a proximal fibular migration and deformity with a prominent fibular head and shortened lateral malleolus [5,20–23]. However, our study revealed continuous shortening of both the proximal and distal fibula over time. In addition, the presence of synostosis was found to inhibit the growth of both the tibia and fibula when compared to the growth rates of those without synostoses. Therefore, these findings suggest that the fibular shortening seen in HME patients appears to show different trends with those observed by fractures or iatrogenic synostosis [21].

Takikawa et al. utilized a longitudinal study design with serial ankle radiographs and reported that an ankle valgus deformity could be predicted in male HME patients with prominent shortening of the fibula and exostoses involvement of both the lateral distal tibia and the medial distal fibula [11]. However, they evaluated a total of 62 ankle radiographs of 33 patients (23 males, 10 females) with a mean age of 11.33 years and did not conduct any statistical analysis related to the passage of time. In contrast, we found no significant fixed effects of sex on the growth of the tibia and fibula and no differences between sex or location of the exostosis on the ankle valgus deformity progression. No significant difference was found in age and sex between our groups, and it is possible that the difference in patient distributions compared to the study of Takikawa et al. may have led to the differing results.

A study evaluating the effect of osteochondroma locations on coronal lower limb malalignment reported that the combined lesions of both the proximal and distal tibiofibular joints had the greatest impact on fibular shortening [10]. In our study, the analysis of the combination lesion was not included, and no significance was found in the location of lesions in relation to fibular shortening, regardless of whether valgus progressed or not. Osteochondromas develop and grow while the physis is still open, and with variations in sex, they grow proportionately to the overall growth of the patient [19]. Therefore, bone age may influence the tibiofibular growth and relationship. Our study investigated the alignment changes over time within a cohort of patients all almost under the age of 10, and, hence, there may be differences between our results and those of the cross-sectional study of Ahn et al. on patients with a mean age of 10–14 years [10].

The initial ankle alignment did not significantly affect the future growth of the tibia and fibula in our study. Despite variations with age, minor valgus alignments of 0–8 degrees for the tibiotalar angle are within the standard range during skeletal development [13]. Noonan et al. investigated the natural history of 38 adult patients, mean age of 42 years, with lower extremity and ankle joint HMEs. Their mean tibiotalar angle was 8.6◦ , half of their cohort (50%) suffered from occasional ankle pain, and 14 of 75 ankles (19%) showed degenerative changes [6]. Although physiologic valgus alignment may be acceptable during the growing period, ankle valgus deformities that persists into adulthood can progress to arthritis and result in deterioration of the ankle joint function. More than 80% of patients with HME are initially diagnosed in the first decade of life. Therefore, careful observation and appropriately timed treatment during the growth period are required to minimize functional impairment and the progression of deformities [4,6].

In a retrospective review study of 113 HME patients, the distribution of osteochondromas was reported to be more likely in the proximal tibia (71%) than the distal tibia, and the case was similar for the proximal and distal fibula (27%) [24]. We also found that the location of osteochondromas occurred more proximally in both the tibia and fibula, a finding which is similar to those previous reported. Therefore, as osteochondromas may arise more proximally, evaluations of full length tibia and fibula radiographs may more reliably predict the progression of ankle valgus than those of ankle radiographs only.

This study has several limitations. First, the duration of follow-up was relatively short, and the follow-up time interval was irregular due to the retrospective study design. A prospective study with extended observations to adulthood and controlled time intervals may provide more reliable information to further our understanding of the progression of angular malalignment of lower extremities during the growth period for HME patients. Second, radiographic evaluation of patients was exclusively performed by teleoroentgenogram, with the center of the beam projection facing the knee. Therefore, there may be an inherent error with this beam projection set-up, and all values should have been compared with simultaneous ankle AP radiographs [25]. Nevertheless, since this study investigated the changes in the tibiofibular relationship over time, depending on the progression of ankle valgus deformity, we focused more on analyzing the overall trends of radiographic changes longitudinally. Last, the accuracy of the observations for the presence of synostosis and evaluation of osteochondroma sizes were insufficient due to the lack of a tomographic examinations. Tomographic scanning has the advantage of accurately demonstrating the condition, location, and synostosis of osteochondroma lesions [20]. However, it is clinically difficult to implement and justify such imaging in the pediatric HME population.

#### **5. Conclusions**

In pediatric HME patients, when valgus deformity of the ankle joint progressed, greater relative tibial growth and significantly relative proximal and distal fibular shortening followed. For the fixed effect of characteristic factors on the tibiofibular alignment over time, tibiofibular synostosis lowered the growth of both the tibia and fibula, whilst initial ankle alignment partially affected the tibiofibular relationship. However, demographic and anatomical factors such as age, sex, location and synostosis were not associated with the progression of ankle valgus deformities. Our observations suggest that pediatric HME patients with an imbalanced tibiofibular relationship, due to greater tibial growth, may require appropriate surgical treatment to prevent the progression of ankle valgus deformities, regardless of predisposing factors.

**Author Contributions:** Conceptualization, J.H.L. and H.W.K.; methodology, J.H.L., H.P. and H.W.K; formal analysis, H.S.L.; investigation, J.H.L. and C.M.R.; data curation, J.H.L. and C.M.R.; writing—original draft preparation, J.H.L.; writing—review and editing, I.R. and H.W.K.; visualization, H.W.K.; supervision, H.W.K.; project administration, H.W.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
