Introduction

Distal tibia physeal fractures are a relatively common physeal injury and are the second most common injury in children after the distal radius physis [1, 2]. Distal tibia physeal fractures account for approximately 11–20% of all physeal fractures [3, 4]. The stronger ligamentous attachments make the physis more vulnerable to injury and more likely to be accompanied by subsequent premature growth arrest, which causes angular deformity and differences in leg length 1–2 years after injury [1, 5, 6]. Growth disturbance is defined as differences in leg length equal to or greater than 1.5 cm or varus or valgus deformity more than 5 degrees [7]. According to different studies, the incidence of growth disturbance varies as between 2 and 40% [8,9,10].

Multiple factors, including the cause of injury, fracture type and location, initial displacement, Salter–Harris classification, number of attempted reductions, open or closed reduction, and residual displacement after the intervention, are thought to be associated with growth disturbance [6, 11,12,13]. However, the factors affecting growth disturbance are still unclear. Therefore, this study aims to assess the factors related to growth disturbance after distal tibia physeal fractures in children.

Patients and methods

Data about children with distal tibia physeal fractures who were treated in Fuzhou Second Hospital Affiliated with Xiamen University between September 2015 to September 2018 were collected. All patient files were reviewed with the approval of the hospital. Patients were placed in the growth disturbance group when the difference in their leg lengths was equal to or greater than 1.5 cm or more than 5 degrees of varus or valgus deformity. Otherwise, they were placed in the no-growth disturbance group.

All patients were further divided into early and delayed operation groups. According to Rang and Salter [14, 15], patients were classified in the early operation group when the time from injury to operation was equal to or less than 7 days, and classified in the delayed operation group when it exceeded 7 days.

The following data were extracted from the database: age, gender, affected side, Salter–Harris classification, initial maximal displacement, the presence of fibula fracture, open or closed reduction, method of fixation, time to surgery, blood loss, duration of operation, and complications.

The inclusion criteria were as follows:

  1. 1.

    Age less than 16 years.

  2. 2.

    Diagnosed with distal tibia physeal fracture.

  3. 3.

    Residual displacement less than 2 mm after surgery, measured by scale with a 2 mm diameter steel ball or K-wire nest to the ankle joint.

  4. 4.

    Follow-up more than 12 months.

The exclusion criteria were as follows:

  1. 1.

    Peripheral vascular diseases.

  2. 2.

    Neuromuscular diseases.

  3. 3.

    Metabolic bone disorders.

  4. 4.

    Pathological fractures.

  5. 5.

    Residual displacement more than 2 mm after surgery.

SPSS 22.0 software was used for all statistical analyses. Descriptive statistics were conducted for continuous variables using mean ± standard error (SE). A comparison test between two independent or more than two groups was made using a two-sided Student’s t-test or one-way analysis of variance at a 5% significance level (2.5% on each side). Descriptive statistics of categorical variables were represented as frequency and percentage. A Chi-square test was used to compare the differences in rates among different groups at a 5% significance level (2.5% on each side).

Results

A total of 143 patients (96 boys and 47 girls), with an average age of 11.19 ± 3.34 (range from 2.8 to 16) years, were enrolled in this study. The follow-up periods averaged 29.34 ± 7.46 months (26 to 61 months). The overall incidence of growth disturbance was 15.39% (22/143); among them, 21 cases had varus or valgus deformities greater than 5 degrees, and 1 case had a limb shortening exceeding 1.5 cm. However, no infection, nonunion, or osteonecrosis was observed in any patient.

The no-growth disturbance group contained 121 patients (84 boys and 37 girls), with a mean age of 11.50 ± 3.20 years, and 68, 51, and 1 cases on the left, right, and bilateral sides, respectively. The average maximal displacement was 5.51 ± 3.18 mm. The growth disturbance group contained 22 patients (12 boys and 10 girls) with a mean age of 9.32 ± 3.56 years, and 12, 10, and 0 cases on the left, right, and bilateral sides, respectively. The average maximal displacement was 5.77 ± 4.89 mm, and the average comprehensive displacement was 5.05 ± 2.38 mm. There was a significant difference in age between the two groups (p = 0.004), but no significant difference in the factors described above (Table 1).

Table 1 The data of patients between no-GD group and GD group
Full size table

According to Salter–Harris classification, patients with type I/II/III/IV fracture were 7/59/9/9 in the normal group and 3/7/4/4 in the growth disturbance group. As transitional fractures, there were 31 Triplane fractures and 6 Tillaux fractures in the no-growth disturbance group, and 4 Triplane fractures and 0 Tillaux fractures in the growth disturbance group (Table 2). More Salter–Harris types III and IV patients were in the growth disturbance group than in the no-growth disturbance group (p = 0.043) (Fig. 1). The SH grade analysis showed that growth disturbance occurred 3/10 (30%) in SH grade I, 7/66 (10.61%) in SH grade II, 4/13 (30.77%) in SH Grade III, and 4/13 (30.77%) in SH grade IV (Table 2).

Table 2 Patients whose fracture type was registered according to Salter–Harris classification and transitional fractures
Full size table
Fig. 1
figure 1

The risk of growth disturbance between Salter–Harris types. (P, Triplane fractures; T, Tillaux fractures)

Full size image

Among the no-growth disturbance group, 27.27% (33/121) of patients also had fibula fractures, while 45.45% (10/22) of patients in the growth disturbance group also had fibula fractures, which was significantly different (p = 0.011). Differences in open or closed reduction, type of fixation, duration of operation, and blood loss were not significant between the two groups. Forty-four cases received K-wires, 76 received screws, and 1 received plates in the no-GD group; 8 cases received K-wires, 14 received screws, and 0 received plates in the GD group. Differences between the two groups were not significant.

There was no significant difference in growth disturbance between the early group (18/123) and the delayed operation group (3/20) (p > 0.05) (Figs. 2, 3).

Fig. 2
figure 2

a An 11-year-old patient, Salter–Harris type II, distal tibia physeal fracture accompanied by fibula fracture, accepted open reduction 17 days after injury, osteotyluss growth can be seen in CT scan showed in the red frame. Growth disturbance has not been observed during 39 month follow-up. b An 11-year-old boy, Salter–Harris type III Distal tibia physeal fracture, accepted open reduction 4 days after injury. Growth disturbance has been observed at 13-month follow-up, the lateral angle of the distal tibia was 98° at 26-month follow-up

Full size image
Fig. 3
figure 3

Flow diagram detailing the methods used to identify study subjects

Full size image

Discussion

Like some fractures, physeal injuries can be complicated by malunion, infections, neurovascular problems, or osteonecrosis. The incidence of growth disturbance after distal tibia physeal fractures was variable based on previously published findings, ranging from 2% [8] to almost 40% [9]. In this study, the overall incidence of growth disturbance was 15.39% (22/143), which was within the range of previous research [8, 9, 16].

Growth disturbance has been related to many different factors, including the cause of injury, fracture type and location, initial displacement, Salter–Harris classification, number of attempted reductions, open or closed reduction, quality of reduction, and the different composition characteristics of patients. However, few articles have reported the effects of age on growth disturbance after distal tibia physeal fractures. Our study found that age was an important factor affecting growth disorders. The average age of the growth disturbance group was significantly lower than that of the no-growth disturbance group. This could be due to the greater growth potential of younger children and a greater effect on growth after a fracture. Additionally, there were 4 Triplane fractures and 0 Tillaux fractures among the 22 patients with malformations, which could occur in older adolescents. This suggests that age was an important factor in growth disorders after distal tibia physeal fractures.

The growth disturbance in grade I was similar to that in grade III/IV, probably because there were so few cases in grade I. Salter–Harris types III and IV fractures displayed a higher growth disturbance, which could be due to injuries to the germinal layer of the physis [17]. Our study also found that there were more Salter–Harris types III and IV patients in the growth disturbance group than in the no-growth disturbance group (p = 0.043) (Fig. 1).

We also found that the growth disturbance group (12/22) (Table 3) was more likely to also have fibula fractures than the normal group (33/121) (p = 0.011). Patients with fibula fractures could have suffered more significant injuries. Significant injuries often manifest as larger initial displacements but can also be accompanied by other fractures.

Table 3 The data of patients in growth disturbance (GD) group
Full size table

Patients were classified either into an early operation group or a delayed operation group based on the methods used by Rang and Salter [13, 14]. They suggested that delayed reduction makes it more difficult to achieve fracture reduction and places the viability of the physis at risk when forceful reduction is required. However, Egol KA reported that delayed reduction showed no evidence of physeal damage, physeal growth disturbance, or radiographic bar formation in a rat model study [18]. Our study included closed or open reduction in 20 delayed operation patients (mean 12.8 ± 6.68 days, range from 8 to 37 days). Differences in growth disturbance were not significant compared with early operation patients (Fig. 2).

This study had some limitations. It was a retrospective study, and the number of cases was limited. Growth disturbance from a physeal fracture is typically evident 2 to 6 months after the injury, but it may not become obvious for up to 2 or more years after the injury. Therefore, follow-ups should be extended to near skeletal maturity. However, Stenroos et al. [19] suggested that routine radiographic follow-up is unnecessary after physeal fractures of the distal tibia in children. We believe this study will significantly improve prognosis in children after distal tibia physeal fractures.

Conclusion

Patients in the growth disturbance group were significantly younger and more likely to be Salter–Harris types III and IV than patients in the no-growth disturbance group. Patients who also had fibula fractures could have a higher incidence of growth disorders.