A Comparative Study of the Modified Baumann’s Angle Between the Dominant and Non-Dominant Sides in a North Indian Pediatric Population

Background The literature points towards the fact that paediatric elbow fractures happen more frequently and have greater variability when contrasted with adults. Between 65%-75% of pediatric fractures involve the upper extremity, and supracondylar humerus is the most common of them all. To know the exact site of injury and to estimate the degree of reduction after manipulation, the modified Baumann’s angle, the Baumann’s angle, the Humero-condylar angle, the Anterior humeral line, and the Radio-capitellar line are the parameters most commonly used. This study was carried out to compare the modified Baumann’s angle between both upper limbs in the paediatric population. Methodology This cross-sectional study was conducted in a tertiary health care centre in Northern India for one year from September 1, 2021, to August 31, 2022. We included pediatric patients in the age group of 3-16 years. Age, sex, weight, height, BMI, secondary sexual characters, and handedness were noted in all the children enrolled in our study. In both the dominant and non-dominant sides, the mean arm length, the forearm length, the inter-epicondylar distance, the clinical carrying angle, the radiological carrying angle, and the modified Baumann’s angle were calculated. Results A total of 113 children were enrolled in the study. The majority of children (71.7%) had dominance on the right side. In both the dominant side and non-dominant side, mean arm length, forearm length, inter-epicondylar distance, clinical carrying angle, radiological carrying angle, and modified Baumann’s angle values were calculated. On evaluating the data statistically, a significant difference between the two sides was observed for all the parameters (p<0.05), except forearm length (p-value -0.954). Multivariate analysis showed that only BMI was significantly negatively associated with modified Baumann’s angle (p=0.016), and only age (0.019) and BMI (<0.001) were found to be significantly associated with the difference in modified Baumann’s angle. Conclusions The findings of this study will be helpful in the management of elbow disorders and their reconstruction following trauma. A significant difference was found in the modified Baumann’s angle between dominant and non-dominant sides, and it also showed a negative significant correlation with arm length, forearm length, and the presence of secondary sexual characteristics. The equations derived in this study will be helpful in the simple derivation of the modified Baumann’s angle and its difference from simple measurements of the upper limb parameters.


Introduction
Elbow fractures have been seen to have a greater incidence and variability in the pediatric age group as compared to the adults [1].Between 65%-75% of fractures in children affect the upper limb [2], and supracondylar fractures of the humerus have the lion's share of 60% out of all upper limb fractures [3].The percentage of elbow trauma shows a rising trend due to a surge in the levels of participation of this particular age group in recreational and competitive outdoor as well as indoor sports [4].Even during the coronavirus disease 2019 (COVID-19) times, due to the lockdown, children found ways of recreational activities within the four walls of their homes, which further increased the incidence as well as the spectrum of elbow traumas.To assess the exact state of injury and to estimate the degree of reduction after manipulation, certain radiological criteria are employed, which include the modified Baumann's angle and the Baumann's angle in the Anterior-posterior view of the X-ray and the Humero-condylar angle, the Anterior humeral line and the Radio-capitellar line in the lateral view of the X-ray [5].The Baumann's angle was found to be a useful indicator of the adequacy of reduction of a displaced supracondylar humeral fracture while managing such patients by closed reduction and overhead traction of the olecranon [6].Hence, it finds its application as an outcome measure for supracondylar fractures of Humerus in children [7].However, there is limited information about the reliability of this measurement, with widely varying opinions in the existing literature.This study was conducted to compare the modified Baumann's angle in the upper limbs of both sides in the pediatric population.

Materials And Methods
This cross-sectional study was conducted in a tertiary-level health care centre in Northern India for one year from September 1, 2021, to August 31, 2022.We included pediatric patients in the age group of 3-16 years who attended our hospital for some other non-traumatic complaints and children who/whose parents volunteered to participate in the study.We excluded children who had a history of orthopedic Trauma, history of orthopedic tumours, history of bone or joint Infections, history of surgery, and history of burns to either of the upper limbs.We also excluded patients who had post-polio residual paralysis, patients with metabolic bone diseases like rickets, Osteogenesis Imperfecta, Fibrous Dysplasia, patients suffering from neurological conditions with abnormal muscle tone due to cerebral palsy, head injury, or motor neuron disease, patients of arthritis like rheumatoid arthritis, juvenile rheumatoid arthritis, congenital orthopedic deformities, or children who had not developed hand preference at the time of enrollment in the study.
Age, sex, weight, height, BMI, secondary sexual characters, and handedness were noted in all the children enrolled in our study.In both the dominant side and non-dominant sides, the mean arm length, the forearm length, the inter-epicondylar distance, the clinical carrying angle, the radiological carrying angle, and the modified Baumann's angle (on anteroposterior (AP) view of X-ray of the elbow joint) were calculated.The clinical carrying angle was measured with full extension of the elbow and full supination of the forearm.The angle between the central axis of the forearm and arm was measured using manual goniometry.The central axis of the forearm was taken as a line joining the midpoint of the inter-epicondylar line to the midpoint of the inter-styloid line at the wrist and the central axis of the arm was taken as a line joining the midpoint of the inter-epicondylar line to the tip of the acromion process of the arm (Figure 1).

FIGURE 1: Measurement of the clinical carrying angle
The radiological carrying angle was measured on a true AP view of the X-ray of the elbow as the angle between the longitudinal axis of the shaft of the humerus and a longitudinal line drawn along the shaft of the ulna.The axes of the humerus and ulna were determined by two central points on the shafts of the respective bones (Figure 2).

FIGURE 2: Measurement of the radiological carrying angle
The radiological carrying angle is shown in yellow.Modified Baumann's angle was measured on a true AP view X-ray of the elbow.A straight line was drawn through the middle of the humeral shaft by taking two central points on the humeral shaft and a second line was drawn perpendicular to the humeral shaft.A third line was drawn along the lateral condyle growth plate.To measure the inter epicondylar distance, the arm was first lifted to the level of the shoulder and the forearm was flexed to 90 degrees to make the humeral epicondyles prominent and easily palpable.A Vernier calliper was used to measure the inter-epicondylar distance.The fixed arm of the calliper was placed on the lateral epicondyle and the movable arm was then adjusted to the medial epicondyle.Forearm length was measured as the distance from the mid-point of a line joining the radial and ulnar styloid to the mid-point of a line joining the medial and lateral epicondyles of the humerus using a standard measuring tape, with the arm flexed to 90 degrees.The arm length was measured as the distance from the tip of the acromion process to the mid-point of a line joining the medial and lateral epicondyles of the humerus using a standard measuring tape.The height was calculated in a standing, erect, anatomical position barefoot from the vertex to the heel using a measuring tape.
The sample size was calculated as 110, considering Baumann's angle in the common range of 71 -800, with a prevalence of 60%, as observed in the study by Awasthi et al. (2017) [8].Data was analysed using SPSS software, version 24.0 (IBM Corp., Armonk, NY).The ANOVA test, the paired t-test, and the independent samples t-test were used to analyse the data.Pearson correlation coefficient was used to evaluate bivariate linear correlation.Linear regression was performed for multivariate analysis.A p-value less than 0.05 was considered statistically significant.
Ethical clearance was taken from the Institutional Ethics Committee of King George's Medical University (Ref No.: 103rd ECM IIB-Thesis/P34, dated 2020 Dec 10) before beginning the study, and written informed consent was taken from the parents of the children and permission was also taken if the age of the child was above or equal to 18 years, agreeing to participate in this study.

Results
A total of 113 children were enrolled in the study in the age group of 3 to 16 years.The majority of them were boys (n=78, 69.03%), while the remaining (n=35, 30.97%) were girls.The mean age of the children was 8.35±3.52 years.Maximum (n=34; 30.1%) children were in the age group of 9-12 years, followed by those in the 6-8 years age group (n-31; 27.4%), 3-5 years age group (n=30; 26.5%) and >12 years age group (n-18; 15.9%).Among boys too, the maximum (n=25; 32.1%) were aged 9-12 years followed by those aged 6-8 years (n=23; 29.5%), 3-5 years (n=17; 21.8%) and >12 years (n=13; 16.7%) respectively.The mean age of boys was 8.62±3.34years.However, among girls, maximum (n=13; 37.1%) were aged 3-5 years followed by those aged 9-12 years (n=9; 25.7%), 6-8 years (n=8; 22.9%) and >12 years (n=5; 14.3%) respectively.The mean age of girls was 7.74±3.88years.Although the mean age of boys was higher as compared to that of girls, this difference was not significant statistically (p=0.225)(Table 1) (Figure 4).On evaluating the data statistically, a significant difference between the two sides was observed for all the parameters (p<0.05)except forearm length (p=0.954).Among boys, a significant difference between the two sides was observed for all the parameters (p<0.05)except for arm length (p=0.063) and forearm length (p=0.954).Among girls, a significant difference between the two sides was observed for the clinical carrying angle and radiological carrying angle (p<0.001)respectively.For none of the other parameters the difference between the two groups was significant (p>0.05)(Table 3).2023   For all the parameters mean values were higher in males as compared to that in females; however, this difference was significant statistically only for arm length.On the dominant side, for all the parameters mean values were higher in males as compared to that in females; however, this difference was not statistically significant for any of these parameters.On the non-dominant side, for all the parameters mean values were higher in males as compared to that in females; however, this difference was significant statistically for any of these parameters (Table 4).

TABLE 4: Comparison of linear measurements and modified Baumann's angle between males and females
Among those aged 3-5 years, statistically significant differences were observed for arm length (p=0.043),inter-epicondylar distance (p=0.024),clinical carrying angle (p<0.001), and radiological carrying angle (p=0.001)respectively.Among those aged 6-8 years, statistically significant differences were observed for clinical and radiological carrying angles only, which were found to be significantly higher in non-dominant as compared to the dominant side.Among those aged 9-12 years, on the dominant side, statistically significant differences were observed for clinical and radiological carrying angles only, which were found to be significantly higher in non-dominant as compared to the dominant side.Among those aged >12 years, mean clinical and radiological carrying angles were found to be significantly higher in the non-dominant as compared to the dominant side while mean modified Baumann's angle was found to be significantly higher in the dominant as compared to the non-dominant side (Table 5).2023

TABLE 5: Age-wise comparison of linear measurements and modified Baumann's angle between dominant and non-dominant sides
Though with increasing age there was a decreasing trend of modified Baumann's angle values, but it was not significant statistically (p=0.150).Boys had higher mean modified Baumann's angle value (18.25±4.25°)as compared to girls (17.06±4.20°)however this difference was not found to be significant statistically (p=0.053).Left-handed children (17.45±4.48°)had lower mean values as compared to right-handed children (18.05±4.17°)yet this difference was not found to be significant statistically (p=0.344).Those having secondary sex characteristics had lower mean values (17.02±4.38°)as compared to those not having secondary sex characteristics (18.58±4.04°)and this difference was also significant statistically (p=0.006).
The difference in modified Baumann's angle between dominant and non-dominant sides did not show a significant association with age, sex, handedness, and secondary sex characteristics (p>0.05)(Table 6).A weak negative yet significant correlation of modified Baumann's angle was observed with arm length, forearm length, and clinical and radiological carrying angles.There was a weak negative correlation of modified Baumann's angle with inter-epicondylar distance too but this was not significant statistically.Among other bivariate correlations, there was a weak positive correlation of arm length with forearm length, inter-epicondylar distance and clinical carrying angle, and a moderate and significant and positive correlation of forearm length with inter-epicondylar distance, clinical and radiological carrying angles.

Modified Baumann's Angle
There is a strong positive correlation between inter-epicondylar distance with clinical and radiological carrying angles and a strong near-perfect correlation between radiological and clinical carrying angles.
There was no significant correlation of difference in modified Baumann's angle between dominant and nondominant sides with arm length, forearm length, inter-epicondylar distance, clinical carrying angle, and radiological carrying angle (p>0.05)(Table 7) (Figure 5).In a multivariate linear regression model where modified Baumann's angle was projected as a dependent variable on independent variables age, sex, BMI, handedness, secondary sexual characteristics, arm length, forearm length, inter-epicondylar distance and clinical and radiological carrying angles only BMI emerged as the significant variable negatively associated with modified Baumann's angle (p=0.016).The model had a low explanatory power (r2=0.096)thus implying that based on this multivariate model changes in modified Baumann's angle could be explained by 9.6% only.
In a linear regression model, where the difference in modified Baumann's angle between the dominant and non-dominant side was considered as a dependent variable on independent variables age, sex, handedness, secondary sex characteristics, arm length, forearm length, inter-epicondylar distance, clinical carrying angle and radiological carrying angle, only age, and BMI were found to be significantly associated with the dependent variable.The projected model had a limited explanatory ability (r2=0.132)(Table 8).(Where: Age is in years; Sex: Male=1, Female=2; BMI is in kg/m 2 ; Handedness: Left=1, Right=2; Secondary sexual characteristics: Present=1, Absent=2; Arm length is in cm; forearm length is in cm; inter-epicondylar distance is in cm; clinical carrying angle is in degrees; and radiological carrying angle is in degrees).

FIGURE 4 :
FIGURE 4: Age profile of the study population

TABLE 7 : Correlation of modified Baumann's angle and modified Baumann's angle difference between dominant and non-dominant sides
Correlation of modified Baumann's angle and modified Baumann's angle difference between dominant and non-dominant sides with different linear and angular measurements (n=226) (Pearson correlation coefficient).