Abstract
Aims
To evaluate if union of clavicle fractures can be predicted at six weeks post-injury by the presence of bridging callus on ultrasound.
Methods
Adult patients managed nonoperatively with a displaced mid-shaft clavicle were recruited prospectively. Ultrasound evaluation of the fracture was undertaken to determine if sonographic bridging callus was present. Clinical risk factors at six weeks were used to stratify patients at high risk of nonunion with a combination of Quick Disabilities of the Arm, Shoulder and Hand questionnaire (QuickDASH) ≥ 40, fracture movement on examination, or absence of callus on radiograph.
Results
A total of 112 patients completed follow-up at six months with a nonunion incidence of 16.7% (n = 18/112). Sonographic bridging callus was detected in 62.5% (n = 70/112) of the cohort at six weeks post-injury. If present, union occurred in 98.6% of the fractures (n = 69/70). If absent, nonunion developed in 40.5% of cases (n = 17/42). The sensitivity to predict union with sonographic bridging callus at six weeks was 73.4% and the specificity was 94.4%. Regression analysis found that failure to detect sonographic bridging callus at six weeks was associated with older age, female sex, simple fracture pattern, smoking, and greater fracture displacement (Nagelkerke R2 = 0.48). Of the cohort, 30.4% (n = 34/112) had absent sonographic bridging callus in addition to one or more of the clinical risk factors at six weeks that predispose to nonunion. If one was present the nonunion rate was 35%, 60% with two, and 100% when combined with all three.
Conclusion
Ultrasound combined with clinical risk factors can accurately predict fracture healing at six weeks following a displaced midshaft clavicle fracture.
Cite this article: Bone Joint Res 2021;10(2):113–121.
Article focus
-
Nonunion drives morbidity following mid-shaft clavicle fractures and is difficult to accurately predict.
-
The aim of this study was to establish the accuracy of sonographic bridging callus to predict ultimate fracture healing at six months post-injury.
Key messages
-
Sonographic bridging callus at six weeks following fracture predicts union in 99% of cases and was detected in approximately two-thirds of all patients.
-
When absent, nonunion occurred in 41% of cases at six months.
-
Further refinement of nonunion prediction can be achieved when combined with clinical markers of poor recovery at six weeks post-injury.
Strengths and limitations
-
This was a large prospective study using a validated ultrasound technique.
-
Conventional ultrasound relies on operator image acquisition and interpretation.
Introduction
There is increasing evidence that acute plate fixation of mid-shaft clavicle fractures does not result in long-term functional advantage over nonoperative management provided that union occurs.1-5 Early operative fixation for only those patients destined to develop nonunion would reduce the overall intervention rate, minimize prolonged delay to treatment, and potentially be a more cost-effective approach.6
The ability to accurately predict nonunion following a displaced mid-shaft clavicle fracture is challenging. Currently, prediction is largely based on patient or fracture characteristics at the time of presentation.7,8 We have recently evaluated all of the available clinical, functional, and radiological data in the early recovery phase following nonoperative treatment and found this improves nonunion diagnostic accuracy. At six weeks post-injury, a combination of a Quick Disabilities of the Arm, Shoulder and Hand questionnaire (QuickDASH) score of ≥ 40, fracture movement on examination, and absence of callus on radiographs were significant predictors on regression modelling.9
Improving the accuracy of this model further is difficult. Radiological callus at six weeks can be found in patients who ultimately go onto nonunion and, conversely, can be absent in patients who progress to union.10-12 Bridging callus, however, accurately predicts union in long bone fractures but is rarely detected before three months on radiograph.13-15 Ultrasound may have the ability to evaluate bridging callus at the fracture site prior to radiographs from as early as three to six weeks following injury.16-18 The application of ultrasound to predict fracture healing has been employed in two studies of tibial fractures to date where it has been shown to detect union prior to radiographs.19,20
The use of ultrasound to detect bridging callus at six weeks following a displaced mid-shaft clavicle fracture may further improve the accuracy to predict fracture healing. The primary aim of this study was to establish the accuracy of sonographic bridging callus to predict ultimate fracture healing at six months post-injury.
Methods
A consecutive series of patients who presented to our trauma unit with a mid-shaft clavicle fracture were recruited over a two-year period. Inclusion criteria included a fully displaced mid-shaft clavicle fracture with no residual cortical contact, with or without comminution (Edinburgh Type 2 fractures),21 age over 16 years at time of injury, and ability to attend for follow-up.
All patients who met the inclusion criteria were referred to a single specialist clinic during the study. On first clinic review with an orthopaedic specialist, patients were consulted on operative and nonoperative treatment options for displaced mid-shaft fractures. The patients who opted for acute fixation following injury were excluded. The decision to undertake acute fixation was based solely on patient request.
Over the study period 139 fractures eligible for recruitment were identified, of which six patients opted for acute plate fixation (4.3%, n = 6/139). Exclusion criteria are summarized in Figure 1. A total of 125 patients undergoing nonoperative management were recruited into the study, of whom 112 patients completed follow-up at six months and were analyzed (89.6%, n = 112/125).
Fig. 1
Flow diagram of patient recruitment and outcome.
The nonoperative management protocol was a sling for three weeks post-injury, after which patients received range of motion and strengthening exercises under the supervision of a physiotherapist using a standardized regime.
Functional assessment
Patients underwent a standardized clinical review at six, 12, and 24 weeks using a questionnaire of symptoms and examination findings (Table I). Self-reported QuickDASH and EuroQol five-dimension questionnaire (EQ-5D) were completed.22
Table I.
Protocol of patient assessment.
| Assessment category | Criterion |
|---|---|
| History | Night pain at rest in bed |
| Unable to dress normally without assistance | |
| Sling no longer required at any timepoint | |
| Returned to work with or without adaptation | |
| Returned to usual activities including hobbies and sport | |
| Examination | Fracture site tenderness (pain on palpation of fracture) |
| Hand to head (palm to forehead without assistance) | |
| Hand to head and elbow to back (active internal rotation of shoulder with elbow on lumbar spine) | |
| Movement at fracture (crepitus or movement of fracture on palpation with passive movement of the shoulder) | |
| Abduct shoulder > 90° in scapula plain with elbow extended) | |
| Patient outcome scores | QuickDASH |
| EQ-5D | |
| Radiograph | Bridging or non-bridging callus present |
| Ultrasound | Sonographic bridging callus present |
-
EQ-5D, EuroQol five-dimension questionnaire ; QuickDASH, Quick Disabilities of the Arm, Shoulder and Hand questionnaire.
Radiological assessment
Radiographs of the clavicle were taken at each clinical appointment, in keeping with standard practice with a single erect anteroposterior view. These were graded prospectively for the presence of callus formation at six weeks in comparison to the radiograph at time of injury. Callus was defined as calcified material at the fracture site with or without clear bridging of the fracture. The overall displacement of the fracture was estimated using an erect calibrated radiograph measuring the distance between the ends of the proximal and distal fragment from the centre of the medullary canal using a validated technique.7
All patients who had ongoing pain at rest or on movement, palpable movement at the fracture, or absent bridging callus on radiograph at six months underwent a CT scan. Union on CT was defined as bridging callus of more than 50% of the cortical diameter between fracture fragments on 3D reconstruction.23 Patients who had ongoing symptoms and nonunion on CT at six months were offered operative exploration and plate fixation.
Ultrasound image details and callus interpretation
At each clinic visit, all patients underwent a standardized ultrasound scan performed by an orthopaedic surgeon with training in ultrasound (JAN). A Sonix L14-5 MHz/38 mm ultrasound probe (BK Medical North, Peabody, Massachusetts, USA) was used and set to 3 MHz to 7 MHz and calibrated to 6 cm depth (a standard setting for superficial musculoskeletal ultrasound evaluations).
Ultrasound interpretation was carried out using a validated technique.24 Sonographic bridging callus was defined as a continuous hyperechoic signal, with a similar echo intensity (EI) to that of the cortical bone, bridging the fracture ends. The fracture site was imaged in both the long and short axis. The probe was carefully rocked, tilted, and rotated at the fracture site to ensure perpendicular visualization of the callus and to avoid anisotropy, which may produce a false anechoic signal (loss of signal) (Figure 2). To avoid confusion with fracture comminution, a complete unbroken view of the bridging callus was required between the fracture fragments without any obvious gapping. For segmental fractures this was required at both fracture sites.
Fig. 2
a) Ultrasound example of six week bridging callus. Cortical surface of bone (vertical arrows), bridging callus (horizontal arrow). Note acoustic shadow below surface of bone. b) Example of anisotropy or loss of ultrasound signal during the same scan sequence. Orientation of the probe that is not perpendicular to the callus results in loss of signal of callus, as shown by the horizontal arrow.
Scans were given a unique code at time of capture. Judgement for the presence of sonographic bridging callus was carried out by the lead author (JAN), who was blinded to the patient’s details, radiograph, and clinical union outcome. The repeatability of this technique was established with a series of 20 scans by four independent reviewers (one musculoskeletal radiologist and three orthopaedic surgeons (CMR, WMO, JAN)), which revealed an intraclass correlation of 0.82 (95% confidence interval (CI) 0.68 to 0.91), giving excellent agreement.
Statistical analysis
Statistical analysis was performed using SPSS version 24 (IBM, Armonk, New York, USA). Data were tested for normal distribution with the D’Agostino and Pearson test. Linear variables were assessed using the independent-samples t-test for parametric data or the Mann-Whitney U test for nonparametric data. Differences between dichotomous data were assessed by using the chi-squared test and odds ratio (OR). Logistic regression was used to determine the influence of variables present at time of injury on six-week sonographic bridging callus detection. Forward and backward conditional regression was used to ensure stability of predictors. Clinically relevant predictors identified on univariate analysis with a p-value of < 0.1 were inserted into a single unadjusted model.
The absence of sonographic bridging callus was modelled with clinical risk factors for nonunion at six weeks post-injury from a published regression model; QuickDASH ≥ 40, fracture movement on examination, and no callus on six-week radiograph.9 This was used to determine if the accuracy of nonunion prediction was improved when combining sonographic callus findings with clinical recovery.
Ethics and source of funding
Ethical approval was prospectively obtained from the local Research Ethics Committee (reference number 06/S1103/51).
Results
The mean age was 41.3 years (SD 17.6, 16 to 73) and 77% were male (n = 86/112). Of the 13 patients lost to follow-up and six patients who chose acute fixation, no evidence of statistically significant differences in demographics (age, sex, smoking status) and fracture characteristics (overall displacement or comminution) were found compared to those who completed the study.
The nonunion incidence was 16.7% in the cohort (n = 18/112) and this was associated with increasing age (p = 0.015, independent-samples t-test), fracture displacement (p = 0.006, independent-samples t-test) and smoking (p = 0.002, chi-squared test) on univariate analysis (Table II). The functional outcome scores were significantly better at six months for patients that united their fractures compared to those that developed nonunion; six-month QuickDASH (6.4 vs 29.7, p < 0.001, independent-samples t-test) and EQ-5D (0.92 vs 0.64, p < 0.001, independent-samples t-test).
Table II.
Patient demographics of the study and six week clinical findings.
| Parameter | Union (n = 94) | Nonunion (n = 18) | p-value | OR | 95% CI |
|---|---|---|---|---|---|
| Patient demographic | |||||
| Mean age, yrs (95% CI) | 39.6 (35.9 to 43.3) | 50.6 (44.2 to 56.9) | 0.015* | N/A | N/A |
| Male, n (%) | 75 (79.8) | 11 (61.1) | 0.086† | 2.5 | 0.9 to 7.3 |
| Smoker, n (%) | 13 (13.8) | 8 (44.4) | 0.002† | 5 | 1.7 to 15.0 |
| Right side, n (%) | 42 (44.7) | 9 (50) | 0.706† | 0.8 | 0.3 to 2.3 |
| Dominant side, n (%) | 44 (46.8) | 10 (55.6) | 0.496† | 0.7 | 0.3 to 1.9 |
| Injury mechanism, n (%) | |||||
| High-energy | 14 (14.9) | 3 (16.7) | 0.848† | 1.1 | 0.3 to 4.5 |
| Fracture findings | |||||
| Comminution, Edinburgh 2B, n (%) | 25 (26.6) | 3 (16.7) | 0.373† | 0.5 | 0.1 to 2.1 |
| Mean fracture displacement, mm (95% confidence interval) | 23.0 (21.6 to 24.4) | 28.0 (24.6 to 31.3) | 0.006* | N/A | N/A |
| History, n (%) | |||||
| Night pain | 44 (46.8) | 15 (83.3) | 0.004† | 5.7 | 1.5 to 20.9 |
| Unable to dress normally | 14 (14.9) | 7 (38.9) | 0.017† | 3.6 | 1.2 to 11.0 |
| Sling still required | 7 (7.4) | 5 (27.8) | 0.011† | 4.8 | 1.3 to 17.3 |
| Have not returned to work | 26 (27.7) | 11 (61.1) | 0.006† | 4.1 | 1.4 to 11.7 |
| Have not returned to usual activities | 70 (74.5) | 15 (83.3) | 0.420† | 1.7 | 0.5 to 6.4 |
| Examination, n (%) | |||||
| Fracture site tenderness | 21 (22.3) | 11 (61.1) | 0.001† | 5.5 | 1.9 to 15.8 |
| Inability to reach hand to head | 5 (5.3) | 5 (27.8) | 0.002† | 6.8 | 1.7 to 26.9 |
| Inability to reach hand to head and elbow to back | 7 (7.4) | 4 (22.2) | 0.054† | 3.5 | 0.9 to 13.7 |
| Movement at fracture | 8 (8.5) | 5 (27.8) | 0.019† | 4.1 | 1.2 to 14.6 |
| Inability to abduct shoulder beyond 90° | 10 (10.6) | 9 (50) | < 0.001† | 8.4 | 2.7 to 26.1 |
| QuickDASH ≥ 40 | 22 (23.4) | 11 (61.1) | 0.001† | 5.1 | 1.8 to 14.9 |
| Patient outcome scores | |||||
| Mean QuickDASH (95% CI) | 27.2 (23.1 to 31.3) | 44.2 (32.9 to 55.5) | 0.002* | N/A | N/A |
| Mean EQ-5D (95% CI) | 0.74 (0.71 to 0.78) | 0.61 (0.46 to 0.76) | 0.020* | N/A | N/A |
| Radiograph, n (%) | |||||
| Callus 6-wk radiograph | 59 (62.8) | 6 (33.3) | 0.020† | 3.4 | 1.2 to 9.8 |
| Callus 12-wk radiograph | 92 (97.9) | 9 (50) | < 0.001† | 9.2 | 4.6 to 18.2 |
| Callus 24-wk radiograph | 94 (100) | 10 (55.6) | < 0.001† | 10.4 | 5.8 to 18.7 |
| Ultrasound, n (%) | |||||
| Sonographic briding callus 6 wks | 69 (73.4) | 1 (5.6) | < 0.001† | 46.9 | 5.9 to 371.1 |
| Sonographic briding callus 12 wks | 82 (87.2) | 1 (5.6) | < 0.001† | 116.2 | 14.1 to 954.2 |
| Sonographic briding callus 24 wks | 94 (100) | 1 (5.6) | < 0.001† | N/A | N/A |
-
*
Independent-samples t-test.
-
†
Chi-squared test.
-
CI, confidence interval; EQ-5D, EuroQol five-dimension questionnaire; N/A, not applicable; OR, odds ratio; QuickDASH, Quick Disabilities of the Arm, Shoulder and Hand questionnaire.
The presence of sonographic bridging callus at six weeks was strongly associated with superior patient-reported outcome measures at each timepoint over the six months post-injury when compared against those patients who lacked it (Figures 3 and 4, Table III).
Fig. 3
Line diagram of sonographic bridging callus (SBC) present and not present and influence on Quick Disabilities of the Arm, Shoulder and Hand questionnaire (QuickDASH).
Fig. 4
Line diagram of sonographic bridging callus (SBC) present and not present and influence on EuroQol five-dimension questionnaire (EQ-5D).
Table III.
Comparison of demographics and functional outcomes of those with and without sonographic bridging callus at six weeks post-injury.
| Parameter | Bridging callus (n = 70) | No callus (n = 42) | p-value | OR | CI 95% |
|---|---|---|---|---|---|
| Patient demographic | |||||
| Mean age, yrs (range) | 36.7 (32.5 to 40.9) | 49.1 (44.6 to 53.7) | < 0.001* | N/A | N/A |
| Male, n (%) | 61 (87.1) | 25 (59.5) | 0.001† | 4.6 | 1.8 to 11.7 |
| Smoker, n (%) | 7 (10) | 14 (33.3) | 0.002† | 4.5 | 1.6 to 12.4 |
| Union outcome | |||||
| Union at six mths, n (%) | 69 (98.6) | 25 (59.5) | < 0.001† | 46.9 | 5.9 to 371.1 |
| Injury mechanism | |||||
| High-energy, n (%) | 7 (10) | 10 (23.8) | 0.049† | 2.8 | 1.0 to 8.1 |
| Fracture findings | |||||
| Comminution, Edinburgh 2B, n (%) | 22 (31.4) | 6 (14.3) | 0.043† | 0.4 | 0.1 to 1.0 |
| Mean fracture displacement, mm (range) | 21.9 (20.4 to 23.4) | 26.9 (24.8 to 29.1) | < 0.001* | N/A | N/A |
| Callus present on six-wk radiograph, n (%) | 24 (34.3) | 2 (4.8) | < 0.001† | 10.4 | 2.3 to 47.0 |
| Mean QuickDASH (range) | |||||
| 6 wks | 23.4 (19.2 to 27.6) | 40.7 (33.6 to 47.8) | < 0.001* | N/A | N/A |
| 12 wks | 8.8 (5.8 to 11.8) | 25.2 (17.4 to 33.0) | < 0.001* | N/A | N/A |
| 24 wks | 4.2 (2.1 to 6.2) | 20.0 (12.2 to 27.9) | < 0.001* | N/A | N/A |
| Mean EQ-5D (range) | |||||
| 6 wks | 0.78 (0.74 to 0.82) | 0.63 (0.54 to 0.71) | < 0.001* | N/A | N/A |
| 12 wks | 0.91 (0.89 to 0.94) | 0.69 (0.60 to 0.79) | < 0.001* | N/A | N/A |
| 24 wks | 0.96 (0.94 to 0.98) | 0.73 (0.64 to 0.82) | < 0.001* | N/A | N/A |
-
*
Independent-samples t-test.
-
†
Chi-squared test.
-
CI, confidence interval; EQ-5D, EuroQol five-dimension questionnaire; N/A, not applicable; OR, odds ratio; QuickDASH, Quick Disabilities of the Arm, Shoulder and Hand questionnaire.
Logistic regression for sonographic bridging callus detection at six weeks post-injury was modelled on age, sex, smoking, energy, comminution, and fracture displacement. Sonographic callus was less likely to be detected with increasing age (p = 0.004), female sex (p = 0.013), simple fractures (Edinburgh 2B1; p = 0.027), smoking (p = 0.009), and greater overall fracture displacement (p < 0.001) (Forward and Backward Conditional entry, Nagelkerke R2 = 0.48)(Table IV).
Table IV.
Logistic regression of detection of sonographic bridging callus at six weeks using ‘forward conditional’ methodology.
| Parameter | p-value | OR | 95% CI |
|---|---|---|---|
| Age | 0.004 | 1.05 | 1.01 to 1.08 |
| Female | 0.013 | 4.39 | 1.37 to 14.08 |
| Simple fracture pattern (Edinburgh 2B1) | 0.027 | 4.05 | 1.17 to 13.96 |
| Smoker | 0.009 | 5.14 | 1.50 to 17.63 |
| Fracture displacement | < 0.001 | 1.16 | 1.07 to 1.25 |
-
Nagelkerke R2 = 0.48 model fit.
-
CI, confidence interval; OR, odds ratio.
Fracture healing and association with six-week sonographic bridging callus
At six weeks following injury, sonographic bridging callus was found in 62.5% of the cohort (n = 70/112), increasing to 74.1% (n = 83/112) by 12 weeks. This was in contrast to radiographs at six weeks, where callus was found in 52.7% of the whole cohort (n = 59/112) but only bridging in 22.3% of patients (n = 25/112).
When sonographic bridging callus was found, 98.6% (n = 69/70) of these patients united their fractures by six months (Figures 5 and 6). Of the patients without sonographic bridging callus at six weeks, 40.5% (n = 17/42) went on to nonunion. One patient who ultimately developed a hypertrophic nonunion was classified incorrectly with sonographic bridging callus. Therefore, when sonographic bridging callus was detected it had a 98.6% positive predictive value (PPV) for union and a 42.9% negative predictive value (NPV) when absent (chi-squared test, p < 0.001, odds ratio 46.9). Alternatively, 73.4% sensitivity to detect union and 94.4% specificity.
Fig. 5
Nonunion with absence of bridging callus as seen in a 37-year-old male. a) Six-week ultrasound showing absent bridging callus. b) Six-month anteroposterior radiograph with atrophic nonunion.
Fig. 6
Remodelling of bridging callus on ultrasound as seen in an 18-year-old male. a) Six-week sonographic bridging callus at fracture site; b) six-week anteroposterior radiograph showing minimal callus formation; c) 12-week ultrasound scan showing remodelling of sonographic bridging callus; d) 12-week anteroposterior radiograph with bridging callus now evident.
None of the three clinical risk factors for nonunion were present in 38.4% (n = 43/112) of the cohort, of which only 4.7% developed nonunion (n = 2/43). The remainder had one or more risk factor present (n = 69/112), and when combined with absent sonographic bridging callus (n = 34/69) the nonunion prediction could be further refined. If just these patients underwent fixation, and assuming union is successfully achieved following surgery, the numbers needed to treat (NNT) to prevent one nonunion can be estimated. If there was absent sonographic bridging callus with one risk factor the nonunion risk was 35% (NNT 2.9), if two clinical predictors were present this increased to 60% (NNT 1.7), and when all three were present the nonunion incidence was 100% (NNT 1.0) (Table V).
Table V.
Absence of sonographic bridging callus in combination with six week clinical risk factors (Quick Disabilities of the Arm, Shoulder and Hand questionnaire score ≥ 40, no callus on radiograph, and fracture mobility on exam).
| Number of risk factors | Observed nonunion, % | Predicted NNT | At risk, % |
|---|---|---|---|
| 3 | 100 | 1 | 2.7 |
| 2 | 60 | 1.7 | 9.9 |
| 1 | 35 | 2.9 | 17.9 |
-
NNT, number needed to treat with plate fixation to prevent nonunion.
Discussion
Ultrasound evaluation of bridging callus may have a useful role following nonoperative management of clavicle fractures to predict healing. In a prospective cohort of 112 displaced mid-shaft clavicle fractures that underwent nonoperative management, we found that ultrasound detection of bridging callus was strongly predictive of ultimate union. When bridging callus was observed on ultrasound at six weeks post-injury, it predicted union in 99% of cases. In combination with a previously validated model of clinical predictors of nonunion, fracture healing estimation can be further refined. To our knowledge, this is the first study to explore the use of ultrasound to predict nonunion following a clavicle fracture, or indeed any fracture which has been nonoperatively managed.
The role of acute fixation for mid-shaft clavicle fractures is contentious. Despite early trials suggesting a sustained benefit of plate fixation over nonoperative management at one year,25-27 increasing evidence suggests that functional outcomes are similar if union occurs.1,2,5,28-30 Therefore, rather than the perceived malunion causing morbidity, it is the development of nonunion that likely accounts for the majority of dissatisfaction and poor function at one year post-injury.31-33 The ability to target those at high risk of nonunion would decrease the morbidity associated with prolonged delay to treatment and reduce unnecessary intervention in those patients who would likely unite with nonoperative treatment.
We have previously shown that a six-week clinical assessment has superior accuracy to predict fracture healing when compared to estimate at time of injury alone, however the model has limitations and relies on radiological callus detection.7 Callus on ultrasound is thought to have an increasingly hyperechoic ultrasound signal as it calcifies and reconstitutes into cortical bone, and this does not appear to be based on dense mineralization as required for traditional radiographs.16,34 The ability to detect callus formation at three weeks following a fracture was first described by Ricciardi et al35 and Maffuli and Thornton17 soon after, who described the lack of sonographic callus in a patient with a humeral shaft nonunion. Following this, Moed et al18,19 reported that ultrasound can predict union prior to radiographs in a series of tibial shaft fractures following intramedullary nailing. They reported that bridging callus could be found at a mean of 6.5 weeks on ultrasound versus 19 weeks on radiographs, with a PPV of 97% to predict union when present. This finding has recently been replicated by an independent group examining a series of tibial fractures postoperatively, where ultrasound was able to detect bridging callus prior to radiographs and this correlated with union.20
Our work is the first to evaluate the prognostic use of ultrasound to identify patients at high risk of nonunion following a clavicle fracture. Six-week sonographic bridging callus was more likely to be absent with increasing age, female sex, absence of comminution, smoking, and increasing fracture displacement. When present, the functional recovery was superior over the first six months following injury. We have previously evaluated our 2D scanning technique between observers and found excellent agreement.24 The detection of bridging callus on six-week ultrasound in that pilot study of 20 patients was 80% sensitive and 100% specific for subsequent union. In this larger cohort of 112 patients, we validated this result with 73.4% sensitivity to detect union and 94.4% specificity. Whether this holds true for other long bone fractures is an interesting premise for future work. The combination of absent bridging callus and clinical risk factors for poor fracture healing at six weeks post-injury enabled further nonunion prediction. Unfortunately, our study was underpowered to evaluate the nonunion predictors specific to this cohort with regression modelling, given the modest number of nonunion events. We therefore used a previously validated model of six-week clinical risk factors for this study.9
The main limitation with conventional ultrasound is the reliance on the operator for image acquisition, and orthopaedic surgeons’ unfamiliarity with image interpretation. Additionally, there is the risk of loss of signal and incorrect image evaluation if an ultrasound beam is not perpendicular to the site of interest (anisotropy). One surgeon (JAN) performed all of the ultrasound scans and image interpretation. Bias was kept to a minimum by grading the anonymous scans independently of the clinical review of patients to avoid potential interpretation bias.
In this study at least one patient was misinterpreted with bridging callus who was later diagnosed with a hypertrophic nonunion at six months post-injury. In equivocal cases, CT evaluation beyond three months may play a role in detect bridging callus when ultrasound cannot make an adequate judgement.23 Specific fracture patterns may also affect the ability to detect bridging callus. A more detailed analysis of sonographic bridging callus and fracture patterns using 3D modelling may enable more accurate image interpretation.
In conclusion, ultrasound evaluation of bridging callus following a displaced mid-shaft clavicle fracture has excellent accuracy to predict fracture healing. When combined with clinical risk factors for nonunion at six weeks post-injury, the absence of sonographic bridging callus may enable targeted operative fixation in select patients.
References
1. Ahrens PM , Garlick NI , Barber J , Tims EM The clavicle trial . Bone Joint J . 2017 ; 99-B ( 16 ): 1345 – 1354 . Crossref PubMed Google Scholar
2. Qvist AH , Væsel MT , Jensen CM , Jensen SL Plate fixation compared with nonoperative treatment of displaced midshaft clavicular fractures: a randomized clinical trial . Bone Joint J . 2018 ; 100-B ( 10 ): 1385 – 1391 . Crossref PubMed Google Scholar
3. Amer K , Smith B , Thomson JE , et al. Operative versus Nonoperative outcomes of Middle-Third clavicle fractures: a systematic review and meta-analysis . J Orthop Trauma . 2020 ; 34 ( 1 ): e6 – 13 . Crossref PubMed Google Scholar
4. Lenza M , Buchbinder R , Johnston RV , Ferrari BA , Faloppa F Surgical versus conservative interventions for treating fractures of the middle third of the clavicle . Cochrane Database Syst Rev . 2019 ; 1 : CD009363 . Crossref PubMed Google Scholar
5. Robinson CM , Goudie EB , Murray IR , et al. Open reduction and plate fixation versus nonoperative treatment for displaced midshaft clavicular fractures: a multicenter, randomized, controlled trial . J Bone Joint Surg Am . 2013 ; 95-A ( 17 ): 1576 – 1584 . Crossref PubMed Google Scholar
6. Nicholson JA , Clement N , Goudie E , Robinson CM Trauma routine fixation of displaced midshaft clavicle fractures is not cost-effective: a cost analysis from a randomized controlled trial . Bone Jt J . 2019 ; 101-B ( 8 ): 995 – 1001 . Google Scholar
7. Murray IR , Foster CJ , Eros A , Robinson CM Risk factors for nonunion after nonoperative treatment of displaced midshaft fractures of the clavicle . J Bone Joint Surg Am . 2013 ; 95-A ( 13 ): 1153 – 1158 . Crossref PubMed Google Scholar
8. Liu W , Xiao J , Ji F , Xie Y , Hao Y Intrinsic and extrinsic risk factors for nonunion after nonoperative treatment of midshaft clavicle fractures . Orthop Traumatol Surg Res . 2015 ; 101 ( 2 ): 197 – 200 . Crossref PubMed Google Scholar
9. Nicholson JA , Clement ND , Clelland AD , MacDonald D , Simpson AHRW , Robinson CM Displaced Midshaft clavicle fracture Union can be accurately predicted with a delayed assessment at 6 weeks following injury: a prospective cohort study . J Bone Joint Surg Am . 2020 ; 102-A ( 7 ): 557 – 566 . Crossref PubMed Google Scholar
10. Mills LA , Aitken SA , Simpson AHRW The risk of non-union per fracture: current myths and revised figures from a population of over 4 million adults . Acta Orthop . 2017 ; 88 ( 4 ): 434 – 439 . Crossref PubMed Google Scholar
11. Robinson CM , Court-Brown CM , McQueen MM , Wakefield AE Estimating the risk of nonunion following nonoperative treatment of a clavicular fracture . J Bone Joint Surg Am . 2004 ; 86-A ( 7 ): 1359 – 1365 . Crossref PubMed Google Scholar
12. Leow JM , Clement ND , Tawonsawatruk T , Simpson CJ , Simpson AHRW The radiographic Union scale in tibial (rust) fractures: reliability of the outcome measure at an independent centre . Bone Joint Res . 2016 ; 5 ( 4 ): 116 – 121 . Crossref PubMed Google Scholar
13. Lack WD , Starman JS , Seymour R , et al. Any cortical bridging predicts healing of tibial shaft fractures . J Bone Joint Surg Am . 2014 ; 96-A ( 13 ): 1066 – 1072 . Crossref PubMed Google Scholar
14. Strotman PK , Karunakar MA , Seymour R , Lack WD Any cortical bridging predicts healing of supracondylar femur fractures after treatment with locked plating . J Orthop Trauma . 2017 ; 31 ( 10 ): 538 – 544 . Crossref PubMed Google Scholar
15. DiSilvio F , Foyil S , Schiffman B , Bernstein M , Summers H , Lack WD Long bone Union accurately predicted by cortical bridging within 4 months . JB JS Open Access . 2018 ; 3 ( 4 ): e0012 . Crossref PubMed Google Scholar
16. Derbyshire NDJ , Simpson A A role for ultrasound in limb lengthening . Br J Radiol . 1992 ; 65 ( 775 ): 576 – 580 . http://www.ncbi.nlm.nih.gov/pubmed/1515893 Crossref PubMed Google Scholar
17. Maffulli N , Thornton A Ultrasonographic appearance of external callus in long-bone fractures . Injury . 1995 ; 26 ( 1 ): 5 – 12 . Crossref PubMed Google Scholar
18. Moed BR , Watson JT , Goldschmidt P , van Holsbeeck M Ultrasound for the early diagnosis of fracture healing after interlocking nailing of the tibia without reaming . Clin Orthop Relat Res . 1995 ; 310 : 137 – 144 . PubMed Google Scholar
19. Moed BR , Subramanian S , van Holsbeeck M , et al. Ultrasound for the early diagnosis of tibial fracture healing after static interlocked nailing without reaming: clinical results . J Orthop Trauma . 1998 ; 12 ( 3 ): 206 – 213 . Crossref PubMed Google Scholar
20. Chachan S , Tudu B , Sahu B Ultrasound monitoring of fracture healing: is this the end of radiography in fracture follow-ups? J Orthop Trauma . 2015 ; 29 ( 3 ): e133 – 138 . Crossref PubMed Google Scholar
21. Robinson CM Fractures of the clavicle in the adult . J Bone Joint Surg Br . 1998 ; 80-B ( 3 ): 476 – 484 . Crossref PubMed Google Scholar
22. EuroQol Group EQ-5D-3L health questionnaire. English version for the UK (validated for Ireland) . 2019 . https://euroqol.org/wp-content/ uploads/2019/10/Sample_UK__English__EQ-5D-3L_Paper_Self_complete.pdf (date last accessed 21 December 2019 ). Google Scholar
23. Nicholson JA , Fox B , Dhir R , Simpson A , Robinson C The accuracy of computed tomography for clavicle non-union evaluation . Shoulder & Elbow . 2019 . Crossref PubMed Google Scholar
24. Nicholson JA , Oliver WM , LizHang J , et al. Sonographic bridging callus: an early predictor of fracture Union . Injury . 2019 ; 50 ( 12 ): 2196 – 2202 . Crossref PubMed Google Scholar
25. McKee MD , Kreder HJ , Mandel S , Canadian Orthopaedic Trauma Society Nonoperative treatment compared with plate fixation of displaced midshaft clavicular fractures. A multicenter, randomized clinical trial . J Bone Joint Surg Am . 2007 ; 89-A ( 1 ): 1 – 10 . Crossref PubMed Google Scholar
26. Mirzatolooei F Comparison between operative and nonoperative treatment methods in the management of comminuted fractures of the clavicle . Acta Orthop Traumatol Turc . 2011 ; 45 ( 1 ): 34 – 40 . Crossref PubMed Google Scholar
27. Melean PA , Zuniga A , Marsalli M , et al. Surgical treatment of displaced middle-third clavicular fractures: a prospective, randomized trial in a working compensation population . J Shoulder Elbow Surg . 2015 ; 24 ( 4 ): 587 – 592 . Crossref PubMed Google Scholar
28. Virtanen KJ , Remes V , Pajarinen J , Savolainen V , Björkenheim J-M , Paavola M Sling compared with plate osteosynthesis for treatment of displaced midshaft clavicular fractures: a randomized clinical trial . J Bone Joint Surg Am . 2012 ; 94-A ( 17 ): 1546 – 1553 . Crossref PubMed Google Scholar
29. Woltz S , Stegeman SA , Krijnen P , et al. Plate fixation compared with Nonoperative treatment for displaced Midshaft Clavicular fractures: a multicenter randomized controlled trial . J Bone Joint Surg Am . 2017 ; 99-A ( 2 ): 106 – 112 . Crossref PubMed Google Scholar
30. Tamaoki MJS , Matsunaga FT , Da CARF , Netto NA , Matsumoto MH , Belloti JC Treatment of displaced Midshaft clavicle fractures: Figure-of-Eight harness versus anterior plate osteosynthesis . J Bone Jt Surg Am . 2017 ; 99-A ( 14 ): 1159 – 1165 . Crossref PubMed Google Scholar
31. Rasmussen JV , Jensen SL , Petersen JB , Falstie-Jensen T , Lausten G , Olsen BS A retrospective study of the association between shortening of the clavicle after fracture and the clinical outcome in 136 patients . Injury . 2011 ; 42 ( 4 ): 414 – 417 . Crossref PubMed Google Scholar
32. Stegeman SA , de Witte PB , Boonstra S , et al. Posttraumatic midshaft clavicular shortening does not result in relevant functional outcome changes . Acta Orthop . 2015 ; 86 ( 5 ): 545 – 552 . Google Scholar
33. Goudie EB , Clement ND , Murray IR , et al. The influence of shortening on clinical outcome in healed displaced Midshaft Clavicular fractures after Nonoperative treatment . J Bone Joint Surg Am . 2017 ; 99-A ( 14 ): 1166 – 1172 . Crossref PubMed Google Scholar
34. Eyres KS , Bell MJ , Kanis JA Methods of assessing new bone formation during limb lengthening. ultrasonography, dual energy X-ray absorptiometry and radiography compared . J Bone Joint Surg Br . 1993 ; 75-B ( 3 ): 358 – 364 . Crossref PubMed Google Scholar
35. Ricciardi L , Perissinotto A , Dabala M Mechanical monitoring of fracture healing using ultrasound imaging . Clin Orthop Relat Res . 1993 ; &NA ( 293 ): 71 – 76 . PubMed Google Scholar
Author contributions
J. A. Nicholson: Conceptualized the study, Performed the scans, Handled the study methodology, Recruited the patients, Wrote and edited the manuscript.
W. M. Oliver: Contributed to the scanning technique, Handled the study methodology, Edited the manuscript.
T. J. Macgillivray: Conceptualized the study, Handled the study methodology, Edited the manuscript.
C. M. Robinson: Conceptualized the study, Handled the study methodology, Recruited the patients, Edited the manuscript.
A. H. R. W. Simpson: Conceptualized the study, Handled the study methodology, Edited the manuscript.
Funding statement
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
ICMJE COI statement
A. H. R. W. Simpson is the editor-in-chief of Bone Joint Research.
Acknowledgements
The authors thank the SORT-IT (Scottish Orthopaedic Research Trust into Trauma) Edinburgh research team members Keri Swain and Debbie Ketchen for their assistance with administration of the study recruitment. Furthermore, the authors thank Dr Fergus Perks, Musculoskeletal Consultant Radiologist, for his assistance in establishing the ultrasound scanning technique.
Ethical review statement
Ethical approval was prospectively obtained for this study with the unique ID 06/S1103/51
Follow J. A. Nicholson @OrthoNicholson
© 2021 Author(s) et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/.
