CPT is a troublesome disease in pediatric orthopedics because of segmental skeletal dysplasia, and is often accompanied by nonunion and tibial deformity. At present, the commonly used surgical treatment methods mainly include intramedullary nail fixation, Ilizarov external fixation, combination of intramedullary nail and Ilizarov external fixation, as well as vascularized fibula transplantation [14, 15]. Because vascularized fibula transplantation tends to form pseudarthrosis, and single internal fixation or external fixation cannot ensure stability to help fracture healing, combination of intramedullary nail and Ilizarov external fixation are preferred clinically in order to obtain a higher fracture fusion rate. 15 patients with CPT were treated by Agashe et al. with combination of intramedullary nail and Ilizarov external fixation, of which, 14 patients had bony union, while 7 patients suffered from ankle valgus deformity . A total of 56 patients with CPT were treated by our hospital in early years by intramedullary nail combined with encapsulated bone graft and Ilizarov external fixation. After long-term follow-up with an average of 8.5 years, it was found that the fracture healing rate reached 89.2%, but with the incidence of ankle valgus deformity of 17.9% . The patients with CPT are prone to the proximal migration of distal fibula and ankle valgus deformity after ankle intramedullary nailing. Therefore, it is recommended to closely monitor the ankle function in the long-term treatment of patients with CPT.
As early as the 1940s, Phemister first proposed the idea of treating children's bone deformity by epiphyseal plate block . Subsequently, Stevens further proposed the concept of "guiding growth", that is to restrain the growth of epiphyseal plate of one side by internal fixation under engineering mechanics, and to retain the normal growth of epiphyseal plate of the opposite side, so as to correct bone angulation [9, 18]. Currently, the commonly-used internal fixation materials for guiding growth mainly include “8” or “U”-shaped tension band screws and epiphyseal plate screws [13, 19]. We conducted an experiment by using “U”-shaped tension band screws, cortical bone screws and hollow screws respectively in 41 patients with postoperative ankle valgus deformity of CPT from December 2010 to July 2019. It was found that the operation was safe and simple, and there was no need for fixation and load limitation. The results showed that both “U”-shaped tension screws and epiphyseal plate screws could gradually correct ankle valgus deformity in patients with immature bone. The comparison of the three kinds of internal fixation indicated that the “U”-shaped tension screw could correct at a faster speed (0.71°/month), compared with hollow screws (0.64°/month) and cortical bone screws (0.61°/month). But there was no statistical difference. Furthermore, comparing the age of the patients in the three groups, it was found that the average age of the patients in the "U"-shaped tension screw group was the smallest. The study suggests that the correction of ankle valgus deformity is significantly related to the age of patients, and young patients often have better correction improvement [9, 20]. Dirscoll et al.  reported that a total of 42 patients with the average age of 10.3 years who suffered from ankle valgus deformity were treated with epiphyseal screws and tension band screws. They found that the average correction rate of epiphyseal screws was 0.55°/month, the average correction rate of tension band screws was 0.36°/month. Because of the early onset of CPT, ankle valgus deformity occurs at a young age after surgical treatment. The average age of “U”-shaped tension screw group was 7.04 years old, and that of hollow screw group and cortical bone screw group was 7.46 years old and 7.33 years old respectively. Therefore, the correction effect was better than that reported by other scholars.
In this study, the self-developed “U”-shaped tension screws were selected, which is an improved version of the widely used Blount-staples. The head of the screw has three shallow indentations, and the transverse part and the longitudinal axis form 24-30° angle. It has the advantages of close fitting the tibia and strong holding force. The direction of “U”-shaped tension screw implantation is required to be parallel to the growth plate to ensure that the depth of implantation is about half of the diameter of the tibia. At the same time, the “U”-shaped screws with appropriate angle are selected, and the shorter side is implanted into the proximal, so as to fit the medial cortical bone better. As for the position of the epiphyseal screws, it is recommended that it should be implanted vertically to the growth plate as far as possible, and placed in the middle of the medial malleolus, while it is noted that it should not be inserted into the joint cavity. After the removal of internal fixation, there was no injury of epiphyseal plate which affected the growth of limbs for all the patients.
The ages and growth rate of different patients are not consistent, so it is difficult to accurately predict the individual correction rate. Therefore, close imaging follow-up is necessary for the patients to receive distal tibial hemiepiphyseal, which should be continued until bone maturity. Hemiepiphyseal with epiphyseal plate screws can lead to complications such as over-correction or screw buried by bone [19, 22]. Out of the 18 patients who were performed with epiphyseal plate screw block, 1 patient had the complication of screw ingrowth into the bone, and 1 patient with cortical bone screws and 1 patient with hollow screws had the loss of correction effect after removal of internal fixation. The tibiotalar angles of the 2 patients with the loss of correction effect were greater than 80°, which could be corrected by the range of podarthrum motion. “8”-shaped tension band plate combined with screw fixation is also the most commonly used hemiepiphyseal material, which is commonly used in the distal femur and proximal tibia. Considering less soft tissue and lack of muscle tissue coverage in the medial ankle, the “U”-shaped tension screws with lower notch were selected. Previously, the Blount-staples were often used to be inserted into the epiphysis of children who were not fully ossified, and with their growth, staples could withdraw [13, 23]. However, due to the increase of the screw width and the indentations of the head, there was no screw withdrawal phenomenon in all 23 patients treated with “U”-shaped tension screw. At the same time, additional concern was required on the wound infection of the patients treated with tension band screws, especially the thinner patients. There was no wound infection in any of our patients treated with “U”-shaped tension screws, but there were 2 cases with obvious postoperative pain in the internal fixation position, and the symptoms disappeared after oral medication and physical therapy.
Temporary hemiepiphyseal by using cortical bone screws, “U”-shaped tension screws, hollow screws, etc., is an effective treatment for postoperative ankle valgus deformity in patients with CPT. At the same time, the "U"-shaped tension screw provides relatively better orthopedic results and has a lower rate of internal fixation complications. However, our study has some limitations. The effect of different age groups on the correction rate was not quantified. Subsequently, more strict grouping and longer-term follow-up should be set up.