Reliability of lateral wall-thickness measurement in trochanteric fractures and parameters associated with lateral wall integrity
Article Category: Brief Communication
Published Online: Mar 20, 2024
Page range: 24 - 29
DOI: https://doi.org/10.2478/abm-2024-0005
Keywords
© 2024 Chindarat Ratanakornphan et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
The AO Foundation/Orthopaedic Trauma Association (AO/OTA) Fracture and Dislocation Classification Compendium introduced a new trochanteric fracture classification in January 2018, the previous edition having been introduced in 2007. The major difference between the 2007 and 2018 classifications is the significance of the lateral wall thickness of the femur. The 2007 classification divides the fractures into the simple pertrochanteric fracture (31A1), the multifragmentary pertrochanteric fracture (31A2), regardless of the lateral wall thickness, and the intertrochanteric fracture (31A3). However, in the 2018 edition, trochanteric fractures are classified as pertrochanteric fractures with intact lateral wall (31A1), pertrochanteric fractures with incompetent lateral wall (31A2), and inter-trochanteric (reverse obliquity) fractures (31A3) [1, 2]. The lateral wall thickness is measured by a distance in millimeters drawn from a fixed point 3 cm below the innominate tubercle of the greater trochanter and angled 135° upward to the mid fracture line (
Figure 1.
Illustration of the lateral wall-thickness measurement (d). It is defined as the distance in millimeter (mm) from a reference point that is located 3 cm below the innominate tubercle of the greater trochanter and angled at 135° upward to the midline between the two cortices of the fracture.
Our study's aims are to determine the reliability of lateral wall-thickness measurement in intertrochanteric fractures and the parameters that predict the lateral wall integrity.
There are 267 femurs and 263 patients with intertrochanteric fractures, who had had surgery in King Chulalongkorn Memorial Hospital between January 2014 and January 2019. The inclusion criteria are patients aged >18 years and the associated 2018 AO/OTA classification 31A1 and 31A2. The exclusion criteria are non-traumatic fractures (N = 1), a previous ipsilateral trochanteric region, associated fractures in the ipsi-lateral femur (N = 3), and the absence of preoperative plain radiography (N = 2). A total of 236 femurs and 232 patients were included in the study. No informed consent was obtained because the research is a retrospective descriptive study. No new imaging has been obtained in this study.
This study was approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University (certificate of approval no. 956/2020).
The medical records and the previous imaging of the patients were retrospectively reviewed. The age, gender, height, weight, BMI, BMD, side of fracture, type of fracture as per the 2018 AO/OTA classification, type of surgery, lateral wall thickness, fracture length, fracture angle, greater trochanteric involvement, and fragment site were collected.
The lateral wall thickness was measured two separate times at least 2 months apart by the radiology resident and orthopedic resident (rater 2,3—T.P., S.J.), the musculoskeletal radiologist (rater 1, C.R.), and the orthopedic surgeon (rater 4, C.A.) in an anteroposterior view of the preoperative hip or pelvis plain radiography. The raters are blinded from each other's measurements.
The fracture length is measured from the length of the straight line drawn between the breaking point of the medial and lateral cortices. The fracture line angle is the angle between the femoral shaft alignment and the straight line between the breaking point of the lateral and medial cortexes. The area of involvement of the greater trochanter is divided into the proximal, middle, and distal one-third by a line from the superior to the inferior aspects of the greater trochanter. The sites of the bony fragment are the greater trochanter, the intertrochanter, and the lesser trochanter [6] as shown in
Figure 2.
(
Figure 3.
Patients with AO 31A1.3 classification intertrochanteric fracture. Lateral wall thickness = 26.34 mm, Fracture length = 6.02 cm, Fracture angle = 47°, Greater trochanter involvement at middle one–third, Fragment at lesser trochanter.
Figure 4.
Patient with AO 31A2.3 classification intertrochanteric fracture (lateral wall incompetence). Lateral wall thickness = 14.22 mm. Fracture length = 8.01 cm. Fracture angle = 22°. Greater trochanter involvement at distal one-third, Fragment at greater and lesser trochanters.
The mean, SD, and independent
There were 89 males (38.36%) and 143 females (61.64%) in the study having a mean age of 77.64 years (SD = 12.84). Of the 236 femurs, there were 115 right-sided fractures (48.73%) and 121 left-sided fractures (51.27%). The lateral wall status was classified according to the 2018 AO/OTA classification; 164 femurs (69.49%) were present in the intact lateral wall group (31A1.2 = 63 [26.69%], 31A1.3 = 101 [42.8%]) and 72 femurs (30.51%) were present in the incompetent lateral wall group (31A2 = 7 [2.97%], 31A2.2 = 30 [12.71%], 31A2.3 = 35 [14.83%]). The height, weight, and BMI had no significant differences between the two groups. In total, 103 cases had BMD data for the total femur and femoral neck, which were not significantly different between the two groups (
Demographic data for the intact and incompetent lateral wall groups
| N | 164 | 72 | |
| Age (years) | 76.98 ± 12.91 | 79.22 ± 12.542 | 0.221 |
| Range (years) | 24–99 | 25–99 | |
| Gender | |||
| Male | 69 (42.4%) | 21 (29.2%) | 0.06 |
| Female | 95 (57.6%) | 51 (70.8%) | |
| Side of fracture | |||
| Right | 82 (50%) | 33 (45.8%) | |
| Left | 82 (50%) | 39 (54.2%) | |
| Height (cm) | 159.42 ± 8.67 | 156.19 ± 8.35 | 0.461 |
| Weight (kg) | 57.00 ± 12.00 | 53.08 ± 10.72 | 0.207 |
| BMI (kg/m2) | 22.32 ± 3.76 | 21.65 ± 3.48 | 0.397 |
| BMD (T score) | |||
| Total femur | −2.26 ± 1.11 | −2.70 ± 0.83 | 0.055 |
| Femoral neck | −2.87 ± 1.09 | −3.06 ± 0.78 | 0.388 |
The causes of fracture are: falling, 223 cases (94.49%) motorcycle accidents, 9 cases (3.81%); and bicycle accident, 4 cases (1.69%). The patients were treated with a dynamic hip screw—41 cases (17.37%), intramedullary nail—143 cases (60.59%), hip hemiarthroplasty—51 cases (21.61%), and plate and screw—2 cases (0.85%).
The mean lateral wall thickness in the intact group was 28.57 mm (SD = 4.87), which is significantly >18.11 mm (SD = 3.69) in the incompetent group (
Fracture parameters for the intact and incompetent lateral wall groups
| Lateral wall thickness (mm) | 28.57 ± 4.87 | 18.11 ± 3.69 | <0.0001 |
| Fracture length (cm) | 6.51 ± 1.00 | 6.83 ± 1.59 | 0.067 |
| Fracture angle (degree) | 40.35 ± 10.32 | 35.71 ± 12.86 | 0.004 |
| Greater trochanter | N | N | <0.0001 |
| involvement | 87 (53.05%) | 39 (54.17%) | |
| Proximal | 64 (39.02%) | 11 (15.28%) | |
| Middle | 13 (7.93%) | 22 (30.56%) | |
| Distal | |||
| Fragment | N | N | |
| Greater trochanter | 34 | 28 | 0.004 |
| Intertrochanter | 25 | 31 | <0.0001 |
| Lesser trochanter | 94 | 61 | <0.0001 |
The inter-rater ICC was 0.944 (95% CI = 0.927–0.957), indicating excellent reliability between the four raters. The intra-rater reliability was good to excellent and was from 0.835 to 0.972 with 95% CI = 0.792–0.978 (
Intra-rater reliability and inter-rater reliability of lateral wall-thickness measurement
| Rater 1 | 0.972 | 0.963–0.978 | 0.944 | 0.927–0.957 |
| Rater 2 | 0.961 | 0.950–0.970 | ||
| Rater 3 | 0.835 | 0.792–0.870 | ||
| Rater 4 | 0.882 | 0.850–0.908 | ||
ICC, Intra-class correlation coefficient.
In univariate binary logistic regression, the parameters with a statistically significant odds ratio associated with lateral wall incompetence were total femur BMD (OR = 0.33), fracture angle (OR = 0.96), greater trochanter involvement at middle (OR = 0.38) and distal (OR = 3.78), and the fragments at the intertrochanter area (OR = 3.54) and lesser trochanter (OR = 3.12) (
Univariate analysis of the odds ratio for the incompetent group
| BMI | 0.95 | 0.198 | 0.880–1.027 |
| BMD T score | |||
| Total femur | 0.33 | 0.02 | 0.129–0.842 |
| Femoral neck | 2.27 | 0.09 | 0.877–5.864 |
| Fracture length | 1.23 | 0.075 | 0.979–1.547 |
| Fracture angle | 0.96 | 0.004 | 0.936–0.988 |
| Greater trochanter | |||
| involvement | 0.38 | 0.001 | 0.182–0.806 |
| Middle | 3.78 | <0.0001 | 1.726–8.258 |
| Distal | |||
| Fragment | |||
| Greater trochanter | 0.79 | 0.583 | 0.341–1.832 |
| Intertrochanter | 3.54 | 0.003 | 1.539–8.133 |
| Lesser trochanter | 3.12 | 0.003 | 1.461–6.667 |
There are 230 cases in the multivariate analysis with the exclusion of the BMD data. The statistically significant odd ratios associated with lateral wall incompetence were fracture angle (OR = 0.95), distal greater trochanter involvement (OR = 9.47), and the fragments at the intertrochanter area (OR = 4.49) and lesser trochanter (OR = 2.6) (
Multivariate analysis of the odds ratio for the incompetent group
| Fracture angle | 0.95 | 0.001 | 0.921–0.978 |
| Greater trochanter | |||
| involvement | 9.47 | <0.0001 | 3.517–25.475 |
| Distal | |||
| Fragment | |||
| Intertrochanter | 4.49 | <0.0001 | 2.046–9.856 |
| Lesser | 2.6 | 0.026 | 1.12–6.036 |
Our study showed that the reliability of lateral wall-thickness measurements in an intertrochanteric fracture was excellent for inter-rater reliability (ICC = 0.944) and good to excellent for intra-rater reliability (ICC = 0.835–0.972). Therefore, the lateral wall-thickness measurement is reproducible per the experience of the raters (in our study, they included an orthopedic surgeon, musculoskeletal radiologist, and residents of the orthopedic and radiology departments). This can be useful in classifying the intertrochanteric fracture group according to the 2018 AO/OTA classification. Chan et al. [8] found that the 2018 AO/OTA intertrochanteric fracture classification had moderate inter-rater reliability in group classification (Cohen's kappa = 0.479) and fair inter-rater reliability in subgroup classification (Cohen's kappa = 0.376). The raters consisted of two orthopedic surgeons and four residents. However, they did not demonstrate the reliability of the lateral wall-thickness measurement. Further studies may be needed to establish the reliability of the 2018 AO/OTA intertrochanteric fracture classification and lateral wall-thickness measurement.
Our study further describes the fracture parameters that are associated with lateral wall incompetence. We found that fracture angle (OR = 0.95), distal greater trochanter involvement of the fracture (OR = 9.47), and fragments at the intertro-chanter area (OR = 4.49) and the lesser trochanter (OR = 2.6) are associated with lateral wall incompetence in intertrochanteric fracture. These findings will help when classifying patients. If the fracture has distal greater trochanter involvement or the fragments at the intertrochanter area or the lesser tro-chanter, it is more likely to be lateral wall incompetence that suggests unstable type of fracture and suitable for intramedullary nail fixation. On the contrary, if the fracture has a high fracture angle, the lateral wall has more chance to be intact. Stable intertrochanteric fracture is currently treated with either dynamic hip screw or intramedullary nail fixation.
Due to the retrospective nature of the study, some plain radiographs had a poor position or inadequate traction, obscuring the fracture line that could cause errors in the measurements and thus, affect the reliability. We reduced the errors by selecting the best possible image, and all the raters measured the same film. We also excluded the BMD data in the multivariate analysis because there are about 66% of the cases with missing data. This could confound the multivariate analysis.
In summary, lateral wall-thickness measurements of the AO/OTA 2018 classified intertrochanteric fractures show excellent inter-rater reliability. Fracture angle, distal greater trochanter involvement, and the fragments at the intertrochanteric region and the lesser trochanter are associated with lateral wall incompetence in intertrochanteric fractures.