Body Weight at Birth: The Only Risk Factor Associated with Contralateral Clavicular Fracture in Patients with Congenital Muscular Torticollis

To date and to the best of our knowledge, there have been limited studies on the risk factor of clavicle fracture combined with congenital muscular torticollis (CMT), despite it being the most common fracture in newborns. So, the aim of this study was to investigate the risk factors associated with clavicular fracture combined with CMT, and its effect on prognosis. In this study, a total of 134 infants with CMT were included. The risk factors associated with clavicular fracture combined with CMT were analyzed. To analyze the correlation between the clinical parameters and the clavicular fracture in patients with CMT, demographic data, such as body weight at birth, maternal age, gender, gestational age, delivery method, sternocleidomastoid (SCM) thickness of ipsilateral side, its ratio between the ipsilateral and contralateral side, and the first visitation date after birth were evaluated. In the results of this study, the clavicular fracture was found in 15 of 134 patients with CMT (19%). In multivariate logistic analysis, the body weight at birth was the only significant parameter for predicting clavicular fracture in patients with CMT (p-value < 0.05). However, there was no significant difference of treatment duration between CMT infants with or without clavicular fracture. In infants with CMT, the area under the ROC curve of the body weight at birth for predicting clavicular fracture was 0.659 (95% CI, 0.564–0.745.; p < 0.05). The optimal cut-off value obtained from the maximum Youden index J was 3470 g (sensitivity: 57.14%, specificity: 75.76%), and the odd ratio of clavicular fracture in patients with CMT increased by 1.244 times for every 100 g of body weight at birth. In conclusion, birth weight appears to be a clinical predictor of clavicular fracture in infants with CMT. More studies and discussions are needed on whether any screening should be recommended for detecting the concurrent clavicular fracture in subjects with CMT.

perinatal torticollis occurring concurrently with neonatal BPI ranged from 2% to 19% 11 . To date and to the best of our knowledge, there have been limited studies on the risk factor of clavicle fracture combined with CMT, despite it being the most common fracture in newborns 12 . Although one previous study showed that clavicular fracture tends to develop on the contralateral side of CMT, the risk factors of clavicular fracture in patients with CMT have not been fully evaluated 6 . Therefore, the aim of this study was to investigate the risk factors associated with clavicular fracture in infants with CMT and its effect on prognosis.
Material and Method ethics statement. This study was approved by the Institutional Review Board of Daegu Fatima Hospital.
Declaration of Helsinki protocols are being followed, and informed consent was obtained from a parent and/or legal guardian.
patients. Between January 2016 and June 2018, subjects who visited our rehabilitation outpatient clinic due to abnormal posture of the head and neck were included. The medical records along with radiological findings were reviewed. Among them, subjects who met the following criteria were excluded: (1) subjects with no specific finding on ultrasonography; (2) subjects who did not undergo plain radiography of the cervical spine and/or clavicles.
CMT was diagnosed when subjects met the following criteria: (1) thickness of the involved SCM ≥ 2 mm greater than that of the contralateral side, along with increased echogenicity on ultrasonography 13 ; and (2) subjects who showed shortening of the unilateral SCM since childhood, ending up with a limitation of passive range of rotation of the chin toward ipsilateral shoulder and/or limitation of passive range of lateral flexion toward contralateral shoulder. Plain radiographs (X-ray) of the cervical spines and/or clavicles were evaluated, using the reference set to any clavicular fracture and any structural abnormalities on cervical spine. Clavicular fracture was diagnosed when the fracture lines and/or callus were detectable by a naked eye on the clavicle on antero-posterior plain radiographs of the cervical spines and/or clavicles 6 and confirmed by radiologists who were specialized at musculoskeletal disease. Therefore, we retrospectively reviewed the medical records of 134 patients with congenital muscular torticollis in the rehabilitation unit.
parameters associated with clavicular fracture in cMt patients. Demographic data, such as body weight at birth, maternal age, gender, gestational age, delivery method, SCM thickness of ipsilateral side, and its ratio between the ipsilateral and contralateral side, treatment duration, in addition to the first visitation date after birth, were collected by reviewing the medical records.
Statistical analysis. To find the difference of the demographic data between CMT patients with and without clavicular fracture, an independent T-test, Fisher's exact test, or chi-square test were performed. In addition, to analyze the correlation between the clinical parameters and the clavicular fracture in patients with CMT, multivariate logistic analysis through forward stepwise selection was then performed. To evaluate the accuracy of predictive factors for clavicular fracture in CMT patients, we performed a receiver operating characteristic (ROC) analysis in each group. All statistical analyses were conducted using MedCalc and SPSS version. 22.0 (IBM, Armonk, NY, USA).

Results
Demographics and concurrence rate of cMt and clavicular fracture. Between January 2016 and June 2018, 449 patients visited our rehabilitation outpatient clinic due to abnormal posture of the head and neck. The medical records along with radiological findings were reviewed. Among them, 315 subjects who met the exclusion criteria were excluded ( Fig. 1). Finally, a total 134 infants with CMT were included in this study. Moreover, the clavicular fracture was found in 15 of 134 infants, with the concurrent rate being 11.19% (Table 1). contralateral involvement of cMt and clavicular fracture. CMT and clavicular fracture occurred on the opposite side of each other in 13 out of 15 subjects (86.7%; Table 2). Table 2 is the contingency table between the side of CMT and the side of clavicular fracture. In chi-square analysis, discordance of side between CMT and clavicular fracture was significant (P = 0.004).
Risk factor of clavicular fracture in infants with cMt. In a comparison of the demographic data between CMT infants with or without clavicular fracture, there was a significant difference in delivery mode (p-value < 0.05) ( Table 1). There was a significant difference in the birth weight between CMT infants with or without clavicular fracture (p-value < 0.05) ( Table 1). However, there was no significant difference in the maternal age, gestational age, SCM thickness ratio, first visitation to the clinic, and gender. In a multivariate logistic analysis, the birth weight was the only significant parameter for predicting clavicular fracture in CMT infants (p-value < 0.05) ( Table 1). In infants with CMT, the area under the ROC curve of the birth weight for predicting clavicular fracture was 0.659 (95% CI, 0.564-0.745.; p < 0.05) ( Table 3). The optimal cut-off value obtained from the maximum Youden index J was 3470 g (sensitivity: 57.14%, specificity: 75.76%), and the odd ratio of clavicular fracture in CMT infants increased by 1.244 times for every 100 g of birth weight (Fig. 2). clinical parameters correlated with treatment duration. There was no significant difference of treatment duration between CMT infants with or without clavicular fracture. (Table 1) In a multivariate regression analysis, the SCM thickness ratio was the only significant parameter for predicting treatment duration in infants with CMT (p-value < 0.05) ( Table 4).

Discussion
The causes of CMT remain contentious to date 2 . Multiple theories exist, including intrauterine crowding 8,14 or fibrosis from peripartum bleeding, vascular phenomenon, primary myopathy of the SCM, and compartment syndrome 15,16 . A difficult birth history has been reported in 30-60% of patients with CMT 17,18 . In a study of 996 patients with CMT, Yim et al. 6 suggested a hypothesis that CMT may likely develop during vaginal delivery. For an effective expulsion of the baby's head during vaginal delivery, the antero-posterior axis of baby's head needs   to be parallel with the antero-posterior axis of the mother's pelvis, by simultaneous internal rotation of both the head and trunk of the baby, in addition to neck flexion 6 . Moreover, external rotation of the shoulder occurs so that the right-left axis of the baby's shoulders becomes parallel to the antero-posterior axes of the mother's pelvis for expulsion of the baby's shoulder 6 . However, Yim et al. 6 suggested that an isolated internal rotation of the baby's head that occurs during head expulsion, instead of simultaneous internal rotation of both the head and shoulder, can cause overstretching and damages to SCM. Moreover, they also suggested that a downward traction of the assistive maneuver, which may facilitate delayed delivery of the baby's shoulder, may cause fracture of the clavicle (contralateral side of damaged SCM), which is in the anterior shoulder. As aforementioned, clavicular fracture is significantly correlated with vaginal delivery and body weight at birth. The suggestions made by Yim et al. and our findings support that clavicular fractures in patients with CMT are associated with difficult vaginal delivery. Interestingly, however, there was no significant difference of treatment duration between CMT infants with or without clavicular fracture. In this study, the only risk factor correlated with treatment duration was SCM thickness ratio. Although other theories aside from difficult vaginal delivery, such as vascular phenomenon, primary myopathy of the SCM, compartment syndrome, hereditary hypothesis, and infection, have also been proposed, the pathogenesis of CMT remains uncertain. Injury to the SCM muscle can occur as a result of muscle disease, such as muscular dystrophy, exposure to myotoxic agents, ischemia, and exposure to hot or cold temperatures 19 . However, focal myopathy in the SCM muscle may be rare, and neonates of uncomplicated pregnancy may rarely have events causing injury to the SCM muscle, such as exposure to hot or cold temperatures and ischemia. In addition, our data, which shows the correlation between clavicular fracture in patients with CMT and body weight at birth, suggests that mechanical injury to SCM by overstretching -as a result of difficult delivery -could cause ischemia, compartment syndrome, and/or hematoma of SCM that have been known to be related to the  Table 3. Multivariate logistic regression analysis associated with clavicular fracture in patients with congenital muscular torticollis.

Figure 2.
In patients with CMT, the area under the ROC curve of the body weight at birth for predicting clavicular fracture was 0.659 (95% CI, 0.564-0.745.; p < 0.05). The optimal cut-off value obtained from the maximum Youden index J was 3470 g (sensitivity: 57.14%, specificity: 75.76%), and the odd ratio of clavicular fracture in patients with CMT increased by 1.244 times at every 100 g of body weight at birth.  www.nature.com/scientificreports www.nature.com/scientificreports/ development of CMT. However, CMT occurs not only in subjects born through vaginal delivery, but also in those born through cesarean section. Therefore, other causes should not be excluded.
The incidence of clavicular fractures in newborns ranges from 0.01 to 1.65 percent, and the incidence of clavicular fracture in patients with CMT in the study of 996 patients with CMT was 2.01% 6,[20][21][22][23] . Considering the prevalence of CMT, the prevalence of clavicle fractures in CMT patients is very low in infants as a whole. However, in our study, the incidence of clavicular fracture in patients with CMT was 11.19%. This is likely due to two reasons: First, we did not perform routine x-ray examinations for clavicular fractures in all CMT children, but performed in patients with suspected clavicular fractures on physical examination, such as decreased Moro reflex, swelling, mass, tenderness, and crepitation of the affected side. However, in our study, only CMT patients who received x-rays were enrolled; the incidence of clavicular fracture in CMT patients seems to be higher than in previous studies. Second, more complicated SCM patients tend to be admitted to our hospital, as they are referred from local primary clinics due to the nature of the medical system in Korea. However, considering the relatively high incidence rate of clavicular fracture in CMT patients and the association between clavicular fracture and increased body weight at birth, a more thorough evaluation seems necessary. In this study, only a conservative treatment was performed in CMT patients with clavicular fracture, except for the incorporation of stretching exercise and home education for CMT. On follow-up plain radiography, however, most of the CMT patients with clavicular fracture showed a tendency to recover well without deformation of the clavicle (Figs 3 and 4).