Abstract
Purpose
To determine whether the restoration of greater sigmoid notch (GSN) congruity differs between osteotomized and simple fractured olecranon treated with tension-band wiring (TBW) using arc center distance (ACD) as a quantitative radiographic indicator.
Methods
We retrospectively evaluated 45 patients who underwent TBW with ring pins, including 23 who underwent olecranon osteotomy and 22 with Mayo type 2A olecranon fractures. Postoperative sagittal computed tomography images were analysed to measure ACD, articular gap, step-off, interfragmentary distance (IFD), and osteotomy angle. GSN was considered congruent when ACD was <2 mm. Subgroup analysis evaluated the association between osteotomy angle and GSN restoration using receiver operating characteristic (ROC) analysis.
Results
Congruent GSN was achieved less frequently in the osteotomy group than in the fracture group (60.9% vs 90.9%, p = 0.019). The osteotomy group showed a higher ACD (1.2 ± 1.1 mm vs 0.5 ± 0.8 mm); however, there were no differences in articular gap, step-off, IFD, and bone mineral density between the groups. In the osteotomy group, the mean osteotomy angle was significantly smaller in the incongruent subgroup than in the congruent subgroup (12.2 ± 3.6° vs 20.7 ± 4.4°). ROC analysis identified a 19.0° cutoff (area under the curve = 0.948) for predicting congruent restoration, with 100% sensitivity and 78.6% specificity.
Conclusion
GSN restoration after TBW was less accurate in olecranon osteotomies than in simple olecranon fractures. A more transverse osteotomy angle (<19°) was associated with incongruent restoration. Maintaining an osteotomy obliquity of approximately 20° may help preserve articular congruity.
Keywords
Introduction
Olecranon discontinuity arises from traumatic olecranon fractures or surgical osteotomies used to facilitate the distal humeral fracture fixation. Among the various fixation techniques used to restore olecranon integrity, tension-band wiring (TBW) is a commonly performed technique with satisfactory clinical outcomes.1,2 However, there are concerns that bone loss proportional to saw blade thickness during olecranon osteotomy procedures could compromise the congruency of the greater sigmoid notch (GSN), especially when wires are tightened in TBW. 3
In a recent study using synthetic bone, Goetti et al. 3 demonstrated that using a pre-drilled locking compression plate for osteotomy-site repair could achieve more precise GSN restoration compared to TBW, as seen on lateral elbow radiographs. Although these findings suggest the potential advantages of locking plate fixation in osteotomy-site repair, they were obtained under experimental conditions, and it remains unclear whether the same phenomenon occurs in patients in clinical practice. Furthermore, although locking plates are increasingly popular, TBW is widely used for olecranon fractures and osteotomy-site repair because of its relatively small implant volume, sufficient stability for elbow range of motion, and low risk of serious wound-related complications.2,4,5
Therefore, this study aimed to determine whether radiographic GSN congruency differs between TBW repair for simple olecranon fractures and that for osteotomized olecranon in clinical practice. Using postoperative sagittal computed tomography (CT) scans and the arc center distance (ACD), a quantitative measure of GSN congruency, we sought to clarify whether osteotomy-related bone loss affects GSN congruency.6,7
Materials and methods
Patients
This study was approved by our institutional review board prior to its initiation. We retrospectively reviewed 22 patients (nine males and 13 females) with displaced, noncomminuted intra-articular transverse olecranon fractures involving the GSN (Mayo type 2A) 8 and 23 patients (10 males and 13 females) who underwent olecranon osteotomy for exposure and open reduction and internal fixation of AO type C distal humeral fracture (C1: eight cases [34.8%], C2: 10 cases [43.5%], and C3: five cases [21.7%]) 9 between February 2016 and January 2021. All patients underwent TBW with ring pins (Jaeil Medical, Seoul, Korea) to restore olecranon discontinuity. 2 These patients were classified into the fracture (simple olecranon fractures) and osteotomy groups (olecranon osteotomy cases).
Bone mineral density (BMD) was assessed in 41 of the 45 patients undergoing dual-energy X-ray absorptiometry (DEXA) within 3 months before and after surgery, except four younger individuals (aged ≥20 and ≤40 years) who did not undergo DEXA.
The exclusion criteria were as follows: (1) age <18 years, (2) any previous fracture of the ipsilateral elbow, (3) systemic bone disease, (4) absence of postoperative elbow CT, and (5) extra-articular avulsion-type olecranon fractures that did not involve the articular surface of the GSN.
Surgical procedures
A single surgeon (MBK) performed all procedures. In both groups, the modified TBW with ring pins was used, as previously described.2,10,11
In the osteotomy group, a distally based Chevron osteotomy was created at the center of the trochlear notch using an oscillating saw with a 0.67-mm blade and completed with an osteotome (Figure 1), offering coronal and rotational stability across the osteotomy site as well as a larger bony contact area.
12
Oscillating saw blade used for olecranon osteotomy, with a digital calliper showing a measured thickness of 0.67 mm.
After completion of fixation of the distal humeral fracture, the osteotomized fragment, along with its attached triceps tendon, was reduced using reduction clamps and fixed with ring pins and wires, as previously described (Figure 2).
2
Arbeitsgemeinschaft für Osteosynthesefragen (AO) Type C distal humeral fracture treated using an olecranon osteotomy approach.
In the fracture group, the fracture site was exposed using the posterior approach. The olecranon fracture was reduced with a reduction clamp and fixed with ring pins and wiring, as previously described.10,11
Measurement
Postoperative CT images were used to measure the ACD, which is a quantitative method for evaluating GSN congruency.
7
All measurements were conducted using INFINITT PACS software (INFINITT Healthcare, Seoul, Korea). Using the multiplanar reconstruction function, the sagittal plane was adjusted to align with the longitudinal axis of the ulna and to pass through the deepest portion of the GSN. Following a prior description,
7
we selected the sagittal CT slice passing through the tip of the coronoid process corresponding to the line that bisects the centre of the trochlear notch in the coronal plane to achieve the longest articular surface arc (Figure 3(a)). Measurement of the arc center distance (ACD) on postoperative CT.
Anatomically, the GSN forms a circular arc in the sagittal plane.7,13 Using the ‘multiple circle’ tool in PACS system, we identified the best-fit circles for coronoid process and olecranon segment; thereafter, we measured the ACD between the respective centres of these circles (Figure 3(b)–(d)). We defined the GSN as congruent when the ACD was <2 mm (Figure 3(b) and (c)) and the GSN as incongruent when the ACD was ≥2 mm (Figure 3(d)).
7
This threshold was determined based on a previous CT analysis of 20 normal elbows, in which the mean contour defect was 0.7 ± 0.6 mm (range, 0–1.7 mm), and all cases demonstrated <2 mm of contour defect.
7
Therefore, 2 mm was considered the upper limit of normal anatomical variation in the GSN, and discrepancies beyond this value were defined as incongruent restoration.
7
Additionally, we measured the ‘osteotomy angle,’ defined as the acute angle formed between the osteotomy plane and vertical reference line drawn perpendicular to the longitudinal axis of the ulnar on the sagittal CT image (Figure 3(c) and (d)), following the definition of Yamazaki et al.
14
The articular gap, step-off, and interfragmentary distance (IFD) were measured (Figure 4). Postoperative sagittal CT image showing measurement of articular gap, step-off, and interfragmentary distance (IFD) at the olecranon. These parameters, in addition to the arc center distance (ACD), were used to assess the quality of reduction.
To evaluate the reduction quality of the distal humerus, the articular step-off, gap, and IFD were measured on sagittal and coronal reconstructed postoperative elbow CT images. In addition, coronal and sagittal alignment was evaluated on postoperative radiographs to determine the overall alignment of the distal humerus. Satisfactory reduction was defined as <2 mm of articular step-off, gap, or IFD, and acceptable alignment was defined as <5° of coronal and sagittal angulation.15–17
Intra- and inter-observer reliability
Two observers, an experienced orthopaedic trauma surgeon with >10 years of experience and a hand surgeon with >5 years of experience, independently evaluated the ACD on postoperative CT images. Both observers were blinded to the patients’ clinical information and each other’s measurements. To assess intra-observer reliability, the same observer repeated all measurements 3 weeks after the first evaluation. Inter-observer reliability was determined by comparing the measurements between the two observers. Intraclass correlation coefficients (ICCs) were calculated for both intra- and inter-observer reliabilities. Based on the ICC results, the mean values of repeated measurements from the senior author were used for further analysis.
Statistical analysis
Descriptive statistics (mean ± standard deviation) were calculated for all variables. Between-group comparisons (osteotomy vs fracture) were performed using either the Mann–Whitney U test or Student’s t-test, depending on the distribution of the data. For categorical variables, the chi-squared tests or exact tests (Fisher’s exact test or Fisher–Freeman–Halton exact test) were used, depending on the expected cell counts and table dimensions. All statistical tests were two-sided, with the significance level set at p < 0.05.
To illustrate the relationship between the osteotomy angle and the ACD, a scatter plot with a fitted linear regression line and regression equation (R2 value) was generated within the osteotomy group. In addition, to evaluate the predictive ability of osteotomy angle for incongruent GSN restoration (defined as ACD ≥2 mm), a receiver operating characteristic (ROC) curve analysis was performed in the osteotomy group. The area under the ROC curve (AUC) was calculated, and the optimal cutoff value was determined using the Youden index to maximise sensitivity and specificity. Statistical analyses were conducted using SPSS software (version 31.0; SPSS, Chicago, IL, USA) and MedCalc Statistical Software (version 23.2; MedCalc Software Ltd., Ostend, Belgium). Scatter plot and distribution figures were generated using Python (version 3.11; Python Software Foundation, Wilmington, DE, USA) with the Seaborn and Matplotlib packages.
Results
Comparison of demographics and measurements between the osteotomy and fracture groups.
ACD; arc center distance, IFD; interfragmentary distance. Values are presented as mean ± standard deviation.
Distal humerus fractures in the osteotomy group demonstrated satisfactory reduction on postoperative radiographs and CT scans in all included cases, with <2 mm of articular step-off, gap, and IFD, and acceptable coronal and sagittal alignment (<5°).15–17
Comparison of measurements between groups
In the osteotomy group, 14 of the 23 patients (60.9%) achieved congruent GSN, whereas 20 of the 22 patients (90.9%) in the fracture group achieved congruent GSN. The osteotomy group showed a significantly larger ACD than that in the fracture group (1.2 ± 1.1 mm vs 0.5 ± 0.8 mm, p = 0.012) [Figure 5]; additionally, the osteotomy group had a more prevalent incongruent GSN than did the fracture group (p = 0.019, OR = 6.4) [Table 2]. There were no significant differences in residual gap, articular step-off, or IFD between the two groups (Figure 5; Table 1). BMD did not differ significantly between the osteotomy and fracture groups (−1.7 ± 0.9 vs −2.1 ± 1.1, p = 0.18). Violin-box-scatter plots showing the distribution of articular contour defect (ACD), gap, step-off, and interfragmentary distance (IFD) between the osteotomy and fracture groups. Each dot represents an individual patient. Violin contours depict the overall data distribution, boxes indicate the interquartile range with the median line, and scatter points show individual measurements. Among the four parameters, ACD was significantly greater in the osteotomy group (p = 0.012), whereas gap, step-off, and IFD did not differ significantly between groups (p > 0.05). Comparison of the GSN restoration status between the osteotomy and fracture groups after TBW. GSN: greater sigmoid notch, ACD: arc center distance; TBW: tension-band wiring.
Subgroup analysis of congruent (ACD <2 mm) and incongruent GSN (ACD ≥2 mm) in the osteotomy group
There were no significant differences in age or sex distribution between the two groups. Additionally, BMD was comparable between the two groups (−1.6 ± 1.1 vs −1.8 ± 0.4, p = 0.645). Similarly, there was no significant difference in the distal humerus fracture classification (C1, C2, and C3) between the two groups (Fisher-Freeman-Halton exact test, p = 0.296). Furthermore, there were no significant differences in distal humerus reduction parameters, including articular gap (0.67 ± 0.63 mm vs 0.82 ± 0.35 mm, p = 0.469), step-off (0.01 ± 0.05 mm vs 0.07 ± 0.13 mm, p = 0.286), and IFD (0.70 ± 0.66 mm vs 0.87 ± 0.40 mm, p = 0.461), indicating comparable reduction quality between the two subgroups.
However, the osteotomy angle showed a statistically significant difference, with the incongruent GSN group demonstrating a significantly lower angle (indicating more transverse osteotomy) than did the congruent GSN group (12.2 ± 3.6° vs 20.7 ± 4.4°, p < 0.001).
A significant negative correlation was observed between the osteotomy angle and the ACD, indicating that larger (more oblique) osteotomy angles were associated with smaller ACD values and thus more congruent restoration of the GSN contour (R2 = 0.48, p < 0.001; Figure 6). Scatter plot showing the negative correlation between osteotomy angle and arc center distance (ACD). Each point represents a single patient. The solid line shows the linear regression line (ACD = −0.13 × angle + 3.46, R2 = 0.48, p < 0.001), and the dashed line indicates the 2-mm threshold for incongruent restoration. ACD = arc center distance.
To evaluate whether the osteotomy angle affected the restoration of congruent GSN, ROC curve analysis was performed. The AUC for the osteotomy angle was 0.948 (p < 0.001) and the optimal cutoff value was 19.0° (Figure 7). When the osteotomy angle exceeded 19.0°, the sensitivity and specificity for predicting congruent GSN restoration were 100% and 78.6%, respectively. Receiver operating characteristic (ROC) curve of osteotomy angle for predicting incongruent restoration of the greater sigmoid notch.
Among patients treated with TBW after olecranon osteotomy, an osteotomy angle >19.0° resulted in congruent GSN restoration in 91.7% of the cases (11/12), whereas only 27.3% (3/11) achieved congruent restoration with osteotomy angles <19.0°. The odds ratio for incongruent GSN was 29.3 for osteotomy angles <19.0° (p = 0.003).
Intra- and inter-observer reliabilities
The inter-observer ICC for the ACD values was 0.920 (95% confidence interval [CI], 0.837–0.964), indicating excellent reliability. The intra-observer ICC was 0.958 (95% CI, 0.901–0.982), reflecting a high level of measurement consistency.
Discussion
Olecranon osteotomy is commonly performed to gain access to the distal humeral articular surface during open reduction and internal fixation of intraarticular distal humerus fracture. 18 A recent synthetic bone study comparing fixation methods for olecranon osteotomy demonstrated that TBW was less effective than a predrilled locking plate in restoring congruent GSN. 3 These experimental findings suggest that a bone defect caused by the saw blade may impair the accurate restoration of the GSN during wire tightening. 3 However, the clinical relevance of this phenomenon in actual patients remained unclear.
In the present study, the articular gap, step-off, and IFD did not differ significantly between the osteotomy and fracture groups, indicating that the osteotomy-based bone defect did not necessarily lead to a measurable gap in GSN continuity following TBW. Nonetheless, the ACD, a parameter reflecting the restoration of the native circular arc of the GSN, was significantly greater in the osteotomy group. This suggests that, although there were no discernible differences in articular gap, step-off, and IFD, osteotomized olecranon exhibited inferior anatomical restoration of the GSN in TBW; additionally, the finding suggests that the ACD can reveal subtle curvature distortions of the GSN undetectable by conventional parameters.
This finding aligns with previous findings indicating that a bone defect caused by the thickness of the saw blade compromises joint congruity during TBW. 3 In our cohort, 90.9% of patients in the fracture group (without a saw blade-induced bone loss) achieved successful restoration of the native GSN arc, whereas only 60.9% of patients in the osteotomy group (with a saw-induced bone defect) achieved such restoration, with a significant difference between both groups. During wire tightening in TBW, saw blade-related bone defect may be sufficiently closed to prevent an articular step-off or gap. However, deformation of the natural curvature of the GSN may occur, resulting in two distinct arcs rather than one continuous circle (Figure 3(d)).
Notably, subgroup analysis of the osteotomy group revealed a significant association between the osteotomy angle and the restoration of GSN congruity. Specifically, a smaller osteotomy angle (i.e., a more transverse cut) strongly correlated with poorer restoration of the GSN. ROC curve analysis identified a cutoff angle of 19.0°, below which the likelihood of incongruent restoration increased markedly (AUC = 0.948, p < 0.001). When the osteotomy angle was <19°, the likelihood of incongruent GSN restoration was 29.3 times higher (p = 0.003).
Mechanically, more transverse osteotomies (<19.0°) may reduce the effective contact area between the osteotomy fragments, resulting in an uneven distribution of wire tension and distortion of the native articular curvature, particularly in the presence of a saw blade defect (Figure 3(d)). This finding aligns closely with the finite-element analysis by Yamazaki et al., 14 which demonstrated that a 20° distally directed osteotomy generated the most favourable compression pattern across the osteotomy and articular surfaces, whereas more transverse osteotomies produced less effective compression toward the articular side. Maintaining an osteotomy obliquity of approximately 20°—consistent with the mean angle observed in our congruent subgroup, the ROC-derived cutoff, and existing finite-element-based biomechanical evidence—appears to represent an optimal target for restoring the native articular curvature and minimizing the risk of GSN incongruence (Figure 3(c)).
This study has several limitations. First, this was a retrospective and cross-sectional study with a limited sample size and lacked long-term clinical outcomes such as range of motion, pain, or arthritic changes. Second, although we hypothesised that saw blade-induced defects contribute to GSN deformation, bone defect size was not directly measured; thus, we could not analyse the dose-response relationship between bone loss and ACD. However, because a saw blade of uniform thickness was used in all cases, we assumed that the magnitude of the bone defect remained constant. Third, although the osteotomy and fracture groups differ in nature, bone quality and reduction quality of distal humerus fracture were comparable; additionally, subgroup analysis showed no significant differences in these parameters between the congruent and incongruent subgroups. Therefore, the comparative analysis was considered feasible; however, it should be interpreted with caution given the inherent group differences. Finally, although the use of a single surgeon minimised variability in the reduction technique, this may limit the generalisability of our results to other surgical practices. Further validation with a larger sample size is needed, and future studies should compare long-term elbow function and post-traumatic arthritis rates between patients with congruent GSN restoration and those with incongruent GSN restoration. Additionally, future studies should quantify osteotomy-related bone loss to determine whether a dose-response relationship exists between bone loss and GSN incongruency.
Conclusion
In TBW repair, restoration of the GSN was less accurate in olecranon osteotomies than in simple olecranon fractures, as reflected by larger ACD values despite comparable articular gap, step-off, and IFD measurements. A more transverse osteotomy (<19°) was strongly associated with incongruent restoration. Maintaining an osteotomy obliquity of approximately 20°—consistent with the mean angle observed in our congruent subgroup, the ROC-derived cutoff, and existing finite-element-based biomechanical evidence—appears to represent an optimal target for restoring the native articular curvature and minimizing the risk of GSN incongruence.
Footnotes
Ethical considerations
This study was approved by the Institutional Review Board of Seoul National University Hospital (IRB No. H-2207-126-1342).
Consent to participate
The requirement for written informed consent was waived by the Institutional Review Board because of the retrospective nature of the study.
Author contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a grant from Seoul National University Hospital funded by Mr. Kang-Hyun Kim. The funding source was not involved in the study design, data collection, analysis, interpretation, writing of the report, or decision to submit the article for publication.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data supporting the findings of this study are available from the corresponding author upon reasonable request.