Redisplacement after Closed Reduction in Pediatric Both-Bone Forearm Fractures: The Role of Initial Fracture Angulation and Cast Index

Abstract

Background:

Both-bone forearm fractures are common in children and often treated with closed reduction and casting, though redisplacement remains a concern. This study investigates whether initial fracture angulation and cast index can reliably predict redisplacement and post-treatment angulation outcomes.

Materials and Methods:

A total of 53 skeletally immature patients with closed both-bone forearm fractures were included. Baseline radiographs assessed fracture location, angulation, displacement, and apex direction. Correlation and regression analyses evaluated the relationship between initial angulation, cast index, and angulation at follow-up. A binary logistic regression model assessed the predictive value of cast index and initial angulation for early redisplacement. Additionally, a receiver operating characteristic (ROC) curve was generated to determine the optimal angulation threshold for predicting redisplacement.

Results:

At 6 weeks, the mean residual angulation was 6.63° (standard deviation = 4.92). Redisplacement occurred in 5 patients (10%) within the first 2 weeks. No significant differences were observed in outcomes based on sex, fracture location, or apex type. Cast index correlated strongly with 6-week angulation (r = 0.619), while initial angulation strongly correlated with 2-week angulation (r = 0.623) and moderately with 6-week angulation (r = 0.478). Initial angulation significantly predicted early redisplacement (P < 0.05). ROC analysis showed an AUC of 0.920 (P = 0.02), with an angulation threshold of 23° yielding 100% sensitivity and 80% specificity.

Conclusions:

Initial fracture angulation is a stronger predictor of early redisplacement than cast index. Angulations exceeding 23° significantly increase the risk of redisplacement. Further studies with larger cohorts and longer follow-up are warranted.

INTRODUCTION

Fractures of the radius and ulna shafts, commonly called “both-bone forearm fractures,” are the third most prevalent type of fracture in children, accounting for 13%–40% of all pediatric fractures.[1,2,3,4] These fractures present varying degrees of severity, ranging from plastic deformation to complete displacement, typically resulting from falls onto an outstretched hand.[5] Traditionally, the primary treatment method has been closed reduction followed by immobilization in a molded cast, often achieving acceptable alignment.[6,7] Various indices have been developed to assess the quality of cast molding and alignment, including the cast index,[8] padding index,[9] and the Canterbury index.[9] The cast index, which is the ratio of sagittal to coronal width from the inside edges of the plaster cast at the fracture site, aims to ensure a well-fitted oval-shaped cast for the forearm.[10]

A notable limitation of cast immobilization is the potential for fracture displacement, reported in range of 7% up to 39% of cases.[11,12] Untreated displaced fractures often result in malunion.[13] Although various risk factors for redisplacement have been identified, there is no consensus on the most critical predictors of treatment failure.[14] The risk of fracture redisplacement has been linked to various factors, including greater initial displacement, less surgeon experience, reduction during evening hours, inadequate initial reduction, and poor cast molding and positioning.[15,16,17,18]

Surgical stabilization is a recommended treatment option for children who do not respond to initial conservative treatment.[6,19] Several studies have indicated that intramedullary nailing is an effective treatment for forearm fractures in children.[20,21] This technique is most effective when applied early in the postfracture period as it becomes increasingly challenging to perform once callus formation begins.[22] It would be highly beneficial for clinicians to identify patients who are likely to fail nonoperative treatment early in the process, during the timeframe when surgical intervention is most likely to succeed.[22] Alternatively, remanipulation can be considered to achieve and maintain the reduction of redisplaced forearm fractures.[18]

Although initial fracture angulation is recognized as a key determinant of treatment outcomes in pediatric forearm fractures, the precise angulation threshold at which surgical intervention should be favored over nonoperative management remains uncertain. In this prospective study, our first objective was to evaluate the influence of fracture-related factors such as cast index, initial fracture angulation, fracture location, and apex type on the primary outcomes of bone redisplacement and angulation at the endpoint follow-up. Additionally, we aimed to determine if there is a threshold for initial fracture angulation that can predict redisplacement following casting interventions in pediatric both-bone fractures.

MATERIALS AND METHODS

This prospective cohort study was conducted at *Blinded hospital name and city* and included 53 pediatric patients presenting with both-bone forearm fractures from July 2023 to January 2024. The study protocol was approved by *Blinded Institution’s Name and IRB Approval No*. Written informed consent was obtained from the parents or guardians of all participants, and permission was obtained from children aged 7 years and older.

Participants and data collection

Inclusion criteria consisted of patients under 12 years of age (skeletally immature) who presented with closed injuries or fractures involving both the radius and ulna. Patients with incomplete fractures (such as greenstick fractures), isolated fractures of the radius or ulna, or open fractures or those requiring surgical intervention (defined as unstable fractures) were excluded at the presentation time. A fracture was deemed unstable if the proximal forearm’s full pronation and supination resulted in the fracture’s redisplacement under fluoroscopic imaging.[23] Additionally, cases involving metabolic bone disorders, genetic conditions affecting bone healing, or those with incomplete follow-up were not included. All patients underwent closed reduction followed by immobilization in a cast above the elbow, maintaining the elbow at a 90° flexion. Wool padding was applied uniformly, and no casts were bivalved.

The initial radiographs were examined to evaluate the characteristics of the fractures, including their location, angle, displacement, and the apex where the fracture occurred [Figure 1]. The cast index was determined using digital calipers on high-resolution images of the post-reduction X-rays. This index is calculated by measuring the ratio of the sagittal to coronal widths at the inner edges of the cast at the fracture site [Figure 2]. A cast index value ranging from 0.6 to 0.8 is deemed optimal for ensuring a proper fit and reducing redisplacement risk.[10] However, some patients did not fall within the desired range.

Figure 1.

A sample pediatric case with both-bone forearm fracture

Figure 2.

Cast index measurement: 50.74/75.44 = 0.67

Outcome measures

Follow-up visits took place at 2 and 6 weeks after the reduction to check the healing and alignment of the fracture. X-rays were taken at each visit to assess the angle and alignment of the bones and identify any complications. The primary outcome of interest was the occurrence of redisplacement during the follow-up period. Redisplacement was defined as coronal angulation >15°, sagittal angulation >20° (30° in children younger than 10 years), or translation exceeding 80% on subsequent radiographs. These thresholds are consistent with definitions used in prior studies of pediatric distal forearm fractures.[10,24] Other things we were checking for included complications such as the nonunion, malalignment, and the need for surgical intervention.

Statistical analysis

Data analysis was performed using IBM SPSS Statistics software, version 29.0 (IBM Corp., Armonk, NY, USA). Correlation and regression analyses assessed the relationships between initial fracture angulation, confidence intervals, and displacement. Spearman’s correlation was utilized to evaluate the associations between cast index and initial angulation, post-reduction angulation at 2 weeks, and post-reduction angulation at 6 weeks. Additionally, a binary logistic regression model was employed to determine how well cast index and initial fracture angulation predict the outcome of displacement. The goodness-of-fit of the logistic regression model was assessed using the Hosmer-Lemeshow test, which confirmed the model’s adequacy in fitting the data. Furthermore, a receiver operating characteristic (ROC) curve was plotted to ascertain the optimal initial fracture angulation threshold for predicting displacement. Odds ratios (ORs) with 95% confidence intervals were calculated to estimate the strength of these associations. A P < 0.05 was considered statistically significant.

RESULTS

A total of 49 participants were included in the final analysis. Figure 3 presents a flow diagram illustrating participant recruitment throughout the study. Among the participants, 77.5% were boys and 22.5% were girls, with ages ranging from 2 to 12 years and a mean age of 7.47 [Table 1]. The majority of fractures occurred in the distal part of the forearm (48.9%), followed by the middle part (42.8%) and proximal fractures (8.3%). Most fractures were attributed to the volar apex (71.4%), while a smaller proportion were associated with the dorsal apex (28.6%). The mean cast index at the reduction time was 0.8 (standard deviation [SD] = 0.09). The study found no significant differences in post-reduction angulation or redisplacement at the 6-week follow-up when stratifying patients by sex, fracture location, or apex type. Detailed findings can be found in Tables 2 and 3.

Figure 3.

Flow diagram of participant recruitment

Table 1.

Patient and fracture-related characteristics

Variables (subgroups)	Number of patients	Age	Cast index	Initial fracture angulation
Sex
Boy	38	6.91	0.8	16.24
Girl	11	7.63	0.81	16.09
Fracture location
Proximal	4	7.75	0.82	11.50
Middle	21	6.52	0.82	17.55
Distal	24	8.25	0.79	16.04
Apex direction
Volar	35	7.37	0.80	16.37
Dorsal	11	7.71	0.82	15.79
Total	49	7.47	0.80	16.20

Variables are presented as mean value

Table 2.

Incidence of redisplacement within the first 2 weeks

Variable	Subgroups	Redisplacement	P
Sex	Boy	5 in 38	0.57a
	Girl	0 in 11
Fracture location	Proximal	0 in 4	0.78
	Middle	3 in 21
	Distal	2 in 24
Apex direction	Volar	3 in 35	1.00
	Dorsal	2 in 11

^aAll statistical comparisons were conducted using Fisher’s exact test

Table 3.

Effect of variables on post-reduction angulation at 6 weeks

Variable	Subgroups	Mean angulation (SD)	Significance
Sex	Boy	7.08 (5.16)	P=0.45a (95% CI: −5.36 to 1.39)
	Girl	5.09 (3.78)
Fracture level	Proximal	6 (4.55)	P=0.36b (95% CI: 3.48 to 9.78)
	Middle	7.81 (4.85)
	Distal	5.71 (5.02)
Apex direction	Volar	6.86 (5.08)	P=0.70a (95% CI: −2.37 to 3.94)
	Dorsal	6.07 (4.65)

Data are presented as mean and SD. Statistical comparisons were conducted using independent samples t-test (a) and one-way ANOVA (b), as appropriate. CI=Confidence interval, SD=Standard deviation

The mean initial angulation of fractures in patients was 16.2°, ranging from 2° to 30° (SD = 9.69). After 2 weeks, the mean residual post-reduction angulation was recorded at 3.84°, ranging from 0 to 15° (SD = 3.94). However, by the 6-week mark, the mean residual angulation had increased to 6.63°, with values ranging from 0° to 19° (SD = 4.92). Notably, fracture redisplacement was observed in 10% of cases (n = 5), necessitating manipulation under anesthesia (MUA) for proper alignment. Regarding cast index, we identified 16 patients with cast index values exceeding the optimal range of 0.6 to 0.8. Among these patients (cast index >0.8), only two experienced redisplacement. Comparison of redisplacement rates between this subgroup and patients with cast index values within the optimal range revealed no statistically significant difference (P = 0.71).

Regression analysis

A correlation regression analysis was performed to investigate the relationship between cast index and angulation, as shown in Table 4. The findings revealed a strong correlation between cast index and post-reduction angulation in 6 weeks and between initial fracture and post-reduction angulation in 2 weeks. Additionally, a moderate correlation was observed between initial fracture angulation and post-6 weeks angulation.

Table 4.

Relationship between baseline variables and post-reduction angulation

Variable	Angulation at 2 weeks	Angulation at 6 weeks
Cast index	No correlation (P=0.22, CC: 0.18)	Strong correlation (P=0.00, CC: 0.63)
Initial fracture angulation	Strong correlation (P=0.00, CC: 0.62)	Moderate correlation (P=0.00, CC: 0.48)

Spearman’s rank correlation test was used for all analyses. CC=The correlation coefficient (Spearman’s rho)

We conducted a binary logistic regression analysis to assess the predictability of the cast index and initial fracture angulation on redisplacement outcomes. The results of the Hosmer-Lemeshow test (χ² = 0.62, P = 1.00) demonstrate that the logistic regression model provides a good fit for the data, ensuring the reliability of the outcomes. This nonsignificant result indicates no substantial discrepancy between the observed values and those predicted by the model. Our analysis revealed that initial fracture angulation is a statistically significant predictor of displacement, with an OR of 1.33 (P = 0.02). Although the cast index suggested a considerable OR and effect size (effect size = 18.86), it did not reach statistical significance (P = 0.07).

Determining optimal initial fracture angulation thresholds for predicting displacement

Initial fracture angulation was found to be a statistically significant predictor of displacement. As a result, to determine the optimal threshold for predicting redisplacement, an ROC curve was constructed. The area under the curve (AUC) was 0.920 (P = 0.02), indicating excellent discriminative performance [Figure 4]. An angulation cutoff of 23° yielded a sensitivity of 100% and a specificity of 80%. At this threshold, the positive predictive value (PPV) was 36%, while the negative predictive value (NPV) reached 100%.

Figure 4.

ROC curve revealed the most sensitive and specific angulation for predicting displacement. AUC = 0.920 (P = 0.02). An angulation of 23° had 100% sensitivity and 80% specificity in identifying displacement cases. AUC = Area under the curve, ROC = Receiver operating characteristic

Complications

All five cases that experienced redisplacement within the first 2 weeks underwent MUA and were followed up for 3 months. Each case achieved bone union, as confirmed by an X-ray. Additionally, during physical examinations, none of the patients reported any loss of range of motion.

DISCUSSION

Our prospective study of 49 pediatric patients with complete both-bone forearm fractures demonstrated that initial fracture angulation is a significant predictor of early redisplacement following closed reduction and casting. Redisplacement occurred in 10% of cases, all within the first 2 weeks. Initial fracture angulation showed a strong correlation with post-reduction angulation at 2 weeks (r = 0.62, P < 0.001) and a moderate correlation with 6-week angulation (r = 0.48, P < 0.001). While the cast index (mean = 0.80) correlated with 6-week angulation (r = 0.63, P < 0.001), it did not independently predict redisplacement in logistic regression analysis. Sengab et al. conducted a meta-analysis of 12 studies with 1256 pediatric patients to identify risk factors for redisplacement after reduction and casting of distal forearm fractures.[17] Redisplacement rates ranged from 9.7% to 35%, with 61% requiring secondary intervention. Significant risk factors included complete initial displacement (OR = 4.69), both-bone fractures (OR = 1.95), and nonanatomical reduction (OR = 0.14), while cast index, patient age, sex, and surgeon experience were not associated with redisplacement.[17] Also, in our study, patient sex, fracture location, and apex direction showed no significant association with redisplacement risk.

Previous studies indicate that a significant proportion of redisplacement cases following initial reduction occur within the first 2 weeks of follow-up.[25] Also, a heightened risk of redisplacement was observed early during nonoperative treatment, with reported redisplacement rates ranging from 7% to 34% within the first 2 weeks following reduction.[3,19,26,27,28,29] Bowman et al. evaluated 282 pediatric both-bone forearm fractures treated with closed reduction and casting, reporting a 51.1% failure rate, mostly within 3 weeks.[28] Multivariate analysis identified age ≥10 years (OR = 2.79), proximal-third radius fractures (OR = 6.81), and ulna angulation <15° (OR = 2.94) as significant predictors of failure (P < 0.001). Furthermore, inadequate initial reduction was associated with an 84% failure rate.[28] In our investigation, five cases (10%) experienced redisplacement and all of them occurred within the first 2 weeks of monitoring. Also, we identified initial fracture angulation as a substantial predictor of redisplacement, while cast index did not demonstrate significance.

Typically, in younger children, fractures closer to the distal end and lesser fracture angulation are associated with more favorable outcomes without surgical intervention.[30,31] Kralj et al. conducted a retrospective study to determine whether positioning the wrist joint to counteract the angulation of fracture fragments is more critical than the cast index in preventing redisplacements in pediatric distal forearm fractures.[32] Patients were categorized based on the cast index (<0.9 vs > 0.9). In both cast index categories, a significantly higher rate of redisplacement was observed among patients whose wrists were positioned neutrally, compared to those whose wrist positioning counteracted the fracture angulation. These differences were statistically significant (P = 0.0278 and P = 0.0302, respectively). These findings suggest that appropriate wrist positioning plays a more decisive role than cast index in preventing redisplacement.[32] In our study, we found that, although cast index is significant, initial fracture angulation may be more critical in determining fracture stability. ROC analysis identified an angulation threshold of 23°, above which the risk of redisplacement significantly increased, with an AUC of 0.920 (P = 0.02), yielding a PPV of 0.36 and NPV of 1.

While initial fracture angulation is a critical factor, the specific degree of angulation at which internal fixation should be prioritized over nonoperative methods remains a subject of ongoing debate.[33] A study by van Delft et al. examined the management of displaced metaphyseal forearm fractures in children, focusing on redisplacement and intervention thresholds.[34] Among 200 patients, reduction and casting in the emergency room succeeded in 83% of cases, with only 5% experiencing secondary displacement. However, 47% of those treated with reduction alone in the operating room had secondary displacement, 80% requiring reintervention with K-wire fixation, which entirely prevented redisplacement.[34] In another study, Eismann et al. retrospectively analyzed 31 pediatric patients with redisplaced both-bone forearm fractures managed by re-reduction.[18] They found that re-reduction significantly corrected angulation to <5°, maintained at union, with 87% achieving radiographic improvement and no complications. Re-reduction was also more cost-effective than surgical stabilization, costing less than half.[18] A recent meta-analysis conducted by Sengab et al. indicated that children with both-bone forearm fractures experience the tremendous benefits from reduction combined with primary K-wire fixation.[17]

Limitations

The present study has several limitations that should be considered when interpreting its findings. The most significant limitation is the relatively small sample size (n = 49), which resulted from our prospective study design and strict inclusion criteria. We included only complete both-bone forearm fractures, excluding greenstick fractures and isolated single-bone fractures. Furthermore, although we measured the cast index, we did not assess other indicators of casting quality, such as the padding index or Canterbury index, which could have provided additional insights into the association between casting quality and treatment outcomes. Finally, the limitations of the ROC analysis should also be acknowledged, particularly considering the small sample size, which may affect the robustness and generalizability of the derived cutoff values.

CONCLUSIONS

Our study suggests that initial fracture angulation is a more dependable predictor of redisplacement in the first 2 weeks than the cast index. Patients exhibiting an initial angulation more significant than 23° showed a markedly increased risk of redisplacement, indicating that closed reduction and casting alone may not suffice to maintain alignment in these cases. We recommend that future studies incorporate larger sample sizes and extended follow-up periods to refine these thresholds and investigate long-term outcomes further.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.